Table of Contents

Chapter 1
INTRODUCTION

Outline OverviewWhat is MIM?

Process Essentials
Where MIM Differs from PM
Unique Aspects of MIM
Historical Evolution
Market and Technology Status
Contrast with Other Technologies
Key Terms
Resources

OVERVIEWWHAT IS MIM?
Metal powder injection molding (MIM) is a shaping
process focused on forming and sintering small metal pow-
ders. It is widely used in the fabrication of complex-shaped,
high-performance components in large production quantities.
The Venn diagram shown in Figure 1.1 illustrates how the
convergence of four main considerations makes for the MIM
sweet spot. An example component is pictured in Figure 1.2,
showing a stainless steel cellphone hinge formed by MIM
cost, performance, complexity, and production quantity are all
evident in this application.
The concept of using plastic-forming equipment to shape
a metal, such as surgical stainless steel, excites most design
engineers, especially when they learn the final properties are
the same as with other metalworking processes. Metal powder
injection molding succeeds as a net-shaping approach, mean-
ing there is little material loss, and in several situations 98%
of the purchased raw metal powder ends up in final products.
Further, MIM enables shape complexity in materials that are
often difficult to fabricate by other processes, such as com-
posites (W-Cu), electronic alloys (Fe-Ni, Fe-Co), hard mate-
rials (tool steel), titanium, stainless steels, nickel-base
superalloys, and tungsten.
This guide is for engineers seeking to design components
that intersect with the MIM technology. The following chap-
ters provide information on the process, design rules, materi-
als, properties, applications, costs, and markets. The MIM
design process combines decisions on the material, compo-
nent geometry, component function, and fabrication process.
It is important to balance economic and technical considera-
tions; hence, beyond engineering specifications this book

Figure 1.2. This cellphone swivel hinge is an excellent example of

Figure 1.1. This Venn diagram shows how the factors of materials the factors involved in identification of a MIM applicationit is
performance, component cost, production quantity, and shape formed from a high-strength stainless steel, is produced in large
complexity intersect to define the target MIM applications. Each of quantities, has considerable shape complexity, and proves to be the
these factors is discussed in this book. least expensive of all alternatives.

1
 Table of Contents
Chapter 1: Introduction
provides information on the cost aspects. Further, winning
technicaleconomic successes are used throughout the book
to illustrate the various concepts. This information is useful
in rationalizing the process with the applications.
So, to the question of this sectionWhat is MIM? Metal
powder injection molding is a net-shaping process that starts
with a small metal powder that is dispersed in a thermoplastic
polymeric binder. Paraffin wax is a common thermoplastic
polymer distinguished by the fact that it can be repeatedly
melted. Most plastics are thermoplastics, and several are used
in binder formulationspolyethylene, polypropylene, poly-
acetal, and ethylene vinyl acetate are examples.
Most common engineering alloys are possible by MIM, but
about 30 alloys dominate the applications. The most popular
alloys are surgical stainless steel (commonly called 17-4 PH,
or American Iron and Steel Institute 630 or AISI 630) and
austenitic stainless steels (AISI 304L and AISI 316L). After
injection molding, the polymeric binder is extracted and the
particles are subjected to a high-temperature heat treatment
termed sintering. During sintering the capillary forces between
the small particles densify the structure to give a shrunken
component that is essentially full density with tensile and
hardness properties equivalent to handbook values.
Alloys can be formed by mixing elemental powders to
match the desired alloy, for example, iron, nickel, and
chromium to form a stainless steel. The powders homogenize
during sintering to deliver the alloy composition and proper-
ties. Another approach is to use prealloyed powder where each
particle contains all of the elements. Ferrous alloys are popular
by MIM, and nearly half of the commercial activity is in stain-
less steels. Complex-shaped, small components (10 g median
mass and 25 mm median size) formed by MIM are widely
used in dental orthodontics, portable computers, automotive
engines, surgical instruments, electronic packages, hand tools,
firearms, sporting devices, and cellular telephones.
Since small powders are costly, one of the keys to economic
success is to apply MIM to more complicated geometries,
those that would require extensive machining. A further gain
comes by adding cores, holes, or other mass-reduction fea-
tures, the opposite to the situation in a machined component.
Thus, MIM excels vs. other fabrication approaches when the
component complexity is high and performance demands re-
quire excellent properties. It wins against casting since it has
better dimensional control and a smoother surface finish. Un-
like machining, MIM is best applied to large production quan-
tities. Accordingly, several situations have reached production
quantities in excess of 10 million parts per year.
PROCESS ESSENTIALS
The clear advantage in MIM is the use of plastic molding
concepts to form structures with properties not possible
from plastics. This idea has its origins in the 1930s when
automotive ceramic spark-plug bodies were fabricated using
2
thermoplastic binders. By the mid-70s metallic products were
offered for commercial use by a few firms. However, up to
the 90s, MIM lacked the required infrastructure for broad
success. As the field grew large, several users stepped in to
ensure proper standards, providing the quality gains important
to widespread acceptance. Today, MIM is essentially a three-
step process involving:
formulation of feedstock from appropriate metal powders
and polymers,
molding of that feedstock into tooling that is designed
for the final part and includes dilation of the size in
anticipation of sintering shrinkage, and
thermal processing of the shaped part to remove the poly-
mer and sinter the powder.
The last step might be accomplished in a single cycle or in
two separate steps of first debinding to remove the polymer,
followed by sintering.
To begin the process, MIM mixes a small powder typically
in the size range between 1 and 40 m with a plastic binder.
The typical ratio is about 60 v/o powder, requiring then about
40 v/o binder. A common binder would be two-thirds paraffin
wax and one-third polyethylene or polypropylene. Depending
on the powder, the mixture constitutes about 94% powder by
weight, but for low-density powders such as aluminum this is
86% powder, and for high-density powders such as tungsten
this is 97% powder. The binders melt at relatively low tem-
peratures, up to 130C (266F), to allow easy mixing and
molding. The combination of powder and binder is termed
feedstock, and feedstock production is a significant business
supporting MIM.
A MIM process flow chart is given in Figure 1.3. The heart
of production is molding, where the heated feedstock is
shaped in a custom-designed tool. Prior to molding, the pow-
derbinder feedstock is heated inside the injection molding
machine to melt the polymer phase. The molding machine
rapidly presses the molten feedstock mixture into the mold
cavity to avoid any premature freezing. During this molding
Figure 1.3. Outline of the key steps in MIMstarting by mixing a
small metal powder with a polymeric binder, followed by molding,
binder removal, and sintering
 Table of Contents
Chapter 1: Introduction

step the powder remains unaffected by the molten binder, and machining. It is most competitive when used to form com-
since the peak temperature is far below the powders melting plicated and durable components, especially when the com-
temperature. After injection into the mold cavity, the pressur- ponents require high properties above those associated with
ized feedstock cools to solidify the polymer, thereby freezing die-cast zinc. Further, MIM is best suited for mass production
the particles into the desired shape. Cooling is usually the slow and is commonly applied to production quantities over
step, so molding cycles often are 10 to 20 seconds. After ex- 200,000 parts per year. Recognition of the wide range of
traction from the mold, the polymer portion is removed by applications, materials, component features, and properties
heat, solvents, or catalysts without disturbing the shaped pow- is now helping to identify new uses that are contributing to a
dera process known as debinding. The final portion of the period of sustained application expansion.
polymer decomposes while heating the shaped body to the
sintering temperature. WHERE MIM DIFFERS FROM PM
Sintering is a particle-bonding heat treatment that naturally Sometimes confusion exists about the difference between
densifies packed powders when they are heated to a tempera- metal powder injection molding and traditional presssinter
ture where atomic motion is active. For most powders, sinter- powder metallurgy (PM). The standard powder metallurgy
ing starts well below the melting temperature. Besides
considerable strengthening, one common result of sintering
small powders is component densification; the sintered com-
ponent is smaller than the molded component. Indeed, in MIM
the final component is about 15% smaller in each dimension
due to annihilation of the space initially filled with binder.
Figure 1.4 provides before and after versions of a molded
component to illustrate sintering shrinkage. After sintering
there is no evidence of the initial powders or polymers. Thus,
performance attributes of MIM products are similar to the
properties cited in common engineering handbooks.
Secondary steps after sintering are very much the same as
applied to all metals. The components can be heat treated,
electroplated, drilled and tapped with threads, and welded.
After sintering there is nothing different in the MIM material
compared with that of standard metallurgical treatments; there Figure 1.5. The key forming step in traditional presssinter powder
are no artifacts from the powders. metallurgy is uniaxial die compaction. The powderpolymer mixture
Metal powder injection molding is a widely accepted net- is fed into the tool cavity by gravity and the upper and lower punches
shape production route that competes with investment casting press the powder with a high pressure, which is followed by ejection

Figure 1.6. Examples of side-wall undercuts and holes possible in

Figure 1.4. Example of sintering shrinkage in a MIM latch MIM, but impossible via die compaction because of the need to
component: the lower version is as-molded and the upper version easily eject the component along the pressing axis (component
is after sintering (components courtesy of Megamet Solid Metals) courtesy of Retco Tool)

3
 Table of Contents
Chapter 1: Introduction
forming step, sketched in Figure 1.5, relies on uniaxial (ver-
tical axis) compaction to press the powder into a green body.
Pressing is a long-established technology for shaping a broad
range of materials into low-height parts. It is the mainstay in
compacting pharmaceuticals pills. The compacted shape is
formed using upper and lower punches working on the pow-
der placed inside a laterally constraining die. Die compaction
is best at forming squat shapes that are easily ejected from the
tooling. The ejection step requires the sides be parallel, a re-
striction in geometry not seen in MIM. Thus, undercuts and
holes perpendicular to the main axis are possible in MIM, as
illustrated by the molded component shown in Figure 1.6.
A typical compaction pressure for PM die compaction is
near 700 to 840 MPa (100,000 to 120,000 psi). Unfortunately,
the proportionately high ejection stresses can crack the weak
pressed powder during removal from the tooling. A minimized
height helps avoid ejection cracks, so as is evident in Figure
1.7, the shape complexity is all in the pressing direction. Little
shape detail is permitted perpendicular to the pressing direc-
tion. Further, because of friction between the powder and
moving tooling, die compaction results in density gradients in
the pressed powder. This contrasts with MIM, which is uni-
form in forming pressure with a lower pressure, meaning
smaller pressure and density gradients. Since density gradients
result in distortion during sintering densification, the homo-
geneous density in MIM gives uniform dimensions after high-
temperature sintering densification. In die compaction, the
typical sintering temperature is lower to avoid distortion that
accompanies densification. Consequently, components fabri-
cated from pressed powder:
are sintered at lower temperatures where sintering den-
sification is avoided,
are often machined after sintering to correct critical
dimensions,
Figure 1.7. A good example of a traditional powder metallurgy
component formed by die compaction, where from a top view there
are internal splines, holes, and gear teeth, but from a side view the
squat component has little complexity (component courtesy of
Metaldyne)
4
if sintered to full density, may have poor dimensional
uniformity, and
if not sintered to full density, then their properties are
degraded by pores.
In short, with die compaction the choice is either full
density giving high properties with poor dimensional control,
or good dimensional control with lower density and lower
properties.
As an illustration, Figure 1.8 compares the scatter in final
dimensions for a die-compacted-and-sintered copper compo-
nent and the same component fabricated by molding and sin-
tering. The scatter is measured by the standard deviation in
dimensions, both as-formed and after sintering. Note that,
after sintering, the MIM component has a significantly lower
standard deviation since MIM forming is hydrostatic, thereby
minimizing green-density gradients. Thus, after sintering to
full density, the MIM process is capable of tighter dimensional
control, because it avoids the density gradients encountered
in compaction. For die compaction, the standard protocol is
to use larger powders vs. MIM and to sinter at lower temper-
atures for shorter times to avoid distortion during sintering
densification.
In powder metallurgy systems, residual pores dispropor-
tionately reduce the particleparticle bonding. Under stress,
fracture occurs by cracks jumping between residual pores.
MIM products are sintered to near-full density while 85% den-
sity is typical in PM. Consequently, the properties fall rapidly
due to a short-circuit failure path associated from pores. As a
consequence, the MIM tensile properties are almost twice that
of presssinter powder metallurgy. Differences in component
shape complexity, density, and performance means that tradi-
Figure 1.8. A comparison of dimensional scatter for molded and
pressed copper powder compacts sintered to full density. This bar
chart shows similar green dimensional scatter, but the pressure
gradients in die compaction result in a larger scatter for the
sintered dimensions vs. the binder-shaped MIM variant (data
courtesy of John Warren)
 Table of Contents
tional presssinter powder metallurgy and MIM do not com-
pete with one another. The performance penalty from pores is
evident in Figure 1.9, which plots the tensile-strength change
with sintered density for a steel (Fe-2Ni-0.5Mo-0.3C) fabri-
cated by the PM presssinter route (sintered at 1,120C or
2,050F). At 15% porosity the properties are all less than half
the full-density values, and impact energy is only 16% of
full density. For comparison, when fabricated by MIM and
subjected to heat treatment, the tensile strength is 2,000 MPa
(290 ksi).
Improved properties from higher densities are driving the
shift to MIM vs. presssinter powder metallurgy. Full density
is increasingly of value, as evident by the trend shown in Fig-
ure 1.10. This plot gives the relative change in product density
for presssinter ferrous powder metallurgy between 1989 and
Figure 1.9. These four curves illustrate how porosity degrades
sintered properties for a die-compacted 4630 steel sintered at
1,120C (2,050F). The impact properties are most seriously
degraded while the tensile strength is less sensitive. In contrast, the
MIM variant is capable of reaching 2,000 MPa (290 ksi) tensile
strength after sintering and heat treatment.
Figure 1.10. The relative trend in final sintered density for ferrous
presssinter powder metallurgy components. These data show the
progressive move to higher performance increasingly forced the
production at higher density levels, an area compatible with MIM
(data courtesy of Diran Apelian).
Chapter 1: Introduction
1999. Clearly, disproportionate growth took place in the
higher density range, which matches with MIM.
A key point in the comparison of MIM with PM is the link-
age of performance to microstructure. Figures 1.11 and 1.12
provide a pore-structure contrast between presssinter 316L
stainless steel and MIM 316L stainless steel, which is sintered
to near-full density. In light microscopy the pores appear black
since there is no reflection. Because the sintered MIM product
is nearly free of pores, the performance characteristics
compete with other metal forming processes. Admittedly, die
compaction results in a lower-cost product, so when it can de-
liver the shape and required properties, it is the economical
choice. As outlined in Table 1.1, PM and MIM only have the
use of metal powders as a common base, but otherwise are
very different. Thus, the two technologies are not competitors,
since they have significant differences in powders, processing
steps, properties, shapes, costs, and applications.
Figure 1.11. A typical sintered microstructure for presssinter
316L stainless steel, showing about 15% porosity and a strength
that is just one-third the handbook value (pores are black)
Figure 1.12. A typical sintered microstructure for a MIM 316L
stainless steel with about 2% porosity, a level comparable to
castings (pores are black), and a strength comparable to the
handbook value
5
 Table of Contents
Chapter 1: Introduction

TABLE 1.1. COMPARISON OF MIM WITH PM

attribute PM MIM
powder size nominal 100 m nominal 10 m
powder shape irregular spherical
polymer addition 0.6 w/o 5 w/o
key forming step die compaction injection molding
pressure state uniaxial hydrostatic
forming pressure 700 MPa (100 ksi) 140 MPa (20 ksi)
sintering temperature 1,120C (2,050F) 1,300C (2,370F)
sintering hold time 20 min 120 min
sintered density 85% 98%
shape complexity one dimensional three dimensional
performance 50% of handbook 100% of handbook
typical sale price per kg $11 $130

Figure 1.13. An example of serrations possible on a firearm

UNIQUE ASPECTS OF MIM
Metal power injection molding produces a wide range of
high-performance, complex-shaped components to net shape,
meaning little or no machining is required to form the desired
final body. Because of the high final density, MIM products
are equivalent to other net-shape fabrication routes.
Besides metals, a range of materials are molded, including
oxides, nitrides, and carbides, as well as several composites.
Further, novel compositions have emerged for customized
applications in lightweight devices or where controlled poros-
ity is desired, and even laminated or functionally designed
structures.
Secondary benefits include high equipment productivity,
high material utilization, good surface finish, and good toler-
ances. For example, producing both internal and external
threads in the molded component is an option, thereby avoid-
ing machining. Also, serrations, waffle patterns, part identifi-
cation numbers, and insignias can be molded directly into the
component. One example is shown in Figure 1.13 where ser-
rations for gripping are molded directly into the component.
Controlled pores are possible, and even stratified pores or
phases can be placed in specific areas in a component to pro-
vide custom designed functionality.
For the component producer, MIM provides a means to
form large production quantities. Some components are pro-
duced at rates exceeding 200,000 per day. On the other hand,
small production runs are possible, with as few as 5,000 parts
per year. However, as with all technologies, the essence
centers on economics. Metal powder injection molding is cost
advantageous for complex shapes when compared with op-
tions such as machining, casting, or forging. This advantage
comes from the elimination of machining or grinding as typ-
ically required for precision components. Also, since runners,
sprues, and damaged moldings can be recycled, material use
is high. This is especially important for costly raw materials
such as refractory metals, titanium, superalloys, and precious
metals. Thus, MIM is a three-dimensional forming technology
that targets discrete components built on a foundation of plas-
component
tic injection molding. About the only limitation is the eco-
nomic availability of small powder. These attributes are illus-
trated by Figures 1.14 through 1.21.
Obviously, MIM has gained much credibility and achieved
broad industrial acceptance. Unlike traditional presssinter
powder metallurgy, where the final structure is porous and the
pores hinder finishing and welding, MIM components do not
have this problem. Further, the properties are competitive with
other metalworking processes: for example, corrosion resist-
ance for MIM components rivals most other processes.
HISTORICAL EVOLUTION
First demonstrations of the PIM technology were per-
formed when plastic injection molding was a new technology.
An early product was the ceramic portion of the automobile
engine spark plug molded in the 1930s and patented in the
40s. Demonstrations with cemented carbides (compositions
based on WC-Co) date to the 60s, and metal powder injection
molding has been in constant use since the 70s. Over the
years, many products and technologies have been developed,
but widespread commercial success did not take place until
the 90s when standardized processes emerged. This long in-
cubation period reflects the fact that
early technologies were proprietary and closely
guarded,
the required small metal powders were not readily
available,
there was no trained workforce,
early equipment was built from scratch with poor
process-control capabilities,
designers did not have design rules and material proper-
ties, and
optimization was impossible, since the scientific basis
for MIM was missing.
Metal powder injection molding was successful in the early
days because of captive operations at IBM, Citizen Watch,

6
 Table of Contents
Chapter 1: Introduction

Figure 1.15. A stainless steel MIM disk-drive rotational

Figure 1.14. A thin-walled, finned titanium component fabricated component with tabs over the undercut grooves (component
by MIM as a pivot for a toy (component courtesy of Epson-Atmix) courtesy of Advanced Materials Technologies)

Figure 1.16. Hexagonal screw holders fabricated for a hand tool Figure 1.17. An example of the shape complexity and
using MIM stainless steel (component courtesy of Advanced perpendicular holes and undercut features possible using MIM
Powder Products)

Figure 1.19. An example of a MIM component with molded slots,

Figure 1.18. An example of the MIM design freedom is seen in this protrusion, tab, and holes, illustrating the possible design
coupling with merged internal channels and perpendicular flexibility
mounting hole and undercuts and identification text all formed in a
single molding step

7
 Table of Contents
Chapter 1: Introduction
Figure 1.20. A microminiature latch used on notebook computers.
It is formed by MIM (photograph courtesy of Kuen Shyang Hwang)
DuPont, Borg-Warner, Brunswick, Remington Arms, Ford,
GTE, Degussa, Seiko/Epson, 3M, Canon, and Swatch. Today
about one-fourth of the MIM firms are still captive, but now
also perform custom molding for other firms.
One of the most significant technical developments was
the computerization of process equipment. Improved control
at all steps has reduced defects and greatly improved preci-
sion. Older firms still tend to rely on secondary operations,
such as coining or machining, to achieve dimensional unifor-
mity, but newer firms have updated controls that provide ex-
cellent mixture homogeneity, closed-loop feedback control in
molding, and precise temperaturetime sintering cycles.
Significantly, the steps required for production of uniform
products are known and essentially the technical evolution
is captured and now attention moves to how to mature the
markets.
MARKET AND TECHNOLOGY STATUS
Over the past four decades the MIM technology has moved
from a curiosity to a widely accepted metal-forming process.
It is one of several net-shape production routes (such as
investment casting), with applicability to most engineering
materials. Initially the technology was closely guarded and
available only via licensing, but the emergence of a supplier
infrastructure generated some commonality in the MIM
community.
Metal powder injection molding is most competitive when
used to form complicated and durable components, especially
when the components require competitive properties. Further,
because of the plastic injection molding underpinning, MIM
is ideal for mass production. Indeed, some production com-
panies do not consider the approach viable for fewer than
200,000 parts per year. On the other hand, aerospace applica-
tions for MIM require only a few thousand parts per year and
the technology has successfully penetrated applications in jet
engines. Recognition of the wide range of applications in
various trade magazines is helping in the identification of new
projects.
8
Figure 1.21. A steel mechanical latch formed by MIM showing a
cantilever arm, undercut grooves, slot, and noncircular through hole
In 2010, the annual sales for MIM products passed $1
billion. It is now applied to a wide diversity of applications.
Regionally, there is a significant difference in emphasis, with
Europe pushing into automotive and luxury applications,
North America pushing into dental, firearm, and medical ap-
plications, and Asia giving emphasis to computer, cellphone,
and electronic applications. Today the market is about 42% to
45% Asia (dominated by Japan, China, Korea, Taiwan, India,
and Singapore), about 24% to 26% North America (mostly
USA) and about the same in Europe, and 4% to 5% the rest
of the world (including South Africa, Brazil, Israel, and such).
Most of the MIM firms are relatively small, and with over 350
production sites and $1 billion in sales, this works out to an
average of $3 million in annual sales per site. About one-quar-
ter of these operations are in-house facilities used to make just
one closely grouped set of productsorthodontic brackets,
handgun components, medical/surgical tools, or semiconduc-
tor-processing tools. The top companies, representing 10% of
the firms, control 80% of the sales. The larger MIM firms that
exceed $20 million in annual sales dominate the technology
in terms of technology evolution and applications develop-
ment.
On a global basis, the approximate breakdown of sales and
production focus by market is given in Table 1.2. These are
global averages, and one must bear in mind that regional
differences are potentially large.
CONTRAST WITH OTHER TECHNOLOGIES
Plastic injection molding is another technology that im-
pacts on MIM. Indeed, once MIM was established, it became
fashionable for plastic injection molders to integrate into metal
and ceramic offerings. The forming machines, molds, and
even molding cycles are similar. Plastic tools can be used to
 Table of Contents
Chapter 1: Introduction

of this book includes a comprehensive set of definitions.

TABLE 1.2. BREAKDOWN OF MIM APPLICATIONS Particlea discrete solid with a size less than 1 mm. Particles
BY MARKETS come in many sizes, ranging up to below the size of a grain
market percent of sales of sand. Engineering particles are measured on a microm-
household and consumer 22 eter (m) scale, which is 10-6 m. For MIM most of the par-
electrical and electronic 21 ticles range from 1 m to 40 m.
automotive 20 Powdera collection of particles that collectively exhibits
hardware and industrial 17 fluid-like attributes, such as flow and an ability to conform
medical 8 to the shape of a container.
firearms 6
dental 3 Bindera polymer mixture that provides lubrication and
military 2 strength to the powder during molding. The binder is crit-
aerospace * ical to feedstock fluidity during molding and to the strength
business machines * of the molded compact. The binder is decomposed and
cellular telephones * evaporated during debinding. Binder formulations gener-
computer devices *
cutting tools * ally rely on molecular-weight paraffin wax or similar mol-
defense * ecules with some higher-molecular-weight plastics. The
hand tools * binder chemistry impacts the debinding process.
instrumentation and sensors * Feedstockthe mixture of powder and binder used in injec-
jewelry and watches * tion molding. Feedstock formulation involves decisions on
microelectronics and optoelectronics *
sporting * the composition, particle characteristics, binder formula-
telecommunication * tion, mixing practice, and ratio of powder to binder. A crit-
ical step is to homogeneously mix the binder and powder
*less than 1%
without damaging the ingredients.
demonstrate feedstock molding. However, after molding all Injection moldinga hydrostatic forming technique applied
similarity ends. Plastics do not have concerns with post-mold- to plastics at relatively low forming temperatures and pres-
ing processing such as sintering and the 15% dimensional sures, and adapted for shaping MIM feedstock.
shrinkage. Greena term from early work on ceramic forming. It is
Other differences between MIM and plastic molding relate widely used to designate the shaped body after molding
to the properties and economics. In plastic molding the resin but prior to debinding or sintering.
(raw material) is a large cost factor. But the debinding and sin- Debindinga step in MIM between molding and sintering
tering steps in MIM are significant expenses not seen in plas- where the majority of the binder is extracted by heat, sol-
tics. This shifts the costing metrics away from material costs vent, catalysis, or other techniques. The debinding tech-
in plastic molding to processing costs in MIM. nique varies with binder formulation. Thermal debinding
As for applications, many fields use both plastic and MIM is the oldest and most common approach, but several op-
components, so this is not substantially different. However, erations rely on alternatives or combinations of methods.
properties are very different, starting with the component Brownlike green, a designation for the component during
mass. Metals are higher in density when compared with plas- processing, in this case when it has been debound, but not
tics, and they have thermal, electrical, magnetic, mechanical, sintered.
and other properties not attainable in plastics. Indeed, plastic Sinteringthe thermal process that bonds and densifies the
molding and metal molding are complementary; MIM is only molded powders. It increases the compact strength nearly
applied when lower-cost plastic moldings cannot deliver the 200-fold via atomic-level events. Most of the mechanical,
needed properties. The useful strength of polyethylene formed magnetic, or other properties of a MIM part are developed
by plastic molding is less than 20 MPa (3 ksi), while MIM in the sintering cycle. It is akin to firing a clay pot after
steels are stronger by a factor of 20 to 100. Only fiber-rein- forming on the pottery wheel.
forced specialty polymer systems (such as Kevlar-epoxy) MIM (metal powder injection molding)the term for form-
compete at these property levels (at costs 10 to 100 times ing metallic components in a plastic molding machine
those typical to polymers). Although the two technologies using polymer binders and metal powders.
seem similar, the engineering properties possible by MIM are Net-shapea compact manufactured to final density and di-
outside the reach of contemporary plastics. mensions without the need for machining. By definition
MIM is a net-shape process.
KEY TERMS
To this point we have encountered a few terms central to RESOURCES
MIM, so some key definitions are given here. The last section R.G. Cornwall and R.M. German, An Analysis of the Powder

9
 Table of Contents
Chapter 1: Introduction

Injection Molding Industry Global Market, Advances in Pow-

der Metallurgy and Particulate Materials - 2001, Metal Pow-
der Industries Federation, Princeton, NJ, 2001, pp. 4.114.16.
R.M. German, R&D in Support of Powder Injection Mold-
ing: Status and Projections, International Journal of Powder
Metallurgy, 2007, vol. 43, no. 6, pp. 4757.
R.M. German and A. Bose, Injection Molding of Metals and Ce-
ramics, Metal Powder Industries Federation, Princeton, NJ, 1997.
Y. Liu, Demand for Precision Components Set to Drive Chi-
nese PIM Manufacturing into the Mainstream, Powder In-
jection Moulding International, 2007, vol. 1, no. 4, pp. 1725.
V. Piotter, T. Benzler, R. Ruprecht and J. Hausselt, Replica-
tion of Micro Components by Different Variants of Injection
Molding, Microsystem Technologies, 2004, vol. 10, pp.
547551.
K. Rodiger, H. Van Den Berg, K. Dreyer, D. Kassel and S.
Orths, Near Net Shaping in the Hardmetal Industry, Inter-
national Journal of Refractory Metals and Hard Materials,
2000, vol. 18, pp. 111120.
D. Whittaker, Powder Injection Moulding Looks to Automo-
tive Applications for Growth and Stability, Powder Injection
Moulding International, 2007, vol. 1, no. 2, pp. 1422.

10
 Table of Contents

Chapter 2
PROCESS OPTIONS

Outline Process Overview

Background
FeedstockPowder Plus Binder
Tooling and Materials
Molding
Debinding
Sintering
Novel Options
Attributes and Advantages
Secondary Operations
Process Simulations
Vendor Differences
Resources

PROCESS OVERVIEW
Background
Most engineers think of injection molding as a forming
technology devoted to plastics, useful for material such as
polyethylene, polystyrene, and glass-filled nylon. On the other
hand, they know that metals such as steel, ceramics such as
alumina, cemented carbides, and other inorganic materials
have several desirable properties when compared with plastics.
The metals and ceramics are stronger, tougher, and stiffer, and
offer significant latitude in electrical and thermal conductivity,
magnetic response, and wear behavior, and can operate to very
high temperatures. Powder injection molding is a synergistic
combination of the performance properties associated with
ceramics and metals and a shaping option that is associated
with plasticsinjection molding. High performance, shape
complexity, and low cost are attributes that indicate a good
application of powder injection molding. Indeed, most appli-
cations for powder injection molding are in metals, especially
stainless steelswhat is termed MIM (metal powder injection
molding).
In both MIM and PIM, small powders are molded in the
desired shape using a thermoplastic binder during the shaping
stage. A common thermoplastic is paraffin wax (also termed
candle wax). It softens and melts on heating and then freezes
on cooling, and repeats this behavior with each heating cycle
without degradation. Compare this with a thermosetting ma-
terial such as epoxy, which can only be heated once and
remains rigid since heat causes the polymer chains to link
together.
After molding with a thermoplastic binder, the shaped
powder is sintered to attain the target properties. A schematic
of the process is given in Figure 2.1. The polymeric binder
lubricates and bonds the particles during the shaping process.
Subsequently, the binder is extracted in a process termed de-
binding where the polymer is dissolved or decomposed by
burning, technically known as pyrolysis. This leaves a porous
powder structure that looks like the molded shape, but it is
weak. However, when the powder is heated to a high temper-
ature, the particles bond and densify due to capillary attrac-
tion, in what is called sintering. There is a tremendous strength
gain in sintering, to the point that the sintered solid is essen-
tially indistinguishable in strength from wrought metal. So un-
like in plastic injection molding, where the plastic is the
object, in MIM the plastic or polymer is only a transient phase,
added to help the metal particles flow and fill out the mold.
The MIM process has many variants, reflecting different
combinations of powders, binders, molding techniques, de-
binding routes, and sintering furnaces. However, in all variants
there are three key technology areas:
feedstock: powders, binders, mixing
molding: rheology, tool design, machine operation
thermal processing: debinding, sintering, heat treatment

11
 Table of Contents
Chapter 2: Process Options

FeedstockPowder Plus Binder
Feedstock is the mixture of powder and binder used for
molding. Figure 2.2 is a scanning electron microscope picture
of a typical stainless steel powder fabricated by gas atomiza-
tion for MIM. It is nearly spherical and exhibits a range of
particle sizes that allows good fitting together of the particles.
Actually, there are many ways to form the small particles used
in MIM, ranging from vapor condensation to oxide reduction.
The best choice changes with material and some of the inher-
ent material attributes, such as melting temperature and reac-
tivity, as well as the desired particle size, particle shape, alloy,
purity, and cost. For example, it is very difficult to melt tung-
sten due to its high melting point (3,410C or 6,170F) so hy-
drogen reduction of a milled tungsten oxide is appropriate,
but for copper it is common to form a melt and to spray that
melt into droplets that solidify as particles (atomization).
A few example MIM powders are compared in Table 2.1.
More than 800 different powder variants have been used in
MIM, so the few examples in this table are simply for illus-
tration purposes. The common name is given in the first col-
umn, and the nominal composition is listed next. The third
column gives the fabrication process. (For those seeking
details on the powder fabrication routes, typical powder char-
acteristics, and standard tests to measure the characteristics

Figure 2.1. Process outline for MIM, where the powder and binder
phases are combined by premixing and intense hot mixing, then Figure 2.2. A scanning electron micrograph of a typical gas-
granulated into feedstock. This feedstock is molded, debound, and atomized stainless steel powder used in MIM. These particles are
sintered to give the final shape about 15 micrometers across, but intentionally have a range of sizes

TABLE 2.1. EXAMPLES OF POWDERS USED IN MIM

material nominal composition, fabrication method median particle apparent density,
w/o size, m g/cm3
copper 99.5 Cu water atomization 14 3.3
iron 98.3 Fe carbonyl decomposition 4 2.6
steel Fe-0.9 C carbonyl decomposition 5 2.7
molybdenum 99.8 Mo oxide reduction 3 2.1
nickel 99.8 Ni carbonyl reduction 8 3.6
316L stainless Fe-19 Cr-9 Ni-2 Mo gas atomization 11 4.1
630 stainless* Fe-17 Cr-4 Ni-4 Cu water atomization 20 3.3
titanium 6-4 Ti-6 Al-4 V gas atomization 32 2.1
tool steel Fe-12 Cr-1.5 C gas atomization 11 4.3
tungsten 99.5 W oxide reduction 2 3.0
*commonly designated 17-4 PH stainless steel

12
 Table of Contents
should see the textbook by the author; R. M. German, Powder
Metallurgy and Particulate Materials Processing, Metal
Powder Industries Federation, Princeton, NJ, 2005.)
In the powder data, the median particle size corresponds to
the 50% point on the cumulative particle-size distribution and
it is also termed the D50 or D50 size. At the median size half
of the powder mass corresponds to particles smaller than the
stated size and half of the powder mass corresponds to particles
larger than the stated size. This is different from the mean or
average particle size, and is different from what would be seen
in a microscopic picture of the particles. If the number of par-
ticles is counted from a micrograph, then it would appear the
size is smaller than from the particle-mass distribution. It takes
many small particles to equal the mass of one large particle.
Hence, invariably mass-based powder statistics skew toward
the larger particles. In the last column, the apparent density
tells how the powder packs when poured into a container with-
out vibration. For example, copper had a theoretical density of
8.9 g/cm3, but the water-atomized small cooper powder used
in MIM packs to 3.3 g/cm3 or about 37% of theoretical. Often
suppliers of metal powders for MIM include more exact details
in the data sheet accompanying the powder.
Common binders in MIM are waxpolymer mixtures.
When heated, these binders provide a low-viscosity glue-like
liquid for transporting the powders into the mold cavity. At
the end of the molding cycle, the cooled binder freezes to pro-
vide sufficient strength to resist damage during ejection. A fa-
vorite binder system is 65% paraffin wax, 30% polypropylene,
and 5% stearic acid. It is fully molten near 130C (266F), but
this temperature changes with the molecular weight of the
polypropylene. Stearic acid is a polar molecule that attaches
to the metal powder, yet is mostly the same chemical structure
as paraffin wax. It helps connect the powder and binder and
further eases ejection from the mold. Other polymers such as
polyethylene, polyacetal, and ethylene vinyl acetate are used,
and even cellulose, water, beeswax, carnauba wax, or peanut
oil are used in some binders. The binder system always con-
sists of two or more components; the stronger polymer is
called the backbone. The idea is to remove the wax phase of
the binder, while leaving the polypropylene to hold the parti-
cles in place as the compact is heated to the sintering temper-
ature. All of the backbone is removed during heating before
the temperature passes over about 550C (1,022F).
Mixing involves first melting and then dispersing the pow-
derbinder feedstock. It often starts at room temperature to
mechanically disperse the phases and, as heat is applied, the
molten binder is forced between the particles. Mixers come in
two basic classes, batch and continuous. The batch mixers are
reserved for smaller lots, but up to 700 kg (about 1,500 lbs.)
at a time is mixed in the largest batch mixers. Batch mixers
involve a rotary paddle or blade that splits the molten feed-
stock repeatedly over a mixing time from 30 min to a few
hours. More typical to production is a continuous mixer. The
Chapter 2: Process Options
powderbinder cold-mixed feedstock is fed into a twin-screw
mixer where two threaded shafts rotate inside a barrel to shear,
compress, split, and recombine the feedstock as it moves into
a final extrusion step. When the mixture is improperly mixed,
the feedstock inhomogeneities carry over to become molding
defects; thus, high shear is required in mixing to force the
binder molecules between all particles without contamination.
In formulating the feedstock, a typical binder content is
near 40 v/o of the mixture; for steel, that corresponds to about
6 w/o binder, but for a low-density ceramic such as alumina it
will be 14 w/o, while for a high-density metal such as tungsten
it will be less than 3 w/o. Feedstock formulation balances sev-
eral considerations. Sufficient binder is needed to fill all voids
between the particles and to further lubricate particle sliding
during molding. Too much powder makes the feedstock stiff
and difficult to mold, but too much binder leads to distortion
during thermal processing. To find the right ratio of powder
to binder requires a simple mixing experiment. For example,
in the kitchen we can mix baking flour with water or milk, as
illustrated in Figure 2.3. Flour is a powder and milk acts like
a binder. The mixture is a liquid at low concentrations of flour,
and will flow, but as more flour (powder) is added, the mixture
becomes viscous like a paste. Too much flour makes the mix-
ture a solid that will not flow. As the Goldilocks Principle says,
there is a balance that is just right.
Best molding success is attained when the hot feedstock
mixture viscosity is near 100 Pas (1,000 poise). As a bench-
mark, this is thicker than most paints (which are small parti-
cles such as titania dispersed in binders) and more like
toothpaste (which also consists of small oxide particles such
as silica, calcium phosphate, alumina, or calcium carbonate
in a binder formed from cellulose or gum dissolved in water),
but not like cold ice cream.
The final step in feedstock preparation is to form granu-
lated pieces such as shown in Figure 2.4. The pellets are
loaded into a molding machine via the hopper. Premixed feed-
stock is available from several companies. Accordingly, in
MIM, two variants exist with respect to feedstock:
older and larger MIM operations that formulate and
mix their own feedstock
newer or smaller MIM operations that purchase feed-
stock from a supplier
The situation is comparable to buying premixed cookie
dough as opposed to making cookies from scratch. There are
merits to both approaches, but the time, expertise, and capital
cost of the mixing equipment often leads to a decision to pur-
chase feedstock. More than a dozen feedstock suppliers exist,
each with a range of materials.
A homogeneous feedstock mixture is needed to maintain
process control. Because feedstock viscosity is sensitive to
composition, inhomogeneities impede uniform flow in the
mold cavity. At higher temperatures the polymer thermally
expands more than the powder, effectively increasing the rel-
13
 Table of Contents
Chapter 2: Process Options

ative polymer content and lowering the viscosity. Thus, injec-

tion molding is performed at temperatures over the binder
melting point. When the mixture is subjected to shear, the par-
ticles tend to concentrate, align, and streamline, leading to an
effective lowering of viscosity as shear rate increasessimilar
to race cars drafting behind each other to reduce drag. Because
of this effect, it is possible to see a 10-fold decrease in viscos-

Figure 2.4. Pellets of granulated copper feedstock ready for use in

the molding machine

ity for every 10-fold increase in shear-strain rate; this is

termed shear thinning. Unfortunately, this attribute means the
filling of the tool cavity must be carefully controlled to avoid
separation of ingredients or incomplete mold filling.

Tooling and Materials

The tool cavity or mold for MIM is constructed as an en-
(a)
largement of the final part. The space taken up by binder in
the feedstock is removed during sintering densification. Thus,
the tool cavity is positively dilated to compensate for that
shrinkage. For example, Figure 2.5 is a picture of a sintered
medallion resting on the tool cavity used for molding, illus-
trating the shrinkage. Usually about 15% sintering shrinkage

(b)
Figure 2.3. Demonstration of how powderbinder rheology varies
with the ratio of powder to fluid phase, in this case using flour
and milk; (a) low solid-particle content giving a fluid that flows, Figure 2.5. A sintered tungsten medallion resting on the tooling
(b) a high solid-particle content mixture that is stiff and not able to insert used for molding, illustrating typical sintering shrinkage, in
flow. MIM relies on a critical viscosity that allows flow in a high this case for a tungsten heavy alloy (components courtesy of Anita
solids loading Hancox)

14
 Table of Contents
takes place in each dimension, so tooling is dilated by a factor
related to the shrinkage. For example, if a 25 mm (1 inch) final
size is desired after sintering and the feedstock has 60 v/o
solid-particle loading, then sintering to 100% final density oc-
curs with 15.66% linear shrinkage in each dimension. For this
case, the starting tool size needs to be 29.64 mm (about 1.18
inch) in size to shrink to the desired 25.00 mm final size. The
tooling dilation factor of 1/(10.1566) = 1.1857 is directly re-
lated to the sintering shrinkage. To summarize:
final size = 25.00 mm
final density = 100%
starting density = 60%
sintering shrinkage = 1 (60/100)1/3 = 0.1566 or 15.66%
dimensional-dilation factor = 1/(1 0.1566) = 1.1857
required initial size = (25.00)(1.1857) = 29.64 mm
sintering dimensional change = 29.64 25.00 = 4.64 mm
confirm sintering shrinkage = 4.64/29.64 = 15.66%.
Although tool dimensions are expanded to allow for sin-
tering shrinkage, angles are generally preserved (except where
gravity and substrate friction during shrinkage have an effect).
However, mold design is much more complicated than simply
dilating the part dimensions. Many critical decisions are re-
quired on features, including the following:
parting line location: where the mold will open to extract
the component
gate size and location: how the feedstock will enter the
mold
taper: whether a small taper can be included to ease ejec-
tion from the tooling
ejector pins: where and how many ejector pins are
required for ejection
slides: whether tool motions are required to add features
perpendicular to the parting line
other motions: for example, whether threads are to be
added using unscrewing motions
Molds for MIM are constructed in a manner similar to how
plastic injection tooling is formed, so MIM vendors largely
rely on the same tool-and-die industry. Mold design and con-
struction takes time, and often this is a slow step. In spite of
many efforts to embrace rapid prototyping and other routes to
quick tooling, the tried and true machining approaches persist.
Machining cuts the mold cavity into a mold base and adds the
slides, gate, runner system, ejector pins, and cooling channels.
It is common to see tool design and tool construction run sev-
eral weeks. In practice most tool sets are first created with
outer dimensions set to the lower end of the component toler-
ance band and the inner dimensions set to the high end. This
allows for final mold size adjustments after first test pieces
are sintered. Once precise shrinkage factors for each dimen-
sion are known, the final mold cavity dimensions are appro-
priately modified by final machining or polishing.
Within a mold, the number of cavities ranges from 1 on up
to 320. A single-cavity tool is satisfactory for low production
Chapter 2: Process Options
Figure 2.6. Calculations for the optimal number of cavities in a
molding operation for components of mass ranging from 1 g (upper
side of the solid band) to 100 g (lower side of the solid band). The
lowest cost depends on the batch size. These calculations are based
on a total-project cost minimization, showing how larger batches
drive MIM toward multiple cavity tooling
quantities. The lowest project cost depends on the number of
parts per year to be produced. In rough terms here are some
typical break points and a plot for components ranging from
1 to 100 g is given in Figure 2.6:
below 100,000 parts per year1 cavity
at 1,000,000 parts per year4 cavities
by 10,000,000 parts per year12 cavities
These are not fixed rules, since many parameters are
involved, but the calculations do illustrate the trade-off
between tool cost and molding productivity. For larger parts
this is not true, since the maximum clamping force available
on the molding machine becomes a limitation.
The materials for tooling are usually hardened steels, such
as P20. When heat treated, this steel easily resists wear from
the feedstock. Harder tool steels and even cemented carbide
inserts are used for tooling for high-production-quantity situ-
ations, the latter requiring electro-discharge machining. Re-
ports from various MIM operations give 2 million shots
between mold refurbishing treatments for a gas-atomized
stainless steel feedstock. Water-atomized stainless steel tends
to have a harder particle surface and nonspherical particle
shape; both factors significantly increase tool wear. However,
this is a significant advantage for metals when compared with
ceramics. Ceramic particles are very hard and angular, a com-
bination that requires mold refurbishing as often as every
50,000 shots.
Molding
The conversion of feedstock into a defined shape occurs in
the molding machine. For each component, a mold is designed
that is oversized to compensate for the sintering shrinkage.
15
 Table of Contents
Chapter 2: Process Options
During molding the feedstock granules are loaded at room
temperature into the hopper of the molding machine. Figure
2.7 is a picture of a production molding machine. It is essen-
tially the same as used for plastic injection molding. Attached
to the molder is robot system for removing the component and
staging it for debinding and sintering. Such an integrated han-
dling and quality system is shown in Figure 2.8 in the form of
an unloading robot and conveyor system.
A cross-section schematic helps to understand the feed-
stock flow through the molder. As sketched in Figure 2.9, the
hopper for loading the pellets feeds into a heated barrel with
an internal screw for stirring, compressing, and melting the
feedstock. The end of the barrel has a nozzle that butts against
the mold during filling. The mold is clamped closed up to the
point of component ejection. It is important for the screw to
heat, melt, and stir the feedstock to ensure homogeneity and
no trapped air bubbles. The screw has a taper that compresses
the hot feedstock to squeeze out any trapped air.
Figure 2.7. A production MIM molding machine. The feedstock
pellets enter the hopper and the molded shape is ejected between
the tool halves which are not installed for this picture (photograph
courtesy of Arburg)
Figure 2.8. An integrated robot and conveyor system to move the
green shapes to a staging area prior to debinding (photograph
courtesy of Arburg)
16
When ready to fill the mold, the hot feedstock accumulated
at the tip of the screw is rammed into the mold cavity. At this
point the feedstock consistency is similar to toothpaste. In
molding, the feedstock flows through the nozzle, at the end
of which it enters the tooling through the sprue, runner, and
gate prior to filling the cavity. Once the mold is filled, the
feedstock is cooled to freeze the powderbinder mixture into
the desired shape. When sufficiently cool, the rigid component
is ejected from the mold and the molding cycle is repeated,
typically once every ten to twenty seconds.
Figure 2.10 is a picture of an open mold located in a mold-
ing machine. This is for a turbocharger rotor, so the tool has
eleven slides that are activated by the pins when the mold
closes; the pins are evident on the right side of the picture.
The turbine part is still in the mold to show the possible blade
shape. A screw, barrel, and mold are shown in Figure 2.11a;
Figure 2.9. A cross section through a MIM molding machine to
show the feedstock pathway into the mold. The feedstock enters the
hopper, feeds along the screw where it is heated, stirred, and
pressurized, and flows through the nozzle into the tooling
Figure 2.10. The open mold inside the molding machine. This is a
turbocharger tool cavity that has several sliding motions to form
the vanes on the turbine which is still in the mold (photograph
courtesy of T&N Technology)
 Table of Contents
normally the screw is inside the barrel and not seen. An
alternative tool design based on a circular mold is given in
Figure 2.11b.
A motor turns the screw to stir the feedstock to ensure uni-
(a)
(b)
Figure 2.11. Mold tooling; (a) is a picture of a mold, screw, and
barrel used in MIM, and (b) shows a circular design for the mold
Figure 2.12. The upper sketch illustrates the rotating screw inside
the barrel prior to mold filling. To fill the mold, as shown in the
lower sketch, the screw becomes a plunger and the check ring
stops the back flow of molten feedstock during the forward motion
Chapter 2: Process Options
form heating while maintaining mixture homogeneity. As
shown in the cross section of Figure 2.12, after heating and
stirring the feedstock the screw acts as a plunger to generate
the pressure needed to fill the die. The screw tip has a check
valve, the action of which transforms the rotating screw into
a forward moving plunger to generate pressure to fill the mold.
Figure 2.13a shows a sketch of the internal flow path, indi-
cating the component (two cavities in this case), gate, runner,
and sprue, and Figure 2.13b is a picture of molded parts from
(a)
(b)
Figure 2.13. The flow path in molding is shown in terms of
(a) a sketch of the internal flow path in the tooling for the molten
feedstock exiting the nozzle, and (b) the sprue, runner, and molded
automotive components aligned to show the cavity in the mold
17
 Table of Contents
Chapter 2: Process Options
Figure 2.14. A molded watch case with the sprue still attached
where the blemish will be hidden by the watch mechanism
Figure 2.15. Short-shot images showing feedstock flow to pro-
gressively fill out a bullet component. The flowing feedstock pushes
air out of the mold through a vent located opposite to the gate
Figure 2.16. Vent details on the segmented bullet mold used for the
images shown in Figure 2.15
18
a two-cavity mold, again with the gate, runner, and sprue.
Figure 2.14 is a picture of the ejected part with the sprue
attached; in this case the gate is located under the watch
mechanism and so is invisible to the consumer.
During molding, the molten feedstock flows into the cavity
from the gate. Figure 2.15 gives a sequence of mold filling
shots, a process that usually takes place in less than a second.
A properly located thin vent allows air to escape from the
mold from the side opposite the feedstock entry point. An ex-
ample vent is shown in Figure 2.16. After mold filling, pres-
sure is maintained on the feedstock during cooling to prevent
void formation or sink marks as the binder contracts volume
on freezing.
Peak temperatures in molding are below 190C. Usually the
tooling is cooler than the feedstock, so the feedstock undergoes
a progressive viscosity increase as it flows into the mold and
gives up heat to the tooling. Accordingly, this increasing flow
resistance requires a compensating pressure to properly fill out
the cavity, achieved by programming the instantaneous screw
position, or force, vs. filling time for each mold. This is a key
point of sophistication in MIM, since lower cost molding ma-
chines lack this feature and do not necessarily ensure every
part is free of voids or sink marks. A hydraulic or electric sys-
tem pushes the screw to deliver the needed control for both
screw position and applied pressure. The peak molding pres-
sure varies with the geometry, but might range from 1.4 MPa
to 60 MPa (200 to 8,500 psi). In a few situations, molding is
performed at low pressures on the order of 0.3 MPa (45 psi).
For a thick section the mold is usually kept cold to reduce
cycle times; however, thin cross sections pose a difficulty in
mold filling since the feedstock freezes before fully filling out
the features. In this case the mold is heated during filling and
then cooled after it is filled. Such mold heatingcooling cycles
slow the molding time.
At the high shear rates typical to flow through the gate, the
powder and binder can separate, creating a rapid viscosity
change due to shear thinning. Separation of powder and binder
is exacerbated by sharp corners or rapid direction changes in
the flow paththe heavier particles do not change direction
as easily as the low-density polymers. When the powders con-
centrate, there is a corresponding rise in viscosity. Since mold
filling is sensitive to feedstock viscosity, separation of the
powder from the binder must be avoided. Likewise, since sin-
tering shrinkage depends on the solids distribution in the
molded body, binder separation from the powder leads to un-
even shrinkage and nonuniform final dimensions, seen as
warped components. Coining is a common means to
straighten the component, but at an extra cost.
Multiple-cavity molding is used to obtain more parts per
molding machine per unit time, and up to 40-cavity molds are
used in production, and one operation even relies on a 320-
cavity mold. Multiple-cavity molding presents some interest-
ing balancing and control problems; hence, the trade-off
 Table of Contents
between the molding problems, tooling cost, and machine pro-
ductivity often results in a compromise of about two- or four-
cavity molds.
Debinding
After ejection from the mold, the binder is removed from
the component, hopefully without damaging the powder struc-
ture. Thermal debinding is the easiest to envision and the most
widely practiced. The molded component is slowly heated to
evaporate and decompose the polymeric binder. For this op-
Figure 2.17. Batch thermal-debinding furnaces for heating of the
molded component to burn out the binder phase and presinter
the particles (photograph courtesy of Advanced Materials
Technologies)
Figure 2.18. Hot-water debinding of injection molded stainless
steel tensile bars, where the polyethylene glycol used as a wax
phase in the binder is dissolved from the molded body in first stage
solvent debinding
Chapter 2: Process Options
tion the majority of the binder is selected to evaporate at lower
temperatures while a higher-temperature backbone polymer
holds the particles in place. In this regard, paraffin wax evap-
orates first, leaving polypropylene to hold the particles in
place. Finally, the residual binder evaporates as the porous
powder structure approaches the sintering temperature. A
common burnout cycle then is at least 12 hours for the com-
bined heating, hold, and cooling. A few batch-debinding fur-
naces are shown in Figure 2.17. Trays of parts are inserted
through the front door and a burnoff stack is evident where
the decomposed polymer is combusted to form water and car-
bon dioxide. Note that many newer operations integrate de-
binding and sintering into a single furnace.
An alternative to thermal debinding is to immerse the com-
ponent in a solvent that dissolves some binder, leaving an
insoluble backbone polymer behind to hold the particles in
place for handling. That remaining backbone binder is ther-
mally extracted as part of the sintering cycle.
Early in the development of MIM the selected solvents
were hazardous chlorinated hydrocarbons such as carbon
tetrachloride. Heptane is popular for debinding paraffin wax
and is widely used in spite of flammability concerns. Since
the 1990s, solvent debinding turned to water-soluble polymers
mixed with insoluble backbone phases. Debinding is achieved
by immersion in hot water as illustrated in Figure 2.18, where
polyethylene glycol is substituted for the paraffin wax.
Another option involves catalytic-phase erosion of a poly-
acetal binder. The polyacetal binder is attacked by nitric acid
vapor during heating, generating formaldehyde that must be
properly handled. A polyethylene-type backbone polymer pro-
vides handling strength during heating to the sintering tem-
perature. The laboratory-scale catalytic debinding unit shown
in Figure 2.19 has a front door that is used to insert the com-
ponents and, once a nitrogen atmosphere is established, nitric
Figure 2.19. A laboratory-scale batch catalytic-debinding oven for
debinding the polyacetal binder systems
19
 Table of Contents
Chapter 2: Process Options

acid is added and the byproducts combusted in the stack at the
top of the unit.
In all of the debinding approaches, the erosion of the binder
is from the outside surface. If the system is heated rapidly, then
internal vapor pockets form and create blow holes, blisters, or
even miniature volcanic eruptions. Generally, slow heating to
600C (1,112F), or higher, over several hours in nitrogen re-
moves the binder and presinters the powder. The heating rate
and time for debinding increase as the section thickness in-
creases. For most metal powders and binders, a rule of thumb
is that debinding can penetrate about 2 mm/h (0.08 in./h).
Since the erosion is from both the front and back (or top and
bottom) surfaces, the time for debinding is approximated by
considering the thickest half-section. For example, a wall that
is 10 mm across requires heating at 1C/min (1.8F/min) to at
least 450C (842F) with a hold of 2.5 h (5 mm divided by 2
mm/h) as long as both surfaces can be accessed during debind-
ing. This says that heating requires about 6 hours, with a peak
hold of 2.5 hours, and about 6 to 8 hours of cooling time, so a
12-to-24-hour cycle is typical.
Sintering
Independent of the initial debinding route, final polymer
extraction usually occurs during heating to the presintering
temperature. Full sintering occurs as the powder compact ap-
proaches its melting temperature. Sintering was used thou-
sands of years ago to form bricks and pottery by firing the
shaped green ceramic body to a high temperature. Today,
the same idea is applied in powder metals, technical ceramics,
cemented carbides, and some polymers. Sintering dramati-
cally changes a powder structure into a solid structure. This
is illustrated in Figures 2.20a through 2.20f, a sequence of
high-magnification cross-section pictures of 17-4 PH stainless
steel powder taken after heating to various temperatures up to
1,365C (2,460F), showing progressive particle bonding and
pore elimination during sintering. In these images the pores
are black.
The goal of sintering is to densify the powders by removing
the void space previously occupied by the binder. Obviously
full binder burnout is required prior to densification, or some
polymer residue will be trapped in the pores. When properly
performed, the sintered microstructure shows no evidence of
the original powders and little to no porosity. For example,
Figure 2.21 is a micrograph taken of tool steel after MIM pro-
cessing to near-full density. This microstructure shows desir-
able small carbides dispersed in the steel structure with a small
amount of residual porosity.
The typical sintering shrinkage is from 12% to 18% on
each dimension, nominally 15%, so the molded component is
oversized to deliver the desired final dimensions. Figure 2.22
illustrates the typical uniform, isotropic shrinkage attained in
going from the molded to sintered body, in this case for a med-
ical-pump housing. However, since gravity acts in one direc-
tion, the component mass gives frictional drag on the sintering
support. As a consequence, the reality is slightly less than
isotropic shrinkage. Thus, the common approach is to leave a
little extra on the mold for final dimensional corrections after
trial sintering-shrinkage measurements.

(a) (b) (c)

(d) (e) (f)

Figure 2.20. A sequence of polished and etched cross sections from an injection molded water-atomized 17-4 PH stainless steel powder, where
the samples were quenched from various points in the sintering cycle; (a) 1,000C (1,832F), (b) 1,100C (2,012F), (c) 1,200C (2,192F),
(d) 1,260C (2,300F), (e) 1,300C (2,372F), and (f) 1,365C (2,489F). The black regions are pores (pictures courtesy of Yunxin Wu)

20
 Table of Contents
Figure 2.21. An example micrograph of the nearly pore-free
structure associated with MIM tool steel after sintering; in this
case the prealloyed powder is sintered to nearly 99% density
(micrograph courtesy of Neal Myers)
Figure 2.22. Shape preservation is evident in this pump housing com-
paring the as-molded condition with that after sintering; the sintered
component is smaller, but retains the details and shape of the molded
component (components courtesy of Phillips Metal Injection Molding)
Figure 2.23. A typical small box furnace for sintering MIM
components up to 1,600C in a protective atmosphere of hydrogen,
nitrogen, or argon
Chapter 2: Process Options
Except for oxide ceramics, sintering is usually performed
in a protective atmosphere or vacuum. Sintering furnaces are
a major capital cost in MIM. Usually they are custom fabri-
cated in size, atmosphere, capacity, and cycles depending on
binder, component size, and peak temperatures. Small box fur-
naces, such as shown in Figure 2.23, reach 1,400C to
1,600C (2,552F to 2,912F) with a protective atmosphere
such as hydrogen, nitrogen, or argon. Many production sin-
tering units are larger vacuum batch furnaces similar to the
two shown in Figure 2.24; Figure 2.24a is a refractory metal
(molybdenum, tungsten) construction and Figure 2.24b is a
graphite construction. Such furnaces can be scaled up to sizes
(a)
(b)
Figure 2.24. Two examples of a batch vacuum-atmosphere
sintering furnace used for production MIM sintering: (a)
refractory metal hot zone, and (b) graphite hot zone (photographs
courtesy of Powder Injection Moulding International, Advanced
Materials Technologies, Elnik Systems, and Key Safety Systems)
21
 Table of Contents
Chapter 2: Process Options
Figure 2.25. Continuous debinding and sintering is possible in a furnace such as shown here (photograph courtesy of CM Furnace)
of 2 m (over 6 ft.) diameter and 3 m (about 10 ft.) long, reach-
ing load capacities up a ton per day. For larger production
quantities it is appropriate to move to continuous furnaces,
where sintering and debinding are combined into a single de-
vice with a sequence of zones that require from 6 to 24 hours
to traverse. Figure 2.25 is a photograph of a continuous fur-
nace capable of 1,600C (2,912F) operation. The central
zones are profiled to hold precise temperatures while the com-
ponents move via a pusher-plate conveyor mechanism from
the cold entry to where they are cooled prior to exit.
Since shrinkage is associated with sintering, precise final
dimensions rely on uniform molding and uniform heating,
such that the component shrinks to the specified final size.
When MIM is properly executed, the sintered compact has the
shape and precision of an injection molded plastic, but the sin-
tered metal delivers performance levels unattainable with
plastics. This is because the sintered microstructure is essen-
tially fully dense, so there is no residue from the original pow-
der. Further, proper cooling in the furnace can impart a heat
treatment to the structure to induce excellent mechanical prop-
erties. The sintered product may be further densified, coined,
plated, or machined to complete the fabrication process.
NOVEL OPTIONS
One fact evident in MIM is how inventive everyone can be
once acquainted with the basic concepts. A large range of op-
tions exist to form novel sizes or shapes, novel material com-
binations, or even new compositions. Here are some of the
options that fall under auspices of MIM:
green machiningA green molded compact is machined
to add extra detail or to remove blemishes after molding.
22
This is an ideal way to evaluate design variations without
committing to tooling changes. Commonly green machin-
ing is used to remove gate blemishes or parting line flash.
controlled porosityImportant powder metallurgy prod-
ucts are created with controlled porosity and controlled
pore size for use as flow-control devices, filters, and
bearings. In addition, controlled porosity is a means to
form air bearings and tools for aeration. If an insoluble
phase is mixed into the feedstock prior to molding, then
the size and amount of that insoluble phase determines
the pore size and porosity in the final sintered object.
Several sacrificial additions serve this function, including
sodium chloride (salt). Other sacrificial phases include
polymer particles that are insoluble in the binder.
green joiningTwo green MIM parts are joined to make
a larger, more complex shape by effectively gluing the
two bodies together prior to debinding and sintering.
Variants on this idea include first molding a core material
and then over molding a second material or combining
two pieces with binder along the bond interface. The lat-
ter is a simple means to form a hollow body using two
halves. Demonstrations have even included ball and
socket creations assembled green.
two-material moldingIn some molding operations it is
possible to use molding machines with two barrels, each
barrel containing a different feedstock material. The two
barrels work in sequence to fill the mold, allowing place-
ment of a different feedstock in different regions. Special
two-barrel molding machines, similar to that shown in
Figure 2.26, allow the tooling to move between the two
nozzles for sequential filling. The first step molds the
 Table of Contents
Figure 2.26. A two-barrel MIM molding machine intended to make
multiple material components by first filling part of the mold with one
feedstock, dilating the mold to make room for the second feedstock,
and then rotating to enable final filling with the second feedstock
(photograph courtesy of Advanced Materials Technologies)
Figure 2.27. A hollow handled ceramic spoon formed using gas
pressure filling to blow the feedstock against the mold walls
(component courtesy Arburg).
Figure 2.28. An example large component molded using a low-cost
tool material. This bronze statue weighs 2 kg and was molded in
silicone rubber using low-pressure molding (component courtesy
of Aesthetic Materials).
Chapter 2: Process Options
inner material, and the second step molds the outer ma-
terial. This approach is used to combine magnetic and
nonmagnetic materials in automotive timing sensors.
gas- or water-assisted moldingA fluid phase can push
on the inside of the feedstock during molding to pressur-
ize a hot feedstock bubble against the mold walls, like
blowing bubble gum into a cavity. Warm water or gas is
pressurized to expand the feedstock bubble. An early
demonstration component was a ceramic spoon fabricated
with a hollow gas-filled handle, shown in Figure 2.27.
large structuresCosts usually increase rapidly with
component size in MIM, so novel means are required to
enable large structures in the 200 g and larger range. One
approach for limited production is to form soft silicone
rubber molds around a model and this silicone rubber
mold is filled with hot feedstock at low pressures. Once
the feedstock has hardened, the body is removed from the
mold, debound, and sintered. Examples of the shape com-
plexity, size scale, and feature preservation are given in
Figure 2.28 which is a picture of a bronze component 120
mm (4.8 in.) high, 2 kg (4.4 lb.) in weight that illustrates
the ability to create undercuts, holes, and thickthin
sections.
microminiatureAt the other end of the size scale are
tiny medical and electronic components produced using
small powders and automated molding. Molds are fabri-
cated using ultraviolet lithography techniques developed
for semiconductor fabrication. Since the parts are very
small, part handling must be automated. An early exam-
ple was a stainless steel biopsy tool with 100 m wall
thickness. The next wave of applications are seen in
Figure 2.29, in this case for microarrays of features, such
as used for blood analysis, chemical studies, and even
ultra-capacitors.
composite materialsDifferent powder chemistries are
mixed into the feedstock to form a particle composite
after sintering. If the added particles are insoluble and
hard, then the sintered structure has enhanced wear re-
sistance. Examples include borides in a stainless steel
matrix, titanium carbide in a titanium matrix, and silicon
carbide in an aluminum alloy matrix. An example par-
ticulate composite microstructure is illustrated in Figure
2.30. Wear trials show a small concentration of hard par-
ticles, in the range of 3 v/o, increases wear resistance
from ten-fold to fifty-fold for many alloys.
high elastic modulusThe addition of a ceramic phase
to a metal is useful for increasing the elastic modulus.
For example, steels have been loaded with titanium
boride (TiB2) to increase elastic modulus. With 40 v/o
boride the steel stiffness increased by 60%. An example
micrograph of a titanium carbidereinforced titanium
alloy is given in Figure 2.31.
heterogeneous microstructuresWhen mixed powders
23
 Table of Contents
Chapter 2: Process Options

are sintered, regions of different composition are retained greatly improved fatigue properties. For example, an
even after full density is reached. Often the sintering time Fe-2Ni-0.5Mo-0.4C alloy formed from mixed powders
is prolonged to homogenize these gradients, but in some gives 20% higher tensile strength with a heterogeneous
cases the heterogeneous (hardsoft) microstructure has microstructure vs. that attained with a homogeneous mi-

Figure 2.29. Microminiature molding is possible in several array geometries as used for bioassay, electronic, capacitor, and small-scale
testing applications (photograph courtesy of Taisei Kogyo)

Figure 2.30. Improved wear properties are possible by the Figure 2.31. A high-elastic-modulus composite formed by MIM;
addition of hard particles in a MIM feedstock. Shown here is the in this case the microstructure corresponds to titanium carbide
microstructure of stainless steel reinforced with tungsten carbide reinforcement in a titanium matrix (micrograph courtesy of
(micrograph courtesy of Anthony Griffo) Timothy Weaver)

24
 Table of Contents
crostructure. An example of the microstructure hetero-
geneity is shown in Figure 2.32, where the difference in
properties point-to-point is tailored to improve fracture
properties.
nitrogen alloyingIt is well known that nitrogen has
similar effects on the strength of steel as carbon. In sin-
tering it is possible to infuse nitrogen into the iron in a
way not possible with casting. Thus, new alloys, such as
extra-high-strength tool steels with carbon-nitrides or al-
loys devoid of nickel (both nitrogen and nickel are
austenite stabilizers) become available from MIM and
nitrogen sintering. One very successful product is based
on forming austenite (face-centered cubic iron) as a non-
magnetic stainless steel for use in watches. The nickel,
which is a common austenite former in stainless steels,
is eliminated and that avoids concern over skin irritation
from the common nickel allergy found in humans.
Figure 2.32. Heterogeneity in a full-density microstructure formed
by mixing two metal powders (nickel and iron) together in the
feedstock. This softhard microstructure is useful for forming
higher toughness materials
Figure 2.33. Computer tomography image showing one-fourth
section in a distended titanium structure looking from the inside. This
component is formed by MIM using a pore-forming agent. The porous
rim is where tissue ingrowth is desired (image courtesy of Eric Baril)
Chapter 2: Process Options
tissue scaffolds A relatively important new development
is the creation of MIM tissue scaffolds for implants in hu-
mans. Usually the metal is titanium and the structure is
molded to about 40% solid density to match human bone,
with an elastic modulus near 10 GPa and strength near
100 MPa (about 15,000 psi). The ideal pore size is about
100 to 200 m, much larger than the particle size. These
scaffolds are treated with hydroxyapatite (calcium phos-
phate) to induce bone growth to serve as anchors for
implants, artificial teeth, or trauma repairs. Early demon-
strations relied on sacrificial pore-forming agents, such
as plastic particles or salt, mixed with MIM feedstock.
When that pore former is removed the resulting void cre-
ates a distended structure with desired pore size and
porosity, such as shown by the computer tomography
scan in Figure 2.33. Another variant relies on a preform
polymer foam that is filled with molten feedstock. When
burned out and sintered the titanium takes on the negative
image of the foam (where there previously were pores it
is now solid and where is was previously solid there are
now pores).
ATTRIBUTES AND ADVANTAGES
Metal powder injection molding is applied in the produc-
tion of complex-shaped, high-performance components for
applications ranging from surgical tools to wireless-commu-
nication components. The technology overcomes the property
limitations inherent to plastics, the shape limitations of tradi-
tional powder compaction, the costs of machining, the pro-
ductivity limits of slower routes such as isostatic pressing, and
the defects, surface finish, and tolerance limitations of casting.
Thus, the best applications for MIM are where plastic molding
would be capable of forming the shape, but plastics lack the
mechanical, thermal, or other properties.
As noted, MIM competes with several shaping technolo-
gies such as sand casting, die and investment casting, machin-
ing, cold isostatic compaction, and slip casting. It is cost
competitive as shape complexity increases, but is often too
costly for simple shapes that might be die pressed or screw
machined. One outstanding attribute is the material range.
MIM excels in producing stainless steel, steels, tungsten, ti-
tanium, copper, aluminum, superalloys, tool steels, electronic
alloys, magnetic alloys, ceramics, and cemented carbides,
many of which are difficult to machine. It has been applied to
advanced materials such as ceramic superconductors, alu-
minide intermetallics, specialty ceramics such as titanates, ex-
otic metals (gold, platinum, rhenium, niobium, iridium, and
silver), and ceramicceramic composites.
Thus, as a simple statement:
Powder injection molding is best applied to the
fabrication of complex, smaller components,
especially those that would otherwise require
multiple assembly or forming operations in
25
 Table of Contents
Chapter 2: Process Options

Figure 2.35. A two-material component formed into an automotive

rotational sensor using both magnetic and nonmagnetic feedstock
(component courtesy of Advanced Materials Technologies)

Figure 2.34. An example of how texture and identification

numbers, or other imprints, are molded directly into a MIM
component; in this case the company founding date is included in
the stainless steel component (component courtesy of Advanced
Materials Technologies)

alternative production routes. It excels when a

large production quantity is desired with high-
performance attributes.
Other attributes are worthy of notice. For example, it is
possible to produce both internal and external threads in the
molded component, avoiding post-sintering machining. Also,
names, logos, and various identification insignias can be
molded directly into the component, as illustrated in Figure
2.34. Furthermore, the matte surface finish is usually an ad-
Figure 2.36. Plot of thermal properties important in semicon-
vantage when compared with other fabrication routes. ductor heat-dissipation devices. This plot of thermal conductivity
Early successes with MIM came by replacing existing and thermal-expansion coefficient shows the value of a tungsten
components at a lower cost with improved reliability. For met- copper materials formed by MIM and how the thermal expansion
als these largely were conversions from investment castings fits with the shaded range desired for microelectronics
and for ceramics they largely came from precision-ground
structures. Today, MIM is moving into a phase where design-
ers have learned how to use the process to their advantage.
Such thinking leads to novel applications where materials, de-
signs, and applications are combined to maximize benefits.
For example, Figure 2.35 is an example of a two-material
component for automotive use. The outside three tabs are
magnetic but the central core is nonmagnetic. This innovation
drives MIM and constantly exploits the advantages of the
technology.
As a further example, components for microelectronic ap-
plicationsheat dissipation and microwave absorptionare
a growing application for MIM. New microcomputers contain
millions of transistors on a single chip, with interconnects of
submicrometer widths. The high computing power and small
chip size generate heat that needs to be removed to maintain
Figure 2.37. Microstructure of a high-thermal-conductivity
computing performance. Thermal fatigue is a problem with composite with low thermal-expansion coefficient. The composite
high-thermal-expansion heat-dissipation materials such as alu- is a mixture of tungsten (lighter phase) and copper (darker phase)
minum, used in early designs. formed by MIM

26
 Table of Contents
Chapter 2: Process Options

Accordingly, MIM is a fabrication route for the creation of

unique designs and materials. Only a few combinations of ther-
mal-expansion coefficient and thermal conductivity exist that
match the requirements. Figure 2.36 plots these two thermal
properties to show the desired thermal expansion coefficient
window. Note the unique position of the MIM W-Cu materials,
which are cost-effective solutions, but are also heavy. This
composition is processed from a mixture of copper and tung-
sten powders, giving the microstructure illustrated in Figure
2.37. It consists of a sintered tungsten grain network with a
high-thermal-conductivity copper matrix. This is yet another
example of the innovation possible in MIM technologies.

SECONDARY OPERATIONS
After the component is sintered, many additional changes
in size, shape, surface finish, or heat treatment are possible.
Since the sintered structure is the same as wrought material,
the response to various secondary fabrication steps is the same
as seen with wrought materials.
Some of the common steps include the following:
coining or cold deformationA sintered compact is
forced to conform to a rigid mandrel or substrate and in
doing so the body is slightly deformed to match the sub-
strate. This allows for proper sizing of slots or holes, or
reduced spread in dimensions. Coining is also used to
flatten thin sections that tend to warp in sintering.
hot deformationThe sintered compact is heated and de-
formed by a rapidly applied loading stroke to ensure
proper size and density, and proper heat treatment (often
the compact is quenched from the hot deformation step).
For a steel, the sintered strength jumps from 500 MPa to
720 MPa after hot deformation. In a few materials hot
deformation is a means to remove residual porosity.
machiningAll common machining operations are ap-
plied to MIM components, often to add threads, under-
cuts, grooves, or other special features that would be
difficult or expensive to place in the tooling. It is a good
means to add drilled holes after sintering, as evident in
the solenoid housing shown in Figure 2.38. Careful ex-
amination shows this body has holes and a staked post
added after sintering.
heat treatmentSintering leaves the material in an an-
nealed condition due to the high temperature and slow
cooling rate. For low-carbon ferrous alloys, this is of lit-
tle consequence, but for high carbon levels there is much
change possible via post-sintering heat treatment. For
MIM materials, heat treatment is possible within the
cooling cycle or by use of a properly designed post-sin-
tering cycle. In some cases a cycle is designed to maxi-
mize strength and hardness, while in other cases the
material is cooled to induce a soft structure. Precipita-
tion-hardened stainless steels especially require cycles
of heat and hold to properly optimize the mechanical
Figure 2.38. An example of value added via secondary operations
in a solenoid valve body, formed from a nickeliron alloy. The
sintered component has machining on the base, added holes,
threads, and a staked core (components courtesy of Advanced
Materials Technologies)
Figure 2.39. An example microstructure from the laser-welded
joint between two MIM components showing excellent bonding
(photograph courtesy of Rene Cooper)
properties.
surface carburizationCarbon is important to attaining
high strengths in steels. A high surface hardness is at-
tained with carbon surface additions using a heating
cycle with an atmosphere containing methane. Surface
carburization cycles result in some loss of dimensions,
so trade-off is required between surface hardness due to
the addition of carbon (carburization) and dimensional
accuracy.
joiningLike other metallic components, MIM compo-
nents are joined by welding, brazing, or even adhesive
techniques. For the most part, MIM materials behave the
same as standard metals. Figure 2.39 is a cross section
through a weld between two MIM stainless steel parts,
showing a refined microstructure with excellent strength.
surface treatmentsSurface treatments such as polish-
ing, coating, painting, cleaning, anodizing, plating,
sealing, and laser glazing are all applied to MIM com-
ponents. Surface-hardening treatments sustain a tough

27
 Table of Contents
Chapter 2: Process Options
core with a hard surface. Electroplating is used for either
improved aesthetics or corrosion resistance. Most of the
electroplated coatings are chromium, nickel, copper,
gold, zinc, silver, or cadmium. Nickel electroplate is a
favorite on instrumentation, firearm, or magnetic com-
ponents, and cadmium was a favorite on many electronic
components, but its use is declining because of environ-
mental problems.
Clearly, after sintering, MIM components behave similarly
to other metallic components. No special handling procedures
or steps are required in the secondary operations for MIM parts.
PROCESS SIMULATIONS
Since the early 2000s, the metal powder injection molding
process has been supported with a variety of process simula-
tions. About eight simulations currently exist, and of those,
six are available as commercial codes. Much difference exists
in the sophistication of the codes, the type of computer re-
quired for the solutions, and the time for the solution. Some
of the codes are written exclusively for MIM, while others are
adaptations of casting codes or plastic injection molding
codes. Only one integrates molding, debinding, and sintering
into a stepwise set of simulations.
The simulations are focused first on the filling stage, and
determining how the hot feedstock enters, flows, and cools in
the mold. These require conservation of mass, energy, and mo-
mentum. Images from the filling stage are useful in determin-
ing the sequence of feedstock flow and filling in the cavity so
that gate and vent are properly located and sized to avoid jet-
ting, incomplete filling, or trapped air. Figure 2.40 is an image
from one such filling simulation, showing via coloration the
time-based advance of the flow front.
Next is the packing stage where the proper pressure is de-
termined for the hydraulic system based on calculation of the
required pressure in the mold. Too much pressure and the com-
Figure 2.40. Computer simulation of mold filling for a copper
fitting, showing the gate location and progressive fill vs. time
(images courtesy of Seong Jin Park)
28
ponent sticks in the mold and will not eject when cooled, while
inadequate pressure results in sink marks where the component
shrinks away from the mold during cooling before ejection.
After packing, the calculations shift to the cooling stage and
the extraction of heat from the molded feedstock. Most of the
solutions are based on finite difference and finite analysis so-
lutions, starting with the input material and mold parameters.
An important supplement takes the subtle density gradients
calculated in the molded green body and imports the gradient
information into subsequent sintering simulations to predict
density, microstructure, and final component size and shape.
Generally a skilled engineer is required to fully benefit from
the process simulations, and many of the MIM shops are too
small to make this commitment. Consequently, hands-on ex-
perience is favored in the smaller shops over process simula-
tions during tool and process design. However, considerable
sophistication exists and a few firms have made excellent
progress in integrating process simulation into the product of-
fering. Reports on savings show that on average the extra cost
for process simulations is regained by faster component deliv-
ery with fewer defects. In an overall cost analysis, the consen-
sus is a 15% cost savings when simulation is used. It slows the
initial phase of process development, but helps avoid surprises
and delays when the tool is placed into production.
VENDOR DIFFERENCES
There is no standard MIM process. Thus, variations in
MIM components occur between vendors. Several factors
contribute to the variation, including differences in starting
powders and binders, processing cycles, equipment design and
operation, impurities, and post-sintering heat treatment.
Within one MIM operation, the property scatter is usually very
small. For ductile systems, the strength will typically have a
standard deviation of less than 20 MPa (about 3 ksi) and the
elongation will have a standard deviation of approximately
1%. Yet when compared between operations, properties that
tend to be associated with crack propagation in impact, frac-
ture, and fatigue vary considerably.
The large sensitivity seen in MIM fracture properties traces
to the interaction of impurities (small contents of oxygen, ni-
trogen, carbon, and semi-metals such as phosphorous, sulfur,
silicon, and boron) and the microstructure during cooling from
the sintering temperature or the final heat-treatment tempera-
ture. Usually the cooling is slow, so impurities migrate and
accumulate on grain boundaries or phase boundaries. The
change in atomic bonding due to the concentrated impurities
substantially alters the bulk properties.
For example, stainless steels can absorb nitrogen in sinter-
ing and this tends to increase strength, but also reduces duc-
tility to half the value without nitrogen. Further, if slowly
cooled, residual nitrogen reacts to from a compound with
chromium with the effect that chromium is removed from its
role in corrosion, leading to significant corrosion problems.
 Table of Contents
Chapter 2: Process Options

Hence, depending on the processing atmosphere and final Society of Powder and Powder Metallurgy, 2002, vol. 49, pp.
density there will be property differences. 841845.
One of the challenges facing MIM is to settle on standard-
J.A. Sago and J.W. Newkirk, The Effects of MIM Processing
ized materials and processes to deliver consistent properties.
Parameters on Variations in Part Weights and Dimensions,
For this reason, when a MIM part is to be used in a dynamic
Advances in Powder Metallurgy and Particulate Materials
loading situation, where fracture toughness, impact toughness,
2005, compiled by C. Ruas and T.A. Tomlin, Metal Powder
or fatigue resistance is a concern, it is imperative to perform
Industries Federation, Princeton, NJ, part 4, pp. 7283.
in-service testing to ensure adequate properties. Performance
testing needs to be under the intended loading conditions and G. Schlieper, Back to Basics: Putting the Finishing Touches
in the intended service environment as a basis for establishing to Metal Injection Moulded Components, Powder Injection
the vendor credibility. Moulding International, 2009, vol. 3, no. 3, pp. 3540.
Thus, to summarize vendor effects, between MIM vendors
T. Shimizu, K. Matsuzaki and Y. Ohara, Process of Porous
there is
Titanium Using a Space Holder, Journal of the Japan Society
little difference in hardness and bulk chemistry
of Powder and Powder Metallurgy, 2006, vol. 53, pp. 3641.
small difference in yield strength, tensile strength, and
ductility L.K. Tan, R. Baumgartner and R.M. German, Powder Injec-
potentially larger differences in impact toughness, tion Molding of Bi Metal Components, Advances in Powder
fracture toughness, corrosion resistance, and fatigue Metallurgy and Particulate Materials2001, compiled by
endurance strength. W.B. Eisen and S. Kassam, Metal Powder Industries Federa-
tion, Princeton, NJ, part 4, pp. 191198.
RESOURCES
E.L. Wolf and S.R. Collins, GTA Welding of AISI 316L
S. Ahn, S.T. Chung, S.J. Park and R.M. German, Modeling
Metal Injection Moulded Components, Powder Metallurgy,
and Simulation of Metal Powder Injection Molding, ASM
2002, vol. 45, pp. 201203.
Handbook, Vol. 22, Modeling and Simulation, 2010, ASM In-
ternational, Materials Park, OH.
S.V. Atre, S.J. Park, R. Zauner and R.M. German, Process
Simulation of Powder Injection Moulding: Identification of
Significant Parameters during Mould Filling Phase, Powder
Metallurgy, 2007, vol. 50, pp. 7685.
M.J. Edirisinghe and J.R.G. Evans, Review: Fabrication of
Engineering Ceramics by Injection Moulding. I. Materials Se-
lection, International Journal of High Technology Ceramics,
1986, vol. 2, pp. 131.
M.J. Edirisinghe and J.R.G. Evans, Systematic Development
of the Ceramic Injection Molding Process, Materials Science
and Engineering, 1989, vol. A109, pp. 1726.
J. Fleischer and A.M. Dieckmann, Automation of the Powder
Injection Molding Process, Microsystem Technologies, 2006,
vol. 12, pp. 702706.
R.M. German, Powder Injection Molding, 1990, Metal
Powder Industries Federation, Princeton, NJ.
R.M. German and A. Bose, Injection Molding of Metals
and Ceramics, 1997, Metal Powder Industries Federation,
Princeton, NJ.
A.D. Hanson and S.C. Perruzza, Optimizing Component De-
signs for Metal Injection Molding, International Journal of
Powder Metallurgy, 2000, vol. 36, no. 3, pp. 3742.
T. Harikou, Y. Itoh, K. Satoh and H. Miura, Joining of
SUS316L by Insert Injection Molding, Journal of the Japan

29
Table of Contents
 Table of Contents

Chapter 3
DESIGN PRINCIPLES

Outline Design Cycles

Recognizing Good Candidates
Qualitative Criteria
Quantitative Criteria
Geometric Details
Effective-Density Concept
Surface Finish
Size Range
Mass Range
Geometric Summary
Features
Flat Face
Closed Features
Holes
Imprints, Protrusions, Ribs
Threads
Complexity
Materials
Properties
Design Support
Resources

DESIGN CYCLES
Presented here is a condensed approach to the identifica-
tion of good-design matches with the process capabilities of
metal powder injection molding. Each cycle involving a new
design or product redesign provides an opportunity for em-
bracing MIM. To ensure success, it is important to translate
performance attributes into quantitative-design requirements
that spell out the form (size, shape), function (properties, ma-
terials), tolerances, and cost factors (primary fabrication route
and secondary operations).
Much experience exists on how to deliver the required per-
formance while ensuring success within the MIM process.
During the design cycle, it is best is to have frequent contact
with the fabrication community to rationalize design goals and
process constraints. If this contact is delayed to the quotation
stage, then experiences and knowledge that might improve
productivity and lower cost will be lost. Conceptually, the ide-
alized design cycle is illustrated in Figure 3.1, wherein peri-
odic input from MIM vendors ensures manufacturability as
the design is refined. On each contact the details increase,
touching on component features, materials selection, tooling
issues, and process identification. In general, costs incurred
during design have a relatively small impact on final product
cost, but decisions made during the design process have a
major impact on production costs.
RECOGNIZING GOOD CANDIDATES
Qualitative Criteria
The strengths of MIM are outlined in a qualitative manner
in Figure 3.2. This Venn diagram shows major process
strengths to help in selecting good candidates. The initial qual-
itative criteria look at production quantities, shape complexity,
material performance, and component cost. Metal powder in-
jection molding is not low-cost, except in those instances

31
 Table of Contents
Chapter 3: Design Principles
Figure 3.1. The classic design cycle involves identification of the prob-
lem followed by a product concept development, on to engineering
conceptualization, design, and final definition. Often the MIM technol-
ogy is only brought in after the design is advanced, but this limits the
cost savings and quality improvements. For best MIM designs, it is
important to have reviews with process experts to incorporate changes
in the design to better advance quality while reducing cost
Figure 3.2. This Venn diagram suggests that most justifications for
MIM result from various intersections of considerations of cost,
shape complexity, production quantity, and performance. In each
region, various example components are named as illustrations;
many of these are pictured in this text
32
where the shape complexity is outside the range typical to
other manufacturing routes. Marked on this diagram are ex-
amples of how several applications position with respect to
these criteria.
Improved justification occurs at the intersections or overlap
regions, and examples are listed in this diagram to illustrate
these opportunities. The center of the diagram, where all at-
tributes are acting, is the most effective target for MIM pro-
duction. It is unfortunate that occasionally projects are
initiated without attention to these factors, only to discover
late in the design cycle that costs are out of line or that alter-
native technologies were better matches. Although simple, the
concept is effective in emphasizing the qualitative criteria.
Quantitative Criteria
The attributes of production quantity, shape complexity,
cost, and performance are reflected by several quantitative
factors. First, we will quantify the production-quantity aspect.
Small quantities, at a few thousand per year, give less op-
portunity to amortize tooling and engineering costs and are
generally discouraged by the production community. Often the
best match is with production quantities from 5,000 per year
to more than 100 million per year, such as for cellphone vi-
brator weights, computer logos, eyeglass hinges, and ortho-
dontic brackets. One example disk-drive latch is shown in
Figure 3.3. It is a small stainless steel component produced at
a rate of about 30 million per year in multiple-cavity tooling.
Soft tooling combined with low molding pressures is one
option for the production of trial quantities. One application
for soft tooling is the sculpted female image photographed in
Figure 3.4. This bronze was produced in an edition of three
replicas using low-pressure molding in rubber tooling. Tool
Figure 3.3. This small stainless steel latch was one of the first
high-production-volume disk-drive MIM components, reaching
production rates exceeding 30 million per year (components
courtesy of Advanced Materials Technologies)
 Table of Contents
wear limits such production to a few hundred parts. At the
level of 5,000 per year there are several reports on success for
jet-engine components, specialty firearm sights, military com-
ponents, surgical tools, and luxury items using soft steel tool-
Figure 3.4. An example of a small-production-quantity MIM prod-
uct in the form of a bronze statue produced using a clay original
with rubber tooling in a limited edition of three (component
courtesy of Aesthetic Materials)
Figure 3.5. Gunsight produced using soft steel tooling, meaning
the mold was not heat treated (component courtesy of Injectamax)
Chapter 3: Design Principles
ing. In the context of steel tooling, soft implies the steel is not
heat treated after machining. Figure 3.5 is an example of a
gunsight fabricated in this manner. When the production quan-
tities are large, the tooling and processing technology are well
positioned to deliver outstanding performance levels.
Shape complexity is the next consideration, and at first pass
it depends on the number of engineering specifications re-
quired to define the component. Metal powder injection mold-
ing is best applied when there are more than 20 specifications
(dimensions, locations, surface finish, and such) on the engi-
neering definition. However, the technology struggles when
there are more than 250 specifications. Components with hun-
dreds of features often require several post-sintering machin-
ing operations to hold tolerances. Figure 3.6 plots the
distribution in dimensional specifications taken from a survey
of several production parts. In this survey the average com-
plexity was 74 features. Clearly a vast majority of production
components have less than 100 features. For example, the link
shown in Figure 3.7 has 30 dimensional specifications and is
Figure 3.6. Distribution in the number of features found on
production MIM components, showing the engineering
specifications tend to center on designs with about 74 features
Figure 3.7. Lower-complexity MIM link with 30 specified features
33
 Table of Contents
Chapter 3: Design Principles
at the lower end of complexity, while the optical housing
shown in Figure 3.8 has 165 dimensional specifications and
is challenging. Components, such as wristwatch cases, as
shown in Figure 3.9, are good examples of the complexity
level where the MIM technology excels.
Material and material properties constitute the third factor
to consider. Many materials are available, but a few dominate
the field. When the material is difficult to machine, such as
tool steels, titanium, ceramics, stainless steel, or nickel-base
alloys, then net-shaping is most beneficial. As production
quantities increase the material cost decreases. Because of
widespread use, stainless steels or oxide ceramics are usually
Figure 3.8. Optical housing with 165 specified features, reflecting
one of the more demanding MIM products (components courtesy
AT&T)
Figure 3.9. Three different watchcase designs all fabricated by
MIM, reflecting good examples of the size, shape, and complexity
where MIM has historically been most successful
34
a good choice from a cost standpoint. The powder-cost issue
is a circular argumentstainless steel powders for MIM are
low cost because of the high use, so the low cost drives many
designs to use stainless steel. For low-melting-temperature
materials the large range of forming options generally make
MIM less economically attractive for materials such as alu-
minum, magnesium, and zinc.
As a simplification, the properties attained after sintering
largely converge to the range found in engineering handbooks.
Yet, new materials are emerging within MIM, custom de-
signed for the process, that are not in the handbooks. These
include laminated (hardsoft, magneticnonmagnetic, and
conductiveinsulator) combinations in a single component,
nickel-free nonmagnetic stainless steels, ultrahigh-strength
steels, foamed metals, and mixed-phase wear-resistant com-
posites. These options provide distinct advantages emerging
from basic process attributes. Properties for some special
materials are given in Chapter 5.
The rest of this chapter provides details on the allowed de-
sign features and restricted options. Subsequent chapters pro-
vide details on tolerance capabilities, process yields, materials,
properties, and costs. To summarize, the key criteria needed
to identify good MIM candidates are as follows, recognizing
that a wide variation exists:
production quantity more than 5,000 per year usually,
often exceeding 200,000 per year
number of engineering specifications or features between
20 and 250 and most often in the 50 to 75 range
availability of many materials, although stainless steels,
oxide ceramics (alumina, zirconia), steels, tool steels,
nickel alloys, iron-nickel soft magnetic alloys, and spe-
cialty alloys of tungsten, titanium, cobaltchromium, and
electronic alloys are most popular
property levels generally rivaling handbook tabulations
for similar compositions and heat treatments
GEOMETRIC DETAILS
This section gives more focus on the component size and
mass attributes. Besides the attributes listed above (production
quantities, shape complexity, and performance), several other
geometric design factors are important in MIM.
Effective-Density Concept
A good MIM design is one that would require much mass
removal if it were formed by machining. Here our concept is
similar to that encountered in the aerospace community, what
they term the buy-to-fly ratio. Often the purchased raw mate-
rial is seven times that of the final component, meaning that
85% of the starting material ends up as machining scrap.
In MIM this is viewed in terms of a low effective density.
Density is mass over volume, so an effective density is calcu-
lated as the component mass divided by its outer envelope
volume (outer embracing volume) from which grinding or
 Table of Contents
Chapter 3: Design Principles

machining would start. The concept is illustrated in Figure verse correlation shows that long items are especially low in
3.10, where the sketched outer volume is slightly more than effective density. For example, a stainless steel soup spoon
15 cm3 for a final mass of 27 g. From these two parameters has a length of about 200 mm (8 in.). Based on this plot, an
the effective density is calculated at 1.8 g/cm3 using the mass effective density near 12% (1 g/cm3 out of 8 g/cm3) would
divided by the outer embracing volume. This value is about merit consideration of MIM. Conceptually there would be
23% of the wrought steel density. To phrase it differently, if much mass and time savings using MIM vs. forging followed
the device is machined from a wrought block, assuming that by machining and polishing (the typical process). No surprise,
even the proper sized 15 cm3 block is available, about 77% of the idea of a MIM spoon such as the stainless steel soup spoon
the initial material would be machining scrap. Such waste of shown in Figure 3.13 has been demonstrated by several firms
machining time and initial material adds to the cost of manu- in recent years. Although other tableware has been demon-
facturing by machining. strated, the MIM spoon proved most attractive.
Many cost-effective MIM components have especially low
effective densities, indicating that machining would be time Surface Finish
consuming and wasteful. To reinforce this point, Figure 3.11 Related to the material loss in machining is a similar con-
plots the effective density distribution from a large collection cern over the time spent in reaching a desired surface finish.
of production MIM steel and stainless steel components (gen-
erally these alloys are similar with a range from 7.7 to 8.0
g/cm3 theoretical density). The MIM components are gener-
ally shapes with a low effective density, typically in the range
of 25% of the material density (2 g/cm3 out of nearly 8 g/cm3).
Thus, historically the injection molded components have gen-
erally avoided wasting of 75% of the initial material.
To further illustrate the design window, Figure 3.12 is a
scatter plot of effective density vs. largest dimension. The in-

Figure 3.11. Histogram plot for steel MIM components showing

the effective-density distribution. The effective density was
calculated by dividing the component mass by the its outer
embracing envelope volume, thereby reflecting the material use as
a basis for comparison with material loss by machining

Figure 3.10. The concept of effective density is illustrated using

a MIM steel component. This component weighs 27 g and if
machined would require an initial volume of material more than
15 cm3 (the volume described by the outlined box that represents
an outer embracing envelope). Accordingly, the effective density is
1.8 g/cm3, meaning that MIM avoided 75% mass loss in machin- Figure 3.12. Distribution of effective density vs. maximum
ing. Another way of stating the concept is that the initial 15 cm3 dimension for production MIM steel components, showing
volume would have a mass of 118 g, so only 22% of the initial clustering toward lower effective densities especially as the
mass is in the final component component maximum size increases

35
 Table of Contents
Chapter 3: Design Principles
Figure 3.13. Stainless steel eating utensils. The soup spoon was
identified as the best candidate for MIM production based on a low
effective density
Figure 3.14. Stainless steel figurine with a high polish on the
surface, showing the potential for high-luster finishes from MIM
technology (photograph courtesy Taiwan Powder Technologies)
36
If a matte surface finish is acceptable, then in MIM polishing
is not required after sintering. This provides a significant cost
benefit over other production technologies. Hard materials
prove difficult and expensive to polish so applications with
the following attributes provide an incentive to use MIM:
components that require materials that are difficult to
machine
designs that hinder coolant access during machining
average surface finishes smoother than 5 m, but not
smoother than 0.2 m
designs where considerable mass is removed in
machining
Lustrous surface finishes are possible by polishing. The
time required to achieve a high polish is still shorter for MIM,
because the amount of material that needs to be removed is
reduced. Thus, as is evident in Figure 3.14, high-polish con-
ditions without surface porosity are possible with a secondary
operation after sintering. The surface finish advantage with
MIM is twofold. First, the time is shortened by the small
amount of material removed after sintering. But more signif-
icant, especially in areas of great aesthetic concern, jewelry,
watches, computer logos, designer eyeglasses, and designer
luggage, is the fact that no subsurface pores are exposed in
MIM. This is not the case in casting, where larger residual
pores can lie just below the surface and are opened in polish-
ing, of course making the component useless.
Size Range
Generally, MIM is applied to smaller components, with
over half of all production designs below 25 mm or 1 in. in
maximum dimension. Based on the attributes of components
that made it to production, MIM is most competitive for com-
ponents shorter than about 125 mm (5 in.). A few ceramic
components up to 400 mm are in production, but these are not
common. For example, Figure 3.15 is a picture of a medical
hand tool that is thin and 250 mm (10 in.) long. The length
distribution typical to MIM is plotted in Figure 3.16. This is
a summary for 220 production parts. It shows most are cen-
tered near 25 mm (1 in.) for the maximum size, but the range
is from 2 mm to 250 mm (0.08 to 10 in.).
The plots of typical size and complexity are based on the
number of designs. If the statistics were weighted by produc-
tion levels (number of parts produced vs. number of designs),
then the very large production of tiny MIM parts, such as
orthodontic brackets, would shift the distribution to even
smaller sizes.
Most of the MIM components tend to be slender with thin
walls. Thinner walls reduce mass and ensure rapid cooling
and heating during the process, so about half of all production
components are 3 mm (0.125 in.) or less in wall thickness.
This is illustrated in Figure 3.17 where the cumulative distri-
bution in wall thickness is taken from measurements on hun-
dreds of production parts. The plot shows two distributions,
 Table of Contents
one is for the size of the thickest wall and one is for the
thinnest wall. The overall median wall thickness is in the 1 to
3 mm range (0.04 to 0.12 in.).
Further, MIM works best when the wall thickness is uni-
form to ensure rapid debinding. The scissor blades from zir-
Figure 3.15. A long, thin MIM medical hand tool that represents a
larger maximum size produced in the technology; in this case the
components are 250 mm or about 10 in. long (components courtesy
of Injectamax)
Figure 3.16. Cumulative distribution of MIM parts in terms of max-
imum dimension. This plot has a median size near 25 mm (1 in.)
Figure 3.17. Wall-thickness distributions in MIM, showing the
cumulative distributions in the maximum and minimum walls,
reflecting a typical value near 3 mm (0.125 in.)
Chapter 3: Design Principles
Figure 3.18. These scissors represent an example of metal powder
injection molding used to form a thin-walled shape that has a
nearly constant wall thickness. The constant wall thickness avoids
distortion stresses and a thin wall shortens heat transfer to make
for shorter cycle times in molding and debinding
conia shown in Figure 3.18 are an example of a case where
the length is much greater than the thickness, and the thickness
is uniform; it is thickness that determines mold cooling time
and debinding time, so thick sections are avoided if possible.
An especially difficult situation is when thick and thin sections
join or are close together. The difference in cooling and heat-
ing rates generates stresses that easily damage the component.
During debinding the thinner section gives up binder first and
is weak, while the thicker section induces a stress that gives
cracking. Thus, the typical ratio of maximum to minimum
wall thickness is from 2 to 3 (the industry average is 2.3).
Other ranges are possible, but require special fixtures during
processing to help avoid distortion.
Mass Range
Component mass is another means to sort out good candi-
dates. Figure 3.19 plots the mass distribution of several hun-
dred MIM steel and stainless steel production components.
The median mass is about 10 g, meaning that half of the de-
signs are below 10 g and half above 10 g. About 10% of the
population comprises small structures with mass less than 1
g. These devices include orthodontic dental brackets such as
pictured in Figure 3.20. At the other end of the design popu-
lation are components with mass in the few hundred gram
range, such as the 450 g (about 1 lb.) centrifuge component
shown in Figure 3.21. Only about 10% of the MIM designs
are more than 100 g. Generally, the most popular industrial
and medical designs are in the 8 to 16 g range.
Clearly, MIM technology excels at forming smaller
components where much mass removal would be required by
machining. It should be noted that some of the very low-mass
devices are produced in enormous quantities, while some of
37
 Table of Contents
Chapter 3: Design Principles

Figure 3.19. Mass distribution for MIM ferrous (steel, stainless Figure 3.22. Scatter diagram of component weight (mass) vs.
steel) components, showing a broad range from fractions of a maximum size for ferrous MIM components, giving a general
gram to nearly 1 kg, with a median mass near 10 g correlation over a considerable size range

the high-mass components are produced in relatively smaller

quantities. Thus, if this plot were redone based on number of
parts produced, vs. number of designs, then the distribution
would skew to even smaller sizes, probably around 5 g as typ-
ical on that basis.
To further define the geometric design window, Figure 3.22
scatter plots the number of features vs. the component mass
for a few hundred ferrous MIM parts. By limiting the analysis
to ferrous parts, the component mass is not affected by the
material density. This plot demonstrates an effective area for
the use of MIM. Shape complexity peaks near 10 g; larger and
smaller components tend to be simpler. Similar plots show
mass and maximum dimension and complexity and size at-
tributes. Note that, in production, the mass variation is over
six orders of magnitude.

Figure 3.20. Orthodontic brackets are one of the more popular
MIM products that have small mass, in the 0.1 g range
(components courtesy of Ortho Organizers)
Geometric Summary
Summary statistics on the design window are given in
Table 3.1. These statistics are based on the number of designs,
showing the median MIM design is just over 25 mm (1 in.) in
maximum dimension and 10 g in mass.
In this sample set some individual components are worthy
of mention. Lager components are in production, such as the

TABLE 3.1. SUMMARY CHARACTERISTICS FOR

FERROUS MIM COMPONENTS
largest component
feature dimension mass
mm g
maximum 193 1,097
minimum 1.5 0.003
mean 34 32
Figure 3.21. Centrifuge cemented carbide component with mass of median 26 8
450 g (almost 1 lb.), an example of a larger MIM part (component deviation 32 103
courtesy of Retco Tool)

38
 Table of Contents
200 mm diameter (8 in.) compressor disk shown in Figure
3.23, but these are specialized, lower-production-quantity
structures. One medical device was 193 mm (7.5 in.) in length
with 43 g mass. On the other hand, a large component at 1,097
g (2.4 lb.) has a maximum dimension of 137 mm (5.4 in.).
Even larger bodies are used in very limited quantities. One
pilot effort was devoted to bronze filters used in forming egg
cartons out of recycled paper pulp. That was targeted at 400
per year at 40 kg each. Handgun bodies with mass ranges up
to 500 g are in production. But, to be clear, most MIM designs
are smaller devices that are lower in mass.
Related to the metals, ceramic injection molding tends to
favor larger components. For example, ceramic pouring
spouts used in steel processing reach up to 17 kg (37 lb.). Also
porous ceramic filters for aluminum casting tend to be large.
However, many factors work against injection molding large
structures in general, and specifically from metals. The tooling
is more expensive as the component size increases. Further,
the processing equipment is larger and more expensive, re-
quires more floor space, runs slower, and incurs a higher de-
preciation charge. Most important, for large structures the ebb
and flow of heat requires more time during molding, debind-
ing, and sintering; thus, process cycles slow as size increases.
The net effect from these several factors is a significant in-
crease in component cost as size increases. Consequently, cost
factors make MIM economically attractive for higher produc-
tion quantities and smaller components. For example, an op-
portunity given considerable attention is the production of
turbochargers. For small automotive units the MIM approach
is 40% lower in cost vs. investment casting and machining.
But for larger, diesel truck units the cost by MIM is close
Figure 3.23. Injection molded compressor disk with outside diame-
ter of 200 mm (8 in.) (picture courtesy of Ford Motor Company)
Chapter 3: Design Principles
to that for investment casting and machining. As a result,
MIM of smaller turbocharger units from nickel alloys is well
advanced, but the larger units are restricted to high-tempera-
ture ceramics such as silicon nitride, where there is no casting
option.
Thus, to summarize the geometric design window, the
quantitative attributes of a successful MIM concept tend to be
as follows:
low effective density where over 75% of the starting
mass would be lost during machining
surface finish of 0.2 m or rougher; matte finishes or in-
tentional embossing or texturing are best, smoother sur-
faces require post-sintering polishing
the typical component in production ranges from a size
of 1 mm (0.04 in.) to 125 mm (5 in.) and generally the
maximum dimension averages near 25 mm (1 in.)
wall thickness tends to be small, with 3 mm (0.125 in.)
being typical, but production capabilities exist from 0.1
mm to 25 mm wall thickness
wall thickness variation tends to be small, with a maxi-
mum-to-minimum wall-thickness ratio in a single com-
ponent averaging 2.3
the typical mass in production is 10 g, but ranges from
0.02 g to almost 1,000 g (2.2 lb.)
FEATURES
Certain features and feature combinations are well aligned
with MIM. Early identification of good matches ensures tech-
nical and economic success. Table 3.2 lists examples of al-
lowed and desirable features, as well as features that require
special efforts. The balance of this section gives details to
these points.
Flat Face
A flat surface on the MIM design is useful to provide sup-
port during sintering. For many parts, the component weight
exceeds the low strength that exists during debinding and sin-
tering, resulting in distortion or even cracking. In measure-
ments during sintering the component is similar to artist clay
in strengthso small features tend to move and uneven sur-
faces tend to flatten.
If a flat surface is not possible on the design, then two op-
tions exist to avoid distortion. The first is to form conformal
setters to hold the component during thermal processing. Each
setter is fabricated from ceramic specifically to the component
geometry, so this is an additional manufacturing expense. An-
other option is to add intentional features for stability during
debinding and sintering, possibly with an added machining
step to remove the added feature after sintering. In some cases
splines or stiffeners not critical to the design are incorporated
into the body simply to provide strength or support. If possi-
ble, they are located where the shape is stabilized, but where
the added feature need not be removed. For example,
39
 Table of Contents
Chapter 3: Design Principles

Figure 3.24 shows the underside of a cobaltchromium indus-
trial housing with added stiffeners. Such an added feature
helps avoid warpage while being located in a noncritical spot.
If a flat face is added to provide support during processing,
TABLE 3.2. NOMINAL MIM COMPONENT DESIGN
CONCEPTS
restrictions
inside closed cavities or undercuts on internal bores
require two pieces
must have round corners and slight draft angles
(0.5 to 2) for ejection
corner radius larger than 0.075 mm (0.003 in.)
features (wall thickness, holes, etc.) generally must be
larger than 0.1 mm (0.004 in.)
always below 20 kg (44 lb.) mass and usually below
0.5 kg (1 lb.)
desirable features
largest dimension typically below 125 mm (5 in.)
gradual section-thickness changes
mass below 100 g (0.2 lb.)
wall thickness typically less than 10 mm (0.4 in.),
most desirable is 3 mm (0.125 in.)
flat surfaces to provide support
allowed design features
shape complexity such as blind and flat bottom holes
noncircular holes
holes at angles to each other
undercuts
tapers
ribs and protrusions
knurled and waffle patterns, identification number or
logo, and surface textures
helical threads, internal threads, and external threads if
needed
then the need for a custom-designed conformal setter is
avoided.
Closed Features
In recent years, what were previously considered to be de-
sign restrictions for MIM have led to design or process innova-
tions. Closed internal cavities are now possible. One route is to
mold over a sacrificial plastic shape. During solvent debinding
the plastic is dissolved out of the green body to leave a hollow
cavity. Figure 3.25 captures such an example, where stainless
steel feedstock is molded over a sacrificial plastic insert. The
plastic then forms the circular internal cavity inside the stainless
steel. The plastic is dissolved out during debinding to produce
the feature evident in the cutaway on the right of this figure. An
alternative is to mold the component in two pieces and then join
the parts either as green bodies or after sintering.
Progress has been made on constructing ball-socket-type
moveable joints using two different feedstock solid loadings.
The conceptual steps are outlined in Figure 3.26. At first a
feedstock with more powder and less binder, termed a higher
solids loading, is used to form the outside geometry, in this
case in two pieces. The high-solids-loading feedstock gives
less shrinkage in sintering. A second feedstock with a lower
solids loading is used to fill the internal cavity. Because of the
low solids loading, this portion shrinks more in sintering. The
combination then produces a ball inside a socket with room
for the ball to move but not pull out of the socket.
Related ideas consider how to form undercut features on
an internal bore. The simplest concept is to use a sacrificial
plastic mounted on the core pin. After molding, the sacrificial
part is dissolved out as part of the solvent debinding cycle.
Previously this was one of the excluded features in MIM,
but with diligence it now can be attained, at a cost, but no
longer is out of reach.

Figure 3.25. Example of adding internal features in a MIM

Figure 3.24. The underside of a cobaltchromium industrial hous-
ing with internal ribs for stiffness during processing (component
courtesy of Mold Masters)
component using over-molding with a sacrificial plastic preform
to create the internal cavity. The final component is shown in the
center, with the plastic preform on the left. The right image has
a partial section through the device to show how the sintered
hollow body replicates the plastic preform (picture courtesy of
Lye King Tan)

40
 Table of Contents
Figure 3.26. Schematic outline on how a ball-in-a-socket joint is
fabricated using MIM: (a) the half cavity is molded from a high-
solids-loading feedstock, (b) two half cavities are bonded to form
the hollow chamber, (c) the lower-solids-loading feedstock is
molded into the hollow chamber, (d) after sintering the inner
material has more sintering shrinkage and can swivel in the
chamber
Chapter 3: Design Principles
Figure 3.27. Microminiature pillar array molded from 316L
stainless steel; after sintering the posts show some rounding
(image courtesy of Ngiap Hiang Loh)
Figure 3.28. Micro features formed by MIM using lithography
techniques, showing a mixer and reactor (images courtesy of
Taisei Kogyo)
Small Details
Sharp corners are usually added by machining or grinding
after sintering. So scissors, scalpels, and knife blades require
post-sintering grinding. However, very tiny features are pos-
sible; for example, Figure 3.27 is a scanning electron micro-
graph of microminiature pillars formed by MIM. Note the
corners are rounded, a natural consequence of sintering. The
equipment and tooling required to form these small features
come from semiconductor fabrication, namely, lithography
and laser ablation to form the tool cavities.
Surface embossing is a related small-feature idea, where the
use of finely detailed tools and small powders leads to mi-
crochannels. A few examples are shown in the scanning elec-
tron micrographs of Figure 3.28. Microarrays are potentially a
very large and significant area for MIM, as small genetic-se-
quence tests and blood-chemistry tests become widely applied.
Holes, Slots, Undercuts
Metal powder injection molding enables a wide variety of
hole, slot, and undercut geometries. Undercuts on the outside
41
 Table of Contents
Chapter 3: Design Principles

TABLE 3.3. TEN DESIGN FACTORS THAT JUSTIFY MIM

1. mass/volumeMIM is excellent at reducing fabrication
costs for components with a low effective density, where
much material loss would occur in machining or grinding,
and where possible gains in productivity are sought, an
attractive target would be a low-effective-density design
(25% if possible) as illustrated in Figure 3.30 Figure 3.29. Two undercut variants, where the one on the right is
2. quantitytooling and set-up costs are hard to justify for preferred in MIM since the tool is much easier to design
low production quantities, so MIM is best when annual
production quantities exceed 5,000, preferably at least
20,000 per year to improve tooling and engineering cost
amortizationexample mass production components
are the collection of tungsten fishing weights shown in
Figure 3.31
3. materialMIM is most attractive for designs that use ma-
terials that are hard to machine, especially multiple-phase
microstructures or those that exhibit high work-hardening
ratesFigure 3.32 shows a tungsten carbide drill tip fab-
ricated by MIM that would be quite costly to machine
4. complexityMIM is best applied to complex geometries
that can be held in a hand with some geometric and
tolerance restrictions; it works best on components that
require multiple axes for indexing if machined to give
differentiation from competitive processesan example
is given in Figure 3.33 in the form of a multiple-featured
gunsight
5. performanceif performance is important, such as in the
tungsten projectile shown in Figure 3.34, then the prop-
erties attained via full-density sintering help justify MIM,
as in heart-pacemaker housings, heart valves, and re-
placement knees, hips, and other implants
6. surface finishthe surface roughness of a MIM compo-
nent largely reflects the initial particle size; accordingly,
controlled textures are possible at almost no cost
penalty, unlike with competitive processesillustrated
by the stainless steel seashell texture in Figure 3.35; on Figure 3.30. MIM component that illustrates a low effective
the other hand, polished surfaces require secondary density, meaning that much time, expense, and material would be
operations wasted by machining, giving MIM a competitive advantage
7. tolerancesall production processes have inherent costs (component courtesy of Advanced Materials Technologies)
associated with tight tolerances, including MIM, so it is
attractive when certain features cause difficulties by al-
ternative production techniques; the electronic package
in Figure 3.36 is an example since it involves a difficult
combination of tolerances
8. assemblybenefits from MIM can be gained by the
consolidation of several parts into a single piece to save
on inventory and assembly costs or plan for selective
secondary operationsan example is illustrated in
Figure 3.37
9. blemishesplan for the production-process blemishes
that must be located in noncritical positions or removed
after fabrication; examples include gates, ejector pin
marks, or parting lines in areas that are not functional;
Figure 3.38 is one case where the molding blemishes are
hidden inside the assembly on a passive surface
10. novel compositionsseek compositions or material
combinations that are difficult via traditional processes,
for example, laminated or two-material structures, and
even mixed metalceramic wear-resistant materials
formed from mixed powders as illustrated in Figure 3.39 Figure 3.31. A collection of tungsten heavy alloy fishing weights
by a cemented carbide nozzle in a steel housing fabricated using MIM (components courtesy of Zhuzhou Luke's
Metal Powder Products)

42
 Table of Contents
Chapter 3: Design Principles

of a component are typical and usually not a difficulty if they

are aligned so as to not interfere with the mold opening. Figure
3.29 illustrates two examples of similar shapes, but with dif-
ferent undercut orientations. If possible, the undercut on the
right would be desirable since it is easier to form. The other
design would require tool motions that add expense. Internal
undercuts require special dissolvable inserts or molding of
pieces that are joined after molding or after sintering.
As a guide, Table 3.3 introduces the points and is coordi-
nated with the accompanying photographs in Figures 3.30
3.39 that illustrate these points. These several pictures show
shapes typically very expensive to form using machining, yet
feasible in MIM. For example, blind holes with flat bottoms
are possible in MIM, yet they are difficult to add by machin-
ing. Generally the blind hole depth should be no more than
twice its diameter. Holes also can be perpendicular and this is
Figure 3.34. Example of high performance is evident in this
antiaircraft tungsten projectile (component courtesy of Patco)

Figure 3.35. Novel use of tooling texture to generate an aesthetic

seashell surface appearance in MIM (component courtesy of
Figure 3.32. Industrial drill tip fabricated from tungsten carbide
Kawasaki Steel)
using MIM, reflecting a material that would be very difficult to
machine

Figure 3.33. Rifle sight fabricated from steel powder. One version
on its side is prior to black oxidation treatment. The complexity of
this MIM part would necessitate multiple machining operations
and the ability to avoid those steps makes MIM cost competitive
(component courtesy of Remington Arms)
Figure 3.36. Example of a high-tolerance, high-reliability Kovar
electronic package with glass-metal sealed lead wires (component
courtesy of Hunan Injection High Technology)

43
 Table of Contents
Chapter 3: Design Principles
especially aided by using D shapes to allow for more seating
area and stronger tooling at the intersection.
A surprise in design comes from added features that nor-
mally are expensive if made by machining. In MIM, the cost
increases with component mass due to longer cycle times and
more powder consumption. Thus, coring a design and adding
extra holes is desirable in MIM but not in machining. Already
the point has been made that nearly constant wall thickness is
Figure 3.37. Component that represents a previously assembled
geometry now molded as a single part, reducing inventory and
manufacturing steps
Figure 3.38. Example of locating a molding blemish in a non-
critical spot, in this case the trimmer blade has two gates on the
stubs located inside the cover (courtesy of Panasonic)
44
desirable, so design changes can be made to better match the
MIM process. As an example, component redesign options
focused on a constant wall thickness are illustrated for some
typical situations in Figure 3.40. Sketches are given for meet-
ing the design objective while improving the wall uniformity.
The best option in each case is a design with constant wall
thickness and simplified tool design. The key point is that
holes, cores, and various efforts to reduce wall thickness are
not detrimental or expensive in MIM.
Figure 3.39. Novel material combination of an inner carbide
nozzle (right) coupled to a stainless steel housing. The assembly is
on the left and a cross section (center) shows the combination of
pieces (component courtesy of Retco Tool)
Figure 3.40. Possible redesigns for better matches to the MIM
process by forming a uniform wall. Shown for several cases are the
initial design generated with minimized machining in mind, a
redesign that moves to uses of cores and rounded corners, and a
better redesign for MIM that keeps wall thickness constant
 Table of Contents
Chapter 3: Design Principles

Imprints, Protrusions, and Ribs

While a hole, slot, or undercut represents missing material,
the antithesis is also possible via MIM. It is common to in-
clude raised numbers or identity logos, protrusions, ribs, and
other surface features such as waffle patterns or serrations on
the molded shape. A few examples are illustrated in Figures
3.413.44. The first shows the addition of a raised logo, the
second shows serrations, the third shows a waffle pattern, and
the fourth shows the addition of many surface dimples.

Figure 3.41. An example of a raised logo on a MIM part

(component courtesy Epson-Atmix)

Figure 3.44. Bumps added to an injection molded dental-implant

anchor

Figure 3.42. Serrations molded into a MIM medical component

Figure 3.43. Textured waffle pattern on a MIM component Figure 3.45. Protrusion options possible in MIM

45
 Table of Contents
Chapter 3: Design Principles

Protrusions are useful in forming assembly attachment de-
tailsfor staking, cold heading, threading, or alignment. To
sustain functionality, while easing molding, Figure 3.45 illus-
trates three different protrusion designsa stabilized design
with support ribs, a tapered design with constant wall thick-
ness, and a dead-end design. The component shown in various
views in Figure 3.46 shows several protrusions that take ad-
vantage of MIM to create functional details. Reinforcing ribs
are effective in providing alignment and support during de-
binding and sintering, so some are added to help preserve
geometry if there is no interference with function.
ing tends to be most cost effective for applications where high
performance is required, especially in medical or dental de-
vices, defense and aerospace systems, sporting goods, indus-
trial components, telecommunication devices, hand tools,
instruments and sensors, cutting tools, automotive engines,
electronic packaging, or marine equipment. This implies the
material is valued for delivering the required performance.

Threads
Both internal and external threads are possible in MIM. It
is most fruitful to add partial threads that do not go all the way
around the component. A flat surface on the parting line is
characteristic of a simple tool motion on a thread. The concept
is illustrated in Figure 3.47. Actual threaded features are
shown on MIM components in Figure 3.48 (external threads
with flats on the parting line) and Figure 3.49 (internal threads
added via motorized unscrewing motions).
The expense for including threads comes from the required
motorized unscrewing motion that must be added to the tool-
ing. It is much easier to form external threads where a flat sur-
face is allowed on the parting line.
From this basis it is possible to sort out good candidates
Figure 3.48. Stainless steel surgical device with a flat on the
for metal powder injection molding based on a series of tests. threads corresponding to the parting line (component courtesy of
Based on these several tests it is possible to determine if a MIM Ecrisa)
component design is suitable for powder injection molding.
Cost is the next consideration. Metal powder injection mold-

Figure 3.46. Example MIM components with spindly characteris-

tics are most successful since the protrusions by other metal-
forming technologies prove taxing (photograph courtesy of Metal
Powder Industries Federation)

Figure 3.47. Threads are easily included in a MIM design if a flat

portion is allowed at the parting line. Obviously these would not be
leak-tight Figure 3.49. Internal threads are evident in this molded body

46
 Table of Contents
Chapter 3: Design Principles

MATERIALS
The component material must be available in the small par-
ticle sizes needed for sintering densification. Commercial
powder production tailored to MIM is a relatively new activ-
ity. Even so, a wide variety of materials are available in the
small particle sizes needed by the process. Globally, hundreds
of firms fabricate metal powders, but usually only a few com-
pete for any given composition, particle size, and chemistry.
For example, nearly 50 firms make titanium powder, but
spherical alloy powders in the size and purity desired for MIM
is only available from three firms.
In MIM, the bulk of the sintered products fall into a few
categories:
ferrous alloyssteels, stainless steels, tool steels, iron
nickel magnetic alloys, and specialty ferrous alloys such
as Invar and Kovar
common oxide ceramicsalumina, silica, and zirconia,
and mixtures of these
tungsten alloysboth tungsten heavy alloys and tung-
stencopper
cermets (such as Fe-TiC), cemented carbides (WC-Co),
and related hard materials
Besides these popular materials, there is a rapidly emerging
activity in special materials, which include aluminum, pre-
cious metals, titanium alloys, nickel, nickel-base superalloys,
molybdenum, molybdenumcopper, and nonoxide ceramics
such as silicon carbide and silicon nitride.
Modified chemistries have emerged to take advantage of
the technology. For example, in stainless steels, the 316L com-
position (Fe-19Cr-9Ni-2Mo, in weight percent) is used fre-
quently, because of its combined strength and corrosion
resistance. Compositions with more chromium allow for more
corrosion resistance and easier sintering, especially composi-
tions that have high levels of molybdenum (up to 6%),
chromium (up to 22%), or silicon (up to 3%). Although more
expensive, they provide properties superior to that attainable
via casting, forging, or machining.
Powder availability is a consideration. However, with
many vendors of small powders almost any material can be
custom fabricated for MIM. For complex chemistries, there
are three possible starting forms:
mixed powder compositionsfor example, iron and
nickel particles are mixed and during sintering they form
an alloy
fully precompounded or prealloyed powdersfor
example, each particle is the same composition with the
desired ratio of ingredients
hybrid systems where a partially compounded master
alloy is mixed with an elemental powderfor example,
a chromiumnickel alloy powder is mixed with an iron
powder to form stainless steel during sintering
Often the precompounded powder is easier to process, but
it is more costly.
There are many available compositions; however, there are
also some limitations. For metals, there is difficulty with
strong oxide formers, so metals such as magnesium are
avoided. Likewise, toxic metals such as lead are avoided. An-
other limitation comes with low-melting-temperature materi-
als, since they are easily fabricated by routes such as blow
molding, thixomolding, or die casting. Generally, a metal that
melts at temperatures over 1,000C (1,832F) is most suc-
cessful, since options like die casting are not available at these
temperatures.
PROPERTIES
More than any other attribute, MIM product testing em-
phasizes mechanical properties. For metals, the tensile prop-
erties of yield strength, ultimate tensile strength, and
elongation are primary concerns. The transverse rupture
strength (sometimes called the rupture modulus) is usually the
focus for ceramics and cemented carbides. Some work has
been performed on creep, fatigue, fracture toughness, and in
a few cases high-temperature tensile properties.
Most MIM products are sintered to near 100% density, so
the load-bearing cross section in the material gives strength
about the same as seen with other fabrication routes. Thus, for
MIM the densities are close to that obtained in cast products,
in the 96% to 100% range.
The tensile properties attained in MIM converge to the
same range customarily found in engineering handbooks. For
example, a direct comparison of MIM nickel-base superalloy
(IN718) tensile properties is given vs. cast and forged material
in Table 3.4. All of the test samples were given the same heat
treatment to develop a high strength. The two entries for MIM

TABLE 3.4. COMPARISON OF TYPICAL TENSILE PROPERTIES OF IN718 SUPERALLOY PRODUCTS

(MIM = metal powder injection molded, HIP = hot isostatic pressed)
process yield strength tensile strength elongation hardness
MPa (ksi) MPa (ksi) % HRC
MIM 1,048 (152) 1,372 (199) 28 41
MIM + HIP 1,054 (153) 1,379 (200) 29 42
cast 917 (133) 1,089 (158) 11 35
forged 1,158 (168) 1,379 (200) 20 36

47
 Table of Contents
Chapter 3: Design Principles
material are without and then with a post-sintering hot isosta-
tic pressing (HIP) cycle. This extra step closes any residual
pores by the simultaneous application of heat and pressure,
but in this case gives only minor property gains. Note how the
MIM product exceeds the cast material and rivals the wrought
material. Since each property has some testing variation, it is
necessary to design at the lower property levels. For MIM
products the variation in properties is small so the design level
is relatively high because of the reduced property scatter.
Another example comes from a common surgical stainless
steel, 17-4 PH (AISI 630). It is widely used in medical and
aerospace applications. It nominally consists of 17% Cr, 4%
Ni, and 4% Cu. When sintered and heat treated to the H1025
condition (aged at 552C or 1,025F), the MIM product typi-
cally has a yield strength of 980 MPa (142 ksi). Detailed tests
between different vendors give some variation from a low of
965 MPa (140 ksi) to a high of 1,040 MPa (151 ksi). As cap-
tured visually in Figure 3.50, when compared to a cast prod-
uct, this range generally is positioned higher than the
comparable range seen in wrought products (hot deformed
Figure 3.50. The yield strength ranges for heat treated 17-4 PH
(AISI 630) stainless steel fabricated by three routespowder
injection molding, casting, and machined wrought material. These
plots show the high and low of the range of reported yield strength
for the three process options
TABLE 3.5. COMPARATIVE ROOM-TEMPERATURE TENSILE PROPERTIES FOR 17-4 PH STAINLESS STEELS
(AISI 630) IN THE H1025 CONDITION
(range of several reports for each)
property MIM product
yield strength, MPa 965 to 1,040
(ksi) (140 to 151)
ultimate tensile strength, MPa 1,030 to 1,190
(ksi) (149 to 173)
hardness, HRC 36 to 38
elongation to fracture, % 8 to 12
48
and machined). Likewise, the elongation to fracture is com-
parable between MIM, cast, and wrought.
To give more details, Table 3.5 compiles the alloy property
ranges for the three different processing conditions. Because
sintering is an annealing process, MIM 17-4 PH stainless steel
in the as-sintered condition has an annealed strength. Avoiding
the heat-treatment step lowers costs, and for many applica-
tions this still provides satisfactory properties. However, op-
timal properties require post-sintering heat treatment. As with
many alloys, post-sintering heat treatments allow manipula-
tion of the strength and ductility.
In a few situations the MIM products fall short of the hand-
book properties, such as for steels that normally obtain strength
via deformation, as by forging. Consider that the sintering step
produces an annealed microstructure; without post-sintering
heat treatment the sintered materials tend to be more ductile
but with a lower strength. From an atomistic view, the forged
product has a very high dislocation density that is missing in
the sintered product, so normally the MIM product is high in
ductility while the forged product is high in strength. It is pos-
sible to compensate for the strengthening effect via changes in
the alloy composition. One good illustration would be MIM
steels with nitrogen strengthening as obtained from the sinter-
ing atmosphere. The novel addition of nitrogen delivers
strength in the 2 to 3 GPa (280 to 420 ksi) range.
To emphasize the key point, Figure 3.51 plots yield
strength of MIM products including aluminum, copper, iron,
low-alloy steels, stainless steels, magnetic alloys, titanium al-
loys, tungsten alloys, nickel-base superalloys, electronic al-
loys, and tool steels. The MIM yield strength is on the y-axis
and the handbook yield strength for the same composition and
heat treatment is on the x-axis. Since MIM is a net-shape tech-
nology, the yield strength is a most important design parame-
ter. The plot shows a strong correlation between the MIM and
handbook strengths, with a statistically significant correlation
of 0.953. Note that 40% of the alloys are stronger as MIM
products. The standard error of the estimated MIM strength
is only 104 MPa (15 ksi) if the handbook strength is known,
a reasonable scatter considering differences in impurities, ven-
dors, or microstructures.
cast product wrought product
830 to 1,000 890 to 1,105
(120 to 145) (129 to 160)
890 to 1,070 1,030 to 1,180
(129 to 155) (149 to 171)
37 to 38 38 to 39
6 to 11 4 to 12
 Table of Contents
Similar correlations exist between handbook properties and
sintered properties for brittle materials. There is no significant
difference from the handbook rupture strength when the com-
positions are injection molded and sintered to near-full density
using MIM. Figure 3.52 correlates the fracture strength (also
termed the transverse rupture strength, measured in three-
point bending) for several ceramics and cemented carbides as
a comparison of sintered properties with handbook values. In-
cluded in this plot are data from alumina, zirconia, silicon car-
bide, silicon nitride, and various cemented carbides. A perfect
correlation would correspond to the straight solid line. About
50% of the molded compositions are stronger than the hand-
book entries. The correlation between the fracture strength
and the handbook is 0.955, which is statistically very signifi-
cant. In other words, if the handbook properties are known for
a brittle material, then injection molding can match those
properties.
Figure 3.51. The handbook yield strength is compared with the
value attained via MIM for several common alloys, showing the
handbook property is a good predictor of that attained via metal
powder injection molding
Figure 3.52. Like the prior correlation for metals, this plot
compares handbook fracture strength for ceramics with the
measured value attained by CIM
Chapter 3: Design Principles
In addition to mechanical properties, other properties such
as machinability, thermal conductivity, wear resistance, and
corrosion resistance are competitive for MIM products. Where
special properties are desired, some of the most successful
MIM materials are those that prove difficult to fabricate via
alternative routes.
Chapter 5 provides much detail on the thermal, corrosion,
mechanical, magnetic, electrical, and other properties of MIM
materials. In general terms, the MIM products nominally
match handbook properties.
DESIGN SUPPORT
The MIM process excels in the production of complex
geometries. Early identification of matches between a design
and the MIM process best ensures success. To understand the
decision points a few key considerations were listed earlier
and several photographs have already emphasized these
points.
The metal and ceramic powder injection molding opera-
tions live and breathe the manufacturing details. These pro-
ducers are aware of technology variants that can benefit a
customer, so they advocate for design changes to improve
yields and lower costs. For example, a long molding cycle
adds to the component cost; roughly two-thirds of the molding
expense depends on the cycle time. A design change that
shortens the molding cycle time, such as reducing wall thick-
ness, is beneficial since it makes the component lighter and
significantly lower in cost. The injection molding vendor will
provide guidance on assembly, part combinations, and mate-
rials. Likewise, with respect to tool cost, the mold can be ma-
chined with a single cavity, two cavities, and so on. Each
cavity adds to the tool expense that must be amortized over
the production run. However, multiple cavities reduce the
variable cost associated with molding. Guidelines exist to help
optimize tool cost vs. component cost decisions.
As treated later in Chapter 6, costs in powder injection
molding reflect the component size, tooling complexity, mold-
ing time, debinding and sintering speed, and other relatively
straightforward factors. While many shapes are possible, cer-
tain features greatly impact fabrication ease. Here are a few
examples:
High-vapor-pressure species are difficult to control so it
is appropriate to avoid materials that evaporate during
sintering; for this reason brass is avoided (since the zinc
evaporates during sintering).
Certain chemistries are used frequently, resulting in
lower powder prices, so consultation on materials selec-
tion can result in better properties at a lower price.
Relying on standard equipment enables a variety of ven-
dors to compete on a project, reducing costs.
Gradual changes in section thickness along the flow path
ensure uniform final dimensions and are more desirable
vs. abrupt changes in section thickness.
49
 Table of Contents
Chapter 3: Design Principles
Holes and cores can be used to reduce mass and cost, and
at the same time help create uniform section thicknesses,
such as illustrated in Figure 3.53.
Feedstock must be able to access all of the component
details; for example, intersecting circular holes are very
troublesome, so making one of the holes D shaped
provides a flat surface for proper sealing as sketched in
Figure 3.54.
Mold-filling simulations are used to ensure all features
are properly filled.
Feedstock must properly cool at the end of the molding
cycle; thick sections have poor heat transfer and prolong
the molding cycle, thereby adding cost; coring is one
means to help improve the cooling rate.
At the end of the molding cycle the component must
eject from the tooling, so undercuts, threads, and textures
can be added, but in certain locations they increase cost
considerably.
Mold life depends on many factors, including the powder
chemistry, particle shape, binder, solids loading, and
molding conditions; in some cases a more expensive
Figure 3.53. This comparative sketch suggests how holes and
cores can be used to reduce mass and component cost, and at the
same time help create uniform section thickness
Figure 3.54. Intersecting holes are more successful in MIM when
the feedstock does not penetrate into the intersection. Shown on the
right is a comparison of circular and D shaped holes, showing
how the flat surface gives proper sealing
50
powder might dramatically reduce tool wear, improve
product quality, and generally prove a better choice.
Molding blemishes can be located in noncritical loca-
tions to eliminate any expensive post-sintering finishing
steps.
For improved dimensional control, the addition of a flat
surface to provide support during thermal processing (re-
moved after sintering if necessary) helps reduce distortion.
Tight tolerances greatly add to the cost and reduce
process yield, so geometric tolerances need to realisti-
cally balance function against cost.
Some features can be included in the part at almost no
extra cost (e.g., part identification numbers); consultation
with an injection molding engineer helps attain the in-
tended function with the least cost.
Secondary operations can be a lower-cost means to meet
specifications vs. suffering a low process yield with a
high number of tight tolerances.
Quality and ability to inspect critical features can be de-
signed into both the part and process.
Typically a few features dominate cost, so early discus-
sions with a MIM vendor at the concept stage helps avoid
difficult constructs.
RESOURCES
M. F. Ashby, Materials Selection in Mechanical Design, sec-
ond edition, 1999, Butterworth-Heinemann, Oxford, UK.
T. Baba, T. Honda and H. Miura, Effects of Homogeneous
and Heterogeneous Microstructures on the Fatigue Properties
of 4600 Steels Produced by MIM Process, Journal of the
Japan Society of Powder and Powder Metallurgy, 1997, vol.
44, pp. 443447.
S.R. Collins, Corrosion Resistance of MIM 316L, Advances
in Powder Metallurgy and Particulate Materials2002,
compiled by V. Arnhold, C.-L. Chu, W.F. Jandeska, Jr., and
H.I. Sanderow, Metal Powder Industries Federation, Prince-
ton, NJ, 2002, part 10, pp. 240254.
G.E. Dieter, Engineering Design, 2000, McGraw-Hill,
Boston, MA.
R.M. German and A. Bose, Injection Molding of Metals and Ce-
ramics, 1997, Metal Powder Industries Federation, Princeton, NJ.
H.O. Gulsoy, S. Ozbek and T. Baykara, Microstructural and
Mechanical Properties of Injection Moulded Gas and Water
Atomized 17-4 PH Stainless Steel Powder, Powder Metal-
lurgy, 2007, vol. 50, pp. 120126.
D. Shetty, Design for Product Success, 2002, Society of Man-
ufacturing Engineers, Dearborn, MI.
P. Suri, B. P. Smarslok and R.M. German, Impact Properties
of Sintered and Wrought 17-4 PH Stainless Steel, Powder
Metallurgy, 2006, vol. 49, pp. 4047.
 Table of Contents

Chapter 4
MANUFACTURING CONSIDERATIONS

Outline Mold Attributes and Limitations

Cavity Design
Parting Line
Sprue, Runner, and Gate
Vent
Temperature Control Passages
Ejector
Slides and Cores
Tool Motions
Tool-Cavity Design
Novel Solutions
Blemishes from Tooling
Geometric-Feature Combinations
Dimensional Tolerances
Tolerance Approaches
Process Yields
Computer Simulation for Defect Avoidance
Quality Plans
Inspection
Quality Tests
Quantitative Techniques
Nondestructive Techniques
Resources

MOLD ATTRIBUTES AND LIMITATIONS
The forming mold for MIM has a large influence on the
product dimensional capability during production. Prior to
molding, the feedstock lacks size and shape specificity. Once
the component is ejected from the tool there remains little ca-
pability to adjust dimensions except with extra cost. So the
mold cavity inherently sets a limit on the dimensional control
of the sintered product. An open mold is shown in Figure 4.1;
the tooling for MIM looks similar to that used in plastic in-
jection molding, although molds are oversized to account for
the large sintering shrinkage. In most instances, the tool set
consists of one to four cavities serviced by tool pieces that
move in coordination with the opening, molding, and closing
actions, such as ejector pins.
Tool design and process design go together and vary be-
tween MIM vendors. This variation is due to differences in
equipment, binder, and powder. Mold design and fabrication
are treated in detail in plastics molding handbooks. In metal
powder injection molding there are additional concerns be-
yond plastic moldingnamely, heat transfer and feedstock
density. The metal particles make the binder much more ther-
mally conductive when compared with plastics. Likewise, the
metal particles increase the feedstock density significantly
over typical plastics. Otherwise, the metal powder injection
molding route uses similar principles as found in plastic mold-
ing. Whole books are available on the latter subject. A few
references are given at the end of this chapter.
Cavity Design
The mold cavity is the heart of the tooling. It is dilated to

51
 Table of Contents
Chapter 4: Manufacturing Considerations
Figure 4.1. Open mold installed in an injection molding machine
with cooling inlet and outlet lines attached to both faces and one
slide actuated by the angled projection
Figure 4.2. As-molded computer heat spreader with sprue, runner,
and gate attached to show the flow passages associated with
formation of the component
accommodate the component shrinkage during sintering.
Around the cavity are the tool pieces needed for opening and
closing the mold, ejecting the component from the mold,
aligning the mold sections, moving inserts, cooling the com-
ponent, and locating the sprue, runner, and gate.
A primary concern in designing a mold is the component
shrinkage. The nominal change in dimensions is determined
by the sintering shrinkage Y, which is calculated from the
feedstock solids loading and sintered fractional density /T,
[ ]
Y = 1
/T
1/3
(4.1)
where is the final sintered density (near 98%) and T is the
theoretical density (100% by definition) for the material (T
is larger than ). The fractional shrinkage Y is the change in a
dimension divided by the original dimension, and this is often
52
expressed as a percentage, near 15%. Since the target is the
final component size, each dimension of the mold cavity is
oversized to accommodate this shrinkage. If the desired final
dimension is LF, then the initial tool dimension is given in
terms of the tool-cavity expansion factor Z,
1 .
Z = (4.2)
1Y
As a consequence, the original tool dimension Lo is dilated
from the specified final component dimension LF by the ex-
pansion factor,
LO = LFZ. (4.3)
As an example, if the sintering shrinkage Y is 0.15 (15%), then
the tool expansion factor Z is given by Equation 4.2 as 1.1764
(determined as 1/(1-0.15)). Thus, for example, to obtain a 12
mm final dimension requires a tool dimension of 14.11 mm
(12 times 1.1764). Shrinkage during sintering is from an initial
size of 14.11 mm to a final size of 12 mm or a linear dimen-
sional change of 2.11 mm, so the measured shrinkage is 2.11
mm divided by the starting size of 14.11 mm, or 0.15 (15%).
Although tool dimensions are expanded to allow for sin-
tering shrinkage, generally the angles are preserved. In prac-
tice most molds are first created with outer dimensions set to
the lower end of the component tolerance band and inner di-
mensions set to the high end of the tolerance band. This allows
for additional machining to adjust mold size to converge to
the final size after test pieces are sintered and measured to de-
termine final corrections.
For a typical mold, the feedstock flow path is from the
molding-machine nozzle into the sprue, along the runner or
branching runners, and through the gate into each mold cavity.
Figure 4.2 shows a molded component with the sprue and run-
ner attached to illustrate the flow path through the gate into
the component. Normally the flow path is surrounded by var-
ious clamping plates, alignment and locating pins, and ejector
components. Many of these mold components are available
Figure 4.3. Parting line on a molded component
 Table of Contents
as standard pre-machined pieces, so only the mold cavity
needs to be custom machined.
Within a mold it is possible to have multiple cavities. The
number of cavities in use across the MIM industry ranges up
to 40, and in one rare case is even higher. A single-cavity tool
is satisfactory for low production quantities, say, in the range
of 100,000 shots per year. Optimization calculations show that
for components ranging from 1 to 100 g, the lowest project
cost (tooling and piece cost) results in the following general-
ization:
for production below 100,000 parts per year1 cavity
for production quantities near 300,000 parts per year
2 cavities
at 1 million parts per year4 cavities
at 3 million parts per year8 cavities
at 10 million parts per year12 cavities
For larger parts, the number of cavities is limited by the
molding machines clamping force or shot size, which are not
critical issues for a small component.
The savings from multiple cavities comes about due to in-
creased molding productivity achieved by reducing the re-
quired number of molding machines. As the number of
cavities increases, the cost of manufacturing the mold in-
creases, but the subsequent molding cost per component de-
creases. However, the optimal quantity of cavities may not be
determined solely by costs. Other limitations include clamp-
ing force, shot volume, maintenance costs, and added compo-
nent dimensional variability. For example, with more than a
few cavities, a 100 g component might exceed the clamping
force. Another factor is nonuniform filling. For very tightly
toleranced components, statistical tests show that the cavity-
to-cavity size effect is measurable. Further, the cavity location
with respect to gravity and cooling channels contributes to
sintered dimensional variation in multiple cavity tooling.
Figure 4.4. This plate-shaped microelectronic package exhibits
flashing along the parting line. Note the crack in the corner above
the flash
Chapter 4: Manufacturing Considerations
These factors result in dimensional variation that decreases
process yield and drive MIM toward single cavities for very
precise products.
Parting Line
The location of the parting line is an important decision in
MIM tool design. The parting line is the trace left on the
molded component surface where the mold sections meet, as
illustrated in Figure 4.3. Cosmetic and functional require-
ments may force parting line locations in less than optimal po-
sitions. Part of the problem is that the tool set slightly deflects
during mold filling due to elastic strain, in spite of close align-
ment and high clamping forces. The deflection varies directly
with the injection pressure and part cross-sectional size, al-
lowing slight feedstock intrusion between the moveable tool-
ing plates.
Although the parting-line mark is a common blemish, an
extreme situation occurs with feedstock intrusion along the
parting line to give flashing. Figure 4.4 is an extreme example
of flashing. Note the corner above the flash also has an un-
healed weld line. Removal of flash after molding is costly and
is avoided when possible. On the other hand, in regions where
labor is inexpensive it is common to have a technician at the
molding machine perform flash removal by hand. An alterna-
tive is to place the parting line in a region where the blemish
is not of concern. A few MIM firms have adopted novel
grooved tool designs that remove the parting line.
Sprue, Runner, and Gate
The sprue connects the molding-machine nozzle to the run-
ner network in the mold that delivers molten feedstock into
the cavity. In many MIM operations sprues are tapered with a
typical diameter near 6 mm and 5 degree taper. When the
mold opens for ejection, the sprue is pulled from the sprue
bushing and usually is recycled. For faster production and the
minimization of regrind, the sprue and runner are kept hot to
avoid reheating and recycling this portion of the feedstock.
This option is called a hot runner system.
From the molding nozzle the sprue leads into the runner,
which then directs the melt into the mold cavity through the
gate. Small runners cause slow filling, so a large runner is de-
sirable, with a typical diameter from 3 to 6 mm. Of course, a
large runner means relatively less of the shot goes into the part
cavity. A circular runner design is most desirable since this re-
duces friction and heat loss during filling, but other designs
are less expensive to construct. Hot runners are available to
keep the flow path heated in the tooling. They use a valve in
the gate to seal the runner between shots and are usually re-
stricted to high-volume production where the additional cost
is justified by productivity gains.
When a mold contains multiple cavities, balanced runner
designs with an equal flow length to each cavity are required
to ensure simultaneous filling of each cavity. Two factors are
53
 Table of Contents
Chapter 4: Manufacturing Considerations
requiredfirst, the flow length from the sprue to gate must
be the same for each cavity and, second, the cooling channels
must be balanced to ensure similar cooling temperatures along
each flow path. A valve can be incorporated into the runner
system to isolate any cavities that might become defective
during production.
At the end of the runner is the gate leading into the mold
cavity. The gate leaves a surface blemish on the part, such as
shown in Figure 4.5. It is a small opening designed to freeze
before the part, runner, or sprue freeze. Ideally, the solidified
gate sustains pressure in the cavity to counteract the shrinkage
associated with feedstock cooling and binder contraction. Too
low a pressure when the gate freezes results in voids or sink
marks, while excessive pressure causes sticking in the mold.
It is desirable to have the gate located on the thick portion of
the component. This reduces heat loss and premature freezing
or short shots, and lowers the pressure required for mold
filling.
Gate size is determined by the filling speed and the com-
ponent thickness. It is desirable to have the gate freeze before
the compact freezes. This requires a gate size smaller than the
component wall thickness. Yet, if the gate is too small, then
the resulting higher feedstock, flow velocity during cavity fill-
ing results in jetting, high tool wear, or powderbinder sepa-
ration. Figure 4.6 is picture of the powderbinder separation
around the gate on the bottom of a MIM electronic package,
most likely due to the gate being too small. On the other hand,
if the gate is too large, then molding is slowed by the need to
use the screw to control pressure on the component during
cooling instead of preparing the next shot. In such a situation
a fan gate or even multiple small gates might be required. If
the gate is on a thin section, then filling is hindered because
of flow resistance and feedstock cooling. Likewise, the blem-
ish from the gate impacts the decision on gate location be-
cause of cosmetics.
Figure 4.5. A gate mark and parting line on a MIM component
with some powderbinder separation near the gate. There is a
slight misregistration of the two mold halves, as evident by the
offset split in the gate
54
Additionally, gate placement should consider minimizing
weld lines. Weld lines form when the feedstock splits and then
rejoins after flow around posts during cavity filling. If the
molding conditions are such that the feedstock is warm, then
the weld line heals. Alternatively, if the feedstock is too cool,
then the weld line does not heal and a defect forms. Figure
4.7 offers two pictures of weld-line defects; one is a scanning
electron micrograph showing such a subtle molding defect in
a green body and the second is an optical micrograph showing
how the defect is enlarged after sintering. A defective weld
line is difficult to detect after molding, but invariably it is ev-
ident after sintering, since the material pulls away from the
crack. Computer simulations help avoid these difficulties by
suggesting either different molding parameters or different
gate locations.
Vent
When a molding cycle starts, the mold cavity is filled with
air. During molding this air is pressurized by the intruding hot
feedstock and forced to escape through a vent. The vent is lo-
cated at the end of the mold cavity opposite the gate. It should
be the last portion of the cavity to be filled. Vents are very thin
reliefs typically 0.015 mm (0.0006 in.) deep, and up to 12 mm
(0.5 in.) wide for large MIM parts. Ideally, the vent leaves no
observable trace on the part surface.
Venting is also necessary to allow the part to be ejected
from the tooling. For example, a shape with a closed end can-
not be ejected without creating a vacuum; otherwise, an addi-
tional ejector pin must be built into the forming tool which is
pulled first to vent the mold.
An alternative is to rely on loose-fitting ejector pins to pro-
vide an opening for venting in the annular gap around the pin.
Another option is to use a stepped mold closure procedure,
where only a light clamp force is applied through the filling
stage, allowing air to vent along the parting line. At the end
of the mold stroke the tool is fully clamped as the feedstock
is packed into the mold.
Figure 4.6. Green microelectronic package showing powder
binder separation near the gate, in this case due to a small gate
and fast cavity-fill rate
 Table of Contents
Temperature-Control Passages
Tubular passages are placed in the tooling to allow the flow
of water or oil for mold-temperature control. For binders that
require low mold-temperatures, it is possible to use refriger-
ants for cooling. Alternatively, the temperature-control pas-
sages might be used to preheat the mold before molding,
which is often necessary with polyacetal and some wax-poly-
mer binders. Subsequently, temperature-control passages are
switched to cooling to remove heat from the component. Note
that the cooling time increases with the square of the compo-
nent section thickness. For usual molding conditions, cooling
times are in the range from 10 to 20 seconds. However, com-
ponents are produced with wall thicknesses of 125 mm
(a)
(b)
Figure 4.7. Weld-line defects in MIM: (a) scanning electron
micrograph of carbonyl iron system molded with a wax-polymer
binder with a small weld-line defect after molding that grows to
become a visible crack after sintering, and (b) optical micrograph
of a weld-line defect that is visible after sintering
Chapter 4: Manufacturing Considerations
(5 in.) with a cooling time in excess of 5 minutes.
Conformal cooling, where the cooling channels are buried
in the mold yet follow the shape of the part, is one option to
reduce the molding cycle time. Here the cooling passages are
not straight holes, but are built to follow the component pro-
file. It is most important to use conformal cooling to fabricate
precise components. By bringing uniform cooling to each re-
gion, the mold temperature gradients responsible for subse-
quent sintering distortion are avoided. Construction of the
conformal-cooling passages is expensive. Alternatively, the
use of high-copper alloys for the mold (such as copperberyl-
lium) helps with heat management during molding, but these
are also expensive solutions and in some cases hard to ma-
chine. Rapid-prototyping approaches to hard tooling provide
one means to incorporate curved and strategically placed cool-
ing passages.
Ejector
Once the part has cooled in the mold it must then be
ejected. The ejection force depends on the contact area be-
tween the tooling and component, tool surface finish, coeffi-
cient of friction, and thermal contraction in the cavity. Also,
adhesive polymer phases in the binder that wet and adhere to
the tool surface can cause sticking. A slight taper in the tooling
greatly aids in reducing ejection force and often just 0.5 is
sufficient. Corners in the cavity are usually rounded for easier
ejection; a radius of 0.2 mm (0.008 in.) is satisfactory, but they
can be as tight as 0.05 mm (0.002 in.).
To accomplish ejection, pins in the mold body move for-
ward with the ejector plate and push the component from the
cavity. Inserts, internal cores, or threads must be retracted to
allow free ejection. Ejector pins blemish the component, since
they concentrate the ejection force on a soft and often still
warm molded shape. Thus, larger pins are desirable to reduce
the stress. Figure 4.8 is a component with several visible ejec-
tor pin marks. The location and number of ejection pins de-
pend on the component size, binder strength, and tooling
complexity. Normally, the pins are placed to impress on non-
critical locations and constitute more than 10% of the pro-
jected compact area.
Slides and Cores
Features oriented perpendicular to the parting plane are
normally included in the mold plates, but features that are ori-
ented on the parting-line plane require slides or cores. These
tool pieces are moved in and out of the cavity on each closing
and opening. Figure 4.9 is a picture of a small tool set de-
signed as an insert into a mold base, with slides that induce
motions along the parting line on closing and opening. Slides
and cores are common for inserting undercuts, steps, and
holes. In a few rare situations, a solid piece is inserted into the
mold on each shot, and manually pulled from the part after
ejection. This obviously lowers tooling cost, but increases the
55
 Table of Contents
Chapter 4: Manufacturing Considerations
Figure 4.8. This magnetic automotive component illustrates slight
surface depressions from the ejector pins. A total of ten ejector pin
marks are evident. The word ejector is just below one and there
are six spaced on the outer flange and four smaller marks on the
raised center region (component courtesy of Advanced Materials
Technologies)
Figure 4.9. Mold insert showing the ejector pins, slides, and
examples of the molded and sintered components
defect rate and labor content in molding, so it is an option gen-
erally restricted to low production quantities.
Figures 4.10 and 4.11 are two examples of features gener-
ated by slide motions. Protruding pins in the tool halves force
motion in the parting plane as the mold closes and opens. This
ensures proper placement of the slide in the cavity on closing
and retraction to avoid interference with ejection. The fabrica-
tion of slides, cores, or unscrewing tool components (for ex-
ample, those required to form internal threads) are expensive
aspects of tool fabrication. Thus, efforts to redesign the com-
ponent to avoid these features are rewarded by lower tool costs.
TOOL MOTIONS
Molding starts with the closure of the mold cavity. As
sketched in Figure 4.12, the nozzle of the molding machine is
56
Figure 4.10. Stainless steel medical hand tool with perpendicular
features formed by slides
Figure 4.11. Mounting bracket with perpendicular raised features,
holes, and slots that require multiple tool motions. Note also the
ejector pin marks on this face
inserted into the sprue bushing. After contact, a premeasured
quantity of molten feedstock is forced into the mold cavity.
Pressure is sustained on the cooling feedstock until the gate
freezes, and after complete cooling the component is ejected.
Between the time when the gate freezes and the part is ejected
while it is still cooling, the screw is turning to plasticate and
meter the next shot of hot feedstock.
The pressure in the molding cavity undergoes a rapid in-
crease as the hot, molten feedstock enters the cavity and that
pressure is held until the gate freezes. Once the gate has frozen
the molding machine no longer controls the cavity pressure.
A pressure trace is plotted vs. time in Figure 4.13, indicating
the pressure loss from the hydraulic-system pressure as ap-
plied to the back of the screw, through the nozzle, and into the
mold cavity. It is the cavity pressure that needs to be con-
 Table of Contents
trolled, but the critical pressure is controlled back at the
hydraulic system. Hence, effort is required to ensure proper
hydraulic pressure, and corrections to the hydraulic pressure,
to achieve repeatable mold-cavity pressures. The shot-to-shot
mass variation is the easiest means to monitor the success of
proper control.
After mold filling, as the feedstock cools, a natural pressure
reduction occurs due to binder thermal contraction. At the
point of ejection the residual pressure in the cavity should be
very low. Otherwise, the component will stick in the mold
cavity. For ejection, pins move from flush positions on the
Figure 4.12. This schematic illustrates the MIM molding cycle. It
starts with the mold open. After the mold is clamped the nozzle is
moved into contact with the sprue bushing to enable cavity filling
and feedstock pressurization. Once the gate has frozen, the
injection unit retracts and the mold opens to eject the solid part.
During the last portion of the cycle, the reciprocating screw is
preparing another charge of molten feedstock for injection
Chapter 4: Manufacturing Considerations
tool walls and push the component from the cavity.
Robot coordination is an important aspect of quality con-
trol and automation. Most modern molding machines integrate
the robot control and coordination with the molding cycle.
Further, simple tests such as mass or size determination are
possible after molding to ensure proper quality control.
A die-cavity-pressure transducer is important to molding
control. This pressure-measurement device is buried in the
tooling at two-thirds of the flow length in the cavity. Once the
transducer detects pressure in the cavity, then molding ma-
chine control reverts to repeat the same cavity pressure profile,
via concomitant adjustments to the hydraulic pressure. This
significantly improves uniformity in packing pressure, result-
ing in a tighter part-mass distribution, which results in more
uniform sintered dimensions. Figure 4.14 is a plot of die-cav-
ity-pressure traces taken over 50 shots. The upper plot corre-
sponds to molding without die-pressure control, showing a
wide variation in pressure (resulting in a wide density and sin-
tered dimensional variation). On the other hand, the lower plot
shows pressure traces for 50 shots with die-pressure control,
obviously giving a more homogeneous product. Reduced
dimensional scatter for the sintered component is the key gain
from closed-loop cavity-pressure control.
TOOL-CAVITY DESIGN
The molds used in MIM are custom machined from
wrought stock. Ease of fabrication and subsequent durability
of the tooling are primary concerns dictating tool material se-
lection. Because the MIM feedstock is more abrasive than
pure polymer, wear resistance is a concern. Consequently,
after machining, the tooling is heat treated or subjected to a
surface-hardening treatment. To avoid component adherence
to the tooling, surfaces should be smooth and scratch free.
Figure 4.13. Pressure traces for the hydraulic system, nozzle, and
mold cavity. In MIM the pressure is controlled at the hydraulic
system, but losses in the system leave the cavity pressure poorly
controlled. It is the cavity pressure that ensures uniform products
57
 Table of Contents
Chapter 4: Manufacturing Considerations

This requires polishing or plating to a mirror finish. For ex-

ample, chromium plating is one means for rebuilding and
restoring a hard, smooth surface on tooling after wear. Various
other surface enhancements reduce wear and improve surface
finish, including tungsten disulfide coatings, electroplating
chromium or nickel phosphate, ion nitriding, salt bath nitrid-
ing, and even boron carbide coating. The choice depends more
on tool-vendor experience than any results from comparative
testing.
Table 4.1 identifies several mold materials and their typical
uses. Since these are ferrous alloys, the density is generally
near 8 g/cm3 (0.28 lbs./in.3) and the elastic modulus is near
200 GPa (29 Mpsi). Depending on the hardness, the materials
generally are high in strength but low in ductility, but can be
heat treated to a high-ductility but low-strength combination.
Thermal expansion is nearly the same for these tool materials
at about 10 ppm/C (5.6 ppm/F) and thermal conductivity is
also nearly the same at about 24 W/(m C). The choice of ma-
terial for mold construction depends on the anticipated number
of molding cycles and the required strength or wear resistance.
Wear concerns lead to the use of tool steels heat treated to at
Figure 4.14. Plots of cavity pressure vs. time during molding for least a Rockwell C hardness of 30.
the same component and feedstock. The upper trace is open-loop Most service failures in MIM tools come from one of two
control where only the hydraulic pressure is controlled, showing sources:
cavity-pressure traces taken over 50 shots. The lower plot
corresponds to pressure traces for 50 shots with cavity pressure
excessive wear due to particle abrasion; controlled via
control, obviously giving a more homogeneous product, as selection of a rounded particle shape, lower filling
confirmed by both sintered mass and sintered dimension tests velocities, and harder tooling
fatigue fracture; controlled by higher tool strength and
lower peak pressures
TABLE 4.1. MIM TOOLING CONSTRUCTION Both arise during extended production. Fatigue endurance
MATERIALS strength and hardness are the two key determinants of tool
life. In several metallic MIM projects, tools have survived for
tool material hardness suggested applications
HRC a few million shots. However, wear issues are greater with
hard powders. In such cases, tool refurbishment might be re-
stainless 420 50 corrosion-resistant
quired every 50,000 shots.
cavities, cores, inserts
stainless 440C 57 wear-resistant small NOVEL SOLUTIONS
inserts, cavities, cores
The construction of the tool cavity is costly. However,
steel 4130 35 general mold plates many situations require trial components to identify problems
steel P20 30 general purpose, hot that might exist in the overall product design or acceptance.
runner, large cavities Thus, trial components are fabricated realizing that the final
design will change. Several rapid-tooling and soft-tool tech-
tool steel A2 58 small inserts, high wear
nologies allow for fast formation of a mold that might only
tool steel CPM 10V 63 high toughness, high be used for a few hundred shots.
wear, high strength
Some of the rapid-tooling solutions include use of easily
tool steel D2 57 high abrasion, small tools machined low-carbon steel or aluminum molds, cast particle-
tool steel Elmax 58 high wear, high corrosion filled epoxy-tool inserts, and reliance on laser-formed
stereolithography steel inserts. Other soft materials, even
tool steel H13 50 large, intricate cavities,
spray-deposited zinc, are occasionally used during tool devel-
high toughness, low wear
opment and component prototyping, because of the forming
tool steel M2 61 core and ejector pins ease. Still, the susceptibility of these materials to scratching
tool steel O1 60 slides and wear plates and damage mandates harder materials for production.
One novel technique used in limited production runs relies

58
 Table of Contents
Chapter 4: Manufacturing Considerations

in areas where there is no functional difficulty. Parting lines

can be placed along component edges to reduce visibility as
sketched in Figure 4.16.
Early in the design process, the designer needs to consider
placement of these blemishes in areas where they will not de-
tract from the operation or aesthetics. For example, Figure
4.17 shows the gating on a watchcase located in an area where
the mechanism will hide the blemish on the final product.
Alternatively, the gate can be removed by grinding or polish-
ing after molding, but prior to debinding and sintering, since
the material is easily removed then.
Ejector pin marks are another of the typical tool blemishes.

Figure 4.15. Rapid-prototype mold insert fabricated for prototype

production. Trials have demonstrated such tools are useful for
thousands of molding cycles. However, final machining and
grinding is required to ensure smooth surfaces and good
dimensional precision

on urethane or silicone rubber tooling. A rubber mold is

formed by pouring the polymer around a master model. The
rubber is hardened (cured) and then cut to form a parting line
and to remove the master. When constrained in compression,
rubber can withstand low pressures so component production
runs up to a few hundred pieces have proven possible.
Rapid-prototyping tools are used in a similar manner. The
stereolithography approach based on selective laser scans pro-
vides a means to generate a tool cavity from polymer or metal Figure 4.16. A desirable option in mold design is to place the
parting line on a location where it is less visible and to eliminate
powder. The tools are reasonably strong and newer composi- any need to clean the parting line from the green or sintered part
tions deliver properties comparable to tool steels. One example
is shown in Figure 4.15. These molds, produced in just five
days, do not meet the rigors of production tooling; the surfaces
tend to be rough. But they are available in short times. Various
trials with rapid tooling have generally demonstrated good
success with up to 20,000 components in one case.
One important alternative to early part production is to use
a large, oversized block of MIM feedstock that is machined
into a green shape for the component. Green machining allows
a few pieces to be formed by high-speed CNC machining
since the feedstock material is easily cut. After debinding and
sintering, relevant testing is possible since the samples are
from the same powder and processing cycle intended for pro-
duction, without the expense of cutting a mold. For very high
production, there are demonstrations using MIM to produce
tooling inserts for use in high-wear applications, effectively
using MIM for production of multiple tool pieces.

BLEMISHES FROM TOOLING

A successful MIM component design must allow for non-
critical placement of the parting line, ejector pin marks, and Figure 4.17. Molded watchcase with the gate located under the
gate. Often the gate is located on a parting line, so these two mechanism; as a consequence no special effort is required to
blemishes go hand-in-hand. In many instances they are placed remove the gate blemish

59
 Table of Contents
Chapter 4: Manufacturing Considerations
The greater the area used for ejection, the shallower the ejector
indents.
GEOMETRIC-FEATURE COMBINATIONS
Metal powder injection molding excels when the shape is
possible via plastic molding, but plastics lack the engineering
properties needed for the application. In the MIM field, toler-
ances and engineering specifications are a subject of much dis-
cussion. As MIM has evolved, new process variants have
contributed to substantial precision and accuracy gains. Every
year the basic process knowledge grows, while the equipment
improves and the feedstock becomes more consistent. Conse-
quently, production capabilities improve over time. Thus, there
is no single process capability, rather a spectrum of capabilities
exist that are improved by about ten years of experience com-
bined with modern computer-controlled processing equipment.
Powder-based technologies have an excellent history of
process improvements. A good example of the progressive
learning is seen in Figure 4.18, which plots the dimensional
variability for cemented carbide (WC-Co) cutting-insert pro-
duction over a half-century. The variability is measured by the
coefficient of variation, a dimensionless means for expressing
scatter; it is the standard deviation divided by the mean. Such
scaling allows comparison of capabilities between large and
small components. In Figure 4.18, the data are for the cutting-
edge distance from the central locating hole, showing progres-
sive improvement over 50 years. Most of the improvements
occurred in the first 20 years, but even today there are gains,
with production tolerances (without secondary operations)
converging toward 10 m on a 14 mm distance (0.0004 in.
out of 0.55 in.).
Similar plots exist for other powder metallurgy technolo-
gies, although none have the history of cemented carbides. In
presssinter ferrous powder metallurgy, the dimensional ca-
Figure 4.18. Dimensional precision improves as cumulative
knowledge grows. This plot gathers coefficient of variation data
from cemented carbide insert production over 50 years. By
implication, MIM should show similar gains as technology,
materials, and equipment improve
60
pability has likewise improved over time. Gains in MIM are
occurring as new production technologies are adopted, but as
long as older equipment and older manufacturing practices re-
main in use there will be a wide distribution in capabilities
across the industry. Efforts to hold close tolerances add to the
production expense, but are offset by increased process yields.
Post-sintering machining is a means to generate MIM
products that match conventional machining tolerances. How-
ever, MIMs greatest cost advantage occurs with net-shape
production, where the part is used directly out of the sintering
furnace, without secondary dimensional corrections.
Design engineers tend to be puzzled by the wide variability
in cost estimated for MIM components. However, part of the
variation comes from the fact that some facilities require coin-
ing or machining to hold tolerances while other facilities can
hold tolerances without secondary operations. The latter ap-
proach is always more cost-effective.
DIMENSIONAL TOLERANCES
Since metal powder injection molding is a hybrid between
plastic molding and sintered powder metallurgy technologies,
the tolerance capabilities are equally a hybrid. Plastic molding
generally has a tolerance range from 0.05 to 0.5 mm
(0.002 to 0.020 in.). Sintered powder metallurgy technolo-
gies generally have a tolerance range from 0.1 to 0.3 mm
(0.004 to 0.012 in.), and sintered ceramics are much larger.
Accordingly, MIM is limited by its underpinningsplastics
and sintered materialsto a general tolerance range centered
on 0.1 mm (0.004 in.) or slightly better with special efforts.
In discussing tolerances there are two issues. The first is
the location of the mean size for the specified featureaccu-
racy. The second issue is the variation about the meanpre-
cision. The further the mean is from the target size, the smaller
the allowed variation or tolerance zone. Adjustments to the
mean size are easier to accommodate in MIM than adjust-
ments to the variability. For example, changes in molding con-
ditions, solids loading, sintering temperature, or sintering time
can be used to relocate the mean size of the product.
Current MIM component production ranges from feature
sizes below 100 m to nearly 1 m (a size ratio of 104), making
it difficult to address tolerances directly over such a wide
range. In some instances, the choice has been to create a table
of tolerance vs. size, showing capabilities as follows:
0.05 mm (0.002 in.) for features below 3 mm (0.12 in.)
0.08 mm (0.003 in.) for features between 6 and 15 mm
(0.25 to 0.6 in.)
0.25 mm (0.01 in.) for features between 30 and 60 mm
(1.2 to 2.4 in.)
An alternative is to track the tolerance as a percentage of
the feature size. The coefficient of variation is used to normal-
ize dimensional variability. Statistically the coefficient of vari-
ation CV is defined as the standard deviation divided by the
mean dimension, often given as a percentage:
 Table of Contents
CV = (4.4)
XM
where is the standard deviation from the group of measured
dimensions and XM is the mean dimension. Assuming accurate
measuring devices, then the concern is the processs capability
to hit a target dimension with low scatter. In a recent study of
microminiature molding with 316L stainless steel, the coeffi-
cient of variation on a 175 m dimension was nearly 3%. In
a separate study, the coefficient of variation increased from
0.5% for a 2.4 mm dimension to 5% on a 0.4 mm dimen-
sion. Clearly there are new challenges for small dimensions.
For the more typical size range encountered in MIM produc-
tion, near 25 mm (1 in.), the general industry capability is 75
m (0.003 in.), or a coefficient of variation 0.3%. This is
generally reflective of the technology once the production
process is tuned to center the dimensional variation on the de-
sired mean size.
In large volume production operations tool wear becomes
an issue usually after a hundred thousand shots per mold cav-
ity. To avoid the expense of mold replacement or refurbish-
ment, the desire is to have a wider tolerance band to allow for
tool wear, say, 0.5%. This ensures a high process yield of ac-
ceptable parts over a longer time. However, in production this
is not sufficient, since the desire is a high yield of acceptable
components within the target dimensional range. The mean
plus and minus one standard deviation only includes about
64% of the product. Many industrial firms quote process ca-
pabilities based on at least six standard deviations (the mean
plus and minus three standard deviations, which includes
99.74% of the product).
In production, there are several sources of variation:
feedstockmixture inhomogeneities and lot-to-lot feed-
stock variations
moldinginherent process variation shot-to-shot asso-
ciated with the molding machine and its sophistication
as evident by the integrated controls
cavitylargely from variations between tool cavities,
systematic filling differences, and cavity wear over time
furnaceassociated with furnace-to-furnace differences
in debinding and sintering
placementvariation due to location differences within
the thermal processing equipment
daynormal daily fluctuations, including operator,
humidity, handling, tool wear
vendorvendor-to-vendor differences
Audits on these factors show vendor differences are the
largest, because all of the other factors shift between vendors.
One consequence is that molds cannot be transferred between
vendors with success. The next largest factor is feedstock vari-
ation, especially if the feedstock is mixed in a batch mixer.
Although the sintering furnace is often blamed for dimen-
sional variations, in reality sintering is an amplifier of earlier
Chapter 4: Manufacturing Considerations
defects. Sintering transforms subtle molding variations into
dimensional scatter. In repeated statistical surveys, molding
accounts for 60%80% of the MIM dimensional variation.
But this variation is not evident in dimensional scatter until
after sintering.
The vendor differences are seen with components that are
produced at multiple sites. Clearly, different operations can
have different mean dimensions, so when components from
different vendors are mixed the total batch will have a larger
variance. Thus, if components are purchased from multiple
vendors, then assembly operations must anticipate a dimen-
sional variation due to differences in equipment, processes,
and even measurement accuracy.
With respect to dimensional variation, Table 4.2 summa-
rizes the situation based on both the typical coefficient of vari-
ation and the best possible in the industry (without secondary
operations). An analysis of data provided by 62 studies on dif-
ferent parts and production processes gives an average coef-
ficient of variation of 0.22% (ranging from 1.25% to
0.04% for dimensions ranging from 0.6 to 33 mm or 0.024
to 1.3 in.). Figure 4.19 plots the distribution from these 62
studies. Beyond 5 mm (0.2 in.) the measured process capabil-
ity generally stabilizes at a coefficient of variation near 0.2%,
but features smaller than 5 mm are difficult to control.
For a dimension over 30 mm, the statistical studies show
a standard deviation of 0.06 mm (about 0.2%). Also, the
ability to hold tight tolerances improves with production
experience. Indeed, in studies involving three mutually per-
pendicular dimensions (length along the molding axis, width
in the parting plane, and height perpendicular to the parting
plane) the ratio of coefficient of variation depends on the flow
direction. For example, in one study, the thickness variation
Figure 4.19. Dimensional coefficient of variation (standard
deviation divided by mean size) reported in 62 MIM studies, repre-
senting different materials, shapes, sizes, and processing condi-
tions. A coefficient of variation of 0.2% is a good reflection of
industry capabilities; for a typical 25 mm size the standard devia-
tion would be 50 m (for 1 in. the variation would be 0.002 in.)
61
 Table of Contents
Chapter 4: Manufacturing Considerations

was 0.7%, while the width variation was 0.4%, and the casting cores used for jet-engine components, the mass ranges
length variation was 0.2%. This was confirmed in a different from 10 to 1,200 g (0.02 to 2.6 lb.), and part lengths range up
study, giving a similar ratio between perpendicular dimen- to 500 mm (nearly 20 in.). These cores are produced with a
sions. Most important, mass variations between shots ac- few critical dimensions held to a coefficient of variation of
counted for 87% of the dimensional variation after sintering. 0.05% at production quantities over 50,000 per week. Un-
The thickness variation reflects pressure repeatability and the fortunately, there is only 80% process yield to these toler-
slight tooling deflection. When molding is performed with ances, so it is proper to say that the coefficient of variation of
closed loop cavity pressure control, the mass variation is much the shipped components is 0.05%, but the actual process has
reduced, resulting in smaller dimensional scatter. an inherent scatter of 0.12%. Other MIM companies report
Another factor in determining dimensional tolerances is shipped tolerances of 5 m and even 3 m on critical
surface roughness. Nominally the tolerance specification can- dimensions, but little information is provided on process
not be any tighter than tenfold the surface roughness. Most yields. Obviously, production cost increases rapidly with the
MIM operations deliver an average surface roughness near 1 tighter tolerances if yield falls.
m. In a survey of the industry, the mean is 0.6 m and the Inherently, tolerance capabilities are limited by surface fin-
standard deviation is 0.3 m, but the median (half better and ish. Accordingly, surface roughness is a concern, especially
half worse) is 0.8 m, so the distribution is asymmetrical. On as MIM has entered consumer and jewelry applications. The
this basis the surface roughness limits the tolerances to prob- sintered surface roughness ranges from 0.2 to 1.6 m or 8 to
ably 10 m at best for MIM. 64 in. This range matches with a ground finish. The limit
An option for tighter tolerances is to machine critical sur- has its origin in both the tooling surface finish and the particle
faces after sintering. Post-sintering machining is often used in size. The impact of particle size is trivial unless the tool is
MIM, ensuring precise final dimensions, but in some cases it smooth; without polished tooling, the part surface simply
adds considerably to the fabrication expense. replicates the tool surface topography, an aspect that is desir-
With newer binder systems and integrated process controls, able for textured component surfaces.
dimensional scatter has improved to a coefficient of variation When the tooling is polished, the particles still create a cob-
equal to 0.05%. For example, in the production of ceramic blestone surface as diagramed in Figure 4.20. To help envision
the as-sintered surface, a scanning electron micrograph of a
MIM stainless steel component is given in Figure 4.21. Note
TABLE 4.2. PROCESS VARIATIONS OBSERVED IN MIM the residual appearance of the particles, now fused together
COMPONENT PRODUCTION
(actual values depend on many site-specific and
by sintering. Thus, smaller (and more expensive) particles are
component-specific factors) needed to form smoother surfaces and this is only justified
when using polished tooling. For very smooth (polished)
feature best nominal or
possible typical
products, it is usually most economical to add a polishing op-
eration after sintering.
absolute dimension 0.025 mm 0.05 mm Likewise, the fabrication of sharp corners and small fea-
(0.001 in.) (0.002 in.)
tures is limited by the particle size. An illustration of this is
angle 0.05 2 given in Figure 4.22. Indeed, if the particles are large com-
density 0.2% 1% pared to the feature, then the particles simply will not fill out
the feature. As a rule of thumb, no feature should be specified
dimension 0.05% 0.3%
to a size that is not at least tenfold larger than the particle size.
flatness 0.1% 0.2% For a sharp corner fabricated from 10 m powder this would
hole diameter 0.04% 0.1% say that the corner radius should be 100 m or larger (0.1 mm
or 0.004 in.). Attempts to form sharp edges on cutting tools,
hole location 0.1% 0.3%
knives, or scissors using MIM have been unsuccessful since
parallelism 0.2% 0.3% the particle size limits the sintered edge retention. This is ev-
perpendicularity 0.1% or 0.1 0.2% or 0.3 ident by the rounded edge on a MIM drill tip as seen in Figure
4.23. Smaller particles are needed to produce small features.
roundness 0.3% 0.3%
surface roughness 0.2 m 0.8 m TOLERANCE APPROACHES
(8 in.) (32 in.) There is a high cost associated with tight dimensional con-
weight 0.1% 0.4% trol. Sorting good and bad components after sintering is one
corner radius 0.5 mm 0.3 mm
approach, where those out of tolerance are discarded or re-
(0.02 in.) (0.012 in.) worked, but the cost is high. More attractive is to take steps
to minimize dimensional variation during production, but that

62
 Table of Contents
Chapter 4: Manufacturing Considerations

too comes with an increased cost. For example, custom setters 4.24 is an example of one such conformal substrate. Since the
can be fabricated for use during sintering, as is the case in pro- component shrinks during sintering, the substrate design must
duction of high-precision casting cores used in airfoil appli- provide support but not restraint. In some cases the substrate
cations. These setters are fabricated from high-temperature is a ceramic peg or grooved tray, yet it can become very com-
zirconia machined to the final contours with diamond grinding plicated and expensive.
techniques. Each injection molded body is then nestled into
the setter to ensure conformal support during sintering. In
other words, the MIM component deformation in sintering is
guided to a precise final shape cut into the substrate. Figure

Figure 4.20. Surface roughness in MIM relates to the particle size.

At the tool wall the powders pack against a flat surface. The result-
ing cobblestone surface reflects the particle size more than the mold
surface finish. Smooth sintered surfaces require smaller particles
Figure 4.22. An illustration of how particle packing into a corner
naturally leads to a rounded corner or edge in spite of a sharp tool

Figure 4.21. Scanning electron micrograph of the surface from a

sintered stainless steel MIM component. The surface roughness
reflects the initial particle size. A few handling scratches are Figure 4.23. Scanning electron micrograph from a cutting edge on
evident on the surface as well. Post-sintering polishing is an option a drill formed using MIM. Note the tip has a rounded character as
to improve surface finish controlled by the particle size

63
 Table of Contents
Chapter 4: Manufacturing Considerations
Figure 4.24. Conformal debinding and sintering substrates are
one means to hold closer tolerances on MIM components that
would otherwise tend to distort during sintering. In this case the
substrate is constructed in a manner to allow for support during
sintering shrinkage (courtesy of Advanced Materials Technologies)
In several situations, the setters effectively control the final
dimensions, but only for certain features and certain orienta-
tions. Features perpendicular to the setter are not as precise.
Hence, some dimensions are well controlled while others are
more scattered. A problem with setters lies in the cost of their
fabrication and subsequent maintenance. The setter must resist
heat distortion and thermal fatigue, and must be chemically
inert through the sintering cycle. Moreover, the setters must
have good handling strength and fracture toughness to resist
damage in a production environment. Materials with such
attributes are expensive and difficult to machineindeed, set-
ters are sometimes fabricated by injection molding ceramics.
On the other hand, some features can only be controlled by
post-sintering steps, including grinding. One option is to
include an extra feature or support in the MIM body known as
a knock-out. These are pieces molded into the component for
strength or alignment during sintering, which are removed after
sintering.
An error budget concept helps capture the cost of highly
toleranced MIM production. Similar to the monthly household
budgetso much for rent, so much for food, and so much for
the automobilean error budget partitions the dimensional
tolerance allowance to various features. If very tight toler-
ances are required on one dimension, then loose tolerances
are used elsewhere to offset the cost implications of the tight
tolerances.
As an illustration, consider a component with size specifi-
cations of, say, 32, 20, 15, 10, and 5 mm. If we estimate the
nominal dimensional precision as a coefficient of variation at
0.3% on each feature, then the total tolerance budget B is
B= log(Di / ti) = log(1/CV) (4.5)
where the counter i runs from 1 to N (N is the total number of
specificationsfive for this example), Di is the size of each
of the N dimensions (such as 32 mm), and ti is the correspon-
64
ding absolute tolerance on that dimension (0.3% of 32 mm is
0.096 mm). Note the tolerance budget is related to the coeffi-
cient of variation CV. For this example, the 0.3% coefficient
of variation gives a target process capability budget of 12.6
(dimensionless). If the 20 mm feature is set to a tolerance of
0.1% and the 10 mm feature is set to 0.2%, then the toler-
ance budget has increased to 13.3. Accordingly, to compensate
a few tolerances need to be relaxed to balance the budget. One
solution would be to increase the tolerance on the remaining
three dimensions to 0.5%. Otherwise, the combination of tol-
erances is outside the MIM capability range. As the number
of features increases, the tolerance budget also increases, like-
wise increasing cost. Complexity concepts as applied to MIM
suggest that increases in either the number of features or the
precision of those features raise cost in a nonlinear manner.
As a final note on tolerances, not all of the tolerance al-
lowance can be spent on the part-to-part, day-to-day, or cav-
ity-to-cavity variations. Usually tooling cannot be constructed
with perfection, especially multiple cavity molds. Thus, some
portion of the allowed dimensional variation is set aside for
tooling. Often this is simply an allocation that requires the tool
tolerance be within 20%30% of the part tolerance. If the part
tolerance is tight, then the cost of tooling increases too.
PROCESS YIELD
Process yield has considerable impact on costs. For every
component there is an estimated process yield built into the
price quotation. For example, an optical connector out of
stainless steel with a 113 g mass was unprofitable when
process yields fell below 84%.
Process loss comes from component defects that mostly
arise in molding and from dimensional-control issues, which
also largely arise in molding. As one industry leader said, you
either make it or break it at the molding machine. Early in a
project, while experience is gained, the yield of acceptable
components might be low. In part this is because several
process adjustments are needed to balance between dimen-
sional attributes, avoiding molding defects, and properly con-
trolling the equipment. Across the industry, long-running
projects reach about 98% yield. Yields as low as 80% are seen
on some tight-toleranced components, but with considerable
scrap and anguish.
Two key steps work to improve process yield. The first is
to rely on prior experience. Computer simulations also help
determine key parameters and control options. Simulation is
most useful in tool and process design, determining how to
avoid processing defects. The second means to improved
process yield is to rely on in situ instrumentation to ensure
embedded process control. Probably the highest payout comes
from mold cavity pressure transducers that ensure each com-
ponent is formed to the same packing pressure, giving uniform
mass and then sintered uniform dimensions. Mold-cavity tem-
perature transducers are likewise important to production.
 Table of Contents
Chapter 4: Manufacturing Considerations

COMPUTER SIMULATION FOR DEFECT

AVOIDANCE
The success of MIM is linked to its ability to generate pre-
cise, complex shapes at the molding machine and then to sin-
ter that shape to the target size and proper specifications. In
turn, this requires quality tooling and process control. Further,
computer software has been used to simulate molding, tool
design, debinding, and sintering. The mold-filling simulations
are advanced to the point where there are several offerings.
Commercial software packages include Moldflow (Moldflow
Corp., Framingham, Massachusetts, U.S.), Moldex3D
(CoreTech System Co., Ltd., Chupei City, Taiwan), PIMsolver
(CetaTech, Sacheon, Korea), SigmaSoft (Sigma Engineering,
Aachen, Germany), and SIMUFLOW (C-Solutions, Inc., Figure 4.25. Computer-simulated mold filling for a gear compo-
Boulder, Colorado, U.S.). Additionally, several research nent and the associated shear stress at various locations during the
first 0.3 seconds (images courtesy of Chien Min Wang)
groups have written customized codes, but these are not
released for public use.
Current computer programs emphasize a combination of
mold filling, packing, and cooling simulations. Figures 4.25,
4.26, and 4.27 are example visual outputs for some of the sim-
ulations. Models are available for the key concernsproper
mold filling, determination of residual stresses, calculation of
temperature distribution and cooling cycles, prediction of de-
fect formation, and support for mold-design issues such as
gate size and placement, and vent size and placement. Current
efforts are adding debinding and sintering to give full process
models. Such packages allow process evaluation prior to any
commitment to manufacturing. For example, cost evaluations
would be possible based on differences in raw materials, the
number of tool cavities, and options on processing cycles. In-
cluded in the current tooling calculations are allowances for
the tool complexity, tolerances, surface finish, and number of Figure 4.26. Computer-simulated filling time for a printer yoke
MIM component. Filling times are recorded at a few points on the
cavities, number of ejector pins and vents, and complexity of images, where the central region fills first. Total filling time is
the parting plane. Debinding and sintering simulations are still 1 second for the outer corners (image courtesy of CetaTech)
largely in development, but some impressive demonstrations
have taken place.
A driver on the sintering simulations is a need to make ac-
curate mold-size predictions based on the dimensional speci-
fication for the sintered part. Often subtle variations exist in
sintering shrinkage that need to be predicted to minimize tool-
construction errors. For example, powder packing can be
nonuniform due to binderpowder separation in regions of
high shear during filling. Further, subtle factors such as grav-
itational forces in sintering and polymer orientation in mold-
ing cause subtle anisotropic shrinkages that contribute to
dimensional variations. As a consequence, many molds are
designed using costly trial-and-error approaches. The desire
is to have computer software incorporate these factors into the
tool design to help place runners, gates, vents, and cooling
paths. Once the shape and gradients, or stresses, are known in
Figure 4.27. Computer-simulated filling time for a turbocharger
the molded body, then finite element computer-simulation geometry, where the feedstock enters the sprue at time 0, reaches
tools allow prediction of final sintered size and shape. each gate at 0.2 s, fills the central region in 0.5 s, and fills the
For these simulations to operate successfully, a database blade tips by 1 second (image courtesy of CetaTech)

65
 Table of Contents
Chapter 4: Manufacturing Considerations
with the material properties is needed as well as component-
specific details and process details, as outlined in Figure 4.28.
In sintering simulations, there is a need for data on heating cy-
cles, furnace design, atmosphere interactions, and sintering
shrinkage vs. timetemperature during heating. Finite element
analysis integrates all of these features into final predictions.
Figure 4.29 illustrates one example of the success. Shown in
this figure is a U-shaped MIM component with a sequence of
predictions based on a water-atomized 316L stainless steel
feedstock in a wax-polymer binder at a solids loading of 53
v/o. The sequence of images shows the following steps:
a) the 3.17 mm (0.125 in.)-thick component layout and the
location of the sprue, runner, and gate (half-circle of
4 mm dia.)
b) filling stage and velocity profiles at about half filled,
using 130C (266F) feedstock and 30C (86F) mold approach MIM, relying on pressure, temperature, and filling
c) volumetric shrinkage of the packed part based on peak
pressure of 157 MPa (23 ksi), ranging from 4.1% to
5.5%
d) freezing distribution after the gate has frozen (100%
frozen layer) with delayed solidification at the center of
the U
e) outline of the cooling channel layout in the mold with
two cooling lines, showing why the center of the U is
slow to cool
f) surface temperature gradients on ejection, with the cen-
ter of the U being hottest
g) subtle green density distribution in the component
h) predicted sintered size with a peak sintered density of
93.4%, and next to this is the actual component, which
did deform as predicted during sintering
Unfortunately, many of the required modeling parameters
are not known with precision and must be estimated. Of
course, the simulation results are no better than the input
knowledge. A poorly controlled MIM practice is not going to
be improved by simulation. On the other hand, for well-con-
trolled operations, computer-assisted design is an inexpensive
way to explore various options before mold production. One
expectation is that these tools will be available over the Inter-
net to help share and pool knowledge, but this is in the future.
QUALITY PLANS
The final phase of MIM production is inspection. Todays
quality efforts span all manufacturing enterprises, so the be-
havior used in MIM is essentially the same as encountered
elsewhere; they are recognized by large, multinational corpo-
rations as evidenced by:
working in partnership with customers and suppliers
customers are the primary focus
providing a professional work environment
keeping key employees involved and trained with
respect to quality objectives.
Leading MIM producers are qualified by the common
66
Figure 4.28. Outline of how computer-assisted engineering tools
calculations to determine density, temperature, velocity, and other
parameters in filling, packing, and cooling. These parameters are
used to identify defects (binderpowder separation, jetting, short
shots, trapped air, or improper ejection) in the molded component
and to support simulations related to debinding and sintering to
predict final component size, shape, and properties
international systems: International Standards Organization,
Total Quality Assurance, Food and Drug Good Manufacturing
Practices, Military Standards and Specifications, or the Auto-
motive Industry Quality Standards. They operate in a manner
consistent with high quality, sustaining on-time delivery of
defect-free products. The ISO 9000 standards are split be-
tween the companies that have design responsibilities and
those that are only fabricators of those designs. The ISO
14000 standards focus on environmental compliance.
At the core of the quality system is inspection, documen-
tation, and assurance that the shipped product conforms to
both customer and internal standards. There are two different
concernsproduction set-up and actual production. The first
is associated with submission of the first articles for customer
approval. During first-article manufacturing, various tests
might be needed for establishing a robust capability, detecting
defects, and calibrating the process. In MIM operations these
include tests such as visual examinations, weight checks, ul-
trasonic examination, X-ray radiographs, and optical profiling,
and resistivity, magnetic, or dimensional checks. Most impor-
tant are uniformity in component dimensions and weight
(mass) with no defects.
During production, periodic samples are measured based
upon a planned sampling frequency. Common inspection tasks
include visual examination for defects or flaws, density and
weight determinations, measurement of final dimensions, and
verification of the desired oxygen, carbon, or other impurity
levels using hardness or chemistry tests. Occasionally, the com-
ponents might be tested to verify mechanical and other proper-
ties. However, there is much ambiguity in current test methods
and a lack of MIM reference materials. Thus, unlike classic
 Table of Contents
Chapter 4: Manufacturing Considerations

(a) (b) (c)

(d) (e) (f)

(g) (h)
Figure 4.29. U-shaped MIM component showing (a) the component layout and the location of the sprue, runner, and gate, (b) filling
stage and velocity profiles simulation at half filling, (c) volumetric shrinkage of the packed part, (d) freezing distribution after the gate has
frozen, (e) outline of the cooling channel layout in the mold, (f) surface-temperature gradients on ejection, (g) green-density distribution in
the component, (h) predicted sintered size and picture of actual part after sintering (images courtesy of CetaTech)

engineering materials, where reference materials are available
for calibration, MIM is without such baseline materials. Con-
sequently, calibration and acceptance criteria are largely set by 2.
the customer and supplier on an ad hoc basiscases are
handled one at a time. Because of this situation, acceptance test
samples are exchanged between the producer and users. 3.
For success in this environment, the MIM fabricator must
be proactive and plan the testing procedures, document and 4.
obtain acceptance of those procedures, and then conform to
the plan. A vendorcustomer plan documents what will be 5.
measured, how frequently, and with what techniques, and cap-
tures samples that are retained as de facto standards. A full
quality system for MIM requires an integrated view that 6.
ranges from first customer interactions to post-delivery
service. As an example, the following quality plan outline was
provided by an automotive component supplier:
1. Organizational, Sales, and Administrative Systems: 7.
system-wide efforts to manage the total process via an
organizational quality policy
Communication, Information, and Documentation
Systems: communication systems to keep the organi-
zation updated on progress
Indicators: provide quantitative objectives and report
on progress
Personnel: provide training and orientation to all
employees
Product and Process Development and Industrializa-
tion: propagate quality standards at the product- and
process-conceptualization stage
Management of Supplier Quality: suppliers provide
conforming materials, equipment, and products with
identified quality levels, including records of incoming
materials
Handling of Nonconformities: incoming material prob-

67
 Table of Contents
Chapter 4: Manufacturing Considerations
lems and complaints from customers are documented
8. Maintenance Plans: equipment and processes remain
in conformity with set standards in a preventative-
maintenance system
9. Quality of Measurement: inspection devices are em-
ployed to assure proper conformance to set standards
10. Control Systems for Products, Processes, and Proce-
dures: the system looks to improve procedures to prop-
agate the quality system throughout the organization
11. Audits: both internal and external audits are performed
to ensure the process is conforming to standards
INSPECTION
Metal powder injection molding has a variety of standards
organized by the regional governing bodies in the U.S.,
Europe, and Japan: Metal Powder Industries Federation,
European Powder Metallurgy Association, and Japan Powder
Metallurgy Association. Firms have joined one of these
regional trade associations to help promote the field and to
write standards.
Data sharing is progressing between regions. Currently,
Powder Injection Molding International best serves as the
conduit for market news and data. Another important compi-
lation is from Inovar Communications, which produces the
International Powder Metallurgy Directory in both print and
online forms.
Quality Tests
In MIM, the input materials include the polymers and pow-
ders. The powder composition is treated by a range of stan-
dards and nominally these are the same as encountered in
traditional metallurgy. For example, the specification for 316L
stainless steel comparison between the MIM powder and
wrought material are nearly the same. However, the minimum
mechanical properties differ slightly due to the allowed poros-
ity in the MIM variant. Current specifications for composition
and sintered properties include the following:
low-alloy steels (three grades and two heat treatments for
4605)
stainless steels (three grades and two heat treatments for
17-4 PH)
soft magnetic alloys (five grades and two impurity levels
on Fe-50Ni and Fe-3Si)
controlled-expansion alloy (one grade)
In addition, various standards cover testing procedures for
mechanical properties, corrosion resistance, magnetic proper-
ties, wear testing, particle size, powder density, powder pack-
ing, powder flow, and microstructure evaluation. Within the
MIM community unofficial tests have emerged, such as use
of the plastic-melt-flow index for viscosity evaluation, use of
witness-slug-shrinkage measurements for sintering-tempera-
ture-uniformity tests, and helium-gas pycnometry for deter-
mination of granulated-feedstock density.
68
For certain medical and aerospace components, assurances
of proper sintering require witness tensile samples in each sin-
tering run. Although there are well established international
metallurgical standards for tensile geometries, some MIM
firms rely on rounded-dogbone shapes. Across the MIM in-
dustry test-sample sizes range from 5 mm to 140 mm length.
Generally, for a 30 to 38 mm gauge length (1.2 to 1.5 in.), the
sample thickness is near 5 mm (0.2 in.). The gauge length as
a ratio to the gauge diameter varies from 5 to 8, which is larger
than used in traditional metallurgy where the ratio is held to
4. For this reason ductility should not be directly compared
with traditional property listings, especially for materials that
neck prior to failure.
The final phase of MIM production is to perform quality
checks to ensure delivery to specification. Many of the MIM
operations have registered their quality systemsISO 9001
for captive production and ISO 9002 for custom production.
Testing might be performed at either the fabrication or use
sites, and with appropriate quality systems the user will accept
the shipment without inspection. This was the situation in the
disk-drive and cellphone industries, where millions of MIM
components were delivered without inspection (nor problem)
based on the MIM vendors quality system. Benchmarks in
the MIM industry are upwards of 50 million parts shipped
without a defect. To achieve such quality levels requires col-
laboration to ensure the producer understands the functional
attributes and delivers only acceptable components. In pro-
duction, the goal is to capture errors as they occur rather than
after fabrication.
Detailed quality checks are associated with first-article
submission, which includes data showing the measured fea-
tures and the specified featuresdimensions, hardness, mi-
crostructure, chemistry, or other attributes. Once a component
moves into production, the inspection criteria need to establish
several factorswhat should be measured, how it is meas-
ured, where it is measured during production, how frequently
it is measured, how many samples are used each time, how
those samples are captured, the lot size associated with the
measurements, and the allowed ranges. Each user industry
typically has standards that guide implementation of these
tests.
An important shop-floor technique is statistical process con-
trol (SPC). It is used at intermediate manufacturing steps to as-
sess drifts from standard conditions. The goal is to keep the
product in an accepted range throughout the MIM process.
Some variability occurs in any operation, but SPC continually
senses drifts that may take a process out of the acceptable range.
Many sources of defects are possible in MIM. A first cate-
gorization is given by the fishbone diagram in Figure 4.30.
Most of the problems come from molding or thermal process-
ing. In a well-controlled process, the unanticipated defects can
be reduced, since there are some obvious sources and cures.
Others problems, such as parting lines and ejection pin blem-

Figure 4.30. Fishbone diagram for injection molding, showing
various quality issues and how they relate to the overall process
ishes, are present in all MIM components and can be antici-
pated in the design stage. These should be incorporated into
noncritical positions in the mold.
Quantitative Techniques
Underlying the SPC concept is the assumption of a distri-
bution in the measured parameter described by the normal or
Gaussian distribution where the spread in readings is de-
scribed by the standard deviation around the mean. The mean
is calculated by adding all observations and dividing the total
by the number of observations. In school this was also called
the average and is given by
1 tecting defects based on ultrasonics, X-ray radiography, or
UA =
N
N
i=1 Ui
which says to add the individual readings Ui from the first to
last and divide the sum by the number of readings N to obtain
the average UA. For example, the average of 7, 8, 11, 21, and
23 is 70/5 or 14.
The standard deviation is calculated as the variation
about the mean UA as
N 2
i=1 (Ui UA)
= . (4.7)
N1
The standard deviation depends on the departure of each read-
ing from the mean. Some MIM components are specified by
3 standard deviations. This implies that essentially all of the
components are within a prescribed range of values, for ex-
ample, from UA-3 to UA+3. As an illustration, if UA = 10
mm and = 0.1 mm, then essentially all production would
range between 9.7 and 10.3 mm.
The process capability Cp is a measure of the ability of a
MIM operation to stay within a specified limit. It measures
the ratio of the process spread in terms of three standard
Table of Contents
Chapter 4: Manufacturing Considerations
deviations and compares this process spread with the tolerance
allowed spread,
UM U A
CP = (4.8)
3
where UM is the maximum (or minimum) control limit spec-
ified by the designer, UA is the production mean, and is the
production standard deviation. Two values result from this
equation based on the maximum and minimum limits for UM.
If the mean production value is centered between the max-
imum and minimum bounds, then the process has a wider al-
lowed variation. However, if the process skews toward one
end or the other, then there is less allowed variation because
UM - UA becomes smaller. The distance from the production
mean to the nearest specification limit is the process capability
designated as Cpk. The value of Cpk can never exceed Cp. High
values of the process-control capability Cpk are desirable and
values of 1.33 or higher are generally most impressive.
Nondestructive Techniques
In any of the inspection procedures, four possible situations
must be recognized. Accepting defect-free components and
rejecting defective components is the goal. The worst case is
to accept a defective component, while the rejection of a de-
fect-free component is wasteful. Thus, careful attention must
be directed to the criteria for both acceptance and rejection
decisions. These are often based on nondestructive analysis.
Although visual inspection is important, much progress is
being made to automate inspection tasks. Tests exist for de-
(4.6)
eddy currents. Closed-loop feedback control is emerging,
where the fabrication step is instrumented and controlled to
make immediate adjustments on a part as it is being formed.
This is most important in molding using die-cavity pressure
and temperature transducers. That form of in situ control
greatly improves quality.
RESOURCES
S. Ahn, S.T. Chung, S.J. Park, and R.M. German, Simulation
of PM Processes: Metal Injection Molding, ASM Handbook
Modeling and Simulation, 2010, ASM International, Materials
Park, Ohio.
S.V. Atre, T.J. Potter, E.G. Woodward, J.A. Thomas, T. Togwe
and R.M. German, Powder Based Rapid Prototyping Tech-
niques: Limitations and Opportunities, P/M Science and
Technology Briefs, 2000, vol. 2, no. 4, pp. 1820.
D.C. Blaine and R.M. German, Sintering Simulation of MIM
Stainless Steel, Advances in Powder Metallurgy and Partic-
ulate Materials2002, compiled by V. Arnhold, C.-L. Chu,
W.F. Jandeska, Jr., and H.I. Sanderow, Metal Powder Indus-
tries Federation, Princeton, NJ, part 10, pp. 255266.
69
 Table of Contents
Chapter 4: Manufacturing Considerations

D.M. Bryce, Plastic Injection Molding Mold Design and Y. Shengjie, Y.C. Lam, S.C.M. Yu, and K.C. Tam, Thermo-
Construction Fundamentals, 1998, Society of Manufacturing mechanical Simulation of MIM Thermal Debinding,
Engineers, Dearborn, MI. International Journal of Powder Metallurgy, 2002, vol. 38,
no. 8, pp. 4155.
P.S. Cracknell and R.W. Dyson, Handbook of Thermoplastic
Injection Mould Design, 1993, Blackie Academic and Profes- B. Smarslok and R.M. German, Identification of Design
sional, London, UK. Parameters in Metal Powder Injection Molding, Journal of
Advanced Materials, 2005, vol. 37, no. 4, pp. 311.
R.M. German and A. Bose, Injection Molding of Metals and
Ceramics, 1996, Metal Powder Industries Federation, Prince- M. Trunec and J. Cihlar, Thermal Removal of Multicompo-
ton, NJ. nent Binder from Ceramic Injection Mouldings, Journal of
the European Ceramic Society, 2002, vol. 22, pp. 2,231
T. Hartwig and T.V. Duong, Miniaturisation of MIM,
2,241.
Second European Symposium on Powder Injection Moulding,
2000, European Powder Metallurgy Association, Shrewsbury, B. Williams, Setting Standards and Getting it Right, Metal
UK, pp. 229234. Powder Report, 2001, vol. 56, no. 12, pp. 1418.
U. Haupt and H. Walcher, Powder Injection Moulding Con-
cepts for Presses and Tooling, Proceedings of the First
European Symposium on Powder Injection Moulding, 1997,
European Powder Metallurgy Association, Shrewsbury, UK,
pp. 7079.
J.L. Johnson and R.M. German, Design Capabilities of
Powder Injection Molded Materials, Journal of Advanced
Materials, 2005, vol. 37, no. 3, pp. 5159.
B. Julien, M. Ste-Marie, R. Pelletier, F. Lapointe and S.
Turenne, Parametric Modeling of MIM As-Sintered Proper-
ties of Stainless Steel 17-4 PH and 316L Using a Statistical
Model, Advances in Powder Metallurgy and Particulate
Materials2010, compiled by M. Bulger and B. Stebick,
Metal Powder Industries Federation, Princeton, NJ, part 4,
pp. 110.
V. Piotter, T. Benzler, R. Ruprecht and J. Hausselt, Manufac-
ture of Micro Sized Structures by MIM and CIM, Advances
in Powder Metallurgy and Particulate Materials1998,
compiled by J.J. Oakes and J.H. Reinshagen, Metal Powder
Industries Federation, Princeton, NJ, part 5, pp. 161168.
H. Rees, Mold Engineering, 1995, Carl Hanser Verlag,
Munich, Germany, 1995.
B.O. Rhee and J.H. Lee, Investigation of the Gate Effect on
the Powder Injection Molding Process, Proceedings of the
2000 Powder Metallurgy World Congress, Part 1, K. Kosuge
and H. Nagai (eds.), Japan Society of Powder and Powder
Metallurgy, Kyoto, Japan, pp. 278281.
D.V. Rosato and D.V. Rosato, Injection Molding Handbook,
1986, Van Nostrand Reinhold, New York, NY.
N. Salk and B. Troger, Precision Manufacturing of Micro
Ceramic Injection Molded Implants, Advances in Powder
Metallurgy and Particulate Materials2010, compiled by
M. Bulger and B. Stebick, Metal Powder Industries Federa-
tion, Princeton, NJ, part 4, pp. 1117.

70
 Table of Contents

Chapter 5
MATERIAL PROPERTIES

Outline Available Materials

Powders
Feedstock
Standardized Chemistries
Cost Attributes
Materials Selection
Key Classifications
Electrical and Magnetic
Corrosion and Biocompatibility
Inertial Properties
Structural Properties
Tensile Properties of Metallic Materials
Rupture Properties of Ceramic Materials
Impact, Fracture, and Fatigue Properties
Elevated-Temperature Properties
Optical Properties
Thermal Properties
High Temperature
Wear
Microstructures
Composite Materials
High-Elastic-Modulus Cermets
Heterogeneous Microstructures
Atmosphere-Stabilized Compositions
Biocompatible Materials
Infiltrated Preforms
Bonded Materials
Controlled-Porosity Structures
Testing Standards
Material Standards
Resources

AVAILABLE MATERIALS reports over 900 formulations in use of different alloys, pow-
Component fabrication by MIM is contingent on powder der types, and particle sizes.
availability. The good news is that a very broad range of com- The powders used in MIM come in a variety of chemis-
positions are available. For example, one feedstock vendor tries, particle sizes, and particle shapes. Most typically, the

71
 Table of Contents
Chapter 5: Material Properties
powders are equiaxed, rounded, and smaller than 20 m in
size, such as shown in Figure 5.1 (nickel alloy powder) and
Figure 5.2 (titanium powder). These powders pack to densities
over 45% of theoretical, often close to 60%. Powder avail-
ability determines which engineering materials can be pro-
duced by MIM; fortunately, most of the widely used
engineering compositions are formed into small particles.
Table 5.1 is a listing of the major chemistries in use.
This tabulation includes a variety of metals and alloys. Be-
sides the metallic powders, a range of oxide ceramics (alu-
mina, zirconia), nonoxide ceramics (silicon nitride, silicon
carbide), cemented carbides (tungsten carbide with cobalt),
and refractory metals (tungsten, molybdenum) are in use. Be-
sides those listed, it is possible to obtain special materials on
a custom basis. Other custom alloys include aluminide inter-
metallics and NiTi shape-memory alloys. Some of the special
Figure 5.1. Scanning electron micrograph of a typical gas-
atomized powder used in MIM, in this case a nickel-based alloy
with a median particle size below 10 m
Figure 5.2. Titanium milled powder used for MIM, showing a
nonspherical particle shape, which contrasts with the high packing
density powder shown in Figure 5.1
72
materials include maraging steel, tool steels, cobaltchromium
alloys, reactive metals (such as titanium, tantalum, niobium,
and rhenium), and special electronic alloys. Many of these are
not used in high production volumes, so they are prepared as
special ordersa factor that negatively impacts price.
For the more complex chemistries, there are three possible
starting forms:
mixed elemental powders
fully compounded or prealloyed powders
hybrid combinations of compounded master alloys and
elemental powders
For stainless steels, all variants are in use: mixed iron,
nickel, and chromium powders; prealloyed powders where
each particle is the same composition; and iron powders
mixed with master prealloyed ironnickelchromium powder.
The less homogeneous the starting powder, the more effort
required in sintering to obtain high final property levels. Thus,
the mixed elemental powder route has the lowest cost while
generally the fully compounded prealloyed powders give the
greatest assurance of meeting property and performance goals.
In the case of a stainless steel, the prealloyed powder sinters
with relative ease. If iron and a prealloyed master alloy pow-
der are mixed to give the same bulk composition, then the
time at the peak sintering temperature must be extended. Cre-
ation of a homogeneous sintered microstructure is even more
difficult when starting with elemental powder mixtures. The
bulk chemistry conforms to the alloy specification and the
powder cost is lower, but the microstructure will not be ac-
ceptable unless sintering is extended for a long time.
TABLE 5.1. ALPHABETICAL LISTING OF METALS
AND ALLOYS AVAILABLE BY MIM
aluminum (Al), aluminum alloys (Al-Sn, Al-Fe, Al-Cu-Mg)
bronze (Cu-Sn)
cemented carbide (WC-Co, WC-TaC/TiC/VC-Co)
cobalt-chromium (Co-Cr-W-C, Co-Cr-Mo)
copper (Cu)
Invar (Fe-Ni)
iron (Fe), ironsilicon (Fe-Si), ironnickel (Fe-Ni), iron
phosphorus (Fe-P), ironnickelcobalt (Fe-Ni-Co)
Kovar (Fe-Ni-Co)
molybdenum (Mo), molybdenumcopper (Mo-Cu)
nickel (Ni), nickel aluminide (NiAl and Ni3Al), nickeliron
(Ni-Fe)
nickel-base (Ni-Cr-Mo)
Sendust (Fe-Al-Si)
steel (Fe-C), copper-steel (Fe-Cu-C), nickel-steel
(Fe-Ni-C), low-alloy steels (Fe-Ni-Mo-Mn-Cr-C)
stainless steel (Fe-Cr, Fe-Cr-Ni, Fe-Cr-Ni-Mo)
superalloy (Ni-Co-Cr-Ti-Al-Mo)
titanium (Ti), titanium alloy (Ti-Al-V, Ti-Al-Nb), titanium
aluminide (TiAl, Ti3Al)
tool steel (Fe-Co-Cr-W-V-C)
tungsten (W), tungstencopper (W-Cu), tungsten heavy
alloy (W-Ni-Fe, W-Ni-Cu)
 Table of Contents
Chapter 5: Material Properties

Powders sity. One major trend is to rely on smaller particle sizes that
The variety of powder chemistries used in MIM requires a sinter more easily and improve surface detail on the sintered
broad array of powder-fabrication techniques. Widely used article.
powder-production technologies include gas atomization, One problem in powder purchasing is that the materials
water atomization, attritioning, carbonyl decomposition, oxide might come with different designations for the same compo-
reduction, and plasma atomization. Each fabrication approach sition; the standards come from unrelated sourcesMilitary
influences the particle size as well as the shape, microstruc- Specifications (MIL SPEC), Aerospace Materials Specifica-
ture, packing density, chemistry, and cost. Sometimes, the par- tions (AMS), American Society for Testing Materials
ticle shape is not ideal for MIM, mainly because of asperities (ASTM), International Standards Organization (ISO), Society
on the particle surface that interfere with particle flow in of Automotive Engineers (SAE), as well as German, U.S., and
molding. In such cases, higher-cost but more-spherical pow- Japanese trade associations. This results in different names for
ders might be a better choice. As consumption increases, there the same material. For example, surgical stainless steel is also
will be more fabrication of powders specifically for MIM, called 17-4 PH stainless steel. It is often referred to as AISI
with a progressive price reduction. 630 in the medical field (American Iron and Steel Institute)
Table 5.2 gives some examples of the metal powders in use, and AMS 5355 in the aerospace field and is called SUS630 in
listing the composition, median particle size, and packing den- Japan. To avoid trademark infringements, several of the pow-
der suppliers give the materials their own names.
TABLE 5.2. CHARACTERISTICS OF SOME TYPICAL
Feedstock
MIM POWDERS
Several companies supply the small powders required for
nominal median apparent MIM. In some cases the company is focused on supplying
material chemistry, particle density,
small powders specifically to this field, while in other cases
% size, m g/cm3
the company has a broad offering and injection molding
aluminum 99.7 Al 30 1.1 grades are a relatively small offering. Annual updated listings
cemented carbide WC-10Co 1.1 3.4
of firms providing different powders are given in technical
copper 99.5 Cu 14 3.3
iron 98.3 Fe 4.1 2.6 magazines, such as Powder Metallurgy, Ceramic Industry, Ce-
ironcarbon Fe-0.9C 4.7 2.7 ramic Bulletin, Metal Powder Report, or International Jour-
ironsilicon Fe-3Si 7.6 3.9 nal of Powder Metallurgy. Additional information is available
molybdenum Mo 2.5 2.1 via Web searches. Trade associations help point to vendors
nickel 99.8 Ni 7.6 3.6
through their Web sitesfor example, visit the sites of the
niobium 97.8 Nb 7.4 3.0
molybdenum 99.7 Mo 4.1 2.7 American Ceramic Society (www.acers.org), European Pow-
stainless steel Fe-Cr-Ni-Mo 11 3.9 der Metallurgy Association (www.epma.com), Japan Powder
stainless steel Fe-Cr-Ni-Cu 21 3.3 Metallurgy Association (www.icon.perf.nagano.jp), and Metal
stainless steel Fe-Cr-Ni-Cu 10 4.2 Powder Industries Federation (www.mpif.org). The Inovar
titanium Ti-6Al-4V 32 2.1
Communications group produces both a printed and electronic
tool steel Fe-12Cr-1.5C 11 4.0
tungsten 99.5 W 3.0 2.9 report on the field that keeps current with changes.
Compounded feedstock consisting of precisely mixed
binders and powders are commercially available for molding.
TABLE 5.3. ALPHABETICAL LISTING OF Feedstock vendors include the firms listed in Table 5.3. Firms
COMPANIES SUPPLYING PREMIXED FEEDSTOCK enter and exit this business frequently, but at any time there
are typically about 10 to 12 sources.
Advanced Metalworking Practices (U.S.)
Atect (Japan)
BASF (Germany) Standardized Chemistries
Di-Ichi Ceramco (Japan) The standards for MIM materials focus on composition,
eMBe (Germany) typical properties, and testing procedures. More details are
Imeta (Germany) given at the end of this chapter. Most compositional specifica-
Inmatec (Germany)
PolyMIM (Germany) tions are duplicates of traditional handbook specifications with
Ryer (U.S.) tolerances similar to those listed by the American Society for
Starfire Systems (U.S.) Testing Materials (ASTM) or similar bodies. However, there
Syroko (Netherlands) are a few chemistries available that are specific to MIM.
TCK (Dominican Republic) Mostly these represent adjustments to standard chemistries to
Tosoh (Japan)
Witech (Japan) assist in processing. For example, extra chromium in a stainless
steel improves corrosion resistance, but also increases cost.

73
 Table of Contents
Chapter 5: Material Properties

The convention is to show compositions in weight percent
(w/o) unless otherwise noted, and in most cases the percent
designation is skipped where there is no confusion. Hence,
Fe-2Ni indicates 98 w/o Fe and 2 w/o Ni. Unfortunately, the
same alloy might have different names depending on which
standard is adoptedfor example, Fe-2Ni alloys are desig-
nated as 2200, 4605, or 4650, often with similar compositions
(formally 2200 has no carbon, but both 4605 and 4650 have
about 0.5% C). Here are a few examples of the different
names given to essentially the same composition. In each case
the name used here is given first:
ironnickel (Fe-2Ni) = FN02, MIM-2200, FN0200, Fe-
2Ni, steel 2200, AISI 4600, FEN-2, MIM-Fe-2%Ni, 2%
nickeliron, FeNi2, Fe-2%Ni
stainless 316L = X2CrNiMo 17 13 2, DIN 1.4404, UNS
S31603, SUS316L, MIM 316L, MIM-316L, MP-S-
0004, SAE316LG, AISI 316L
stainless 17-4 PH = MIL-S-81506, X5 CrNiCuNb 17 4,
DIN 1.4542, SAE J467 (17-4 PH), AMS 5355, UNS
S17400, AISI 630, SUS630, MIM-17-4 PH, 17.4 PH,
and MP-S-0021
A comprehensive cross-index of names is available on var-
ious Web sites. Additionally, the important databases, such as
Cambridge Engineering Selector (Granta Design, Cambridge,
UK), now carry MIM materials and tabulate their properties.
Compositions used in MIM are detailed in Table 5.4 based
on the common designation. Note that in some cases individ-
ual vendors customize the nomenclature or compositions
slightly, but this listing is characteristic of the popular MIM
materials.
COST ATTRIBUTES
Material selection involves a balance between cost and per-
formance. Lower-cost materials are used most frequently. Fig-
ure 5.3 shows how the powder injection molding sales for
2010 partition by material class (note that this figure is based
on material value, not on tonnage), showing that over 77% of
all sales are metals. Ceramic injection molding is about 15%
and carbides constitute 3% of sales. A small amount of PIM
is applied to composites, and about 4% of sales involve mixed
materials or are indeterminate as to material category.
If plotted on a mass basis the carbide sector would shrink.
This is because carbide powders are generally more expen-
sive. Beyond the differences in ingredient costs (nickel vs.
iron, for example), powder price depends on consumption: the
more popular compositions are more economical. This means
that ferrous alloys and oxide ceramics are the lowest-cost and
most frequently used materials. The classic chicken-and-egg

TABLE 5.4. CHEMISTRY SPECIFICATIONS FOR MIM MATERIALS

common name nominal composition (w/o) maximum allowed
impurities, %
aluminum 201 Al-4.6 Cu-0.35 Mg-0.35 Mn 1.0
copper Cu+Ag = 99.5 < 0.05
Fe-50Co* Fe-4850 Co-<2.5 V-<1 Si-<0.05 C <1.0
Fe-2Cu-0.8C Fe-1.52.5 Cu-0.60.9 C <1.0
Fe-2Ni Fe-1.52.5 Ni-<0.5 Mo-<0.1 C <2.0
Fe-2Ni-0.6C Fe-1.52.5 Ni-<0.5 Mo-<0.40.7 C <2.0
Fe-8Ni Fe-6.58.5 Ni-<0.5 Mo-<0.1 C <2.0
Fe-8Ni-0.6C Fe-6.58.5 Ni-<0.5 Mo-0.40.7 C <2.0
Fe-50Ni* Fe-49.550.5 Ni-<1 Si-<0.05 C <1.0
Fe-3Si* Fe-2.53.5 Si-<0.05 C-<0.02 O <1.0
Kovar (F15)* Fe-2831 Ni-1518 Co-<0.5 Mn-<0.2 Cr, Mo, Cu, Si <0.2
stainless 17-4 PH* Fe-15.517.5 Cr-35 Ni-35 Cu-<1 Mn-<1 Si-0.150.45 (Ta+Nb)-<0.07 C <1.0
stainless 304L Fe-1820 Cr-812 Ni-<2 Mn-<1 Si-<0.03 C <0.045 P, <0.03 S
stainless 316L* Fe-1618 Cr-1014 Ni-23 Mo-<2 Mn-<1 Si-<0.03 C-<0.3 O-<0.01 N <1.0
stainless 410 Fe-11.513 Cr-<1 Mn-<1 Si-<0.15 C <0.1
stainless 420* Fe-1214 Cr-1 Mn-1 Si->0.15 C < 0.03 S
stainless 430L* Fe-1618 Cr-<1.5 Mn-<1 Si-<0.05 C <1.0
stainless 440C Fe-1618 Cr-1 Mn-1 Si-0.75 Mo-11.2 C <0.03 S, <0.04 P
steel 2200* Fe-1.52.5 Ni-<1 Si-< 0.5 Mo-<0.1 C <1.0
steel 2700* Fe-6.58.5 Ni-<1 Si-<0.5 Mo-<0.1 C <1.0
steel 4140* Fe-0.91.2 Cr-<0.9 Mn-<0.4 Si-0.350.5 C <1.0
steel 4340 Fe-1.42 Ni-0.71.4 Cr-<0.8 Mn-0.20.3 Mo-<0.35 Si-0.350.5 C <1.0
steel 4605* Fe-1.52.5 Ni-<1 Si-0.20.5 Mo-0.40.6 C <1.0
steel 4620 Fe-1.652 Ni-0.450.7 Mn-0.20.3 Mo-0.170.22 C <1.0
Ti CP grade* Ti > 99.0 <0.2 O
Ti-6-4* Ti-6 Al-4 V <0.2 O
*designates a composition covered by at least one trade or industry specification

74
 Table of Contents
question arises: are they low cost because they are used fre-
quently, or are they used frequently because of a lower cost?
In most cases the materials are used frequently and, as a result,
are lower in cost.
Besides raw material cost, the conversion of the material
into a powder increases cost, usually by about $1 to $2 per
kg. The preparation of feedstock further adds to the price.
High-volume mixing cuts cost, while frequent composition
changes require much equipment cleaning and adds consid-
erably to the cost. As consumption has increased, some of the
powders have undergone dramatic price decreases. To help
understand the advantage of steels and stainless steels, Figure
Figure 5.3. Relative sales of various PIM materials, showing a
dominance by metal powder injection molding at over 77%,
ceramics at 15%, carbides at 3%, and composites at 1%. About
4% of the global PIM industry sales are based on mixed materials
produced and combined in an assembly
Figure 5.4. Scatter plot of how material strength and cost per unit
volume compare for several of the materials employed in MIM
Chapter 5: Material Properties
5.4 plots a generalized clustering of strength vs. relative cost
per unit volume. Most MIM parts are volume constrained, so
obtaining a high strength at a low material cost makes ferrous
alloys most attractive.
Besides material cost, there are differences in processing
steps and expenses such as sintering that complicate cost con-
siderations. This chapter provides data on the materials and
their properties. This will help guide one toward the selection
of the right material for an application. Coupled with material
cost guides, the most effective material becomes evident by
balancing performance attributes and cost. As a leading engi-
neer phrased the situation: ... first you must meet the per-
formance criteria, then materialprocess selection is dictated
by issues of availability, capacity, reliability, quality, and
finally cost. The data presented here help sort through the
options, but in the end, economic considerations usually
dominate. Thus, the MIM industry relies heavily on a few
compositions;
two stainless steels (316L and 17-4 PH)
ironnickelcarbon steels with typically 2% Ni and 0.4%
C, maybe with Mo or Cr
two electronic alloys (Invar and Kovar)
two tungsten alloys (tungstennickelcopper and tung-
stencopper)
cemented carbides (WC-Co compositions).
MATERIALS SELECTION
A first consideration in materials selection comes by eval-
uating previous designs and failures. Historically, the major
reason for component field failures is improper materials se-
lection, including improper specification of the surface treat-
ment and heat treatment. Besides historical data, other
considerations include the function (property requirements)
and economic criteria.
Performance classifications are given in Table 5.5 based
on dominant concerns. Various combinations are possible.
Prior history with a material (both successes and failures) car-
ries the greatest weight in a new design. In situations where
there is no previous design, analysis of the required properties
is combined with handbook searches for candidate materials.
This leads to a preliminary selection to meet the design
constraints.
If the component size is fixed (constrained design), then
the selection is usually based on the lowest cost per compo-
nent. This means the cost per unit mass (most typically what
is quoted) needs to be multiplied by the density to give the
cost per unit volume. For example, a cemented carbide pow-
der is priced at $50/kg while a zirconia powder is $80/kg. The
density of the former is 14.6 g/cm3 and the latter is 6.0 g/cm3;
thus, the volume-based cost is $0.73/cm3 and $0.48/cm3, even
though the zirconia is higher in cost per unit mass. Since many
designs are volume constrained, the cost per component vol-
ume is a key economic factor in materials selection.
75
 Table of Contents
Chapter 5: Material Properties

shortlists of candidate materials that are ranked to identify best

TABLE 5.5. PERFORMANCE CLASSIFICATIONS candidates. If the material does not exist in a MIM powder
AND COMMON MATERIAL CHOICES form, then brace for an expensive solution.
Electrical Properties The exception to this scenario comes from an ability to mix
electrical conductivitycopper, tungstencopper powders to synthesize intermediate compositions. Some in-
electrical insulationalumina, silica, zirconia teresting examples have been developed this way for im-
glassmetal sealingKovar
Environmental Properties proved wear resistance. They include tool steels mixed with
biocompatibilitytitanium, tantalum, cobalt titanium carbide, tungsten carbide, ironmolybdenum boride,
chromium, stainless steel chromium boride, or alumina particles. Such ceramic-rein-
corrosion resistancetitanium, stainless steel, forced metallic structures possess high strength, hardness,
nickel, bronze wear resistance, and resistance to environmental attack.
oxidation resistancesilicon carbide, superalloy,
cobaltchromium
Inertial and Density Properties KEY CLASSIFICATIONS
high densitytungsten heavy alloy, tungsten Electrical and Magnetic
copper Many applications for ceramics rely on their electrical in-
low densityaluminum, titanium sulation properties. Alumina (Al2O3) is very low in electrical
Magnetic Properties
hard magnetic responseironneodymiumboron conductivity and has a long history in microelectronic and
soft magnetic responseiron, ironnickel, computer use. It is a favorite for high-performance insulator
ironcobalt, ironsilicon, nickel applications. Silica (SiO2) is also a suitable electrical insulator,
Mechanical Properties but usually the lowest cost combination is a mixture of alu-
high stiffnesscemented carbide, alumina, mina and silica.
tungsten, tungsten heavy alloy
high strengthheat treated steel, tool steel, Metallic products for electrical applications are character-
stainless steel ized by their resistivity (inverse of conductivity). Both resid-
high-temperature strengthsuperalloy, ual porosity and impurities lower conductivity, but the
molybdenum, tungsten high-temperature process associated with sintering removes
high toughnessmaraging steel, stainless steel, impurities that lower conductivity. For example, MIM Kovar
steel
high operating temperaturessuperalloy has a measured conductivity of near 2.1106 S/m (or a resis-
Optical Properties tivity of 47 -cm) while the handbook value is 2.0106 S/m
colorstainless steel, zirconia, silver, gold, platinum (49 -cm resistivity). Copper fabricated using MIM is al-
transparencyspinel, alumina most the same as cast material, near 50106 S/m. As tabulated
Thermal Properties in Table 5.6, the electrical properties of MIM products are
high thermal conductivitycopper, tungsten
copper, molybdenumcopper comparable to those attained via other fabrication routes.
low thermal conductivityalumina, silica, stainless The MIM process is frequently applied to the fabrication of
steel
high thermal-expansion coefficientaluminum
low thermal expansion coefficientInvar, tungsten TABLE 5.6. ELECTRICAL CONDUCTIVITY OF MIM
copper, molybdenumcopper MATERIALS
Wear Properties material conductivity, 106 S/m
high hardnesscemented carbide, tool steel
wear resistancecemented carbide, tool steel, copper (Cu) 50
cobaltchromium Kovar or F15 (Fe-29Ni-17Co) 2
Invar (Fe-36Ni) 1
iron (Fe) 510
Materials selection usually converges to a few candidates ironcobalt (Fe-50Co) 7
based on the answers to the following questions: ironnickel (Fe-2Ni) 47
What is the components function? ironnickel (Fe-8Ni) 2
What properties will ensure proper function? ironnickel (Fe-50Ni) 12
ironphosphorus (Fe-0.45P) 6
Is there prior history of success or failure? ironsilicon (Fe-2.5Si) 2
What service extremes must it satisfy? ironsilicon (Fe-6.5Si) 1
Are there constraints on the design or material? stainless 17-4 PH 1
Which properties can be adjusted? (Fe-16Cr-4Ni-4Cu-0.3Nb-0.8Si)
What are the quantitative criteria, including cost, for final steel 4600 (Fe-2Ni-0.5Mo) 24
tungstencopper (W-10Cu) 20
selection? tungstencopper (W-50Cu) 33
Tabulations with go/no-go criteria and weighting factors tungsten heavy alloy (W-7Ni-3Fe) 6
help sort through the candidates. From these questions arise

76
 Table of Contents
Chapter 5: Material Properties

soft magnetic components. Most are ferrous alloys, especially
pure iron, and alloys such as Fe 2Ni, Fe 3Si, Fe 0.45P, Fe 0.6P,
and Fe 50Ni. Of these, the Fe 50Ni alloy has a very attractive
combination of high magnetization and low coercive force.
Soft magnetic properties are determined by applying a field
to the material and measuring the response in terms of induced
magnetic behavior. A schematic plot of induced magnetization
vs. applied magnetic field is given in Figure 5.5. During this
test, the applied field is increased from zero and the induced
behavior is quantified at saturation (at, say, 500 Oe or 39,800
A/m), then with no applied field (remnant magnetization), and
finally by measuring the applied reverse field needed to re-
move magnetization (coercive force). The rate of magnetiza-
tion is quantified by the slope of the induced-vs.-applied
Figure 5.5. Magnetization response to an applied magnetic field
for a MIM soft magnetic material. The test starts with a demagnet-
ized sample and progressively increases the applied field, then
decreases the field through the point of reversal. Characteristics of
the material are extracted from this curve during initial magnetiza-
tion (permeability), at saturation, with no applied field (remnant),
and with sufficient field to fully remove the initial magnetization
(coercive force)
magnetization field from the origin to the first tangent with
the magnetization curve, giving the maximum permeability.
Table 5.7 lists the soft magnetic properties. Magnetic per-
formance depends on details of the fabrication cycle resulting
in vendor differences. Generally MIM products meet or ex-
ceed the handbook magnetic properties, since the process
gives an annealed (strain-free) microstructure and the sinter-
ing cycle evaporates volatile impurities that otherwise degrade
the magnetic properties in wrought products.
Corrosion and Biocompatibility
The resistance to environmental attack depends on several
factors, including the test fluid, and details such as aeration,
acidity, halide ion concentration, oxygen content, and contact
with other materials. The corrosion, oxidation, or biocompat-
ibility of a sintered material is caught up in these details and
further depends on the sintered microstructure, heat treatment,
test conditions, and even the measure used to quantify corro-
sion resistance. As a consequence, for any application it is best
to subject representative test samples to the target environment
to assure proper behavior.
MIM components are used as human implants and have
been tested, but not used, for replacement knees and hips.
Antibacterial response has been quantified and one important
finding is that the addition of 3 w/o copper to MIM 304L
stainless steel gives much improved resistance to bacterial
growth. This is one situation where MIM allows modification
of the chemistry to customize the material for the application.
Several biocompatible materials have passed implant testing,
including zirconia, tantalum, titanium and two titanium alloys
(Ti-6Al-4V, Ti-6Al-7Nb), hydroxyapatite, and composites of
titanium and hydroxyapatite. Titanium fabricated by MIM ex-
hibits resistance to stress-corrosion cracking that would nor-
mally attack a wrought stainless steel. Further, MIM 316L

TABLE 5.7. ROOM-TEMPERATURE MAGNETIC PROPERTIES FOR MIM PRODUCTS

maximum remnant coercive maximum
composition magnetization T magnetization T force A/m permeability
Kovar or F15 (Fe-29Ni-17Co) 1.6 0.3 95 1,400
iron (Fe) 1.21.6 1.01.3 120540 4,000
ironcobalt (Fe-36Co) 1.82.3 0.9 200320 1,7001,900
ironcobalt (Fe-50Co-2V) 1.52.2 1.31.4 120140 3,3009,000
ironmolybdenum (Fe-5Mo) 1.6 1.3 145 2,900
ironnickel (Fe-2Ni) 0.81.5 0.81.5 80200 2,0003,700
ironnickel (Fe-8Ni) 0.9 0.5 200
ironnickel (Fe-50Ni) 0.61.6 0.83.0 10120 17,00047,500
ironphosphorous (Fe-0.45P) 1.4 1.2 80
ironsilicon (Fe-3Si) 1.42.0 1.11.3 5580 6,0008,500
ironsilicon (Fe-6.5Si) 1.3 1.2 24 4,000
Permaloy (Ni-17Fe-2Mo) 1.5 0.5 85 77,000
Sendust (Fe-10Si-5Al) 1.01.4 105 15,800
stainless 17-4 PH (Fe-16Cr-4Ni-4Cu) 0.61.4 0.30.6 1,4003,200 40200
stainless 430L (Fe-17Cr-1Mn) 1.12.2 0.6 140155 1,500
steel 2200 (Fe-2Ni-0.5Mo) 1.42.0 0.8 85120 2,0003,700

77
 Table of Contents
Chapter 5: Material Properties

stainless material has been qualified for implant use. Gener- resistant than wrought stainless steel in both solutions.
ally tests in immersion situations show much benefit; for ex- There are recognized MIM vendor effects on corrosion.
ample, artificial saliva testing of a nickel-free MIM stainless This was demonstrated in exhaustive testing of MIM 316L
steel for dental use shows a 100-fold improvement over stainless steel from various vendors, showing from half to
wrought stainless steels. twice the corrosion rate vs. wrought. For example, in salt-
Evaluation tests are widely varied and range from short im- spray testing (ASTM G85) the wrought sample corroded at
mersion in chlorine bleach, to salt-spray tests lasting weeks, 0.0019 mm per year (0.075 mil per year), while the MIM sam-
and up to two months immersion in acids. From the metals, ples ranged from 0.0009 to 0.0036 mm per year (0.035 to 0.14
stainless steels and titanium are usually selected for corrosion mil per year). Hydrochloric acid tests (as per ASTM G31) re-
resistance, although a few applications exist for pure nickel. sulted in the MIM materials being attacked at 0.35 mm per
Impurities degrade corrosion resistance, but sintering tends to year (14 mil per year), which was nearly half the 0.67 mm per
evaporate impurities; hence, MIM corrosion properties often year (26 mil per year) for wrought 316L. Ferric chloride im-
are better than observed with alternative processing routes. mersion (ASTM G48) proved the most aggressive test, and in
The major effort on corrosion testing of MIM products has that test some of the MIM 316L samples gave up to 10 times
been directed at stainless steels. Stainless steels via MIM the corrosion rate of wrought. When all of the test results were
come as austenitic, ferriticmartensitic, or precipitation-hard-
enable grades. They contain at least 12% chromium. The 300-
grade austenitic stainless steels are used for consumer items
(the classic is 18-8 stainless used in cutlery with 18%
chromium and 8% nickel) due to good corrosion resistance,
excellent toughness, and nonmagnetic characteristics. Of all
stainless steels, the precipitation-hardened stainless grade 17-
4 PH (AISI 630) is the most widely used and tested in MIM.
Its mechanical properties can be adjusted via post-sintering
heat treatments. This allows its use in products ranging from
magnetic sensors to surgical tools. However, the low
chromium content does not fully prevent corrosion, as evident
in Figure 5.6, which shows dissolution on the surface of a
MIM sample after a week of immersion in nitric acid. The
400-grade stainless steels have lower alloying levels, but can
be fabricated to high-strength levels, making them useful for
components that include automotive turbochargers and knife Figure 5.6. Corrosion on the surface of a MIM 17-4 PH stainless
steel after immersion for a week in 20% nitric acid, showing some
blades. The 400-grade stainless steels generally have the low- localized surface attack
est corrosion resistance of the MIM stainless steels.
Corrosion in stainless steels is measured using salt-spray
and immersion tests. The immersion route examines discol-
oration and weight loss during prolonged exposure to solu-
tions that include artificial sweat, artificial saliva, boiling
water, copper sulfate, nitric acid, hydrochloric acid, sulfuric
acid, chlorine bleach, and ferric chloride solution. For some
immersion tests, electrochemical potentials are applied to de-
termine the passive film breakdown and onset of pitting.
When corrosion resistance is a major concern, 316L is the
most common MIM stainless steel, but if higher strength is
needed, then 17-4 PH is the typical choice. Tests on 17-4 PH
stainless fabricated by MIM illustrate the corrosion resistance.
Figure 5.7 plots the 200 h weight loss for two MIM materials
compared with wrought material tested in solutions of chlo-
rine bleach and ferric chloride. The corrosion test results are
given in units mass loss divided by sample surface area, in Figure 5.7. Corrosive weight loss after 200 h immersion in
chlorine bleach or ferric chloride for two different MIM processing
g/m2. Approximately 1 g/m2 converts to 0.0055 mm per year routes applied to 17-4 PH stainless steel. Compared with the
or 0.2 mil per year of penetration or dissolution attack. For wrought material, the MIM materials are very successful in
this testing, the MIM stainless steel products were more withstanding both corrosion environments

78
 Table of Contents
Chapter 5: Material Properties

equally weighted, some of the products ranked significantly W-Cu-Co, and W-Cu) by MIM have been applied to several
better than wrought. Differences correlated to final density applications requiring high inertiamilitary projectiles,
and impurity levelsa higher density and lower impurity shaped-charge liners, golf-club weights, medical radioisotope
level benefitted corrosion resistance. shipping-and-handling containers, birdshot, bowling ball
One important issue in MIM stainless steel corrosion traces weights, fishing weights, cellphone vibrator weights, disk-
to vacuum sintering, as used by some MIM firms. Prolonged drive counterbalance weights, and military fragmentation de-
heating in vacuum results in chromium evaporation from the vices. For higher tungsten contents the density ranges up to
component surface. It is chromium that provides corrosion re- 19 g/cm3. On the other hand, aluminum and titanium are used
sistance, so its loss (it has the highest vapor pressure of the in applications requiring low densities.
iron, nickel, molybdenum, and chromium making up 316L) A plot of elastic modulus vs. density is given in Figure 5.8
is detrimental to corrosion. Also, any oxygen contamination
(which is not specified in the 316L composition) leads to a
significant loss of corrosion resistance. Chromium forms a TABLE 5.8. DENSITY AND ELASTIC MODULUS FOR
MIM MATERIALS
compound with oxygen on the component surface, removing
it from its passivating role. Tests for stress corrosion found no elastic
significant difference from wrought materials. Thus, MIM material density modulus
g/cm3 GPa
materials have the potential to perform superiorly to wrought
material in corrosive environments. cemented carbide (WC-10Co) 14.2 593
Oxidation of MIM materials is no different from the cobaltchromium or F75 7.9 240
(Co-28Cr-4W-3Ni-1C)
wrought counterparts, so long as the sintered density is over copper (Cu) 8.5 130
95% of theoretical (to avoid oxidation through pores). In the copper steel (Fe-2Cu-0.8C) 7.7 190
section dealing with mechanical properties, recommended Hastelloy X (Ni-Cr-Fe-Mo-Co) 8.3 210
maximum-use temperatures are given for several materials. Invar (Fe-36Ni) 8.0 205
With respect to oxidation resistance at 500C (932F), there Kovar (F15) (Fe-29Ni-17Co) 7.8 117
Mo-15Cu 10.0 280
is a wide offering, but also large price difference. For many stainless 17-4 PH 7.8 195
of the metallic species, oxidation behavior is unaffected by (Fe-16Cr-4Ni-4Cu-0.3Nb)
MIM. Materials for oxidation environments include silver al- stainless 316L (Fe-17Cr-12Ni-2Mo) 7.6 190
loys, titanium alloys, tungsten alloys, and nickel alloys. stainless 430L 7.5 190
Nickel-base superalloys are widely recognized for their high- (Fe-17Cr-1Mn-1Si-1Ni)
steel 2200 (Fe-2Ni-0.5Mo-<0.1C) 7.6 190
temperature-oxidation resistance. Ferritic stainless steels are steel 2700 (Fe-7Ni-0.5Mo-<0.1C) 7.6 190
lower in cost with somewhat reduced oxidation resistance, but steel 4650 (Fe-2Ni-0.4Mo-0.5C) 7.7 195
are commonly used in automotive applications. So, from a titanium 6-4 (Ti-6Al-4V) 4.4 110
pure performance consideration, stainless steels provide the tungstencopper (W-20Cu) 16.0 290
best performance per unit cost. tungsten heavy alloy (W-5Ni-2Fe) 17.7 380
tungsten heavy alloy (W-6Ni-4Cu) 17.0 280
tungsten heavy alloy (W-7Ni-3Fe) 17.0 360
Inertial Properties
Materials fabricated by MIM exhibit physical properties
such as density, elastic modulus, heat capacity, and other at-
tributes that generally are close to values listed engineering
handbooks. Since the systems are sintered to nearly full den-
sity there is no effect from the processing route. For inertial
devices, such as counterbalance weights, sporting devices, and
vibration weights, the density and elastic modulus of several
compositions are given in Table 5.8. For comparison the hand-
book values for the same compositions are also listed, where
differences between the two are generally within the test error.
In several cases the Poissons ratio was measured along
with the elastic modulus, with the finding that the value is usu-
ally the same as given in the handbooks. This makes sense,
since the elasticity and Poissons ratio both measure the
atomic spacing dilation with stress, and that is unaffected by
the injection molding route. Figure 5.8. Scatter plot of elastic modulus vs. density for a variety
Tungsten heavy alloys (W-Ni-Cu, W-Ni-Fe, W-Ni-Co, of powder injection molded materials

79
 Table of Contents
Chapter 5: Material Properties

TABLE 5.9. ROOM TEMPERATURE TENSILE PROPERTIES OF MIM ALLOYS

(HT = heat treated, HIP = hot isostatically pressed, density is given as a percentage of the alloy theoretical density)
yield tensile fracture
density strength strength elongation hardness
material (designation) composition, w/o % MPa MPa %
aluminum Al-1Sn 95 100 150 25 24 VHN
aluminum 6061 + 2Sn (HT) Al-2Sn-1Mg-0.6Si-0.3Cu-0. 97 75 160 9 90 VHN
15Mn-<0.7Fe
cobaltchromium F75 Co-28Cr-4W-3Ni-1C 99 550 880 4 25 HRC
cobaltchromium F75 (HIP) Co-28Cr-4W-3Ni-1C 100 560 1,010 30 25 HRC
copper Cu 93 30 145 23 43 VHN
Hastelloy X (HT) Ni-22Cr-18.5Fe-9Mo-1.5Co-0.6W 98 303 675 74 30 HRC
HB alloy Ni-28Mo-5Fe 97 1,230 1,230 0 32 HRC
Inconel 718 (HIP, HT) Ni-19Cr-18Fe-5Nb-3Mo-1Ti-0.4Al 100 1,055 1,380 29 42 HRC
Invar Fe-36Ni 98 240 425 40 65 HRB
Invar, modified Fe-32Ni-5Co 96 285 440 40 65 HRB
iron Fe 96 100 230 40 65 VHN
iron aluminide Fe-12Al 96 385 580 2
ironcobalt Fe-49Co-2.5V-1Si 97 140 205 1 80 HRB
ironmolybdenum Fe-5Mo 98 210 410 34 66 HRB
ironnickel Fe-2Ni 96 190 345 30 55 HRB
ironnickel Fe-8Ni 95 310 430 21 80 HRB
ironnickel Fe-42Ni 99 250 490 43 59 HRB
ironnickel Fe-50Ni 96 170 420 20 50 HRB
ironphosphorus Fe-0.6 P 99 260 280 2 80 HRB
ironsilicon Fe-3Si 99 390 530 24 80 HRB
ironsilicon Fe-6.5Si 99 375 375 0 37 HRC
Kovar (F15) Fe-29Ni-17Co 98 300 460 25 65 HRB
maraging steel (HT) Fe-18Ni-9Co-5Mo-0.5Ti-0.1Si 96 1,600 1,640 2 47 HRC
nickeliron Ni-20Fe 91 470 31 53 HRB
niobium superalloy Nb-10W-10Ta 98 315 440 25 20 HRC
Nitinol NiTi 97 500 670 4
spinodal Cu-5Ni-8Sn 98 310 470 6
stainless 17-4 PH Fe-16Cr-4Ni-4Cu-0.3Nb-0.8Si 96 750 900 10 25 HRC
stainless 17-4 PH (HT) Fe-16Cr-4Ni-4Cu-0.3Nb-0.8Si 96 1,090 1,185 6 35 HRC
stainless 17-4 PH (HIP) Fe-16Cr-4Ni-4Cu-0.3Nb-0.8Si 100 1,103 1,137 13 38 HRC
stainless 304L Fe-18Cr-8Ni 97 240 480 35 85 HRB
stainless 316L Fe-17Cr-12Ni-2Mo-2Mn 96 175 520 50 67 HRB
stainless 316L duplex Fe-21Cr-9Ni-3Mo-2Mn 95 230 540 43 80 HRB
stainless 318 Fe-22Cr-6Ni-2Mn-1Si 97 230 590 29 89 HRB
stainless 410 (HT) Fe-11Cr-0.5C 95 1,240 1,520 5 40 HRC
stainless 410L (HT) Fe-11Cr-0.1C 95 410 650 5 20 HRC
stainless 414L Fe-13Cr-3Ni 97 980 1,360 5 40 HRC
stainless 420 (HT) Fe-13Cr-1Mn-1Si 96 1,200 1,380 1 44 HRC
stainless 430L Fe-17Cr-1Mn-1Si-1Ni-<0.05C 93 230 390 25 68 HRB
stainless 440C Fe-17Cr-1Ni-1Mn-1C 96 410 620 2 43 HRC
stainless 440C (HT) Fe-17Cr-1Mn-1C 98 1,560 1,600 <1 58 HRC
stainless PANACEA Fe-17Cr-10Mn-3Mo-1N 99 670 960 35 25 HRC
steel 1010 Fe-0.1C 97 175 400 19 30 HRB
steel 1020 Fe-0.2C 96 185 380 23 67 HRB
steel 1060 Fe-0.6C 97 260 580 25 80 HRB
steel 2200 Fe-2Ni-<0.5Mo-<0.1C 96 125 290 40 45 HRB
steel 2700 Fe-7Ni-<0.5Mo-<0.1C 95 225 415 26 69 HRB
steel 4140 Fe-1Cr-0.4C 97 390 580 15 18 HRC
steel 4140 (HT) Fe-1Cr-0.4C 93 1,240 1,650 2 46 HRC
steel 4340 Fe-2Cr-1Ni-1Mn-0.4C 96 480 620 6 20 HRC
steel 4340 (HT) Fe-2Cr-1Ni-1Mn-0.4C 96 1,400 1,600 2 48 HRC
steel 4605 (HT) Fe-2Ni-0.4Mo-0.5C 96 205 440 15 62 HRB
steel 4605 (HT) Fe-2Ni-0.4Mo-0.5C 96 1,480 1,655 2 48 HRC
steel 4640 (HT) Fe-2Ni-1Mo-0.4C 97 1,400 2,000 3 30 HRC

80
 Table of Contents
Chapter 5: Material Properties

TABLE 5.9. ROOM TEMPERATURE TENSILE PROPERTIES OF MIM ALLOYS

(HT = heat treated, HIP = hot isostatically pressed, density is given as a percentage of the alloy theoretical density)
(continued)
yield tensile fracture
density strength strength elongation hardness
material (designation) composition, w/o % MPa MPa %
steel ironchromium Fe-1Cr-0.5C 94 600 10 90 HRB
steel ironcopper Fe-2Cu-0.8C 95 700 10 92 HRB
steel ironnickel (HT) Fe-2Ni-0.5C 94 1,230 1,230 1 45 HRC
steel ironnickel (HT) Fe-2Ni-0.5Mo-0.6C 96 1,000 1,200 2 55 HRC
steel ironnickel Fe-2Ni-0.9C 96 450 650 9 90 HRB
steel ironnickel (HT) Fe-7Ni-0.5C 95 1,420 1,460 1 46 HRC
steel ironnickel Fe-8Ni-0.5C 97 260 410 25 75 HRB
steel ironnickel (HT) Fe-8Ni-0.6C 97 1,100 1,300 2 50 HRC
titanium Ti 98 500 620 22 95 HRB
titanium (HIP) Ti 100 700 800 25 195 VHN
titanium -5-2.5 Ti-5Al-2.5Fe 95 960 4 39 HRC
titanium -6-4 Ti-6Al-4V 98 800 900 17 35 HRC
titanium -6-7 Ti-6Al-7Nb 97 710 800 11 270 VHN
tool steel M2 (HT) Fe-6W-5Mo-4Cr-2V-1C 99 1,000 1,100 1 62 HRC
tool steel SKH10 Fe-15W-2Mo-5V-8Co-4Cr-0.7C 96 3,200 3,200 0 70 HRC
tungstencopper W-10Cu 98 530 540 1 280 VHN
tungsten heavy alloy W-8Mo-8Ni-2Fe 100 1,115 20 440 VHN
tungsten heavy alloy W-5Ni-2Cu 98 900 1,050 10 35 HRC
tungsten heavy alloy W-4Ni-1Fe 99 300 500 15 30 HRC
tungsten heavy alloy (HT) W-4Ni-1Fe 99 650 1,000 20 50 HRA
tungsten heavy alloy W-5Ni-2Fe 100 660 930 30 25 HRC
tungsten heavy alloy W-6Ni-2Fe 98 500 800 20 25 HRC
Udimet 700 Ni-18Co-15Cr-5Mo-4Al-3Ti 100 910 1,340 14 400 VHN

to demonstrate the range of common MIM materials,
illustrating how the ceramics and cemented carbides generally
are much stiffer than the steel, stainless steel, titanium, copper,
and other MIM metals. A concentration occurs near 8 g/cm3
and 200 GPa (29 Mpsi) due to the large number of ferrous al-
loys produced by MIM. Note the very high elastic modulus
possible with cemented carbides at a density near 14 g/cm3.
Structural Properties
Structural properties are associated with withstanding
deformation when loaded. Besides hardness, the most typical
attribute is derived from a strength test, with tension being
typical for a metal and compression being typical for a
ceramic. In rapid loading, the resistant to crack propagation
is a major concern, and that is measured by toughness or fac-
ture toughness. For cyclic loading, attention turns to the fa-
tigue strength, which is always much lower than the tensile
strength. The mechanical properties attained in MIM are gen-
erally close to those found in handbooks.
Tensile Properties of Metallic Materials
A tension test pulls the material to fracture and, besides the
elastic behavior, it gives strength and ductility. Some typical
room-temperature tensile properties are compiled in Table 5.9.
The typical specimen has a rectangular or circular cross
section 46 mm (0.20.25 in.) across, and the standard gauge
length is 2538 mm (11.5 in.). The samples are pulled slowly
to avoid adiabatic heating. Ductile metals are characterized
by yield strength, ultimate tensile strength, fracture elonga-
tion, hardness, and reduction in cross-sectional area. Included
in the table are the densities as a percent of the theoretical for
the composition. Heat treatments provide a means to adjust
the properties, and hot isostatic pressing is occasionally used
to reach full density as noted by HT or HIP in the table. Con-
siderable variation is possible in these cycles, so only one ex-
ample is listed, typically the cycle giving the highest hardness.
When post-sintering heat treatments are used, the tensile
strength increases, typically resulting in a lower ductility. In
most cases the MIM products are fairly close to the strength
and hardness of the wrought counterparts, and in some cases
are stronger. For example, the tensile properties of wrought
and MIM Inconel 718 compare as follows:
wrought typical yield strength = 760 MPa (110 ksi)
MIM typical yield strength = 1,055 MPa (153 ksi)
Likewise the MIM ultimate tensile strength and elongation
to fracture were attractive when compared to wrought mate-
rials. Indeed, the MIM elongation was nearly 6-fold the spec-
ified minimum for wrought material.
As another comparison, note that the average handbook
properties for Invar are a yield strength 240315 MPa

81
 Table of Contents
Chapter 5: Material Properties

(MIM gives a yield strength near 240 MPa or 35 ksi), tensile

strength 445555 MPa (MIM is slightly lower with a typical
tensile strength near 425 MPa or 62 ksi), and 30% 45% elon-
gation (MIM materials are often in the 40% range). Thus, the
two materials are comparable with a small property decrement
for MIM from residual porosity. Another factor is the grain
size, since grain growth occurs during sintering, and larger
grains in the MIM product give a lower strength.
With respect to property scatter, for ductile systems the
strength measured over repeat samples typically has a stan-
dard deviation of less than 20 MPa (3 ksi) and the elonga-
tion has a standard deviation of approximately 1%. When
different operations are compared, most properties are similar,
but a few, such as impact toughness, vary considerably.
Between MIM vendors there is little difference in hardness,
maybe a small difference in tensile strength, but often a large Figure 5.9. Plot of tensile strength and fracture elongation for
difference in ductility. Likewise, for ceramic injection molded MIM 316L stainless steel as reported by several vendors,
illustrating some of the processing cycle effects on properties
materials the hardness should be similar between vendors,
with some difference in strength, but potentially a large dif-
ference in fracture toughness and Weibull modulus. Since
there is no standard MIM process, the user needs to verify
properties. Figures 5.9 and 5.10 illustrate a typical property
variation. These are scatter plots of vendor-reported tensile
strength and elongation to fracture. The first set is for 316L
stainless steel, which is somewhat sensitive to porosity, grain
size, and residual impurities. Hence, depending on the pro-
cessing atmosphere and final density, there will be property
differences. The second plot, in Figure 5.10, is for 17-4 PH
stainless steel. It is one of the most popular alloys, yet exhibits
a small property variation between vendors.
Characteristically, MIM delivers tensile properties compa-
rable to those found in handbooks when the composition and
heat treatment are the same. Although there is a variation, stan-
dard properties can be used for design purposes and the final
Figure 5.10. Tensile strength and fracture elongation as reported
specification should state the minimum acceptable properties. by several MIM vendors of 17-4 PH stainless steel, again
reflecting variations in processing and microstructure
Rupture Strength of Ceramic Materials
Related to MIM, ceramics formed by CIM are well docu-
mented for mechanical properties. For ceramics, the typical
compression fracture test gives the rupture strength. Trans-
verse rupture is the most common test for measuring strength
in brittle materials. The test is shown in Figure 5.11 and typ-
ically the sample thickness is half the width, and the width is
half the span between the two lower supports (often it is 6 mm
thick, 12 mm wide, and 24 mm lower span giving final length
of 32 mm to allow overlap0.25 by 0.5 by 1.25 in.). This flat
test sample is broken by three-point bending. Figure 5.12
shows molded tensile and transverse rupture bars with runner
and sprue attached. An alternative to the transverse rupture
tests is to form a flat disk and to compress it on edge to failure.
Figure 5.11. Transverse rupture strength testing relies on a flat
This is often called the Brazilian test. sample, often in the 46 mm thickness range. The sample geometry
Injection molded ceramics are comparable to handbook requires the test span to be twice the width and the width to be
values with respect to mechanical properties. In most instances twice the thickness (dimensional ratio of 421)

82
 Table of Contents
Chapter 5: Material Properties

ture toughness) values of 10 and 14 MPa-m1/2, respectively.

Handbook values are similar, but can range up to 20% higher
for these two compositions.
Almost all MIM materials exhibit strain rate hardening
the faster they are strained or tested, the stronger the material.
Testing on 95% dense Fe-2Ni shows the yield strength in-
creases from 128 MPa at a testing strain rate of 0.001 per sec-
ond up to 867 MPa at 8,000 per second strain rate. Also a
small grain size gives a stronger material, a factor that goes
beyond the typical composition specification.
Rather than perform tests at various strain rates, it is more
Figure 5.12. Molded test bar clusters that includes a flat tensile
typical to examine the fracture energy in an impact test. For
sample and a transverse rupture sample metals the impact toughness is usually measured using a
notched Charpy bar. Figure 5.13 illustrates both the notched
the same powders are used. For example, tungsten carbides
deliver hardness in the range 8992 HRA, comparable to the
expectation for the same compositions and grain sizes using
die compaction. These carbides deliver transverse rupture
strengths in the 2,0002,600 MPa range (290377 ksi).
Table 5.10 collects the transverse rupture strength data for
several ceramics. Here failure occurs without deformation, so
the tabulation gives the average fracture strength and, where
reported, the Weibull modulus. The Weibull modulus meas-
ures the distribution in fracture conditions. A high Weibull
modulus is desirable. For routine ceramics the Weibull mod-
ulus is between 8 and 14, but with special effort values of 20
and even 35 have been achieved using injection molding.
Figure 5.13. Notched and unnotched Charpy impact bars along
Impact, Fracture, and Fatigue Properties
with the impact hammer device used to measure fracture energy.
A few materials have been tested for fracture toughness, Most of the metallurgical community uses 10 mm by 10 mm cross
and in those cases the MIM values are similar to handbook section with a 45 V notch, but some MIM firms report unnotched
values; for example, WC-9Co and WC-12Co give KIc (frac- data, and in some cases the sample might be half sized as well

TABLE 5.10. ROOM-TEMPERATURE RUPTURE STRENGTH PROPERTIES OF INJECTION MOLDED CERAMICS

AND CARBIDES (compositions in w/o)
density strength Weibull hardness
material % MPa modulus VHN
96% alumina (Al2O3-4SiO2) 98 275 1,7501,900
98% alumina (Al2O3-2SiO2) 98 275
99% alumina (Al2O3) 98 300450 9 1,2002,000
99.9% alumina (Al2O3) 100 550 1,2001,400
aluminatitanium carbide (Al2O3-30TiC) (HIP) 100 700 2,000
aluminazirconia (Al2O3-20ZrO2) 97 400600 1,800
cemented carbide (WC-8Co) 100 2,000 1,500
cemented carbide (WC-10Co) 100 1,5002,600 12 1,3001,700
cemented carbide (WC-7Co-1TaC) 100 2,1002,200 1,700
cemented carbide (WC-7Ni) 100 2,000 10 1,900
porcelain 95 50
SiAlON (Si3N4-5Y2O3-5MgAl2O4) 98 370 12 1,700
silicon carbide (SiC) 98 350410 810 2,100
silicon nitride (Si3N4-8Y2O3) 98 350800 1535 1,600
silicon nitridesilicon carbide(Si3N4-10SiC) 98 900
zirconia (ZrO2 -3Y2O3) 95 200800 12 1,200
zirconiaalumina (ZrO2-20Al2O3) 98 900

83
 Table of Contents
Chapter 5: Material Properties

and unnotched samples along with the test device. In tradi-
tional presssinter powder metallurgy the notched impact bars
are low in toughness. The stress concentration at the notch in-
duces crack propagation though the pores. Because presssin-
ter powder metallurgy relies on unnotched Charpy impact bars
to obtain higher values, MIM followed along. As a further
complication, some of the measurements are performed on
subsize (one-half or one-fourth cross-sectional area) un-
notched samples and sometimes the impact energy is then mul-
tiplied by 2 or 4, but this is invalid. Unfortunately, notches and
section thickness influence the fracture energy as illustrated
for 17-4 PH stainless steel in Table 5.11. Clearly, the reported
value needs to identify the sample size and notch condition.
Caution is required in comparing impact toughness values,
since most of the data are not comparable to standard impact
TABLE 5.11. SAMPLE GEOMETRY EFFECT ON
CHARPY IMPACT FRACTURE ENERGY FOR MIM
17-4 PH STAINLESS STEEL, HEAT TREATED TO THE
H900 CONDITION
sample size notched, J unnotched, J
full size 2.3 1.5
(10 mm x 10 mm) (1.6 ft.lb.) (1.1 ft.lb.)
half size 17.9 14.2
(5 mm x 10 mm) (13 ft.lb.) (10 ft.lb.)
toughness data. In light of that caution, Table 5.12 summarizes
the impact properties reported to date, with notations on the
test conditions.
Based on impact toughness, low-carbon steels are effective
in rapid-strain-loading applications for strength levels in the
700 MPa (100 ksi) range. If corrosion and impact toughness
are concerns, then stainless steels are the most common ma-
terial choice, but at a higher cost.
Fatigue data for MIM products depend on the heat treat-
ment and surface finish. For example, 17-4 PH stainless steel
in the as-sintered condition has a fatigue strength of 280 MPa
(41 ksi), but after heat treatment almost doubles to reach 517
MPa (75 ksi). The tests on MIM products are summarized in
Table 5.13. Improved fatigue strength is possible by case hard-
ening or shot peening the component surface. In one compar-
ison, 4640 steel was fabricated from mixed elemental (iron,
nickel, molybdenum, and graphite) powders and also from
prealloyed powder. The mixed elemental powders gave a het-
erogeneous microstructure which was beneficial to fatigue.
The hardsoft structure interrupts fatigue-crack growth, re-
sulting in a higher endurance strength.
Generally, the properties in fatigue improve significantly as
pores are eliminated, as is evident by MIM cobaltchromium
subjected to a post-sintering HIP treatment. The notched fa-
tigue endurance limit at 107 cycles is 210 MPa (30 ksi) as com-
pared with 240260 MPa (3538 ksi) for cast products.

TABLE 5.12. CHARPY IMPACT TOUGHNESS PROPERTIES FOR MIM ALLOYS (see caution in text)
impact energy
alloy notes * nominal composition J (ft.lb.)
maraging steel U-notch Fe-18Ni-9Co-5Mo-0.5Ti-0.1Si 35 (25)
maraging steel U-notch Fe-18Ni-9Co-5Mo-0.5Ti-0.1Si 5 (4)
stainless 17-4 PH (HT) standard test Fe-16Cr-4Ni-4Cu-0.3Nb-0.8Si 1.5 (0.1)
stainless 17-4 PH (HT) subsize, notched Fe-16Cr-4Ni-4Cu-0.3Nb-0.8Si 14 (10)
stainless 17-4 PH (HT) subsize, no notch Fe-16Cr-4Ni-4Cu-0.3Nb-0.8Si 18 (13)
stainless 316L standard test Fe-17Cr-12Ni-2Mo-2Mn 30 (22)
stainless 316L subsize, notched Fe-17Cr-12Ni-2Mo-2Mn 10 (7)
stainless 316L subsize, notched Fe-17Cr-12Ni-2Mo-2Mn 27 (20)
stainless 316L subsize, no notch Fe-17Cr-12Ni-2Mo-2Mn 190 (140)
stainless 410 standard test Fe-11Cr-0.5C 44 (32)
stainless 420 subsize, no notch Fe-13Cr- Mn-1Si->0.15 C 40 (30)
stainless 430L subsize, no notch Fe-17Cr-12Ni-2Mo-2Mn 150 (111)
steel 2200 subsize, no notch Fe-2Ni-0.5Mo-<0.1C 135 (100)
steel 2700 standard test Fe-7Ni-0.5Mo-<0.1C 48 (36)
steel 2700 subsize, no notch Fe-7Ni-0.5Mo-<0.1C 175 (130)
steel 4140 U-notch Fe-1Cr-0.7Mn-0.3Mo-0.4C 17 (13)
steel 4140 (HT) subsize, no notch Fe-1Cr-0.7Mn-0.3Mo-0.4C 75 (56)
steel 4605 standard test Fe-2Ni-0.4Mo-0.5C 14 (10)
steel 4605 subsize, no notch Fe-2Ni-0.4Mo-0.5C 70 (50)
steel 4605 (HT) subsize, no notch Fe-2Ni-0.4Mo-0.5C 55 (40)
*standard Charpy test is a notched sample with cross section 10 mm by 10 mm with 2 mm deep 45 notch (0.4 in. by
0.4 in. with 0.078 in. notch); U-notch sample has 2 mm wide and 2 mm deep rounded notch in 10 mm by 10 mm
sample: subsize samples are reduced in cross section to 5 mm by 10 mm (0.2 in. by 0.4 in.) and may lack the stress
concentration notch

84
 Table of Contents
Chapter 5: Material Properties

TABLE 5.13. MIM ALLOY FATIGUE ENDURANCE

STRENGTHS FOR 50% SURVIVAL OF 10 MILLION
CYCLES
50%
composition survival stress
cobaltchromium or F75 420 MPa (61 ksi)
(Co-28Cr-4W-3Ni-1C) (HIP)
ironnickel (Fe-7Ni) 219 MPa (32 ksi)
stainless 17-4 PH (HT) 517 MPa (75 ksi)
(Fe-16Cr-4Ni-4Cu-0.3Nb-0.8Si)
stainless 17-4 PH (HIP and HT) 517 MPa (75 ksi)
(Fe-16Cr-4Ni-4Cu-0.3Nb-0.8Si)
stainless 316L 305 MPa (44 ksi)
(Fe-17Cr-12Ni-2Mo-2Mn)
steel 2700 (Fe-7Ni-0.5C) 237 MPa (34 ksi)
steel 4340 (Fe-2Cr-1Ni-1Mn-0.4C) 500 MPa (72 ksi)
steel 4640 (mixed powder) 575 MPa (83 ksi) Figure 5.15. Plot of the maximum recommended operating
(Fe-2Ni-1Mo-0.4C) temperature in air for some common materials fabricated by
steel 4640 (prealloyed powder) 475 MPa (69 ksi) powder injection molding
(Fe-2Ni-1Mo-0.4C)
superalloy IN 718 (HIP and HT) 452 MPa (80 ksi)
(Ni-19Cr-18Fe-5Nb-3Mo-1Ti-0.4Al) ductility as measured by the elongation to fracture with more
superalloy IN 718 385 MPa (55 ksi) than 25% elongation up to 900C. In a similar manner, hot
(Ni-19Cr-18Fe-5Nb-3Mo-1Ti-0.4Al) tensile tests on an iron aluminide (Fe with 1113 w/o Al) show
the strength at 600C is 60%90% of the room-temperature
Elevated-Temperature Properties strength, depending on composition. Alumina CIM products
Both automotive and jet-engine applications have stirred have been used at temperatures over 1,500C (2,732F) with-
interest in high-temperature uses for MIM materials, resulting out failure in NASA space processing with hold times up to
in a variety of property measurements. Tensile tests on MIM 10 h.
17-4 PH stainless steel give a 16% strength loss from room Products fabricated by MIM are used up to standard upper
temperature to 218C (425F) and about the same relative operating temperatures, which for metals is usually limited by
ductility loss. The scale of this property loss for the MIM oxidation or loss of hardness (especially for heat-treated
product was about the same as observed in wrought samples. steels). Figure 5.15 plots the general maximum operating tem-
The metal powder injection molded 316L stainless steel peratures for a few materials operating in air. The distinct ad-
undergoes a decline in strength as test temperature increases, vantage of ceramics over metals is very evident. For tungsten
as plotted in Figure 5.14. While the MIM 316L stainless is or molybdenum a much higher temperature is possible in an
losing strength, formally termed thermal softening, it retains inert or reducing atmosphere, but they are limited to 600C
(1,112F) in air. The highest temperature capabilities in MIM
products are with the oxide ceramics, such as silicon carbide,
alumina, and molybdenum disilicide (MoSi2).

Figure 5.14. Thermal softening data for MIM 316L stainless steel
showing a decrease in yield strength on heating. For comparison,
wrought 316L stainless steel has a yield strength of 250 MPa at
200C and that strength declines to about 100 MPa at 800C
Optical Properties
Optical properties of concern include surface finish and
color. Surface finish is dictated by the powder size: smaller
powders give smoother sintered surfaces. Most MIM products
have a dull matte finish after sintering, but can be polished to
a high luster using standard techniques. For applications sen-
sitive to external appearance (golf club heads, watchcases,
jewelry, eyeglass components, cellphone hinges, cosmetic
cases, computer logos, and camera pieces) this final polishing
step is a routine secondary operation. The surface roughness
after sintering averages near 0.8 m (32 in.). With small
powders this can be improved to 0.4 m (16 in.). Smoother
surfaces are obtained by post-sintering finishing steps, with
0.02 m (0.8 in.) being obtained on watchcases, fountain

85
 Table of Contents
Chapter 5: Material Properties
pens, and jewelry. Figure 5.16 is an example of a MIM stain-
less steel watchcase taken to a high polish after sintering.
For oxide ceramics, the normal opaque color is a white,
ivory, light yellow, or tan (depending on composition and im-
purities). Zirconia can be doped to create an array of brilliant
colors ranging from black to white, and including red, yellow,
green, and blue shades.
Porcelain fabricated by CIM is sintered to a white condi-
tion with a transparent glass glaze. For optical transmission
the ceramic must be sintered to optical transparency or semi-
transparency. This implies sintering to a pore-free condition
with low impurity levels. Alumina (Al2O3) containing small
concentrations of magnesia (MgO) has controlled grain
growth during sintering and is injection molded to form high-
intensity lighting components for sodiummercury vapor
lamps. Such a ceramic is shown in Figure 5.17, where letter-
ing is visible through the sintered tube. Other ceramics are
sintered to similar high-translucency levels, including spinel
(MgAl2O4). Hot isostatic pressing is often required to obtain
the pore-free density necessary to obtain translucency.
Other than intentionally colored ceramics, most materials
are unchanged from that color established by nature. Gold, sil-
ver, and platinum alloys give colors that are unaffected by the
Figure 5.16. Example of the final polish and surface luster possi-
ble on MIM stainless steel, in this case illustrated by a watchcase
86
fabrication process. Aluminum and titanium are colored using
either heating or electrochemical treatments. Steels are ther-
mally or chemically treated to form a black magnetite surface.
Tungsten carbide delivers a lustrous dark gray used for watch-
cases. Alumina (Al2O3) doped with titanium carbide (TiC)
gives a deep black color, while alumina doped with chromia
(Cr2O3) delivers a pink color and, when sintered to translu-
cency, gives a ruby red color. However, zirconia is the most
popular of the colored ceramics. Applications include luxury
pens, watchcases, buttons, and decorative hardware. Figure
5.18 is a picture of a luxury watch with a face and band formed
using over one thousand CIM black zirconia segments.
Translucent oxide ceramics are possible with excellent sin-
tering technologies. Typically a low-impurity, zero-porosity
condition is required. Other than lighting-lamp envelopes, in-
jection molding has not penetrated much of this market.
Figure 5.17. Example of a translucent sintered alumina as used in
high-temperature lighting
Figure 5.18. Luxury wristwatch formed using thousands of small
black CIM zirconia pieces
 Table of Contents
Chapter 5: Material Properties

Thermal Properties
Four thermal properties are of concernmelting range,
heat capacity, thermal-expansion coefficient, and thermal con-
ductivity. The latter two are important to heat-dissipation ap-
plications in microelectronic packaging, where a high thermal
conductivity is desired with controlled thermal expansion. The
thermal expansion needs to match with silicon or dielectric
alumina used as a circuit substrate.
Melting temperatures for MIM alloys are essentially the
same as for other fabrication routes. Hence, if the handbook
melting range for 316L stainless steel is 1,365C1,420C
(2,490F2,590F), then the MIM product will exhibit the
same range. Exceptions occur when sintering additives are
used to lower the onset of melting. For example, boron is
added to stainless steel to lower the sintering temperature, but
it also reduces the onset of melting to 1,230C (2,246F).
Heat capacity is the amount of energy stored in a material
as its temperature is increased. It relates input energy to tem-
perature change on a unit mass basis. The British thermal unit
(BTU) measured the relation between heat and 1F tempera-
ture change for one pound of water. Today, heat capacity is
measured using J/(kg K). There is no substantial difference
between MIM materials and handbook properties. For exam-
ple, MIM Kovar has a heat capacity near 500 J/(kg K) which
is near the handbook value of 460 J/(kg K).
Table 5.14 collects data on the thermal properties of heat-
dissipation MIM materials. The desire is to reduce the ther-
mal-expansion coefficient while maximizing the thermal
conductivity. Semiconductors dissipate heat and must be
mated with high-thermal-conductivity materials to operate
properly. However, the semiconductor has a low thermal-
expansion coefficient, which mandates the MIM material like-
wise have a thermal-expansion coefficient in the 5710-6/K
range. Note, thermal expansion is measured in 10-6/K (kelvin)
or parts per million per K, sometimes abbreviated as ppm/K,
where ppm stands for parts per million. If the thermal expan-
sion of the heat-dissipation material increases beyond this
range, then the assembly will fail due to thermal fatigue.
When power is applied, the device expands due to heating,
but contracts due to cooling when turned off. Onoff cycles
create differential strains that induce failure over time if the
thermal expansion coefficients are different. Thus, thermal-
management materials are constrained with respect to the
allowed thermal-expansion coefficient, but maximized with
respect to thermal conductivity.
Figure 5.19 is a scatter plot of the metal, ceramic, and com-
posite PIM materials associated with thermal management ap-
plications. Generally, the thermal conductivities of MIM
materials are comparable to other forming techniques. A high-
thermal-conductivity material with a desirable low thermal-
expansion coefficient is W-10Cu. The high density of a
tungsten system makes it best for stationary applications, such
as network server and base-station cellphone systems. The low
density of the aluminum nitride comes with a higher cost that
restricts its use to portable electronics. Tungstencopper is
Figure 5.19. Plot of thermal-management materials associated with
heat dissipation in microelectronics, showing the relative placement
of several options with respect to the thermal-expansion coefficient
and thermal conductivity. Because of silicon and the ceramic sub-
strates thermal expansion is constrained, while the highest possible
thermal conductivity is sought, leading to strong interest in tung-
stencopper composites. However, for low-density portable appli-
cations (such as in laptop computers or control circuits for hybrid
automobiles) aluminum nitride is selected in spite of a high cost

TABLE 5.14. THERMAL PROPERTIES OF MIM THERMAL MANAGEMENT MATERIALS

tensile elastic thermal thermal
density strength modulus conductivity expansion
material g/cm3 MPa GPa W/(m K) 10-6/K
copper (Cu) 8.9 100 130 250 13
Invar (Fe-36Ni) 8 300 205 20 5
Kovar or F15 (Fe-29Ni-17Co) 8.0 480 200 17 6.6
molybdenumcopper (Mo-15Cu) 10 210 280 170 7
molybdenumcopper (Mo-20Cu) 9.9 200 280 145 6.5
tungstencopper (W-10Cu) 17 500 340 209 6
tungstencopper (W-20Cu) 16 560 290 247 7
tungstencopper (W-30Cu) 14 420 260 260 11

87
 Table of Contents
Chapter 5: Material Properties
electrically conductive allowing its use for electromagnetic
shielding as well. For high-power semiconductors, these ma-
terials accommodate a 50% increase in operating power over
standard microelectronic packaging materials. Hence, MIM
microelectronic packages are used in radar, cellphone, fiber-
optic, and other high-power units.
Wear
The hard MIM materials exhibit wear resistance similar to
that observed for the same materials fabricated by alternative
routes. Consequently, injection molded cemented carbides are
employed for applications in oil-well drilling, cement-dis-
charge tubes, metal cutting and drilling, stone cutting, and
diesel-engine fuel injectors. Wear resistance is determined by
hardness, as well as application parameters that include sliding
velocity, contact forces, surface roughness, lubrication, tem-
perature, and operating atmosphere. The most attractive solu-
tions are cemented carbides, which are difficult to machine,
hence the advantage of MIM.
Figure 5.20 scatter plots the distribution of wear materials
in terms of hardness and density. At the low-density end, sil-
icon carbide is most attractive, while cemented carbides are
the best choices if density is not an issue. When cost is con-
sidered, silicon carbide is less attractive and WC-Co compo-
sitions by MIM are used.
One option for improved wear resistance and toughness is
to mix metal and ceramic powders to form a composite. Tita-
nium carbide in a steel matrix is an example of such a com-
posite. Other examples include the following:
a 30-fold reduction in dry abrasive wear using 3 v/o cal-
cium difluoride (CaF2) mixed with tool steel powder
a 20-fold reduction in abrasive-wear resistance using 3
w/o of titanium nitride (TiN) mixed with 440C stainless
steel powder
Figure 5.20. Scatter plot of wear materials showing hardness
(VHN = Vickers hardness number) vs. sintered density. The high
hardness of the cemented carbides and low cost make them most
attractive if density is not an issue. On the other hand, CIM
alumina provides an economical solution at lower densities
88
SINTERED MICROSTRUCTURES
Sintered MIM structures are close to full density and look
similar to that observed with other fabrication routes. The
most telling signs that a component was formed by powder
injection molding are from external blemishes, such as ejector
pin marks, gate marks, and parting lines.
Several final microstructures are given in Figures 5.21 to
5.24. These include the following:
Figure 5.21 is a cemented carbide hard material
(WC-10Co)
Figure 5.22 is a two-phase microstructure of MIM
titanium
Figure 5.23 is a 17-4 PH stainless steel formed from
water-atomized powder
Figure 5.24 is boride reinforced stainless steel cermet
A low residual porosity is characteristic of these micro-
structures. As noted above, if it is properly performed, there
is no feature at the microstructure level to indicate the MIM
process. Accordingly, microstructure preparation and etching
steps are the same as employed elsewhere. This contrasts with
the related technology of press-and-sinter powder metallurgy
which suffers from a high residual porosity that is easily dis-
turbed by standard microstructure preparation steps.
Figure 5.21. Sintered microstructure of an injection molded and
sintered WC-Co cemented carbide hard metal (photograph
courtesy of Mu-Jen Yang)
 Table of Contents
Figure 5.22. Microstructure of injection molded titanium after
sintering for two hours at 1,400C (2,552F) in vacuum, a
temperature that produces two crystal phases (photograph
courtesy of Anthony Griffo)
Figure 5.23. Microstructure of 17-4 PH stainless steel after
sintering in hydrogen at 1,365C using water-atomized powder.
The dark spherical pores are trapped steam from a high-
temperature reaction between oxygen (from water atomization)
and the hydrogen sintering atmosphere
Figure 5.24. Cermet microstructure formed by MIM based on a
hard chromium boride dispersed in a stainless steel matrix
(photograph courtesy of Anthony Griffo)
Chapter 5: Material Properties
COMPOSITE MATERIALS
One advantage of MIM is the ability to mix different pow-
ders to create composites. Insoluble phases are combined to
customize propertiesfor example, a soft jewelry alloy gives
improved wear resistance when a small quantity of oxide ce-
ramic is added in the powder stage. Other composites are de-
sired because of wear resistance and the general ability to
customize properties. A few specific examples are given
below of composites fabricated using injection molding, usu-
ally relying on mixed powders.
High-Elastic-Modulus Cermets
Following on the idea of adding a ceramic phase to im-
prove hardness, higher contents of hard phases increase the
elastic modulus. A well-known example comes from the com-
bination of tungsten carbide with cobalt, which is very stiff
when compared to cobalt640 GPa vs. 211 GPa (93 Mpsi vs.
31 Mpsi). Likewise, steels have been loaded with stiff phases
such as titanium boride (TiB2) to increase elastic modulus.
This boride is 2.5 times stiffer than steel, so a composite with
40 v/o boride is stiffer than steel by a factor of 1.6.
Heterogeneous Microstructures
Early in MIM it was common to mix elemental powders
to form alloys during sintering. Since diffusional homogeniza-
tion might not be complete, the sintered microstructure con-
sisted of pockets rich in each individual chemistry, giving
heterogeneous mechanical properties at the microstructure
level. Surprisingly, in some cases the bulk properties were im-
proved by the heterogeneity. Phase heterogeneity has been
used to develop harder, tougher, and more wear-resistant ma-
terials. Table 5.15 directly compares the properties of forged
material with the MIM homogeneous and heterogeneous ma-
terials, and further compares to traditional powder metallurgy
(press-and-sinter) steel. The heterogeneous MIM product is
nearly as strong as forged steel.
Another demonstration comes from Fe-2Ni-0.5Mo-0.4C
steel sintered to different levels of microstructure homogene-
ity. Data from fatigue testing show an impressive gain from
the heterogeneous microstructure:
fatigue endurance limit for homogeneous MIM product
= 475 MPa (69 ksi)
TABLE 5.15. COMPARATIVE TENSILE PROPERTIES
FOR DIFFERENT MIM STEEL MICROSTRUCTURES
tensile
density, strength, elongation,
condition g/cm3 MPa %
forged 7.8 2,115 (307 ksi) 11
MIM homogeneous 7.5 1,790 (260 ksi) 6
MIM heterogeneous 7.4 1,980 (287 ksi) 5
press-and-sinter 7.2 1,440 (209 ksi) 2
89
 Table of Contents
Chapter 5: Material Properties
fatigue-endurance limit for heterogeneous MIM product
= 575 MPa (83 ksi)
Accordingly, selective heterogeneities are used to improve
properties.
Atmosphere-Stabilized Compositions
During sintering the process atmosphere penetrates be-
tween the particles to alter the chemistry. Since adsorption or
de-adsorption are temperature dependent, composition control
during heating allows for selective changes in the material
properties.
There are strengthening benefits from nitrogen absorption
for steels. However, ferrous alloys normally do not take ad-
vantage of this process, since nitrogen is unstable in molten
metals. Nitrogen absorption peaks at intermediate tempera-
tures. For bulk components, this means that nitrided surfaces
are possible, but overall bulk strengthening is not realistic
the nitrogen diffuses only into the surface regions. In MIM
the opportunity is to add nitrogen during sintering. This is
used in MIM tool steels to form carbo-nitrides, giving a final
strength of 3,000 MPa (425 ksi) with 3% elongation.
Similar reactions are seen in nickel-free, nitrogen-strength-
ened stainless steel. Some humans are nickel-sensitive as ev-
ident by skin rashes around jewelry. Austenitic stainless steels
with 8%10% nickel (such as 304L and 316L) can induce
these reactions. Nitrogen is an austenite stabilizer that allows
removal of nickel in a corrosion-resistant composition. The
alloy has the desirable nonmagnetic character with the benefit
of no nickel, while exhibiting twice the strength and 20-fold
more corrosion resistance when compared to 316L stainless
steel. A direct comparison of properties is given in Table 5.16.
Obvious uses for this alloy are in watch, dental, and surgical
components where prolonged wet contact with skin would
lead to nickel release and an allergic response.
Some metals are hydride formers, including titanium, zir-
conium, tantalum, and niobium. This means they form com-
pounds with hydrogen, in the same way that oxygen forms a
compound with hydrogen (water). The compounds are very
different in properties from the metals and often are consid-
ered for hydrogen-storage media, since the hydridedehydride
TABLE 5.16. COMPARATIVE PROPERTIES OF
NITROGEN-STABILIZED MIM STAINLESS STEEL
AND WROUGHT 316L STAINLESS STEEL
nitrogen- wrought
property stabilized MIM 316L
nitrogen content, w/o 12 <0.05
density, g/cm3 7.6 8.0
hardness 30 HRC 68 HRB
yield strength, MPa (ksi) 670 (97) 170 (25)
ultimate strength, MPa (ksi) 960 (139) 480 (70)
elongation to fracture, % 35 45
90
reaction is reversible. Normally sintering of these metals must
avoid hydrogen, relying on inert gas or vacuum to avoid re-
actions. However, most hydrides are not stable at high tem-
peratures. Thus, high-temperature sintering in hydrogen is
possible without degradation of the metal. This facilitates the
use of hydrogen to remove oxygen from the component dur-
ing heating. Cooling in hydrogen leads to hydride formation
and a brittle material, but processing atmosphere changes are
used prior to cooling to avoid forming the hydride. In a novel
concept used for titanium, cooling is performed in nitrogen to
produce a gold-colored titanium nitride for watches, jewelry,
and decorative items. Similar uses of the process atmosphere
have been applied to cobaltchromium, aluminum nitride, and
silicon nitride compositions.
Biocompatible Materials
Hydroxyapatite is a biocompatible ceramic (calcium phos-
phatebased compound) with widely recognized compatibility
with bone. Unfortunately, hydroxyapatite is very weak. To pro-
vide MIM implants with tailored strength and biocompatibil-
ity, several ideas are based on mixed powders to form
composites. One composition is based on a mixture of
hydroxyapatite and titanium and this is accepted for implants,
such as posts for affixing replacement teeth. Proper combina-
tion of porosity and the two phases allows matching the
strength and elastic modulus of bone, nominally with 100 MPa
(14.5 ksi) strength and 20 GPa (2.9 Mpsi) elastic modulus.
Much current attention is on MIM tantalum or titanium to
form a porous structure for tissue ingrowth. Variants are used
for replacement knees, hips, and shoulders where tissue at-
tachment is desired. The MIM porous surface structure is at-
tached to a dense replacement component to provide pores for
tissue attachment.
Infiltrated Preforms
Ceramics are low-thermal-expansion materials, while met-
als are higher in thermal and electrical conductivities. Com-
posites are formed by first molding a porous ceramic skeleton
which is subsequently infiltrated with molten metal. The ce-
ramic skeleton provides the desired low thermal-expansion co-
efficient while the metal provides a high thermal conductivity.
Early commercial products relied on silicon carbide (SiC) with
copper or aluminum as the infiltrant. Two applications have
gained notice: electronic heat sinks and disk-brake calipers.
An advantage is that sintering does not have to densify the ce-
ramic powder, so larger powders are used in molding. A few
other demonstration systems are listed in Table 5.17.
Figure 5.25 is the cross section microstructure of a molded
Invar preform after infiltration with copper; this composite
targets applications requiring a low thermal-expansion coef-
ficient and high thermal conductivity.
Current efforts are looking to promote a reaction between
the infiltrant and the porous preform. One concept uses a dia-
 Table of Contents
Chapter 5: Material Properties

TABLE 5.17. COMPOSITES BASED ON MOLDED

PREFORMS FOLLOWED BY METAL INFILTRATION
composite sintered metal
preform infiltrant
Al2O3-Al Al2O3 Al
AlN-Cu AlN Cu
Invar-Ag Invar Ag
Invar-Cu Invar Cu
SiC-Al SiC Al
SiC-Si SiC Si
steelbronze Fe-0.8C bronze
W-Cu W Cu
WC-Cu WC Cu
Figure 5.26. Demonstration of applying MIM concepts to form a
bonded structure, in this case wood particles are mixed with
polymer to form a thermoplastic feedstock that is molded, but no
debinding or sintering is required

black to protect against oxidation and ultraviolet light are

Figure 5.25. Cross section near the top of a sintered porous Invar
(Fe-Ni alloy) infiltrated with copper to fill the pores without
inducing sintering shrinkage, giving a high thermal-conductivity
phase in a low thermal-expansion body
mond preform which is infiltrated with molten silicon. Reac-
tions at the diamondsilicon interface produce silicon carbide
(SiC) as a hard intermediate phase. Thus, the final product
consists of diamonds in shells of silicon carbide, bonded by
solidified silicon. A wide variety of compositions become pos-
sible, but so far only a few have reached the trial stage.
Bonded Materials
The injection molding of a glass or metal-particle-filled
polymer is common in plastics. Automotive tires with carbon
probably the most common filled polymers. MIM is a deriva-
tion of filled-polymer molding, where now the filler is that
target and the polymer is a transport vehicle. The bonded
composites build from MIM, for cases where no debinding or
sintering is required.
Examples include the iron powder bonded into magnetic
components. The structures are magnetic, but not electrically
conductive, useful attributes for automotive alternators since
eddy current losses are avoided, giving higher efficiencies.
Another is the amorphous Fe14Nd2B compound bonded to
form hard magnets. Other production materials include ce-
ramicpolymer systems used for bathroom fixtures, polymer-
bonded wood flour used to form toys, bronze bonded with
polymer for figurines and musical instruments, copperpoly-
mer systems for heat dissipation, and tungsten polymer mix-
tures for fishing weights and practice ammunition. Figure 5.26
is a picture of a molded wooden piece as a demonstration of
a bonded structure. These novel structures where shape com-
plexity is combined with non-plastic materials find a broad
array of applications.
Controlled-Porosity Structures
Filters are open-pore structures used to remove debris from
fluids. Other applications for controlled-porosity structures
are also used in fluid-control surfaces, sound absorbers, air
bearings, spargers, batteries, insulators, and capacitors. There
are two forms of porosity:
open porositymeaning gas or liquid can pass through
the body
closed porositymeaning the pores are not connected to
the free surfaces
The creation of pores is possible by two routes. The first
involves dissolution of nitrogen (or other soluble gas) into the
binder prior to molding. During molding dissolved gas nucle-
ates into bubbles that become stable pores, similar to how

91
 Table of Contents
Chapter 5: Material Properties

With respect to mechanical properties, several test bar

Figure 5.27. Foamed stainless steel shown in cross section, where
the amount and size of the internal dark pores are determined by
how much nitrogen is dissolved into the feedstock polymer prior to
molding. During cooling the nitrogen comes out of solution and
nucleates pores inside the body that are retained during sintering
geometries are used in MIM. The flat tensile bar pictured in
Figure 5.28 is easily formed by MIM using low-cost tooling,
so it is a favorite. After sintering, depending on the initial
solids loading and sintering shrinkage, the thickness ranges
from approximately 4 to 6 mm cross section and 20 to 30 mm
gauge length is common (about 0.25 in. thick and 1 in. gauge).
It is often molded in a cluster with other test samples, such as
a transverse rupture bar or impact bar or corrosion test coupon.
Figure 5.12 showed an example of a bending bar (useful for
transverse rupture strength, hardness, density, and corrosion
testing) formed at the same time as the tensile bar.
Another tensile test sample option is a round bar 38 to 90
mm (1.5 to 3.6 in.) long with a typical diameter 45 mm
(0.160.2 in.), shown in Figure 5.29 as it is being extracted
from the molding machine.
A hybrid tensile bar design is flat but rounded. It is shown
in Figure 5.30 after sintering and after bending. To avoid weld

styrofoam is formed. If the bubbles are large, they remain dur-

ing sintering. Usually the bubbles are in the 10 m size range
and the porosity is adjustable from 10% to 40%. However, the
pores are only present on the component interior, so the sur-
face is dense. Figure 5.27 is the microstructure after sintering
for a 316L stainless steel molded using the gas-bubble
approach to foam the product. A first application is in lighter-
weight stainless steel cosmetic cases and variants have been
explored for precious metal jewelry, where the amount of pre-
cious metal is reduced to make the same volume component, Figure 5.28. Flat tensile bar that is used for MIM strength tests
without the need to dilute the gold or silver level.
The other way to form pores is to include a pore former or
space holder in the feedstock. These are usually inert particles,
such as salt. The size and amount of inert particles determines
the final porosity and pore size. Most desirable are particles
that are easily decomposed during the sintering cycle; ammo-
nium carbonate is historically a favorite, although camphor,
stearates, and nylon have been used successfully. With high
concentrations of pore formers of the proper size, it is possible
to form highly distended structures. If polymer fibers are used,
such as chopped nylon fishing line, then pore alignment is
possible. Such alignment is useful since the pores are mono-
sized and fluid flow is directionally controlled.

TESTING STANDARDS
Industry standards for MIM have been formed by the re-
gional trade associations such as the European Powder Met-
allurgy Association (Shrewsbury, UK) and Metal Powder
Industries Federation (Princeton, NJ). The standards cover
testing procedures for the powders and sintered materials,
composition ranges for popular materials, and property levels Figure 5.29. Example of the molding operation for forming a
for sintered materials. totally round tensile sample

92
 Table of Contents
line defects that might cause premature fracture out of the
gauge length, the pin holes are often drilled after sintering.
Strength is based on yield and tensile values, recognizing
that designers focus on the yield strength since component de-
formation is usually not acceptable in service. Tensile testing
is performed at a slow strain rate to avoid adiabatic heating in
the manner diagramed in Figure 5.31. The elastic modulus is
estimated from the initial stressstrain response. Fracture
elongation, and the sample reduction in area at the point of
fracture, gives a guide to ductility.
Impact bars range from square cross sections 10 mm by 10
mm by 55 mm length (0.4 in. by 0.4 in. by 2.2 in.) to subsize
samples of one-half or one-fourth the cross-sectional area. In
some cases the samples lack the standard 2 mm deep 45
notch. Several variants are in use. Some of the bars are
notched while others are unnotched, as illustrated in Figure
5.13. This unnotched variant also can have 10 mm by 10 mm
or 5 mm by 10 mm cross sections. The ASTM impact standard
Figure 5.30. Popular MIM tensile bar that is a hybrid of the round
and flat geometries. It is for strength and ductility tests. Shown is a
sintered sample and one twisted to demonstrate ductility
Figure 5.31. Schematic illustrating how a tensile sample is
clamped and pulled to fracture. Instrumentation for load and
displacement during the test allow calculation of elastic modulus,
yield strength, tensile strength, facture strength (not normally
reported), reduction in area, and fracture elongation
Chapter 5: Material Properties
calls for 10 mm by 10 mm cross section with a notch. But
some data are for unnotched bars, and furthermore are from
subsize samples. This leads to serious confusion as illustrated
earlier in Table 5.11 for MIM 17-4 PH stainless steel. The
smaller sample size delivers the higher fracture energy both
with and without a notch.
Detailed instructions exist for magnetic property testing,
where the convention is to use a 25 oersted maximum applied
field (as per ASTM A 773). Other data might include surface
finish, microstructure (including phases and their distribu-
tions), grain size, porosity or density, hardness, wear resist-
ance, corrosion resistance, chemical composition, and various
other parameters.
Special concern is given to corrosion testing. Corrosion
testing can be simple using immersion tests. One corrosion
test applied to MIM stainless steels is to immerse a sample
5 mm by 10 mm by 55 mm (0.4 in. by 0.4 in. by 2.2 in.) in a
2% sulfuric acid solution at room temperature for 1,000 h.
However, it is faster to rely on a 6 min immersion in a copper
sulfate solution and to inspect for copper plating or other at-
tack (ASTM F 1089). Other tests for corrosion include boiling
water immersion for 30 to 180 min (ASTM F 1089). One very
effective test, commonly used but not standardized, is to im-
merse the sintered component in chlorine bleach and observe
for discoloration over one minute.
MATERIAL STANDARDS
MIM compositions are treated at several levels. The first
level deals with the nominal composition, generally expressed
by a designation such as 316L. At the next level are the de-
tailed chemical specifications for the major ingredients. It has
a nominal composition specification of 10%14% nickel,
16%18% chromium, 2%3% molybdenum, up to 2% man-
ganese, and up to 1% silicon. These allow for reasonable com-
position variations, often in the range of 1%2% on the
major alloying ingredients. The next level of specification
treats the impurities, and for 316L the key limitation is in the
carbon level, which is limited to less than 300 ppm (parts per
million) carbon.
Impurities are important to MIM, for example, as the car-
bon level decreases in 316L, the corrosion resistance im-
proves. Most of these specifications are borrowed from
wrought alloys. Unfortunately, wrought alloys do not antici-
pate some of the MIM options. If strength is a concern, then
just a specification of less than 300 ppm C for MIM 316L is
inadequate, since new impurities arise in MIM. Two important
cases are oxygen and nitrogen, both possible impurities from
sintering atmosphere contamination. Since the wrought ma-
terials are always low in N and O, these two impurities are ig-
nored in the specifications. For MIM this is an unsafe
assumption and a few horror stories result. In one case the in
specification stainless steel components rusted prior to ship-
ment. Thus, even the MIM industry compositional specifica-
93
 Table of Contents
Chapter 5: Material Properties
tions make assumptions that are not always valid. For exam-
ple, 17-4 PH stainless steel sintered in pure hydrogen atmos-
phere delivers a sintered tensile strength of 980 MPa, but that
strength is for no added nitrogen during sintering. If a small
level of nitrogen is in the sintering atmosphere, the sintered
strength is 10% higher, with a concomitant loss of ductility.
For many compositions there are no industry-accepted
standards. MIM component producers have written material
standards through regional trade associations. More rigorous
standards often come from the users based on application data.
This is especially true in automotive, medical, and microelec-
tronic industries. Otherwise, the material standards are derived
for other industries. Some relevant specifications for MIM
products are found in ASTM B 883 and MPIF 35.
In the ASTM and MPIF property specifications, both typ-
ical and minimum values are given. The typical values are
based on samples performed by several companies and ex-
press the average behavior. The minimum values are based on
the lowest values from the samples. Typical implies an indus-
try self-monitored average. Minimum standards anticipate
production values will be higher.
For MIM alloys, the composition specifications include the
following steels:
2200, Fe-based with 1.5%2.5% Ni, 0.5% max Mo,
0.1% max C
2700, Fe-based with 6.5%8.5% Ni, 0.5% max Mo,
0.1% max C
4140, Fe-based with 0.8%1.2% Cr, 0.2%0.3% Mo,
0.3%0.5% C, 1% max Mn
4605, Fe-based with 1.5%2.5% Ni, 0.2%0.5% Mo,
0.4%0.6% C
Soft magnetic alloys are as follows:
2200, Fe-based with 1.5%2.5 % Ni, 0.1% max C
Fe-3Si, Fe-based with 2.5%3.5% Si, 0.05% max C
Fe-50Ni, Fe-based with 49%51% Ni, 0.05% max C
Fe-50Co, Fe-based with 48%50% Co, 2.5% max V,
0.05% max C
430L, Fe-based with 16%18% Cr, 1% max Mn, 0.05%
max C
The low-thermal-expansion alloys used for instrumentation
and electronic packing include Kovar and Invar, but the only
one reaching specification status is the F-15 alloy based on
iron with 29% Ni, 17% Co.
Likewise specifications for stainless steels include the
following:
316L, Fe-based with 10%12% Ni, 16%18% Cr, 2%
3% Mo, 0.03% max C
420, Fe-based with 12%14% Cr, 0.15%0.4% C
430L, Fe-based with 16%18% Cr, 0.05% max C
94
17-4 PH, Fe-based with 3%5% Ni, 15.5%17.5% Cr,
3%5% Cu, 0.15%0.45% (Ta+Nb), 0.07% max C
Most of the metallic systems allow for up to 0.6% or 1%
silicon (Si) and 1% of other total contaminants and some in-
tentionally allow for up to 2% manganese (Mn). No oxygen
or nitrogen levels are included in the specifications.
RESOURCES
Materials Standards for Metal Injection Molded Parts, 2010,
Metal Powder Industries Federation, Princeton, NJ.
Standard Specification for Metal Injection Molding (MIM)
Ferrous Alloys, ASTM International, West Conshohocken,
PA.
Standard Test Methods for Metal Powders and Powder
Metallurgy Products, 2010, Metal Powder Industries Federa-
tion, Princeton, NJ.
T. Beck, J. Schneider and V. Schulze, Characterisation and
Testing of Micro Specimen, Microsystem Technologies,
2004, vol. 10, pp. 227232.
Metals Handbook Desk Edition, second edition, 1998, J.R.
Davis (ed.), ASM International, Materials Park, OH.
F.H. Froes and R.M. German, Cost Reductions Prime Ti
MIM for Growth, Metal Powder Report, vol. 55, 2000, June,
pp. 1221.
R.M. German and A. Bose, Injection Molding of Metals and
Ceramics, 1997, Metal Powder Industries Federation, Prince-
ton, NJ.
V.A. Hackley and C.F. Ferraris, Guide to Rheological Nomen-
clature: Measurements in Ceramic Particulate Systems, NIST
Special Publication 946, January 2001, National Institute of
Standards and Technology, Gaithersburg, MD.
K.S. Hwang and M.Y. Hsiao, Mechanical Properties of Fe-
Ni Injection Molded Parts Made from Composite Powders,
Powder Injection Molding Symposium 1992, P.H. Booker,
J. Gaspervich, and R.M. German (eds.), Metal Powder Indus-
tries Federation, Princeton, NJ, 1992, pp. 5366.
V. Josef and L.K. Tan, Thermal Performance of MIM
Thermal Management Device, Powder Injection Moulding
International, 2007, vol. 1, pp. 5962.
W.S. Lee and C.H. Hsu, Dependence of Impact Deformation
Behaviour of Fe-2Ni Sintered Alloy on Strain Rate and Sin-
tered Density, Materials Science and Technology, 2006, vol.
22, pp. 1,4311,443.
Y. Masai, M. Ishida, S. Tanaka and K. Nishiyabu, Evaluation
on Mechanical Properties and Functionalities of Anti-Bacteria
and Resist-Corrosion for Cu-Added Stainless Steel
 Table of Contents
Chapter 5: Material Properties

(SUS304L) by MIM, Journal of the Japan Society of Powder

and Powder Metallurgy, 2002, vol. 49, pp. 834840.
T.J. McGrady, Chemical Analysis of Powder Metallurgy
Products, Advances in Powder Metallurgy and Particulate
Materials2002, compiled by V. Arnhold, C.-L. Chu, W.F.
Jandeska, Jr., and H.I. Sanderow, Metal Powder Industries
Federation, Princeton, NJ, part 11, pp. 9298.
H. Miura, Fabrication of High Performance Sintered
Materials Through MIM, Proceedings of the 2000 Powder
Metallurgy World Congress, Part 1, K. Kosuge and H. Nagai
(eds.), Japan Society of Powder and Powder Metallurgy,
Kyoto, Japan, 2000, pp. 315319.
H. Miura, T. Masuda, T. Ogasawara and Y. Kankawa, High
Performance Injection Molded Ti Compacts, Journal of the
Japan Society of Powder and Powder Metallurgy, 2002, vol.
49, pp. 825828.
T. Osada, H. Miura, Y. Itoh, M. Fujita and N. Arimoto,
Optimization of MIM Process for Ti-6Al-7Nb Alloy Pow-
der, Journal of the Japan Society of Powder and Powder
Metallurgy, 2008, vol. 55, pp. 726731.
P. Suri, B.P. Smarslok and R.M. German, Impact Properties
of Sintered and Wrought 17-4 PH Stainless Steel, Powder
Metallurgy, 2006, vol. 49, pp. 4047.
B. Williams, Setting Standards and Getting it Right, Metal
Powder Report, 2001, vol. 56, no. 12, pp. 1418.
W. Wu, R.M. German, D. Blaine, B. Marx, and C. Schlaefer,
Effects of Residual Carbon Content on Sintering Shrinkage,
Microstructure and Mechanical Properties of Injection
Molded 17-4 PH Stainless Steel, Journal of Materials
Science, 2002, vol. 37, pp. 3,5733,583.
H. Zhang and R.M. German, Sintering MIM Fe Ni Alloys,
International Journal of Powder Metallurgy, 2002, vol. 38,
pp. 5161.

95
Table of Contents
 Table of Contents

Chapter 6
COSTS

Outline Typical Project Costs

Tooling Cost
Tool Life
Equipment Costs
Component Costing
Competitive Forces
Design Effects on Processing Costs
Resources

TYPICAL PROJECT COSTS
In dealing with MIM production costs, two related con-
cepts are introduced for this discussion:
manufacturing cost is the cumulative expense incurred
in the production of a product; it includes the purchase
of raw materials, as well as the expenditures for
energy and labor, as well prorated items such as mainte-
nance, depreciation, insurance, and taxes
sale price reflects the value of the component to the
user; it is the amount of money exchanged for the
fabricated component and normally includes the manu-
facturing cost and allocations for a return on investment
and risk
Ideally, cost and price should be related, but that is not
necessarily the case. During the 1980s, MIM products were
selling at a price that sometimes was just 25% of the manu-
facturing cost; bankruptcy was a consequence of this pricing
philosophy. Today, MIM is profitable and the manufacturing
cost is generally below the selling price.
Cost analysis is an important aspect of engineering design.
One common approach is simply to send out a component de-
sign and solicit several responses from vendors. That usually
results in several responses, with a large variation in cost and
a list of vendor-desired exceptions or changes to improve
process yield. This chapter focuses on understanding what de-
termines the production costs, realizing that individual com-
pany policies on profits, taxes, interest payments, dividends,
and risk combine to generate a range of price quotes. We rec-
ognize price variations as a natural part of life. For example,
on a local highway repair, ten construction firms bid on the
project. The ratio of the highest to lowest bid was 1.48, and
the lowest, and winning bid, was $3.26 million. The variation
reflects differences in internal costs and capacity factors, as
well as profit and risk factors. The same is true in MIM, but
in some cases the ratio of highest to lowest bid seems to be
larger, up to a factor of 8.
Chapter 9 contains a detailed analysis of cost factors as a
basis for spreadsheet calculations. It starts with tool cost, and
moves sequentially through feedstock cost, batch-size effects,
unit-manufacturing costs, and analysis of sensitivity factors
for several aspects. The mathematics presented there are com-
patible with straightforward spreadsheet calculations and a
few software packages are available that enable rapid cost es-
timation.
In MIM, the most straightforward approach to estimate
component cost is to sum the costs accumulated at each man-
ufacturing step, starting with tool fabrication and extending
to secondary operations and final inspection. The increments
are as follows:
tool cost
feedstock cost (powder, binder, mixing, or purchased
feedstock)
unit-fabrication costs (mixing, molding, debinding, sin-
tering)
secondary, finishing, inspection, and packaging costs
facility overhead burden
profit, tax, interest, and risk aversion
Here the emphasis is on understanding how MIM costs
depend on design factors such as part complexity, material
selection, and component size. As background to this
approach, note that estimation procedures for tool cost are
well established in plastic injection molding, providing a
sound basis for MIM. Indeed some of the tool vendors have
online tool-cost estimators. However, the source of the tooling

97
 Table of Contents
Chapter 6: Costs

is a significant factor, since tool costs for MIM shops in India,
Thailand, and China are 10% of those in Germany, North
America, and Japan.
Raw-material cost is another factor and it is dominated by
the alloy and source. For example, the price of MIM-grade
prealloyed powder in China is half of that in Germany. Poly-
mers used as binders tend to be a low cost factor and final
price is nearly insensitive to the polymer formulation. Unit
costs in molding, debinding, and sintering depend on the fa-
cility and installed equipment; generally these costs vary con-
siderably between operations.
Probably the weakest cost aspect for MIM is in secondary
operations. Some firms have embraced secondary steps such
as coining, but others rely on external services for machining,
heat treating, and coining. Accordingly, secondary costs vary
considerably, and largely this is where the larger MIM opera-
tions have a competitive advantage.
Once the fabrication cost is understood for a component,
the sale price is estimated based on profit, tax, interest, and
risk factors that vary between sites. These factors are well es-
tablished in the related fields of powder metallurgy, ceramics,
cemented carbides, and plastics molding.
For the core manufacturing steps, unit operation calcula-
tions are developed for each step to accumulate a total fabri-
cation cost. An outline of the costing model is given in Figure
6.1. It shows that several decisions and factors, beyond mate-
rial, enter into the calculations. Some industry-wide cost par-
titions are given in Table 6.1 to provide a rough sense on
factors impacting cost during MIM production. Although
there are recognized differences in all cost factors, including
labor, it is interesting that labor is not a dominant factor; MIM
succeeds in both expensive and low-labor-cost regions.

TOOLING COST
Tool cost for MIM is based on principles encountered in
plastic injection molding. Although many models exist, the
most accurate approach assumes purchase of a standard mold
base. It then calculates the time needed to machine the com-
ponent features into a single cavity. For multiple cavities, the
Figure 6.1. Detailed flow chart for part costing via MIM, showing machining time for the first cavity is used to estimate the time
how several factors enter into the overall calculation. The detailed and cost for each additional cavity, recognizing each cavity
model based on this flow chart is presented in Chapter 9 encounters a lower cost due to increased experience. As a first
approximation, the second cavity is constructed for about
70%75% of the first cavity cost, and four cavities are less
TABLE 6.1. RANGES OF TYPICAL MIM PRODUCTION
COSTS AS PERCENTAGE OF TOTAL COST than three times the cost of a single cavity.
Final tool cost is calculated from the total construction time
cost item range multiplied by a shop hourly rate, with additions for profit and
direct labor 10%30% administrative costs. This approach assumes the tool is correct
powder costs 8%20% on the first cycle, so it does not include a provision for rework.
feedstock preparation 5%10% In reality, tool shops are averse to risk when fabricating a tool
tooling amortization 2%6%
molding 6%15% that is outside their experience base; thus, a risk aversion fac-
debinding 3%10% tor is often introduced to allow for possible errors. Likewise,
sintering 5%10% current business levels impact tool quotations as evident by
interest and depreciation 5%9% the mapping of cost vs. deliverybusy shops are both costly
secondary operations 4%45% and have a long delivery time.
inspection 1% 4%
cost of yield loss 1%20% Detailed quantitative cost analysis of tool cost builds from
general, administrative, sales expenses 8%20% the assumption that a standard mold base is purchased and
converted into the customized design required for the compo-

98
 Table of Contents
nent molding. The mold base might cost $2,000. It needs to
be modified to include the specific cavity and related sprue,
runner, and cooling channel. Major cost additions are associ-
ated with hot runner systems.
To estimate the cavity-machining cost, the tool geometry
and features determine the time and then the tool shop hourly
rate is multiplied to estimate the cost. In addition, purchased
tool materials and special features are added to the cost. For
example, unscrewing devices, such as those required for in-
ternally threaded features, add significantly to the machining
time, upwards of 250 hours each. Moreover, these motions are
relatively slow and add to the molding cycle time. For exam-
ple, a small 2 g stainless steel part with an internal thread re-
quires 25 seconds molding-cycle time, although cooling in the
cavity is quite fast. Likewise, besides unscrewing devices, side
pull motions add at least 65 hours to the tool manufacturing
time. So the number of cores, ejector pins, and overall com-
plexity and tolerances lead to an estimated machining time
that feeds into the estimated cost. Complexity is based largely
on the number of features, including surface protrusions,
bumps, patches, holes, and depressions. Costs escalate rapidly
with component complexity. Thus, each feature in the com-
ponent is assessed for a typical machining time.
Beyond the time to machine the features, there is additional
time required to machine the parting plane. In some cases the
parting line is curved, and this significantly increases the fab-
rication cost. In other words, a flat parting line is most typical
and least expensive, while complex parting planes described
by curvatures and steps add to the tool expense.
Tooling tolerances are more stringent than part tolerances,
and in many estimates the tool tolerance is held to less than
20% of the component allowance. Accordingly, considerable
time is used to fabricate a precise tool for a tightly toleranced
component, especially when the number of features is large.
The specified tool surface finish incurs a cost, especially
if a polished tool is required. In a similar manner, texturing,
lettering, or insignias on the tooling add to the tool fabrication
time and cost.
Machining cost is determined by adding the individual ma-
TABLE 6.2. COMPARISON OF LABOR RATES IN
SOME MIM COUNTRIES
country base rate, $/h relative increase, %/y
Germany 58.50 2.5
France 47.81 5.4
Sweden 41.73 2.6
Japan 38.63 0.4
U.S. 33.00 2.1
Spain 31.62 3.8
Korea 21.10 12.4
Mexico 3.30 3.6
China 3.00 7.6
Chapter 6: Costs
chining times and multiplying by the machining rate. A critical
factor is the toolmakers hourly rate for mold fabrication,
which might be in the $45 per hour range, but can vary con-
siderably. For example, a recent comparison of union wages
around the world is shown in Table 6.2. With a 20-fold labor
rate difference it is difficult to quote a single value for any of
the manufacturing parameters, including the toolmakers
hourly rate. Besides the listed countries, some recent MIM ef-
forts have set up in India, Singapore, Hungary, and Thailand,
and there are other successful operations in Taiwan, Israel,
Italy, Brazil, and South Africa.
The tooling cost depends on several factors. Recent MIM
tool costs ranged from $2,500 to $75,000. Further, besides
complexity and features, mold costs vary with geographic re-
gion. To control mold cost, it is best to specify multiple cavi-
ties when there is a compelling economic benefit, since
tooling is a significant initial cost. Likewise, for lower pro-
duction volumes it is less expensive to machine threads after
sintering than to make an expensive tool.
TOOL LIFE
Because tooling is a large initial cost, extension of the mold
life is important to the economics of MIM. Initial tooling cost
must be amortized over the production quantity, so a long tool
life is advantageous.
As a generalization, MIM is not cost effective in hard tool-
ing for production quantities below 5,000 parts per year. Using
soft tooling on the other hand, some ceramic and cemented
carbide components are produced in lots as small as 200.
Most MIM shops can operate for 100,000 shots without
tool maintenance. Several coating technologies extend tool
life, especially in resisting wear from harder and more abra-
sive particles. The coatings include intermetallic compounds,
boride layers, hard nickel-phosphorus electroplates with em-
bedded polytetrafluoroethylene particles, hard sputter coat-
ings, and various spray, vapor, or reaction interfaces. Although
they all add expense to tool creation, there are possible gains
over time for larger production quantities.
EQUIPMENT COSTS
Capital equipment creates a cost burden for MIM in the
form of depreciation. Across the industry arises a unit manu-
facturing cell concept roughly consisting of 1 mixer, 4 mold-
ers, and 2 furnaces. Of course, in about 28% of the operations
there is no mixer, since that portion of the MIM industry pur-
chases feedstock. Most operations depreciate these devices in
5 to 7 years. This means the purchase price is subtracted as
an expense in yearly increments, independent of the business
for the year. Some example production equipment costs are
given in Table 6.3 for common purchases.
The sintering furnace tends to be the most expensive item.
For low-volume production, an output of nearly 60 kg per day
(132 lb. per day) is possible from a $300,000 furnace, and
99
 Table of Contents
Chapter 6: Costs

about three times this (300 kg per day or 660 lb. per day) from
a $900,000 furnace. Continuous furnaces are more expensive,
but the capital depreciation on a per part basis is lower because
of higher productivity.
Besides the devices directly associated with production,
additional items are needed for quality control, materials test-
ing, research and development, and general manufacturing
support. Table 6.4 provides an assessment of the equipment
options and associated capital equipment needed to start a
pilot operation. In addition to the equipment, there is a larger
need for operating cash during the time when the operation is
stabilized.
COMPONENT COSTING
The cost for producing a MIM component is scattered, re-
flecting differences between production sites in terms of ca-
pacity, equipment, technology, labor rates, and raw materials
consumption. Many MIM operations do not have the broad
experience needed for accurate costing calculations. Signifi-
cant errors occur by improperly estimating process yield, re-
work rates, cycle times, furnace-loading factors, daily
capacities, employee-benefit costs, mold-maintenance costs,
and depreciation schedules.
Chapter 9 details an approach to component costing com-
posed of several stages. The first stage is associated with the
material from which the component is fabricated. The second
stage is associated with the operating environment. The third
stage is associated with unit operations and the determination
of cost allocation per piece at each step. Finally, risk and pro-
duction balancing are added cost factors. This information will
not be repeated here.
As an illustration of the cost partition between various con-
tributors, two example components are introduced here.
Golf clubs by MIM are a favorite topic. Many demonstra-
tion club styles have been fabricated (driver, wedge, and put-
ter) and a few reached production status, including putters out
of titanium, stainless steel, and zirconia. Drivers were fabri-
cated out of titanium, but the costperformance combination
was not satisfactory to reach production. First consider the
cost breakdown for a MIM stainless steel sand wedge, a case
where MIM is not competitive vs. investment casting.
Figure 6.2 is a photograph of the target wedge design. The
cost factors involved in MIM production of a sand wedge are
detailed in Chapter 9. For this analysis, assume the wedge is
formed from 17-4 PH stainless steel to a final weight of 200 g.

TABLE 6.3. CAPITAL EQUIPMENT COSTS FOR

MAJOR PRODUCTION EQUIPMENT TABLE 6.4. APPROXIMATE COSTS FOR PILOT
OPERATION EQUIPMENT
device cost range
planetary or batch mixer with $25,000$35,000 equipment cost range
heater mixing equipment $25,000$125,000
continuous mixer with $135,000$400,000 (optional)+
temperature control batch weighing and dry $5,000$30,000
30 ton molder $25,000$60,000 mixing (optional)+
80 ton molder $75,000$100,000 2 to 3 molding machines $150,000$360,000*
200 ton molder $95,000$150,000 (total)
granulator (sprue and runner $4,000$9,000 granulator and mold- $16,000
recycle) temperature control
3-axis miniature robot $24,000$50,000 2 to 4 debinding solvent or $25,000$800,000
4-axis machine robot $60,000$100,000 thermal retorts
sprue picker $6,000 experimental furnace $25,000$80,000
mold-temperature-control unit $6,000 small batch vacuum/ $120,000 (consider
solvent debinding (nontoxic, $5,000$50,000 atmosphere furnace toll sintering)
nonexplosive) pilot-production vacuum furnace $250,000
solvent-debinding tank $100,000$400,000 full-scale vacuum furnace $600,000
(controlled substance) integrated continuous furnace $500,000 (optional,
supercritical solvent-debinding $250,000$450,000 second year)
reactor inspection equipment $180,000*
catalytic-debinding furnace $25,000 and up testing equipment (carbon, $60,000
thermal-debinding furnace $20,000$40,000 hardness)
laboratory-scale tube sintering $25,000$35,000 design CAD system, support $150,000*
furnace, batch system
laboratory-scale sintering $45,000$80,000 office equipment and other $200,000*
furnace, batch devices
pilot-scale batch vacuum furnace $135,000$250,000
large-scale batch vacuum $500,000$600,000 + optional expense that can be avoided if feedstock is
furnace purchased vs. self-mixed
continuous furnace $400,000$1,000,000 * for a firm already practicing plastic injection molding,
(atmosphere) these devices already exist

100
 Table of Contents
Based on a production volume of 200,000 per year and 98%
process yield, the cost breakdown is shown in Figure 6.3, as-
suming industry typical values. The dominant feature is the
powder cost at $1.87 per part due to the component mass. The
mixing and binder cost per part is $0.23 and the material loss
in processing makes about half of the total manufacturing cost
assignable to the feedstock cost. The feedstock cost is domi-
Figure 6.2. Candidate golf club head (sand wedge) considered
from MIM out of 17-4 PH stainless steel with a mass of 200 g
Figure 6.3. Pie chart showing how cost is partitioned for the
fabrication of a 200 g stainless steel sand wedge, where the high
material cost is a dominant factor; the estimated manufacturing
cost is $4.53 each
Figure 6.4. Final-price pie chart for the 200 g stainless steel sand
wedge, corresponding to a price of $6.12 each
Chapter 6: Costs
nated by the powder price. Because of the component size and
shape, debinding is slow (20 hours) and batch sintering-fur-
nace loading is only 20% efficient due to conformal setters,
so the 18-hour sintering cycle contributes 25% of the manu-
facturing cost. Thus, total cost for MIM production of the sand
wedge is estimated at $4.53. The molding tool is optimized at
a single cavity with a one-time cost of nearly $25,000. If
amortized over 200,000 parts, then the mold combined with
the processing cost give a per part total of $4.77.
Based on this manufacturing cost, Figure 6.4 shows the
sale price breakdown, where the above costs are supplemented
by overhead and profit, making the final price per part $6.12
with a 10% profit margin. This reflects an industry low of just
over $30 per kg while the industry average is $133 per kg ($60
per pound). For reference, a MIM putter out of stainless steel
went into production at 66,000 per year at a slightly higher
price, giving credibility to these estimates.
Figure 6.5. Cellphone stainless steel knuckle component with 2 g
mass considered in the cost calculations
Figure 6.6. Cost partition for the production of a 2 g stainless steel
cellphone component, showing the relative contributions from each
of the factors to the final price of $0.45 each
101
 Table of Contents
Chapter 6: Costs
As another example, consider a 2 g stainless steel cellphone
component shown in Figure 6.5. Details of the calculations
are given in Chapter 9 for this component. It is fabricated
using MIM at production rates near 1 million per month. The
calculated component price was $0.45 each after reaching full
production (equivalent to $225 per kg) and this is close to the
value quoted for the project. The breakdown of this price is
illustrated in Figure 6.6. Note that the price per unit mass at
$225 per kg makes this a more expensive component com-
pared to the sand wedge. For this shape, the complexity made
MIM the best-choice production route. Curiously, the original
attempt was to form this part out of titanium, but the added
cost from even inexpensive titanium powder doubled the part
cost; thus, stainless steel was used in production.
Across the MIM industry there is considerable variation in
costs, markets, and production approaches. However, surveys
show that most of the successes are priced near $1 per part.
But prices vary considerably. Yet surveys show that the typical
MIM component is selling at from $133 per kg. However,
some of the smaller 0.1 g components equate to pricing near
$600 per kg. Considerable variation exists in the price based
on different production factors.
COMPETITIVE FORCES
For metals there are a host of alternative routes for the fab-
rication of a given design. Thus, competing technologies
greatly influence the viability of MIM on any project. Table
6.5 compares the advantages and limitations of various net-
shape technologies against which MIM competes. Users say
that the decision to use MIM over competing technologies
usually starts with cost, but includes issues of shape capabil-
ities, productivity, surface finish, and precision. The cost ad-
vantage for MIM vs. other forming techniques increases as
the number of machining, grinding, or finishing steps grows,
especially for small components. Further, the cost of machin-
ing increases dramatically as the surface finish improves.
Rough ground or turned surfaces can be produced at just 25%
of the cost to form the smooth surfaces associated with as-sin-
tered MIM structures. If an as-sintered surface finish is ade-
TABLE 6.5. COMPARISON OF MIM VS. OTHER NET-SHAPE TECHNOLOGIES
attribute MIM powder metallurgy
density 98%
strength 100%
magnetic response 100%
surface finish 0.40.8 m
wall thickness 100.1 mm
complexity high
design flexibility high
production rate high volume
material range high
nominal mass range 0.003 g1 kg
102
quate, then MIM has a distinct cost advantage. However, for
large, simple shapes, MIM routinely proves unattractive. As
early as 1988 demonstration stainless steel parts of 12 kg were
produced, but proved unsuccessful economically. The author
has personally been involved in the production of components
reaching up to 40 kg, establishing the technical feasibility, but
alas financial feasibility was lacking.
Each shaping technology is best suited to a particular com-
bination of materials, tolerances, sizes, shapes, and properties.
For example, casting techniques excel in shape complexity,
but lack in surface finish and dimensional precision. On the
other hand, at high production quantities, machining tech-
niques suffer with material wastage and production costs. All
technologies are evolving, making for dynamic competition.
For every component that MIM captures, there is a competing
technology working to win that component on the next design
iteration, resulting in an intense competition between tech-
nologies. Cost is usually the critical parameter in selecting a
net-shaping technology.
DESIGN EFFECTS ON PROCESSING COSTS
Early consultation between the designer and component
producer leads to recommendations that will lower the pro-
duction costs. These range from use of thinner cross sections
to combinations of parts into a single piece. The best analysis
protocol relies on asking many questions concerning the can-
didate design. There are subtle but important changes that im-
prove process yield and reduce tooling costs, and these are
generally beneficial. In any design, a few features add sub-
stantially to the cost. Custom MIM producers help avoid
costly mistakes by suggesting design changes to reduce man-
ufacturing costs.
The approach to cost reduction builds from the 8020 rule.
Which 20% of the design features control 80% of the cost?
The answer provides focus to cost-reduction efforts. Surface
finish is a good case. If the surface specification is too smooth,
then it adds considerable finishing cost after sintering. A so-
lution might be to intentionally design for a rippled or textured
surface that avoids polishing, yet has a suitable aesthetic
casting machining
85% 95%99% 100%
50% 98% 100%
70% 95% 98%
2 m 3 m 0.22 m
2 mm 5 mm 2 mm
low medium high
medium medium low
high volume low volume low volume
medium medium medium
0.1 g10 kg 1 g up 0.1 g up
 Table of Contents
Chapter 6: Costs

appeal. A few places to find 20% of the features that control RESOURCES
80% of the cost include: S.V. Atre, C. Wu, S.J. Park, R.M. German, C.J. Hwang and
nonstandard materials: use common materials if possible R. Zauner, Technical and Economic Comparison of Micro
and graduate to a stronger or more corrosion-resistant Powder Injection Molding, Proceedings Powder Metallurgy
material with a lower cost if it is a widely used alloy; for World Congress, Korean Powder Metallurgy Institute, Busan,
example, in one diesel-engine application the change Korea, 2006, pp. 4546.
from 15-5 PH stainless to 17-4 PH stainless resulted in
G. Boothroyd, P. Dewhurst, W. Knight, Product Design for
improved corrosion resistance and a lower cost
Manufacture and Assembly, second edition, 2002, Marcel
combine parts where possible to eliminate assembly, in-
Dekker, New York, NY, pp. 359379.
ventory, and time spent in molding: one automotive part
was molded from two stainless steels, one magnetic and G. Dieter, Engineering Design: A Materials and Processing
one nonmagnetic, saving in joining costs Approach, third edition, 2000, McGraw Hill, Boston, MA.
look for purchasing quantities that enable better cost ef-
R.M. German, Medical and Dental Applications for Micro-
ficiencies at the MIM producer: for example, if the an-
miniature Powder Injection MouldingA Roadmap for
nual requirement is small (under 20,000 pieces), it is
Growth, Powder Injection Moulding International, 2009,
better to purchase once per year
vol. 3, no. 2, pp. 2129.
components with uniform wall thicknesses are easier to
fabricate and can be fabricated with higher tolerances A.C. Goodrich, Evaluating Technology Strategies for the
and more uniform properties: remember, thick sections, Manufacture of Near-Net-Shape Metal Automotive Compo-
on a component slow the molding cycle and add cost nents, P/M Science and Technology Briefs, 2002, vol. 4, no.
plan for quick ejection from the tooling: it is beneficial 3, pp. 510.
to place undercuts, threads, and textures in locations
R.D. Stewart, Detailed Cost Estimating, Mechanical Engi-
where they do not delay ejection
neers Handbook, second edition, M. Kutz (ed.), 1998, Wiley
place molding blemishes in locations where they are not
Interscience, New York, NY, pp. 2,1172,142.
critical, thereby eliminating any expensive finishing
steps
try to incorporate a flat surface for component support
during debinding and sintering: this avoids the cost of
special fixtures
tight tolerances add greatly to the cost and need to be
critically analyzed to ensure they are justified
This final point is one of great discussion in the MIM field.
Tight tolerances are possible. However, the concomitant in-
crease in equipment sophistication (cost), lower yields, or the
reliance on expensive secondary operations all dramatically
increase component cost. A normal, or Gaussian, distribution
in dimensions typically is associated with MIM. If the toler-
ances are tight, then less of the product will be within speci-
fication. Accordingly, sorting is required to separate those in
specification from those not acceptableand this adds cost
because of the extra step. More of an impact is from the ex-
pense associated with production of the rejected components,
an expense that must be captured in the final shipped material.
One case of a fiber optic component reports only 7 out of
1,000 were in specification. Fortunately, the price for each
was sufficiently high (it was a military sale, sort of like the
$600 toilet seat), so the product was profitable.
All MIM products are eventually limited in terms of di-
mensional precision. As-sintered dimensional scatter is re-
duced by post-sintering machining, but with added production
steps and costs. Early discussions with the vendor provide one
means to ensure the lowest-cost approach to tight tolerances.

103
Table of Contents
 Table of Contents

Chapter 7
APPLICATIONS

Outline Overview
Aerospace
Automotive
Business Machines
Cellphones
Computer Devices
Cutting Tools
Defense
Dental
Electrical and Electronic Components
Firearms
Hand Tools
Hardware and Industrial
Household and Consumer
Instrumentation and Sensors
Jewelry and Watches
Medical
Microelectronics and Optoelectronics
Sporting
Telecommunication
Summary
Resources

OVERVIEW cobaltchromium alloys, titanium alloys

This chapter shows some of the many applications for
metal powder injection molded components. Although a broad
variety of examples are cited here, still the array is far from
comprehensive. Instead this chapter establishes the diverse
uses already existing for MIM.
The products produced using MIM are formed using a
range of materials. Even so, a few materials dominate the
field. Almost all of the current production is based on the
following groupings in terms of primary functional character-
istics, where the example alloy systems are listed in approxi-
mately decreasing use order:
biocompatibilitystainless steels, cobaltchromium
alloys, titanium alloys
corrosion resistancestainless steels, nickel alloys,
electrical conductivitycopper
electronic packagingironnickel, Kovar, tungsten
copper, Invar
heat dissipationcopper, tungstencopper, aluminum
high density, high inertiatungsten heavy alloys
high-temperature servicesuperalloys, refractory metals
magnetic responseironnickel, ironsilicon, iron
structurallow-alloy steels, stainless steels, tool steels
wear resistancecemented carbides, tool steels, cobalt-
chromium alloys
Several applications have reached maturity, while others
are just starting to grow. Early fields to adopt MIM were the
firearms (triggers, trigger guards, sights) and dental (ortho-
dontic brackets), followed by electronic and consumer prod-

105
 Table of Contents
Chapter 7: Applications

the decision to use superalloys (nickel-based or cobalt-based)

TABLE 7.1. EVALUATION OF COMPONENTS IN VARIOUS or titanium alloys. By design these alloys remain strong at
APPLICATIONS high temperatures (which is why they appeal to aerospace).
application lower range upper range approximate But this same resistance to high-temperature deformation
area $/kg $/kg $/cm3 makes them difficult to sinter densify. High-temperature use
biomedical 200 100,000 35 implies a material designed to remain strong as temperature
aerospace 100 20,000 13 is increased, while sintering relies on thermal softening so that
sporting 30 10,000 2 capillary forces between the particles induce densification;
automotive 7 200 0.3 superalloys naturally resist sintering densification so sintering
marine 3 20 0.06
packaging 2 10 0.01
construction 0.3 3 0.008

ucts, after which it spread to everything from golf clubs to jet

engine components. A later development is the use of MIM
for microminiature surgical tools and biomedical implants.
Depending on the life-critical aspects, application areas
assign a different value to engineered components, mostly due Figure 7.1. Niobium alloy rocket nozzle, one of the first aerospace
to higher performance expectations. Even if the component is applications for MIM
the same, still valuation varies by field. This is one way to
explain the $800 toilet seat and $400 monkey-wrench for the
military. Table 7.1 provides a sense of the evaluation spectrum
in terms of price per mass and per unit volume. On this basis
the hierarchal areas that are the highest-value applications for
MIM would seem to be the medical and dental fields. Most
recently MIM has made significant penetration in aerospace-
component production, another area where there is a high val-
uation for performance.
Remember, the key aspects for success are situations where
shape complexity, material performance, component cost, and
production quantity intersect. On a valuation basis, it is no
surprise to see continued emphasis on developing applications
in medical, dental, and aerospace areas. On the other hand,
the automotive, firearm, and sporting devices will continue to
use MIM for small and relatively costly components.

AEROSPACE
Two early MIM components attracted much attention
because they were for aerospace applications, an area where
presssinter powder metallurgy fails to qualify. One was a nio- Figure 7.2. Flow body with mass over 1 kg, formed from
bium rocket nozzle and the second was a seal ring for a Boe- superalloy IN 718
ing commercial airliner.
The niobium alloy (Nb-10W-10Ta) rocket nozzle shown
in Figure 7.1 was fabricated by Parmatech (Petaluma, Cali-
fornia, U.S.) for Rocketdyne (Canoga Park, California, U.S.).
This single-piece nozzle was coupled to a two-piece injector
that was assembled by electron beam welding. The design was
for satellite and missile control using liquid propellants, where
during operation the nozzle-wall temperature exceeds 1,375C
(2,500F).
In the aerospace applications the material and process need
to be qualified for the use temperature, stress, and environ- Figure 7.3. Sintered (left) and molded (right) swirler component
ment. Accordingly, the extreme environments often lead to (photograph courtesy of Latitude Manufacturing)

106
 Table of Contents
cycles are long. For flight-critical uses, the components are
hot isostatically pressed to full density, since residual porosity
is unacceptable.
In the past decade, several other MIM components have
moved into use for aerospace applications. Prior to accept-
ance, extensive test data are required, such as fracture-tough-
ness, fatigue-life, and stress-rupture properties. Determination
of these properties under a representative range of tempera-
tures, stresses, and environments requires expensive testing,
so companies do not embark on this application area without
substantial resources. Products realized to date are valves,
pumps, actuators, hydraulic bodies, nozzles, and components
Figure 7.4. Top and bottom views of a superalloy IN 718 butterfly
valve plate (photograph courtesy of Honeywell Engine and
Accessories)
Chapter 7: Applications
near the hot sections of jet engines. Figure 7.2 is a picture of
a flow body after sintering and machining; it is molded at
1,630 g (3.6 lb.) from gas-atomized IN 718 nickel-base super-
alloy powder, but after machining the mass is 1,130 g (2.5 lb.).
Similarly, Figure 7.3 is a picture of a swirler aerospace com-
ponent in both the molded and finished conditions. It was sub-
jected to extensive testing and found to have excellent
properties.
Another MIM success is the butterfly valve plate pictured
in Figure 7.4 in both the top and bottom orientations. This
component is fabricated by Precision Castparts Advanced
Forming Technology (Longmont, Colorado, U.S.) for Honey-
well Engine Systems and Accessories (Tempe, Arizona, U.S.)
out of the IN 718 alloy. The component is 27 g in mass and
37 mm in outer diameter (1.5 in.). Properties depend slightly
on the powder size, oxygen content, carbon content, hot iso-
static pressing cycle, and heat treatment cycle. Nominally, the
properties by MIM are as follows:
density = 8.22 g/cm3
yield strength = 1,000 MPa (143 ksi)
ultimate strength = 1,330 MPa (193 ksi)
elongation = 29%
hardness = 42 HRC
fatigue-endurance limit = 380 MPa (55 ksi)
Surprisingly, even without hot isostatic pressing the IN 718
tensile properties are nearly the same, although the fatigue-
endurance limit improves by 20% with hot isostatic pressing.
On this basis, HIP will remain a part of the fabrication process.
The MIM fabrication of high-reliability components for
demanding aerospace uses is the subject of much current
research. One example component is depicted in Figure 7.5.
It is a superalloy turbine designed for a small jet engine for
taxi aircraft. As with many MIM aerospace components, it
has been tested extensively but not yet taken to production.
Figure 7.5. Turbine disk formed using a nickel-base superalloy
(photograph courtesy of Oak Ridge National Laboratory)
107
 Table of Contents
Chapter 7: Applications

AUTOMOTIVE
The use of MIM in automotive applications started in the
late 1980s with Chrysler lock components and Buick turn-sig-
nal levers. Since then a large number of automotive MIM
components have reached production, and many early suc-
cesses have passed out of production. Other successes are
found in transmission synchronizers, airbag sensors, electric-
seat gears, interior knobs, spark-plug bodies, turbochargers,
fuel injectors, valve lifters, heat dissipaters, and engine sen-
sors. Some early examples are collected in Figures 7.6 through
7.10; in order, these are an ignition key component, timing
ratchet, airbag actuator, airbag-tilting sensor, and magnetic
position sensor. Figure 7.11 is a convertible-roof top clip
(shown green in Figure 2.13b) fabricated by GKN Sinter Figure 7.8. 17-4 PH stainless steel firing initiator for deploying
Metals (Bad Langensalza, Germany) for Volkswagen. the airbag in an automotive crash (component courtesy of Key
Use of metal powder injection molding for automotive Automotive Group)

Figure 7.6. Stainless steel MIM pivot component for automotive Figure 7.9. 17-4 PH stainless steel pivot, used for sensing an
key, showing the component and the final key assembly automobile crash (component courtesy of Hunan Injection High
Technology)

Figure 7.7. Integrated ratchet and shaft formed from 316L Figure 7.10. Rotational sensor formed from a magnetic outer stain-
stainless steel (component courtesy of Advanced Materials less steel and nonmagnetic inner stainless steel for positional sens-
Technologies) ing (component courtesy of Advanced Materials Technologies)

108
 Table of Contents
Figure 7.11. Convertible-roof top clip molded from Fe-2Ni, the
14 g device is shown green in Figure 2.13b
Figure 7.12. High-temperature alloy turbocharger components:
(a) adjustable vanes, (b) adjustment ring, and (c) vane lever block
(photographs courtesy of PIM Korea)
Chapter 7: Applications
applications required the industry to gain production capacity
and quality systems needed to handle automotive orders. For
example, in the early 1990s, Ford Motor Company had an
engine-tappet component that required a production rate of
300,000 per day. That project alone would have required about
one-fourth of the total industry capacity at the time. On the
other hand, today several MIM products are produced at levels
of 100,000 per day.
As the industry matured, automotive applications grew dra-
matically to reach nearly $150 million in sales for 2010. This
market is dominated by three countries, with about 30% of
the production in Japan, 20% in the U.S., 40% in Germany,
and 10% elsewhere. Significant production has moved to
Korea, Singapore, Spain, and Hungary, as MIM becomes an
established technology. For example, Volkswagen diesel tur-
bochargers for German automobiles are designed by a U.S.
firm and MIM fabricated by a Korean firm at production rates
of 100,000 parts per day.
High-temperature automotive MIM components include
turbochargers, stators, rotors, disks, ducts, and shrouds. Figure
7.12 shows a collection of the turbocharger vanes, adjustment
ring, and vane lever block fabricated by PIM Korea (Gyoun-
gasn-city, Gyoungbuk, Korea). The key gains have been by
converting investment casting components to MIM with a
lower cost. Another example is the burner cover shown in
Figure 7.13. It is fabricated by Schunk Sintermetalltechnik
(Thale, Germany) for Webasto automotive heating systems
using 316L stainless steel. As is evident, the MIM automotive
applications are in areas that require shape complexity and
where competing processes have difficulty producing the
shape.
Figure 7.13. Burner cover for automotive heating systems formed
from 316L stainless steel to a final mass of 43 g
109
 Table of Contents
Chapter 7: Applications
Alternatively, several automotive applications rely on high-
strength, low-alloy (HSLA) steels. The steels are heat treated
after sintering to maximize mechanical properties. Automo-
tive structural steels are heat treated to tensile strengths near
1,800 MPa (260 ksi), with at least 3% elongation, hardness
up to 50 HRC, impact energy over 100 J/cm2, and fatigue
endurance strength up to 500 MPa (85 ksi). For transmission
components, the rolling-contact-fatigue strength must exceed
2,000 MPa (290 ksi). The heat-treated steels include 17-4 PH
and Fe-1Cr-0.2Mo-0.8Mn, with carbon and silicon for hard-
ening. Higher alloying levels are required for applications
involving extreme stress and wear, up to 8% Cr, 2% Mo, and
1% Ni. One example is an all-terrain brake component with
overall length of 40 mm and mass of 11.6 g. Two variants fab-
ricated from 17-4 PH are shown in Figure 7.14 as fabricated
by NetShape TechnologiesMIM (Solon, Ohio, U.S.). The
Figure 7.14. Stainless steel brake fitting for an all-terrain vehicle
(photograph courtesy of NetShape TechnologiesMIM)
Figure 7.15. Steering-system clevis formed from a low-alloy
nickel-steel with a final mass of 65 g (photograph courtesy of
Metal Powder Industries Federation)
110
component has a wall thickness of just 0.7 mm along the stem
in some regions. Besides the dimensional challenges, it is
required to resist corrosion and withstand hydraulic-fluid leak-
age. The small-diameter hole proved easier to fabricate with
MIM than with other technologies.
A related high-strength automotive development is the
steering clevis pictured in Figure 7.15, fabricated by Precision
Castparts Advanced Forming Technology (Longmont, Col-
orado, U.S.) for Delphi Saginaw Steering Systems (Saginaw,
Michigan, U.S.). The ironnickel steel component weighs 65
g and is fabricated to the following properties:
density = 7.5 g/cm3 (less than 4% porosity)
yield strength = 1,450 MPa (210 ksi)
tensile strength = 1,680 MPa (245 ksi).
Variable-valve rocker arms are another example of high-
performance MIM automotive components. These were first
used by Honda (Tochigi, Japan), fabricated by Nippon Piston
Ring (Shimotsuga-gun, Tochigi, Japan), from a heat-treatable
low-alloy steel (Fe-1Cr-0.7Mn-0.2Mo-0.2Si-0.2C). The com-
ponent was processed to a tensile strength of 1,100 MPa (160
ksi) and fatigue-endurance strength of 300 MPa (44 ksi). The
version shown in Figure 7.16 is a variant fabricated by Schunk
Sintermetallwerk for BMW.
An example of MIM automotive growth is found in fuel
injectors. The molded versions were first introduced by Cum-
mins Engine for heavy-duty diesel engines, but are now used
in gasoline engines. The fuel injectors rely on high strength
in a hard, corrosion-resistant, and wear-resistant material.
Shapes vary, but generally are in the 10 to 12 mm (under half
in.) length range, with diameters near 6 to 7 mm (quarter in.),
with flanges and tapered internal bores that are 100 to 150 m
in diameter (0.004 to 0.006 in.). Inner dimensions are added
by post-sintering grinding. Variants are fabricated from pre-
cipitation-hardened stainless steels and ceramics.
Metal powder injection molding is being qualified for
under-the-hood thermal-management applications: for exam-
Figure 7.16. Variable-valve rocker arms fabricated using a heat-
treatable steel (photograph courtesy of Powder Injection Moulding
International)
 Table of Contents
ple, in the control systems for hybrid-vehicle power-switching
units. Copper is the leading candidate, where an internal vapor
chamber enables heat-pipe cooling by evaporation and con-
densation inside the MIM device.
BUSINESS MACHINES
Business machines are a mature application for metal pow-
der injection molding. This area was early to adopt the tech-
nology, largely for printers, postage machines, typewriters,
and dot-matrix printers. Figure 7.17 shows an early MIM
Epson dot-matrix printer head. It was formed using 17-4 PH
stainless steel and was one of the first components produced
using continuous sintering.
Another early success was in the tilt ring used in an electric
typewriter printhead based on Fe-2Ni; production by MIM at
IBM reached 1.5 million units. Similarly, postage machines
relied on hard materials for wear components. Figure 7.18 is
Figure 7.17. Dot-matrix printer head formed from 17-4 PH
stainless steel. This was one of the early business machine
successes for MIM (component courtesy of Raymond Weich)
Figure 7.18. Pitney-Bowes postage meter MIM tool-steel
component used to handle the wear and corrosion associated with
postage tape
Chapter 7: Applications
a tape cutter formed from tool steel for a Pitney-Bowes
postage meter. Most of the applications used Fe-Ni steels
formed from mixed carbonyl iron and nickel powders.
Recent interest has come from computer equipment fabri-
cators for inkjet-printer nozzles and laser-printer components.
Even the metallic logo for Hewlett-Packard and Apple com-
puters relied on MIM. Opportunities continue to exist in var-
ious handheld office machines, magnetic components, and
optical components, but as mechanical drives are eliminated
from office devices the opportunities for MIM will be small.
CELLPHONES
Mobile telephones are now produced at quantities of 1
billion per year. For many years MIM has played a key role.
Applications include hinges, swivels, knuckles, covers,
lanyard loops, side buttons, supports, and vibrators. Some
designs rely on at least four MIM parts. A few early product
examples are shown in Figure 7.19: a thin-walled metallic
cover, knuckles or corner pieces that enable the flip open
action to answer the phone, and a camera phone swivel.
Recently pressures to increase features and reduce the size
have pushed MIM into microminiature components. Two
examples fabricated by Taiwan Powder Technologies
(Taoyuan, Taiwan) are shown in Figures 7.20 and 7.21. The
first is the mount for the audio jack on a smart cellphone and
the second is termed a hanger, also fabricated from 17-4 PH
stainless steel.
One of the more taxing campaigns involved the MIM fab-
rication of a flip slider and hinge barrel, pictured in Figure
7.22 along with the assembled handheld unit. These stainless
steel MIM components are used because of their low cost that
still offers three-dimensional shape complexity in a thin struc-
ture. The design positions the clamshell phone cover so it
slides down and flips open in a single motion. Since compo-
nent alignment was critical, coining was performed after sin-
tering. These sets were produced at rates of 6 million per
month by several firms, including Advanced Materials Tech-
nologies (Singapore) and Parmatech (Petaluma, California,
U.S.) for Motorola (Libertyville, Illinois, U.S.). Both compo-
nents are from 17-4 PH stainless steel with the following
attributes:
final density = 7.6 g/cm3 (less than 2% residual porosity)
yield strength = 1,090 MPa (158 ksi)
ultimate strength = 1,186 MPa (172 ksi)
elongation to fracture = 6%
hardness = 30 HRC
A high strength is required in this design since the corners
were very thin. A special difficulty came from impact tests
based on simulating a dropped phone. In production, the com-
ponents were measured for several critical tolerances and
especially were monitored for perpendicularity of 0.2 mm
(0.008 in.) at two points. Corrosion resistance was mandatory
as well, so steel was not acceptable. Indeed, the parts were
111
 Table of Contents
Chapter 7: Applications

Figure 7.20. Audio jack mount formed from 17-4 PH stainless

steel (pictures courtesy of Taiwan Powder Technologies)

Figure 7.21. A microminiature component for cellphones termed a

hanger, also fabricated from 17-4 PH stainless steel along with the
location of the device (photographs courtesy of Taiwan Powder
Technologies)

Figure 7.19. Example MIM components applied to cellphones;

(a) cell phone cover, (b) stainless steel knuckles, and (c) pivot for Figure 7.22. Cellphone shown along with the stainless steel flip
camera phone swivel (components courtesy of CetaTech and slider and hinge barrel fabricated by MIM (photograph courtesy of
Advanced Materials Technologies) Advanced Materials Technologies)

112
 Table of Contents
chrome finished prior to assembly. Data from Motorola
reported 95% MIM process yield. Most of the loss occurred
during debinding and sintering. For comparison, the cost to
fabricate this component by alternatives was five times more
than via MIM.
COMPUTER DEVICES
Since the early 1980s the computer hard drive has under-
gone a dramatic size decrease and capacity increase. Today
Figure 7.23. Hard disk drive pivot hub formed from 316L stainless
steel (photograph courtesy of Parmatech)
Figure 7.24. Cold plate fabricated by MIM from copper to
maximize heat exchange using an array of 279 pins (photograph
courtesy of Acelent Technologies)
Chapter 7: Applications
hard drives are present in music players, automobiles, global
positioning systems, and industrial tools. As the size shrank,
a wealth of opportunities for MIM arose to include balancing
weights, latches, actuators, connectors, and other mechanical
or magnetic components. In the early 2000s production accel-
erated with small disk drives that critically relied on MIM
components. Following along, other computer devices
adopted the MIM technology for connectors, latches, pivots,
sockets, and heat sinks.
An early MIM computer disk drive component is shown
in Figure 7.23 fabricated from 316L stainless steel. It is known
as a pivot hub, fabricated by Parmatech (Petaluma, California,
U.S.) for Hewlett-Packard. It was used in storage and posi-
tioning mechanisms for high-density disk drives. The sintered
density was 7.5 g/cm3 (about 5% porosity) and the tensile
strength was 500 MPa (73 ksi). However, aluminum postage-
stamp-sized disk drives for handheld Apple iPod devices
emerged and displaced the need for stainless steel.
Due to semiconductor heating during use, heat-dissipation
application arose for MIM. Early designs included high-ther-
mal-conductivity pin arrays to help cool computer chips. The
pin array was formed out of copper or other high-thermal-con-
ductivity materials. Designs targeted a thermal conductivity
of at least 200 W/(m K) and a thermal-expansion coefficient
of 6 ppm/K. More recent designs moved to pin arrays from
copper, such as shown in Figure 7.24.
One difficulty in serving the computer field is the rapid
progression in designs that make it difficult to keep a compo-
nent in production for an extended period. A six-month prod-
uct life cycle is common. However, the savings realized by
MIM vs. casting are significant; machining a casting to com-
puter industry tolerances is 22 times more expensive than via
MIM. For cost and property reasons, MIM has become the
process of choice for small, complicated metallic components.
Thus, it is routinely used in laptop, notebook, and other small
devices.
CUTTING TOOLS
Hard materials for wood and metal cutting are usually
molded from cemented carbides, but a few rely on cermets or
tool steels. Cermets are mixtures of metals and ceramics (TiC-
Ni, for example) and cemented carbides are also known as
hard metals (WC-Co). About 1% of the tungsten carbide cut-
ting inserts are produced by injection molding. The use of
MIM for tungsten carbide cutting tools is relatively small,
since it only amounts to about 4%5% of the commercial
MIM production. MIM products are used in woodcutting
routers, metal drills, end mills, metal-cutting inserts, and drill
tips.
An example of MIM from WC-Co is shown in Figure 7.25.
These are concrete abrasive tips. Figure 7.26 shows two
metal-cutting tools; the first is an end mill with titanium
nitride coating and the second is a rotary cutter. The drill
113
 Table of Contents
Chapter 7: Applications

shown in Figure 7.27 consists of a replaceable molded carbide

tip. The picture includes the final assembly, the titanium
nitride coated tips, and the molded bodies prior to sintering.
A recent activity has been production of super abrasives,
where diamond is included in the composition. One of the first
successes is in wire saws for stone cutting. Diamonds are
mixed with a hard tool-material alloy in the feedstock phase.
Sintering is performed below the decomposition temperature
of the diamond. This product is formed as bead shapes which
are placed on a steel wire to form a wire-cutting saw used in
slicing granite, marble, and other stone. Figure 7.28 is a pic-
ture of one of the beads and the assembled wire saw fabricated
by MIMITALIA (Vado Ligure, Savona, Italy).
Some of the MIM drilling and cutting tools are quite
expensive. Although the mass is small, the value is high. For
example, one 4.6 g drill piece from WC-Co is priced at $65.
In many of the wear applications, the MIM fabrication route
is invisible since the performance characteristics are the same
as attained with other fabrication routes.
Besides the cutting area, a related component is found in
semiconductor wire bonding where hard materials are
required to avoid wear. Several MIM plants are focused on
fabrication of these microelectronic wire-bonding tools using
precision tools with small dimensions, some as small as about
100 m across.

DEFENSE
Defense applications were one of the early success areas
for MIM. An early product was the 316L stainless steel stabi-
lizer fin shown in Figure 7.29. The fin was produced at three
MIM facilities in California for Ford Aerospace and Commu-
nications. A final coining treatment was used to straighten the
fins after sintering.
Defense applications include handguns, weapon sights,
missile fins, fragmentation projectiles, bullets, practice ammu-

Figure 7.25. MIM tungsten carbide wear tips used in the impact Figure 7.26. Cutting tools formed using tungsten carbide
abrasion of cement (photograph courtesy of Powder Injection compositions include the (a) end mill and (b) woodcutting tool
Moulding International) (components courtesy of SECO Tools and CetaTech)

114
 Table of Contents
Chapter 7: Applications

nition, penetrators, missile thrusters, fasteners for missile ship-

ping containers, and rotors for arming devices. Figure 7.30
shows an example of a 316L stainless steel safe-and-arm rotor
fabricated by FloMet (DeLand, Florida, U.S.). It is fabricated
to the following properties:
density = 7.6 g/cm3 (about 5% porosity)
yield strength = 170 MPa (25 ksi)
tensile strength = 515 MPa (75 ksi)
elongation = 50%
hardness = 67 HRB
Figure 7.31 is another example of a defense component,
fabricated for the U.S. Army by Polymer Technologies
(Clifton, New Jersey, U.S.). This 50 mm by 75 mm part is for
a classified application. Tungsten projectiles and fragmenta-
tion shot are also popular defense products. Figure 7.32 is a
picture of a fragmentation cone from a tungsten heavy alloy.
Similar examples are fabricated at several locations, including
tungsten heavy alloy antiaircraft projectiles designed to break
into pieces on impact. This is a precursor to frangible tungsten
ammunition fabricated using MIM. Another success was in
the thermal mount for phase-array radar systems formed using
tungstencopper and various sonar imaging devices.

Figure 7.28. Saw-wire bead (a) and wire-cutting tool (b) formed
using diamond super-abrasive compositions developed specifically
for MIM (photographs courtesy of Nick Williams of Powder
Injection Moulding International)

Figure 7.27. Drill system with replaceable MIM cemented carbide

tips; shown here is the complete drill and green and sintered tips
with a titanium nitride coating (photograph courtesy of Figure 7.29. Stainless steel stabilizer fins, an early MIM defense
Kennametal) product

115
 Table of Contents
Chapter 7: Applications

DENTAL ing for the process, such as the devices shown in Figure 7.34.
A mainstay application for MIM has been orthodontic These are 17-4 PH stainless steel parts for a self-aligning tooth
brackets. These have been in continuous production since the positioning system. The pieces are produced at quantities up
mid-1980s. Example brackets are shown in Figure 7.33, illus- to 600,000 per week by FloMet (DeLand, Florida, U.S.) for
trating the waffle pattern on the bottom and the wire attach- Ormco Sybron Dental Specialties (Orange, California, U.S.).
ment posts and grooves on the front. Previously, these were One bracket and one slide go on each tooth with the hook used
made from welded bands. The small MIM brackets allow in some cases. Another new design is shown in Figure 7.35.
direct attachment to each tooth. The predominant materials
are stainless steels (316L and 17-4 PH), but the offering has
expanded to include nickel-free compositions such as cobalt
chromium alloys, nitrogen-stabilized austenitic stainless steel,
and most recently nitinol (NiTi), titanium, and transparent alu-
mina.
Several larger companies currently produce dental brackets
at production volumes exceeding 250,000 brackets per day.
Recent designs have taken full advantage of MIM by design-

Figure 7.32. Tungsten fragmentation tips with internal serrations

formed using MIM

Figure 7.30. Safe-and-arm rotor for explosive arming, formed

from 316L stainless steel (photograph courtesy of Metal Powder
Industries Federation)

Figure 7.33. Approximately 0.1 g stainless steel orthodontic

brackets fabricated by MIM with attachment posts and waffle
pattern mounts (components courtesy of Ortho Organizers)

Figure 7.31. Steel defense component formed by MIM used as a Figure 7.34. New design orthodontic brackets with self-aligning
retrofit device on military vehicles (photograph courtesy of designs (photograph courtesy of Metal Powder Industries
Polymer Technologies) Federation)

116
 Table of Contents
Figure 7.35. Orthodontic bracket designed for the MIM process
that includes rounded surfaces, funneled wire entrance, and mesh
bondable pad (hidden) for maximum adhesion (image courtesy of
Ortho Organizes)
Figure 7.36. Endodontic tips such as shown here are for root canal
treatments. They are formed using MIM with the generation of
different burr sizes and shapes and patterns; (a) tool with a length
of approximately 35 mm, and (b) scanning electron micrograph of
the working tip (photographs courtesy of CetaTech)
Chapter 7: Applications
This is from Ortho Organizers (Carlsbad, California, U.S.)
which makes its own devices by MIM. The 17-4 PH stainless
steel design comes in 12 combinations and includes an embed-
ded code for easy identification. The corners are rounded to
ensure patient comfort, and the wire entrance is funnel shaped
for easy insertion. The bottom pad is optimized for adhesion.
A typical mass is about 0.1 g. Each bracket costs on
average $0.35 to $0.60 to fabricate. Since much of this
industry is captive, it is difficult to place a total value on MIM
sales, but generally the estimates are under $100 million.
Growth in this application depends mostly on the rate that
modern dental care concepts spread to developing countries.
Dental tools have been a more recent success for MIM. The
devices include working tips for root canal and dental cleaning
use. The latter include ultrasonic agitation with a central hole
in the scaler to flush the tip using pressurized liquid. Endodon-
tic tools are fabricated by MIM with small bumps on the
working tip for root canal treatments. Figure 7.36 is a picture
of one of several tip geometries with a scanning electron
micrograph of the small bumps on one tool. These are formed
by CetaTech (Sacheon, Gyeongam, Korea) for B&L Biotech
(Asan Si, Gyeonggi-Do, Korea). The material is 17-4 PH. Pro-
duction is in single-cavity tooling with typical production
rates of 1,000 parts per batch. The overall device is 0.9 g and
it is fabricated to 20 m tolerances (0.0008 in.). Maintaining
a sharp projection during processing is a significant difficulty,
since sintering tends to round the features. The final treatment
is to coat the device with titanium nitride. Related devices
have similar geometries with different tapers, bump sizes, and
diameters customized for various endodontic treatments.
Other applications include dental implants out of titanium,
sometimes with intentional porosity to form a scaffold for
tissue ingrowth. Some of these have features as small as 20
to 50 m on devices that might be as long as 18 mm. Figure
7.37 is an example microminiature tooth implant structure
fabricated by MIM, where the porous titanium tissue attach-
Figure 7.37. Scanning electron micrograph of a titanium implant
with porous titanium band on a solid post for enhanced tissue
attachment (image courtesy of Eric Baril, National Research
Council of Canada)
117
 Table of Contents
Chapter 7: Applications
ment region is evident in the threaded portion, while the
remainder of the sintered structure is dense. Actually, the core
is solid and the porous region is only on the outside as a band.
ELECTRICAL AND ELECTRONIC COMPONENTS
Electrical connectors and various insulator or conductor
components for electrical service have been produced by
MIM since the mid-1980s. One of the early applications was
in multiple-layer circuits, where molybdenum powder was
injected into alumina laminates to form co-fired circuit pat-
terns. Today, most of the MIM applications are for packaging
devices, such as the hermetic (glassmetal sealed) device
shown in Figure 7.38. In addition, electronic applications
include electronic hardware (latches, locks, hinges) and
mounts fabricated from stainless steel, with 316L being a
favorite. Insulators are commonly fabricated using powder
injection molding from ceramics such as alumina.
Figure 7.38. Three stages in the production of an electronic
package, showing the Kovar box formed by MIM, the glassmetal
sealed leads, and the final gold electroplated device (components
courtesy of Hunan Injection High Technology)
Figure 7.39. Hearing-aid receiver package formed from an iron
nickelmolybdenum alloy (photograph courtesy of Metal Powder
Industries Federation)
118
Electronic packaging relies on low-thermal-expansion-
coefficient materials, such as Kovar, Invar, or tungsten
copper, mounting ceramics such as alumina, or pure metals
such as aluminum, copper, molybdenum, or tungsten. The
choice depends on functional aspects, and typically
if thermal expansion is a dominant concern, then the
choice is Invar;
if glassmetal sealing is a dominant concern, then the
choice is Kovar;
if heat dissipation is a dominant concern, then the choice
is tungstencopper.
A MIM electronic package is shown in Figure 7.39. It is a
hearing-aid receiver capsule fabricated by FloMet (Deland,
Florida, U.S.) for Starkey Laboratories (Eden Prairie, Min-
nesota, U.S.). This thin-walled component is formed using a
nickelironmolybdenum alloy to provide a magnetic shunt
for the hearing aid to separate the receiver signal from the tele-
coil signal. The internal undercut especially made MIM an
optimal approach so the cost reduction in shifting to MIM was
50% when compared to the previous deep-drawing-and-
annealing approach.
FIREARMS
Handguns, shotguns, rifles, and military weapons use sev-
eral metallic components fabricated by MIM. Typical firearm
components are sights and sight mounts, handgun bodies,
safety levers, sling attachments, catches, firing pins, maga-
zines, triggers and trigger guards, and slide blocks. Steels are
typical and most of the production quantities are under
100,000 per year for several years. After sintering, the steels
are heat treated and coated with protective black oxides to
inhibit rusting. Figure 7.40 depicts a steel trigger guard
Figure 7.40. Trigger guard for muzzleloading rifle out of steel
(photograph courtesy of Metal Powder Industries Federation)
 Table of Contents
formed by Megamet Solid Metals (Earth City, Missouri, U.S.)
for Modern MuzzleloadingKnight Rifles (Decatur, Alabama,
U.S.) to tolerances of 0.12 mm (0.005 in.). The 88 g compo-
nent is processed to achieve 650 MPa (94 ksi) tensile strength.
A most impressive development has been use of MIM to
Figure 7.41. Police handgun components fabricated from a iron
nickel steel using MIM: (a) two views of handgun and (b) spindle
and other parts (photographs courtesy of Beijing Jarain Powder
Injection Technology Limited)
Figure 7.42. Housing block for a Colt 45-caliber handgun
fabricated using MIM from 420 stainless steel (photograph
courtesy of Metal Powder Industries Federation)
Chapter 7: Applications
fabricate whole handgun bodies. Pictured in Figure 7.41 are
the handgun body and cylinder and other parts for a police
revolver fabricated for the Chinese police forces by Beijing
Jarain Powder Injection Technology Limited Company
(Huairou District, Beijing, China). The material is a mixture
of carbonyl iron and nickel powders, molded using a thermo-
plastic feedstock, thermally debound, and sintered in vacuum
to full density.
Another firearm component is the 420 stainless steel hand-
gun housing block shown in Figure 7.42, fabricated by Par-
matech (Petaluma, California, U.S.). It is used in a 45-caliber
handgun to hold the spring mechanism to give the desired slid-
ing action. It is a good challenge to MIM, since it has under-
cuts, wings, through holes, blind holes, and both thick and thin
sections. Some specifics on the component are as follows:
final density = 7.7 g/cm3 (about 2% residual porosity)
yield strength = 1,500 MPa (218 ksi)
ultimate strength = 1,800 MPa (261 ksi)
hardness = 50 HRC
A firearm device for military use is pictured in Figure 7.43,
fabricated by Megamet Solid Metals (Earth City, Missouri,
U.S.) for Colt Canada (Kitchener, Ontario, Canada). It is a
4140 low-alloy steel folding front grip used on military rifles.
Sintering is in a nitrogen atmosphere to improve strength.
Post-sintering machining is used to ream and tap the holes.
The component is heat treated to a hardness of 4550 HRC
and a manganese phosphate finish is added after heat treating.
A related development is taking place in the fabrication of
ammunition. Metal powders such as tungsten are molded to
form practice bullets and green bullets for sites where lead
contamination is not acceptable. Environmental concerns on
both practice ranges and waterfowl habitats have led to a ban
Figure 7.43. Front grip on a military rifle generated using MIM
from a 4140 steel (photograph courtesy of Metal Powder
Industries Federation)
119
 Table of Contents
Chapter 7: Applications
Figure 7.44. Birdshot fabricated by MIM from a tungsten heavy
alloy (components courtesy of Springfield Munitions)
of lead products. Springfield Munitions (Kersey, Pennsylva-
nia, U.S.) uses MIM to produce the high-accuracy birdshot
from tungsten, pictured in Figure 7.44, relying on very high
cavity counts to enable production of billions of shot pellets
per year. The company private-labels the products. The com-
position is a tungstenbronze composite that proves very
resistant to corrosion and is fabricated to a density 12 g/cm3,
higher than lead, giving tighter shot patterns than traditional
ammunition.
HAND TOOLS
Early demonstrations of MIM hand tools occurred in con-
sumer devices, including some high-visibility brands, such as
the popular Sears Craftsman socket wrenches and locking
pliers. Several shapes were produced by MIM for the various
multi-purpose hand tools, sort of the Swiss Army knife
approach to hand tools. These include screwdrivers and
wrenches, and feature combinations as illustrated in Figure
7.45. This 220 g component has a length of 122 mm (4.8 in.)
with tolerances of 0.2 mm (0.008 in.). It is fabricated from
tool steel at production rates of 20,000 per year. The compo-
nent is subjected to sandblasting and polishing as final oper-
ations. It is fabricated by Hunan Injection High Technology
(Changsha, Hunan, China).
The most demanding hand tool applications are associated
with professional service, such as tools used on construction
sites. Several professional-tool firms use MIM components
for high-durability wrenches, pliers, wire cutters, and pipe
threaders. To match the target mechanical properties, the met-
als are either tool steel, high-strength stainless steels, or heat-
treatable steels. After sintering, these alloys are heat treated
to tensile strengths from 1,400 MPa to 2,000 MPa (roughly
200 to 300 ksi).
Figure 7.46 is a picture of hand tool fabricated by FloMet
(DeLand, Florida, U.S.) for Silvent (Boraes, Sweden). The
120
Figure 7.45. Multi-purpose hand tool fabricated from tool steel
(photograph courtesy Hunan Injection High Technology)
Figure 7.46. 316L stainless steel compressed-air nozzle to control
air flow and noise over a range of operating environments
(photograph courtesy of Metal Powder Industries Federation)
nozzle controls the flow of compressed air while reducing
noise in a range of temperatures and corrosive environments.
The component is formed from 316L stainless steel to deliver
the following properties:
final density = 7.6 g/cm3 (about 3% porosity)
yield strength = 172 MPa (25 ksi)
ultimate strength = 517 MPa (75 ksi)
elongation to fracture = 50%
hardness = 67 HRB
The hollow plenum chamber is laser sealed after sintering.
In other instances, firms have been successful in avoiding the
laser-sealing treatment, resulting in further cost savings.
HARDWARE AND INDUSTRIAL
Considerable opportunity exists for metallic injection
molded products for hardware applications, such as machine
parts, components for pumps, and spray nozzles. Good suc-
cess is seen in stainless steel, bronze, and copper for plumbing
valves, stems, fittings, locking mechanisms, hinges, latches,
decorative fixtures, and fittings such as pictured in Figure
7.47. One example is the 17-4 PH stainless steel lock-cylinder
 Table of Contents
components shown in Figure 7.48. These are fabricated by
Kinetics Climax (Wilsonville, Oregon, U.S.) for Black &
Decker Hardware (Lake Forest, California, U.S.). The locking
bar, pin, and rack operate in a lock cylinder that contains a
total of 11 MIM parts. They are fabricated to the following
specifications:
sintered density = 7.7 g/cm3 (essentially full density)
yield strength = 730 MPa (106 ksi)
ultimate strength = 900 MPa (131 ksi)
Related hardware components are found in the industrial
control lock pieces (barrel pin, square, barrel boss, and cover)
that form safety switches in machine tools, shown in Figure
7.49. The components were initially formed from 316L stain-
less steel with a high ductility, but are also formed from 17-4
PH stainless steel for higher strength. They are molded and
sintered to a 7.5 g/cm3 minimum density by Advanced Mate-
rials Technologies (Singapore). Molding incorporates multiple
side core inserts in the tooling. This enables the customer to
Figure 7.47. Example of a hardware stainless steel fitting
fabricated using MIM
Figure 7.48. Lock-cylinder components fabricated using MIM
from 17-4 PH stainless steel (photograph courtesy of Metal
Powder Industries Federation)
Chapter 7: Applications
have superior-performance locking devices and allows 30%
cost saving over other processes. Testing includes long-dura-
tion (300,000 rotation) tests and 48-hour salt-spray corrosion
test. The high mechanical strength from MIM 17-4 PH is a
desirable attribute for this application.
Applications in plumbing are a similar story. A 17-4 PH
stainless steel shower-faucet stem product became famous in
the MIM community because of its production challenges.
Figure 7.50 shows the component. This shower safety-valve
system worked to prevent scalding due to changes in water
pressure. It was fabricated by Parmatech (Petaluma, Califor-
nia, U.S.) for Moen (North Olmstead, Ohio, U.S.). MIM was
used to form the oversized blank which was then machined to
final 5 m tolerances on automated machining centers.
Typical mass was 40 g for the stem part and the sintered den-
sity was greater than 7.63 g/cm3.
Figure 7.49. Components for industrial safety switches in machine
tools, including the control lock: barrel pin, square, barrel boss,
and cover (photograph courtesy of Advanced Materials
Technologies)
Figure 7.50. Moen shower safety-valve components from 17-4 PH
stainless steel (photograph courtesy of Metal Powder Industries
Federation)
121
 Table of Contents
Chapter 7: Applications
Figure 7.51. Copper air conditioner Y fittings formed by MIM:
(a) shows the size range and (b) shows some of the shape details
(photographs courtesy CetaTech)
Figure 7.52. Handcuffs include different stainless steel MIM
components (photograph courtesy of Metal Powder Industries
Federation)
122
Copper plumbing fixtures are fabricated by MIM when tra-
ditional hydroforming is unsuccessful. Examples are the Y
copper fittings pictured in Figure 7.51 used in air conditioners.
These fittings are formed by CetaTech (Sacheon, Gyeongam,
Korea) for its customer, a large multinational firm that domi-
nates the portable air conditioner market. A range of sizes are
in production stepping from 6.35 mm (0.25 in.) to 31.75 mm
(1.25 in.) barrels. The dimensional coefficient of variation is
0.2% with a sintered density exceeding 8.7 g/cm3 giving less
than 3% residual porosity. Production is in 4-to-8-cavity tool-
ing at quantities of 35,000 per month.
One of the more recognizable MIM components is shown
in Figure 7.52, fabricated by SSI Technologies (Janesville,
Wisconsin, U.S.) for ASP (Appleton, Wisconsin, U.S.). The
handcuffs consist of several MIM parts that include a lock
pawl, bow, side and center links, and main links, formed from
three different stainless steel compositions. The links are sin-
tered to 7.5 g/cm3 density (under 5% porosity) to give a tensile
strength of 540 MPa (78 ksi).
Industrial spray nozzles for spraying paint, insecticides,
water, or abrasives have long been a success for MIM.
Depending on the fluid, these can be formed from alumina,
stainless steel, or cemented carbide. Hard materials are used
for high-reliability or high-erosive-wear components, such as
cement discharge nozzles. Figure 7.53 is a picture of cemented
carbide nozzles formed by Retco Tooling (Dallas, Texas,
U.S.).
Other example MIM components for hardware or indus-
trial use include a static mixer consisting of interwoven helical
blades fabricated from a cobaltchrome alloy for corrosion
and wear resistance, vending-machine-lock parts, and
mechanical latches, pulleys, and alignment pieces.
Figure 7.53. Hard and wear-resistant tungsten carbide spray
nozzles formed using MIM (components courtesy of Retco Tooling)
 Table of Contents
Chapter 7: Applications

HOUSEHOLD AND CONSUMER Similar early consumer products were the toothed grinding
Powder injection molding has found broad applications in portion of a pepper mill, parts for sewing machines, cutlery,
household, personal care, and consumer markets. Examples and hand-held nutcrackers.
include fingernail clippers, hair-trimming shears, cosmetic Figure 7.57 is a picture of a small, portable paper-hole
cases, toy trains, gears for electric toothbrushes, luggage punch consisting of several pieces exclusively fabricated
clasps, luxury fountain pens, eyeglass-frame hinges, can open-
ers, nutcrackers, pepper grinders, camera buttons and
switches, electric knives, pot handles, kitchen knives, wine-
bottle cork extractors, and food blenders. This is one of the
largest application areas for MIM and most of the components
are formed from stainless steel. Two examples are pictured in
Figures 7.54 and 7.55. The first shows a tripod pivot base
formed using MIM (Mimest S.p.A., Pergine Valsugana, Italy),
with the foreground version in titanium and the background
version in stainless steel. The second example is a toy train
where the stainless steel wheels are formed using MIM.
An early consumer application is shown in Figure 7.56.
These are 0.34 g (maximum dimension 6 mm or 0.25 in.)
stainless steel gears for an electric toothbrush. They were pro-
duced by Schunk Sintermetalltechnik (Giessen, Germany) for
Figure 7.55. Toy train that uses wheels fabricated via MIM
Braun Division of Gillette (Kronberg im Taunus, Germany).
Production exceeded 30,000 per day using 16-cavity tooling.

Figure 7.56. Small stainless steel gears used in a Braun electric

toothbrush

Figure 7.54. Tripod mounts fabricated using MIM from titanium in

the foreground and stainless steel in the background (photograph Figure 7.57. Portable stainless steel paper-hole punch designed
courtesy of Powder Injection Moulding International) for home use (component courtesy of Metalor)

123
 Table of Contents
Chapter 7: Applications
Figure 7.58. A bronze wine-stopper head molded as a limited-
edition gift
using MIM stainless steel components. This device was fab-
ricated by the Metalor Division of Iscar (Mizra, Israel) for
Minitec (Stuttgart, Germany) out of 17-4 PH stainless steel to
a final hardness of 38 HRC.
One of the authors personal favorites is shown in Figure
7.58, a bronze top for a wine stopper. This was produced as a
limited-edition item for sponsors of the Penn State sintering
center; it now comes in real handy living in California.
INSTRUMENTATION AND SENSORS
Sensors, sensor bodies, precision measurement compo-
nents, and scientific-apparatus components fabricated by
MIM are generally low-volume, but high-value products.
Applications include sensor mounts and soft magnetic com-
ponents associated with industrial-process control, such as
fluid-control systems, and automotive devices in the engine,
airbag, emission, and brake systems. Automation systems are
Figure 7.59. Solenoid-valve body fabricated by MIM (component
courtesy of Onishi Light Industries)
124
another large user of MIM, especially for robot components,
and this is a particularly strong activity in Japan where it now
comprises 4% of the MIM market.
Soft magnetic materials in current use include pure iron,
Fe-2Ni, Fe-3Si, Fe-6Si, Fe-0.4P, Fe-0.6P, Fe-50Ni, Fe-36Co,
Fe-50Co, and Fe-49Co-2V. Because of a high magnetic per-
meability, low coercive force, high flux density, and high
Curie temperature (temperature above which magnetic
response is lost), ferrous alloys rich in nickel and cobalt are
important for several applications. High-nickel alloys have a
useful combination of saturation magnetization and coercive
force. High-permeability alloys consisting of nearly equal pro-
portions of iron and nickel are used in housings for microwave
systems, actuators, valves, and other sensor components. An
example solenoid housing formed by MIM is shown in Figure
7.59. The alloy has desirable response over a wide frequency
range (212 GHz). Small variations in the composition and
porosity greatly shift performance; thus, the full-density char-
acter of MIM is necessary.
Ironsilicon alloys are selected for alternating-current
applications where low core losses are important. Core losses,
from hysteresis and eddy currents, are high for iron due to its
low electrical resistivity. Silicon is an effective additive for
increasing the electrical resistivity and reducing the hysteresis
loss in applications involving alternating current. Thus, MIM
Fe-Si alloys are used in high-speed printers, solenoids for con-
trol mechanisms and automatic valves, magnetic amplifiers,
relays, and small motors. An Fe-Si-Al, known as Sendust, is
used in magnetic card readers such as on subways. Several
demonstrations have moved into pilot production, but the
designs are not sophisticated and that means the economic
gains are not sufficiently compelling to take them into full-
scale MIM production.
JEWELRY AND WATCHES
Jewelry is a field where MIM has an advantage because of
its high material utilization. As precious metal costs increase,
much attention has been given to durable (but less expensive)
materials such as tungsten, tungsten carbide, niobium, tita-
nium, and tantalum. Likewise, new production technologies
are under consideration. The definition of what is a precious
metal moves to lower-cost materials as the price of gold, plat-
inum, and silver increase.
Early demonstration jewelry pieces formed by MIM
included tie clasps, pendants, earrings, and engagement rings.
Some were formed in precious metals, mostly silver alloys.
However, some pink platinum objects were demonstrated. The
tolerances required by the jewelry industry are lower than
found in most components and are easily satisfied by MIM.
Another advantage is the potential of the metal powder injec-
tion molding route to produce multiple-material components
containing layered metals to form color layers. However, the
infrastructure is small, with only a few suppliers of powder or
 Table of Contents
feedstock. Thus, only one factory is totally focused on jewelry
by MIM.
To reduce cost, but preserve aesthetics, hollow and foamed
metal technologies are being evaluated for jewelry applica-
tions, such as cosmetic cases. Larger objects with a low mass
are possible with internal pores. One approach is to use gas-
assisted injection molding which blows a gas bubble inside
the feedstock, pushing it against the mold to form a hollow
structure. Another approach involves dissolution of gas into
the feedstock and subsequent nucleation of pores to foam the
feedstock into the die.
The global watch industry has typical annual sales over
$10 billion. Watches are a major user of MIM around the
world for the production of bands, clasps, and cases. For many
years watch-component production accounted for a significant
segment of MIM. During the 1990s the MIM production of
watch components grew at 30% per year. However, the field
reached saturation and is less important as manufacturing
moves to low-cost regions. Today the largest MIM factories
focused on watch components are located in China, but some
activity remains in Singapore, Japan, Taiwan, France, and
Figure 7.60. Example watch components fabricated using MIM
Figure 7.61. Luxury watch with links, spacers, bezel, and clasp all
formed by powder molding, and even the crystal is a spinel
ceramic sintered to optical transparency
Chapter 7: Applications
Switzerland.
Figure 7.60 shows examples of MIM bezel, winding
weight, and watch-band links, all formed by MIM while Fig-
ure 7.61 is a picture of a final product, where almost all of the
components are from powder injection molding, including the
titanium clasp. In many cases the watch companies have
elected to use internal production. These are low-stress com-
ponents, but require a high surface polish. Thus, to avoid
pores, the sintered densities are close to theoretical.
Materials used for watches range from stainless steels to
zirconiatitanium carbide composites, and include cemented
carbides and titanium. Production is generally performed in
highly automated, large facilities operating 15 to 25 molding
machines around the clock. Watch bands are formed as links
that are assembled after sintering, with nominal tolerances of
0.3%. For more critical dimensions, machining is performed
after sintering, such as the bezel mount and some of the criti-
cal fittings associated with watertight construction.
MEDICAL
Medical component fabrication by MIM is growing rapidly.
The field is already a leading application area in North Amer-
ica. Early successes were associated with surgical hand tools.
For example, the video imaging and manipulator surgical con-
nector shown in Figure 7.62 is a good example of the shape
complexity possible. This stainless steel component was fab-
ricated by MIMECRISA (Santander, Spain). Surgical tools
rely on MIM since it can produce the shape with no surface-
connected pores, as required for contact with blood and other
fluids.
Figure 7.62. Stainless steel connector for use in arthroscopic
imaging surgical treatments (component courtesy of MIMECRISA)
125
 Table of Contents
Chapter 7: Applications
Microminiature medical devices are quite successful using
MIM. Concepts call for surgical manipulators with maximum
dimensions less than 5 mm. For example, Figure 7.63 is a pic-
ture of MIM stainless steel components used in heart surgery
fabricated by Advanced Powder Products (Philipsburg, Penn-
sylvania, U.S.). The device is about 2.5 mm long.
Small MIM parts support efforts to reduce the incision size,
benefitting the patient because of faster surgical recovery and
shorter time in the hospital. In a similar manner, Figure 7.64
shows two components, one a manipulator and the other a
biopsy cup. These are formed out of 17-4 PH stainless steel
by Hunan Injection High Technology (Changsha, Hunan,
China) at a rate of 5 million per year. Fabrication is to 20 m
tolerances (0.0008 in.) out of 3 mm (0.12 in.) length and com-
ponent mass reaches to as low as 0.02 g. Some of the new
robotic and small incision surgical tools are heavily dependent
on MIM to form the many links that make up the manipulators.
One robotic surgery device requires 16 different MIM parts.
A factor enabling MIM growth are insurance regulations
Figure 7.63. Small stainless steel pieces used in heart surgery
(picture courtesy of Advanced Powder Products)
Figure 7.64. Manipulator and biopsy tip fabricated by MIM from
stainless steel for use in microminiature medical treatments geared
to minimally invasive surgery (photograph courtesy of Hunan
Injection High Technology)
126
moving health care toward short stays and single-use tools.
The resulting cost pressures justify MIM as a solution. An
example surgical device is shown in Figure 7.65; it is a 17-4
PH stainless steel articulator gear for surgical stapling fabri-
cated by Parmatech (Petaluma, California, U.S.). The device
functions as the drive and locking mechanism for the device
head. This component costs just 30% of its machined prede-
cessor since it is produced net shape with no secondary oper-
ations. The device is sintered to the following properties:
final density = 7.65 g/cm3 (1% porosity)
yield strength = 730 MPa (106 ksi)
ultimate strength = 900 MPa (130 ksi)
hardness = 25 HRC
Beyond surgical tools, MIM applications include implants,
ranging from replacement bones, replacement joints (knee and
hip), and various tissue affixation devices. There is even an
effort to qualify new magnesium MIM alloys for absorbable
stents. Most of the medical instrument applications are satis-
fied by stainless steels, usually 304L, 316L, 420, 440C, or 630
(17-4 PH). However, about 10% of the parts and 25% of the
value will be in titanium alloys. Comparisons of the mechan-
ical, corrosion, and biocompatibility attributes show MIM is
competitive in titanium.
Other options are open to MIM, such as tantalum and
cobalt-base, but currently stainless steels are dominant.
Figure 7.65. Articulator gear formed from 17-4 PH stainless steel
for use on a surgical stapling device (photography courtesy of
Metal Powder Industries Federation)
 Table of Contents
Figure 7.66. Tungsten heavy alloy radiation shield formed by MIM
for use in delivery of radioisotopes for medical treatments
Figure 7.67. MIM-produced 316L stainless steel implant used in
arthroscopic surgical repair of torn rotator cuffs (photograph
courtesy of Metal Powder Industries Federation)
Figure 7.68. Implanted MIM titanium pump for delivery of
treatment drugs to internal organs (component courtesy of TiJet
Medizintechnik)
Chapter 7: Applications
Besides surgical tools, tungsten heavy alloys are injection
molded to form radiation shields for syringes used in the
delivery of medical radioisotopes, such as pictured in
Figure 7.66.
The newest application wave is in implant devices, such as
the pin shroud shown in Figure 7.67. This small pin is fabri-
cated by Kinetics Climax (Wilsonville, Oregon, U.S.) for
ArthoCare (San Juan Capistrano, California, U.S.) out of 316L
stainless steel. It is used as an implant in arthroscopic surgical
repair of torn rotator cuffs by securing a sutured tendon to the
shoulder bone. The typical properties are as follows:
sintered density = 7.85 g/cm3 (less than 1% porosity)
yield strength = 200 MPa (29 ksi)
tensile ultimate strength = 540 MPa (78 ksi)
hardness = 69 HRB
It replaced an assembly made by wire electrodischarge
machining three parts that were then laser welded to form an
assembly.
Applications in implants are emerging with new standards
for MIM implant materials. Both active and inactive structures
are planned, some allowing ingrowth of tissue into large
pores. An important option is to match the elastic modulus
and strength of bone using a highly distended titanium struc-
ture. This important development requires time to achieve
product qualification. Figure 7.68 is a titanium implant port
system fabricated by TiJet Medizintechnik (Kiel, Germany).
For the past few years injection molded ceramics have been
used for heart pacemakers and defibrillators, and stainless
steel implants are now in routine use. However, if the MIM
stainless steels are vacuum sintered, then there is a concern
with chromium evaporation from the surface and loss of pas-
sivity. Thus, care is required to ensure the sintered stainless
steel both meets bulk chemistry specification and is not sus-
ceptible to surface attack. For long-duration implants the pre-
ferred metal is titanium, and it is already used in heart
pacemakers and implanted chemotherapy devices.
For MIM an opportunity is to form a high-porosity titanium
scaffold and to treat this structure with hydroxyapatite
(calcium phosphate of approximately the composition
Ca10(PO4)6(OH)2) on the titanium. The hydroxyapatite
induces bone ingrowth while the titanium provides strength
and toughness that matches bone. The most suitable compo-
sitions have nearly 60% porosity. Sintered strengths are close
to bone, typically near 100 MPa. Because biocompatibility
test results have been very favorable, several components are
moving into MIM production.
MICROELECTRONICS AND OPTOELECTRONICS
High-performance MIM thermal-management materials
are used to move waste heat from computer chips, Internet
servers, cellphone network stations, personal computers,
workstations, and laptop computers. These applications
emphasize low thermal expansion, high thermal conductivity,
127
 Table of Contents
Chapter 7: Applications
or glassmetal sealing attributes. Early devices used ferrous
alloys to absorb electromagnetic interference while extracting
waste heat. An example is shown in Figure 7.69, molded by
Precision Castparts Advanced Forming Technology (Long-
mont, Colorado, U.S.) for AT&T (Murray Hill, New Jersey,
U.S.). Now the packaging is tailored to a low-thermal-expan-
sion coefficient and high thermal conductivity. Some of the
packages are large, and difficulty in machining the large pack-
ages leads to MIM.
The use of MIM low-thermal-expansion Fe-Ni and Fe-Ni-
Co alloys is well established. A recognized alloy with low
coefficient of thermal expansion (CTE) is Invar, with a nom-
inal composition of Fe-36Ni. This alloy exhibits anomalous
thermal-expansion behavior near room temperature (CTE in
near 1.710-6 per C). Alloying with 5%17% cobalt further
lowers the expansion coefficient. It is used to match the
thermal expansion of various semiconductor or glassmetal
sealing materials, such as the electronic package shown in
Figure 7.69. Early success at forming electronic packages, with a
threaded end machined after sintering (components courtesy of
AT&T)
Figure 7.70. Example of a microelectronic package molded from a
controlled-thermal-expansion alloy
128
Figure 7.70. One concept consists of a glassmetal sealing
Kovar package with a W-Cu insert to better dissipate heat.
Components for computer-networking equipment and
Internet servers are MIM targets, and include both electronic
and optical devices. Hermetic housings are formed from low-
thermal-expansion alloys and the transparent optical compo-
nents are fabricated from oxide spinels. Precise dimensions
are necessary because of the required alignment in assembly.
A typical design consists of a small box with a mounting
flange penetrated by several feed holes. Aluminum by MIM
has been considered in a few situations, but the material
proves too soft and is too expensive for most applications.
To overcome the thermal-conductivity barrier, heat-pipe
designs by MIM have arisen. A material such as copper is
molded with intentionally open-pore interior regions, formed
using larger particle sizes and low solids loadings. During sin-
tering pores are intentionally created on the interior while a
dense outer shell forms based on selective placement of the
two feedstocks. A fluid in the pores provides enhanced heat
transport, reaching to 2,000 W/(m K), or about tenfold over
dense MIM materials.
With respect to optical components, MIM has several suc-
cesses, including fiber optic connectors out of stainless steel.
An example optical MIM part is shown in Figure 7.71. This
flagstaff nose is used as an external connector for high-
Figure 7.71. Fiber optic transceiver external connector for high-
performance system, molded from 17-4PH stainless steel
(photograph courtesy of Advanced Materials Technologies)
Figure 7.72. 2 g optical-component seat for use in communication
systems formed from 304L stainless steel (photograph courtesy of
Hunan Injection High Technology)
 Table of Contents
Chapter 7: Applications

performance fiber optic transceivers. It is formed using 17-

4PH stainless steel. The complex design, which is almost
impossible to form as a one-piece component via conventional
processes, enables it to achieve optimum electro-magnetic
shielding. Strength is achieved in a sinter-hardening cycle and
a final 0.5 m gold coating is added after sintering. The pro-
ducer, Advanced Materials Technologies (Singapore), reports
a cost reduction of 40% using MIM at an annual production
volume of 250,000 per year. The final properties are as
follows:
final density = 7.5 g/cm3 (3% porosity)
yield strength = 731 MPa (106 ksi)
ultimate strength = 897 MPa (130 ksi)
elongation = 8%
hardness = 27 HRC
Another example MIM optical communication component
is pictured in Figure 7.72. The device is 2 g in mass and is fab-
ricated from 304L stainless steel. Dimensions are held to 0.02
mm during production at 50,000 per year. This example is from
Hunan Injection High Technology (Changsha, Hunan, China).

SPORTING
First use of MIM in sporting was in the early 1990s with
the production of running spikes for track shoes. MIM track
spikes from titanium alloyed with iron with strength of 887
MPa and 4.8 g/cm3 density were used by sprinter Leroy Burrell
in 1991, leading to a 100 m time of 9.88 seconds. Subse-
quently, sporting-shoe companies have investigated MIM for
cleats and spikes, including for baseball and golf shoes. In par- Figure 7.73. Titanium putter used to demonstrate titanium MIM,
allel, there were efforts to fabricate darts and other weighted and the ability to form golf clubs: (a) bottom view and (b) top view
sport devices by MIM, including the sole plate for golf clubs. (component courtesy of Epson Atmix)
Callaway, Taylor Made, Cobra, Never Compromise, Cleveland
Golf, and others relied on tungsten weights to adjust the center
of gravity.
In the 1990s there were several efforts to form golf clubs
via MIM. This included putters, sand wedges, and drivers. The
materials included stainless steel, titanium, and tungsten. Fig-
ure 7.73 shows a putter fabricated from titanium by Epson
Atmix (Aomori-ken, Japan). Although the potential for using
MIM in golf is exciting, widespread MIM production has been
elusive. One early success was a putter designed by Ping that
reached production levels of 66,000 per year. However, at $6
each MIM was unable to match investment casting costs.
Another goal has been to use MIM to form hollow driver
golf clubs at a target cost of $15 each. Wedges and other flatter
structures are targeted at costs in the $8 range. Several com-
panies have participated in demonstration trials. Technical fea-
sibility has been demonstrated, but economic benchmarks
have been missed. Meanwhile, one operation is working on
golf clubs injection molded from ceramic and infiltrated with
Figure 7.74. Tungsten fishing weights fabricated by MIM with
molten aluminum to form a high-stiffness, low-weight struc- internal channel and identification information (components
ture. This will provide a lower density than titanium, but at a courtesy of Tungsten Heavy Powders and Zhuzhou Luke's Metal
higher expense. Powder Products)

129
 Table of Contents
Chapter 7: Applications
Fishing weights shifted to MIM when lead was banned.
Most are based on tungsten, some are simply bonded tungsten
(molded but not debound or sintered), and others are sintered.
Figure 7.74 is a picture of four tungsten fishing weights made
by MIM. Note the hole, indent, and internal lettering included
in each piece. This is a 22 g mass.
Hunting knives, arrow tips, and other sharp objects used
outdoors are formed using MIM. Figure 7.75 shows a novel
hunting knife. A 17-4 PH stainless steel arrow tip is shown in
Figure 7.76. The latter is fabricated by Smith Metal Products
(Lindstrom, Minnesota, U.S.) for Trophy Taker (Plains, Mon-
tana, U.S.) to a final 7.6 g/cm3 density. Most of these rely on
stainless steels and the designs initially were copies of tradi-
tional shapes; more recently the blades have included holes
to lower mass and reduce cost. In traditional machined blades,
the addition of holes increases manufacturing time and cost,
but in MIM the holes mean less powder is used and this
reduces cost.
A related development is the production of practice bullets
using MIM. The efforts are driven by a desire to eliminate lead
Figure 7.75. MIM hunting knife with several holes used to reduce
mass (photograph courtesy of Powder Injection Moulding
International)
Figure 7.76. MIM hunting-arrow head and the component parts
(photograph courtesy of Metal Powder Industries Federation)
130
from target shooting areas, leading to various steelpolymer,
copperpolymer, and tungstenpolymer compositions. Tests
have been conducted with three variantspowderpolymer
bullets, debound and presintered bullets (not very strong), and
fully sintered bullets. The first choice is the lowest cost. Initial
orders in the area are hundreds of millions per year, and if
accepted by consumers could reach billions of parts per year.
TELECOMMUNICATIONS
MIM components are widely used in telecommunication
devices for everything from cellphone-vibration weights to
microelectronic mounts. The vibrator weights are a lower-
technology product that reached enormous production quan-
tities at very low prices. An example weight is shown in
Figure 7.77, fabricated from tungstennickelcopper by
Advanced Technology and Materials (Beijing, China). These
weights are selling for less than $0.05 each. Other telecom-
Figure 7.77. Eccentric tungsten heavy alloy cellphone weight to
provide vibration when an incoming message arrives (component
courtesy of Advanced Technology and Materials)
Figure 7.78. Electronics hardware fabricated by MIM where a
combination of heat dissipation and electromagnetic shielding are
required in this 32 g housing (photograph courtesy of NetShape
TechnologiesMIM)
 Table of Contents
munication applications for MIM include mounting hardware,
hinges and latches, antenna mounts, housings, and compo-
nents associated with cellphone transmission stations. The
latter are tungstencopper compositions for high thermal
conductivity.
Electronics hardware for telecommunication systems is
another application for MIM, where the combination of heat
dissipation and electromagnetic shielding favor metals. Figure
7.78 is a picture of a 32 g (32 mm by 24 mm by 10 mm or
about 1.3 in. by 1 in. by 0.4 in.) housing formed by MIM. The
thinnest section is below 1 mm wall and the thread is added
as a secondary operation. This part was fabricated by Net-
Shape TechnologiesMIM (Solon, Ohio, U.S.) in 4-cavity
tooling at production rates exceeding 100,000 per year.
Fiber optic connectors rely on MIM stainless steel, but
ceramic versions are more popular. Various alignment devices
Figure 7.79. An optical transceiver housing molded from 17-4 PH
stainless steel. It is used in ultrahigh speed transceiver for
networking and telecommunication systems (photograph courtesy
of Advanced Materials Technologies)
Figure 7.80. Large telecommunication mounting plates formed by
MIM with gold coating (photography courtesy of Advanced
Materials Technologies)
Chapter 7: Applications
are formed using MIM, such as the clip shown in Figure 7.79.
It is fabricated out of 17-4 PH stainless steel, with gold plating
after sintering, by Advanced Materials Technologies (Singa-
pore) for Agilent Technologies (Singapore). The stainless steel
provides rigidity and heat dissipation.
Most of the opportunities depend on expansion of cell-
phone networks. The base stations are heavy users of high-
thermal-conductivity materials such as tungstencopper. The
future worldwide build of telecommunication systems will
continue to drive this application area into higher production
quantities with lower unit costs. Heat plates, such as shown
in Figure 7.80 after gold plating, are an example application.
Variants included high-thermal-conductivity copper, porous
copper with internal heat pipes, tungstencopper with lower
thermal expansion, and Invar with electroplated gold coatings.
SUMMARY
For those people long focused on powder metallurgy, one
of the common remarks is on how MIM differs from the stan-
dard presssinter powder metallurgy process. The differences
are very evident with respect to almost all aspects:
MIM powders are small and sphericaltraditional powder
metallurgy relies on much larger, irregular powders
MIM materials are dominated by stainless steelstradi-
tional powder metallurgy largely relies on iron and steel
compositions
MIM relies on a high polymer content (5 to 10 w/o)
traditional powder metallurgy relies on just 1/10 the
polymer content
sintered MIM densities are comparable to cast metals
traditional powder metallurgy usually is only 85% of
cast
properties in MIM rival handbook valuestraditional
powder metallurgy properties often are 50% of hand-
book values
MIM applications are almost everywheretraditional
powder metallurgy is focused on automotive applica-
tions
The dispersion of application fields is both good and bad
news. The good news is that MIM tends to ride the offsetting
economic ups and downs of various markets. So when one
area stalls, MIM has footholds in other areas that grow. During
the terrible economic times of 200809, fear drove firearm
sales and, unlike other fields, MIM experienced a significant
growth. The bad news is simply the same situation. It requires
much effort to stay abreast of several markets. When all of the
eggs are not in one basket, then considerable work is required
to keep up with such a range of customers, markets, and
design swings.
As a final aspect in MIM applications it is noted that sev-
eral artistic items are being fabricated using the technology.
Figures 7.81 to 7.83 show that not everything is precisely
engineered for function, but some of the MIM products are
131
 Table of Contents
Chapter 7: Applications
Figure 7.81. Ceramic pots with decorative MIM top handles
(photograph courtesy of Taiwan Powder Technologies)
Figure 7.82. Limited-production chess set created out of bronze
using a MIM approach (photograph courtesy of Mark Pilato)
Figure 7.83. Polished stainless steel chopstick rests made by MIM
(photograph courtesy of Taiwan Powder Technologies)
132
attractive. In this category are designer eyeglass logos, lug-
gage buckles and clasps, and various medallions. Clearly,
MIM is a robust forming technology with many applications.
RESOURCES
N. De Cristofaro and A. De Nicolo, A Case Study of MIM
Components for Automotive Applications, Second European
Symposium on Powder Injection Molding, 2000, European
Powder Metallurgy Association, Shrewsbury, UK, pp. 215219.
R. Dwivedi and J. Fauerbach, A New Method for Production
of Micro Porous Metals, P/M Science and Technology Briefs,
2001, vol. 3, no. 2, pp. 2225.
R.M. German, Medical and Dental Applications for Micro-
miniature Powder Injection MouldingA Roadmap for
Growth, Powder Injection Moulding International, 2009,
vol. 3, no. 2, pp. 2129.
J.L. Johnson, Mass Production of Medical Devices by Metal
Injection Molding, Medical Device and Diagnostic Industry,
2002, November, pp. 4853.
M. Kimura, Powder Metallurgical Manufacture of a Rocker
Arm for Variable Valve Timing Mechanism by MIM Process,
Ninth Case Studies on New Product Development, Japan Pow-
der Metallurgy Association, Kyoto, Japan, November 2000,
pp. 3544.
M. Kneuwer, K.H. Wichmann and H. Minhardt, Processing
of Tungsten and Molybdenum Copper Housing for Microelec-
tronic Packaging via Metal Injection Moulding, Second
European Symposium on Powder Injection Molding, 2000,
European Powder Metallurgy Association, Shrewsbury, UK,
pp. 221228.
T. Moritz and R. Lenk, Ceramic Injection Moulding: A
Review of Developments in Production Technology, Materials
and Applications, Powder Injection Moulding International,
2009, vol. 32, no. 3, pp. 2334.
B.N. Nobrega, W. Ristow and R. Machado, MIM Processing
and Plasma Sintering of Nickel Base Superalloys for Aero-
space and Automotive Applications, Powder Metallurgy,
2008, vol. 51, pp. 107110.
J.A. Sago, H. Chen, M.W. Boradley and J.K. Eckert, Devel-
opment of Cobalt-Chromium MIM Alloys Suitable for Med-
ical Device Applications, Advances in Powder Metallurgy
and Particulate Materials2009, Metal Powder Industries
Federation, Princeton, NJ, part 4, pp. 111117.
M. Sonderegger, B. Unternahrer and A. Oberli, Application
of the MIM Technology for SWATCH Irony Watch Cases and
Watch Components, Second European Symposium on Pow-
der Injection Molding, 2000, European Powder Metallurgy
Association, Shrewsbury, UK, pp. 235242.
 Table of Contents
Chapter 7: Applications

T.A. Tomlin, Metal Injection Molding: Medical Applica-

tions, International Journal of Powder Metallurgy, 2000, vol.
36, no. 3, pp. 5357.
D. Whittaker, Powder Injection Moulding Looks to Automo-
tive Applications for Growth and Stability, Powder Injection
Moulding International, 2007, vol. 1, no 2, pp. 1422.
R. Zauner and M. Scheerer, Qualification of Material
DIN1.4404 (316L) for Liquid Hydrogen Storage Applica-
tions, Powder Injection Moulding International, 2007, vol.
1, no. 4, pp. 6062.

133
Table of Contents
 Table of Contents

Chapter 8
MARKET CONSIDERATIONS

Outline Industry Structure

Overview
Historical Shifts
Production Levels
Sales Breakdown
Competitive Technologies
Financial Performance
Geography
Materials
Applications
Emerging Opportunities
Applications
New Concepts in MIM
Industry Changes
Future Trends
Resources

INDUSTRY STRUCTURE
Overview
An outline of the MIM industry structure and interactions
is diagramed in Figure 8.1. The central actors in the industry
generally focus on one of the following sectors:
ingredient supplierspolymers, powders, and ingredi-
ents for either self-mixing or intermediate feedstock pro-
duction; globally there are approximately 40 firms that
provide most of the MIM powders, although about 400
firms supply metal powders of various chemistries, par-
ticle sizes, particle shapes, and purities; for example, in
titanium about 4 companies out of 40 suppliers make the
powders needed for MIM
feedstock productionpurchase raw ingredients and for-
mulate mixtures for sales to molding firms; globally
there are 12 feedstock suppliers
molding firmsboth custom and captive molders that
total to more than 260 MIM operations; about one-third
are captive and make parts for themselves, but many of
the captive firms also perform custom fabrication, so
83% of all parts production is categorized as custom Figure 8.1. Diagram of the industry relations in the MIM field.
manufacturing The central activity is at the parts fabricator, which could be either
thermal processingfirms that own sintering furnaces a custom or captive operation

135
 Table of Contents
Chapter 8: Market Considerations

and debinding equipment that provide toll services; cur- stainless steels
rently 5 firms are active in this area and most are associ- special ammunition that included birdshot, armor pierc-
ated with furnace fabricators ing and frangible bullets
designerslargely systems designers at large multina- turbocharger rotors out of nickel alloys for small sports
tional firms that intersect with the MIM industry, but a cars
few independent designers are available to handle ad hoc Since each of these MIM operations had a single field of
projects focus, little was done to grow that portion of the industry.
equipment suppliersfirms that design and fabricate However, in more recent years growth in MIM has come with
custom furnaces, molders, mixers, debinding systems, the shift to custom molding which services a wide variety of
robotic systems, and other capital devices such as testing applications. The custom molding firms have joined together
devices; the majority of molding machine sales are from in efforts to advance the industry, via collaborative marketing
6 firms, furnace sales are from 8 firms, mixer sales are efforts, promotion of material standards, publicity via annual
from 4 firms, so about 20 firms constitute the key equip- awards, and sharing of business data. Although declining,
ment suppliers captive molding still remains an important part of the MIM
consumablessuppliers of process atmospheres, chem- industry.
icals, molds, heating elements, and sintering substrates
adjunctsincluding researchers, consultants, design Historical Shifts
advisors, conference organizers, trade association per- As an introduction to market progress in metal powder
sonnel, magazine editors, and patent attorneys injection molding, Table 8.1 compares the global characteris-
Component production is the central activity. It is split tics as captured in 2000 and 2010. This table includes data
between internal and external products, referred to as captive from start-up ventures as well as long-established manufac-
and custom molders. Likewise it is supported by two parallel turing firms. For many years the MIM field sustained com-
supply routes, depending on the decision to self-mix or to pur- pound annual sales growth of 22% per year with a 34% per
chase premixed feedstock. An example captive molder would year increase in the number of operations. In recent years the
be a firearms company that uses MIM to fabricate some of the growth rates have become more modest and have stabilized
safety, trigger, or sight components. On the other hand, custom near 8% per year in North America, but continue at a 30% per
molders also can make these same components, but just as year pace in Asia.
well can be involved in several application areas as deter- Although a large number of companies are active in MIM,
mined by their customer base. at any one time some are simply in an evaluation mode. These
As outsourcing increases for multinational firms, custom are designated as level 0 firms. Characteristically they consist
fabrication grows. Accordingly, MIM fabrication facilities of a small team (two or three people) and one or two molding
owned by large firms such as IBM, AMP, and GTE, which machines, usually purchase feedstock, and might even rely on
were early adopters, shifted to purchasing components from toll sintering. After initial exploration, many such efforts are
captive molders focused a variety of application areas. Some terminated or production is transferred to an outside vendor.
of the early captive applications included the following
examples: TABLE 8.1. OVERALL PERSPECTIVE ON THE
orthodontic brackets out of stainless steel or cobalt WORLDWIDE MIM INDUSTRY
chromium alloys
feature 2000 2010
business-machine components for postage meters and
typewriters number of companies 400 265
watch components including weights, bezels, cases, total annual sales $500 million $1,000 million
total number of employees 3,500 6,000
bands, and clasps mean MIM employment 9 22
camera components that included switches and buttons per operation
firearm steel parts such as trigger guards, sights, gun bod- annual sales divided by $143,000 $167,000
ies, and safeties total employment
carbide and tool-steel cutting tools such as wood router number of installed molding 560 1,000
machines
bits, end mills, and metal-cutting inserts number of installed 430 800
electronic packages for electronic systems from glass sintering furnaces
metal sealing alloys number of installed mixers 236 322
personal care items such as hair trimmers from tool steel typical annual sales per $892,000 $1,000,000
medical hand tools for special surgical operations molding machine
MIM production for internal 31% 17%
rocket engines from specialty materials such as niobium use
automotive airbag-actuator components from hardenable

136
 Table of Contents
This is evident by the decrease in number of MIM firms from
2000 to 2010 while the overall field grew.
Production Levels
Most of the firms advance through stages. Initially the
focus is on evaluation of the MIM technology to assess for
opportunities. Some of the evaluation efforts advance to pilot
production, leading to a small facility designed to satisfy a
few special applications or customers. Thus, in metal powder
injection molding there is a natural progression starting with
the evaluation efforts, moving to pilot production, the next
step being small-scale production (characterized by single-
shift operation), and eventually graduating to full-scale pro-
duction characterized by round-the-clock production. In short,
these are segmented as follows:
level 0evaluation mode, involving a small technical and
business effort designed to learn more about MIM;
usually includes some in-house trials, not necessarily
designed for production, but for learning the process,
testing new concepts and designs, and often to pro-
vide marketing and engineering with a sound basis
for dialog with custom fabricators; there are no
measurable sales and no profits (effectively a net loss
or R&D charge at this stage)
level 1pilot mode, represents a company with some capa-
bilities, where they are examining possible products
and developing new customers; usually equipment
has been purchased and molds are typically single
cavity; often characterized by one or two molding
machines and one or two furnaces; profitability is
marginal as the operation learns key steps in stabi-
lizing its process and qualifying early products
level 2small-scale production mode, represents a company
that has gone into production, usually with a single
shift, maybe graduating to some multiple-cavity
Figure 8.2. Schematic illustration of financial performance as a
MIM firm grows from level 0 evaluation mode into full-scale level
3 production. The uneven profit comes from the incremental
burdens that arise with each step in manufacturing growth. A good
example is when a new furnace is installed, the expenses increase
until the unit is made productive
Chapter 8: Market Considerations
molds, typically based on a few customers or even
captive products; often production quantities are low,
so equipment utilization is in the 10% to 30% range
(8 hours per day for 5 days per week, or 40 out of
168 hours); feedstock use in typically less than 5 tons
per year and might be purchased to avoid mixing
equipment purchase
level 3full-scale production mode, represents a company
with a history of constant parts production, charac-
terized by production reaching 24 hours per day up
to 7 days per week; facility utilization often in the
80% or higher range; other characteristics show from
80 to several hundred employees and sales exceeding
ten million dollars per year; this group dominates
sales and profits
Figure 8.2 illustrates the typical maturation cyclic behavior
among these four levels by plotting expected profits vs. sales
(in thousands of dollars) from start-up to full production. Most
firms start in level 0 with a fixed annual expense of $500,000
or more, and rise to break-even with about $2 million in cap-
ital investment and $500,000 in annual sales. At that point
they transition to a level 1 pilot effort. Unfortunately,
increased costs hinder profitability even during sales growth.
Every time staff or equipment is added to increase production,
the costs incurred offset the sales gains.
As business expands, eventually profitability emerges.
However, there can be profit dips even during growth as extra
equipment and extra production capacity are added (cash flow
can be an issue due to the time lag between equipment pur-
chase and productive operation of the equipment due to long-
duration installation and training cycles). Finally, sustained
profitability occurs in level 3.
As should be evident, MIM operations range from large,
automated factories with up to 60 molders, to small shops with
little more than a Web page and a single molder. In some cases
the start-ups are plastic molders that purchase feedstock and
outsource debinding and sintering. They shift between plastic
molding and metal molding as required by sales. As a conse-
quence, across the MIM industry a large disparity exists in
productivity, capacity, and capabilities.
Figure 8.3 plots the relative population of operations at
each level while Table 8.2 gives a statistical sense of the oper-
ating characteristics at each level. Here Prados Principle is
evident, where 20% of the MIM operations dominate the
industry, producing 80% of the parts. Further, the top segment
employs about two-thirds of the workforce, while owning a
vast majority of the production capacity.
Sales Breakdown
Globally, MIM generated $860 million in component sales
for 2009 and entered 2010 with more than 8% growth rate.
The growth history for global MIM sales is traced in Figure
8.4. This figure gives both the total powder injection molding
137
 Table of Contents
Chapter 8: Market Considerations

TABLE 8.2. COMPARISON OF MIM OPERATIONS AT VARIOUS MATURITY LEVELS (2010 statistics)
attribute level 0 level 1 level 2 level 3
characteristic evaluation pilot small-scale full-scale
% of operations 14 27 29 30
% of sales 0 4 7 89
average sales per employee per year nil $60,000 $125,000 $250,000
% of total powder use nil 10 7 83
% of total employment 7 12 12 69

MIM and PIM tend to track similar growth curves.

Figure 8.3. Current partition of MIM operations based on
production level, where the characteristics at each level are
described in the text. Of the approximately 260 firms active in
MIM, about 30% are operating at level 3 and they dominate
all aspects of the industry statistics
Figure 8.4. Historical annual sales for powder injection molding
(PIM) as the upper curve, and for metal powder injection molding
(MIM) as the lower curve, traced from the first statistical survey
in 1986
field (upper curve) and the MIM field (lower curve) from the
mid-1980s when statistics were first assembled. During peri-
ods of intense aerospace programs the PIM values shoot up
since ceramic casting cores are in great demand. Otherwise,
A significant difficulty arises with respect to determining
sales due to two factors. First, sales data are closely guarded
by MIM firms. Second, the value of internal products is diffi-
cult to estimate. Internal transfer values may have little relation
to trade prices. Accordingly, there is an uncertainty in sales
values when trying to capture this portion of the industry.
Across the industry, the production of raw materials for
molding (powders, binders, and feedstock) accounts for about
15% of the commercial revenue, or about $150 million in
2010. Powder consumption is approaching 8,000 metric tons
per year. Feedstock is sold at a significant premium over the
cost of the raw ingredients. Even so, global feedstock sales
are probably $50 million out of the $150 million. But less than
30% of MIM parts production is performed using purchased
feedstock.
The majority of larger operations produce feedstock inter-
nally to lower cost. Additionally, feedstock was unavailable
when many of the early (and now large) firms entered the
field. On the other hand, newer MIM firms have avoided the
feedstock issues by purchasing precompounded materials.
When feedstock is produced in-house, there is still the neces-
sity to purchase powder and polymers, and a further need to
purchase the proper mixing device.
Capital equipment for MIM includes mixers, molders,
debinding devices, robots, and sintering furnaces. The annual
sales for these devices undergo large variations, partly due to
industry growth and used equipment availability. On average
the annual capital investment ranges near 8% of sales, or about
$80 million. Some of this is for new operations, but the major-
ity is for plant expansion or modernization (automation).
Likewise, the R&D investment in MIM is running under
5% of sales. Most of this is for process improvements, such
as increased yield or cost reductions, so that work is per-
formed within the production facilities. A few university pro-
grams support the industry, largely through government
funding, and these focus on new materials, improved proper-
ties, computer simulations, and novel applications. For exam-
ple, porous titanium scaffolds for biological tissue ingrowth
is a very active research topic.
Toll processing, debinding and sintering, is a small activity
with a few firms offering this service. These might be for
materials where proper furnaces do not exist, or for cases

138
 Table of Contents
where large orders exceed in-house capacity, or as part of
equipment evaluation prior to purchase. Less than a dozen
firms support toll processing, but total sales in this area are
relatively small, probably well under 1% of MIM sales.
Today more than 360 firms practice PIM, and about 80%
of those practice MIM. Most of the firms have multiple appli-
cation areas. Of these firms, many are beyond pilot produc-
tion, but only about 100 use their facility 24 hours per day.
Figure 8.5 estimates of the sales distribution for MIM opera-
tions. This bar chart has a logarithmic scaling to the annual
Figure 8.5. Distribution in MIM company size based on annual
sales volume; note, the x-axis is not linear but jumps by approxi-
mately a factor of 3x for each category. The largest firms are over
$30 million but under $100 million in annual sales. The mode or
most common company size is $1 to $3 million in annual sales
Figure 8.6. Major application areas for MIM based on annual
sales for 2009. Historically, fields such as watch production
dominated MIM in the early years, while medical and automotive
are now large application areas. Industrial and consumer
applications tend to be low profile, but constant, areas
Chapter 8: Market Considerations
sales, jumping by a factor of 3 along the lower axis. As of
2010, there is no MIM firm at $100 million in annual sales.
The largest ceramic injection molding firm reached $91 mil-
lion in sales one year, but subsequently retrenched. On the
other hand, only a few full-time operations do less than
$1 million in annual sales. The majority of the industry
generates sales per year in the $1$30 million range. These
operations tend to focus on a few materials and typically have
40 customers.
With respect to market segments, Figure 8.6 provides a par-
titioning of 2010 sales based on major application areas. In the
1990s, major growth occurred in orthodontic brackets, com-
puter disk drive components, consumer products, firearms,
telecommunication devices, and watch components. By the
late 1990s, consumer items reached very impressive produc-
tion rates, but often the products were short lived, remaining
in production for one to two years and then going out of style.
Today, automotive, industrial, and other longer-life products
have taken up MIM and provide more production stability.
In some application areas MIM is in decline. A good exam-
ple is in computer hard disk drives. On the other hand, a recent
surge has come from automotive and medical applications,
while the firearm and general mechanical devices have
remained strong. Some fields, such as firearms and orthodon-
tic brackets, are nearing maturation while aerospace applica-
tions are still at the early adoption stage. New applications are
emerging in hand tools, plumbing hardware, biomedical sys-
tems, computer heat sinks, medical diagnostic components,
and microelectronics packaging. As a result there is a con-
stantly shifting distribution of applications. This is a barrier
since the small firms have difficulty serving so many markets.
COMPETITIVE TECHNOLOGIES
Competitive technologies are the best benchmark for meas-
uring the value of MIM. Process yield and material utilization
are immediate advantages when compared with subtractive or
machining processes. In replication processes, such as press
sinter powder metallurgy, there tends to be less waste, result-
ing in a lower cost. In summary, the successes for MIM come
when key attributes are realized vs. the competitive technolo-
gies, such as:
overcoming the property limitations inherent to plastics
expanding shape capabilities beyond those possible using
stamping, forging, and fine blanking
exceeding the property and shape-range limitations
inherent in presssinter powder metallurgy
providing a lower cost when compared with machining
providing productivity levels not attainable with other
full-density approaches such as isostatic pressing and slip
casting
avoiding the defects, surface-finish limitations, and
tolerance limits associated with casting
Metallic powders are more expensive when compared with
139
 Table of Contents
Chapter 8: Market Considerations
ingot or bar stock, as might be used for machined components.
Consequently, the raw material cost difference must be offset
by savings through the elimination of machining. Figure 8.7
is a schematic outlining the interplay between component
complexity and production volume for several forming
approaches. This roughly identifies where each forming
approach is most successful. For a given complexity, this plot
suggests the greatest economic merit for MIM comes with
high production quantities.
Advantages for MIM occur at lower quantities when the
Figure 8.7. Conceptual map of production quantity on the y-axis
and shape complexity on the x-axis, illustrating the relative
positioning of MIM with respect to competitive technologies
Figure 8.8. Annual production quantity of a million components
showing how MIM costs becomes most attractive vs. competitive
approaches when there is more complexity to the device
140
material is expensive or difficult to process by alternative
techniques. Titanium fits this niche because it is very difficult
to machine. As an illustration of the underlying cost curves,
Figure 8.8 plots the unit cost vs. shape complexity for a batch
size of 1 million units. It is the lower slope of the MIM curve
that makes it attractive. These partitions between fabrication
approaches are for illustrative purposes, since many factors
impact actual demarcations.
Metal powder injection molding suffers when the dimen-
sional precision is higher than a coefficient of variation of
0.2%. However, if only a few dimensions require final
machining, then MIM can be cost effective in forming a near-
net-shape component. The mold costs about the same as in
plastic tooling, but both are expensive, so the production
quantity needs to be sufficiently large to justify mold cost.
Figure 8.9 illustrates the relative tool cost for MIM and several
alternative net-shape technologies. At the lower production
levels, metal powder injection molding has a disadvantage.
For mass production situations the low piece price is attrac-
tive. Consequently, some firms have innovated means to cir-
cumvent cost barriers at low production quantities, using
lower molding pressures (lower viscosity binders) and softer
tooling. These solutions give production quantities under
10,000 per year. As a benchmark, common high-pressure
molding tools easily produce 100,000 or more components
per year.
FINANCIAL PERFORMANCE
About every year a market report is issued on the MIM
industry, documenting the key actors, applications, markets,
and financial performance. This short introduction is not
intended to displace those reports, but should provide a sum-
mary on the key indices.
In reviewing the financial performance, one complicating
factor is in determining where to assign value. An example of
Figure 8.9. Comparison of typical tooling costs for a variety of
net-shape technologies, for comparison with powder injection
molding
 Table of Contents
TABLE 8.3. CHARACTERISTICS OF THE TOP 25 MIM
OPERATIONS
attribute metric
average employees per site 85
average molders per site 17
average furnaces per site 8 Singapore, and India, with a few smaller operations in
average mixers per site 2 Malaysia and Thailand. The Americas are largely based on
average sales per site $25 million
median annual powder consumption 180,000 kg
per site
median parts shipped per year 23 million
median part weight 6g
median profit as percent of sales 14%
median sale price per piece $0.85 each
median sale price per kg $143
median annual sales per employee $300,000
median annual sales per molder $1.5 million
median annual sales per square meter $6,140
of space ($560 per ft.2)
the problem is illustrated by the orthodontic bracket. The sale
price can vary by 800% depending on where value is assigned.
With a direct fabrication cost of $0.35 each, the industry trade
price would be near $0.65. This is the price a manufacturer
would sell the bracket to a distributor. However, if the same
bracket is fabricated by an internal operation and polished,
packaged, and promoted, then the sale price is several times
higher, near $6 to $10 each to the orthodontists. The price par-
ents pay the orthodontist is significantly higher. For MIM
industry statistics, the evaluation is based on the number of
brackets sold times the trade price. A few firms report produc-
tion rates at more than 100,000 brackets per day. Thus, the
worldwide sales of MIM dental brackets are nominally at $78
million per year, and definitely between $65 million and $100
million per year.
To better appreciate the productivity possible with full-
scale operations, Table 8.3 gives a statistical profile of the
operating characteristics of the top 25 MIM production oper-
ations. This group is two-thirds custom, one-third captive. In
terms of installed capacity, they control 20% of the mixers,
26% of the molding capacity, and 19% of the sintering fur-
naces. From a financial view, they account for 63% of the
annual industry sales, 42% of the industry employment, and
81% of the industry profit.
Productivity is measured by sales per employee and sales
per molding machine. The most telling statistic with respect
to overall performance is the sales per employee. In 2010, the
industry-leading firms approached an impressive $300,000
per employee per year. Ratios such as parts per year per
molder are other good indicators of operational productivity.
Geography
A market breakdown based on geography (Americas,
Europe, and Asia) and materials (metals, ceramics, and
Chapter 8: Market Considerations
cemented carbides) helps understand the technology variants.
Figure 8.10 plots the relative fabrication value in the three
regions based on 2009 sales. Outside these three regions, there
are a few MIM operations in other parts of the world, includ-
ing Israel, Russia, Saudi Arabia, Australia, Brazil, and South
Africa. Asia is dominated by China, Japan, Taiwan, Korea,
activities in the U.S., but include a few in Canada and Mexico.
In Europe, the dominant actor is Germany (53% of opera-
tions), followed by Switzerland, Spain, France, UK, Nether-
lands, Belgium, Czech Republic, Hungary, Ireland, Israel,
Italy, Portugal, and Austria. Much of the new-company cre-
ation is in regions such as India (5 firms) and Brazil (4 firms),
with a few others in Australia, South Africa, Dominican
Republic, Puerto Rico, and Russia.
Materials
With respect to materials, the bar chart in Figure 8.11 gives
the value of current MIM sales by material class. Most of the
stainless steel products are 316L or 17-4 PH, but several other
400 and even 300 grade alloys are in use. The nickel alloy
sales are dominated by higher-temperature alloys for tur-
bocharger components for use in small automobile engines.
Steels are usually low-alloy types, with small contents of
refractory metals such as chromium, molybdenum, or others
that give high strength in a post-sintering heat treatment.
Related to the steels are the ironnickel alloys, some intended
for low-carbon magnetic applications, and others with carbon
for higher strengths after heat treatment. Tungsten, usually
with alloying additions of copper, nickel, and iron, is largely
used in heat sinks, military ammunition, fishing weights, radi-
ation shields, and vibrators for cellphones and similar devices.
With respect to emerging materials, there is much interest
in titanium, and some use of precious metals, such as gold
Figure 8.10. Relative MIM sales by geographic region for 2010.
Historically North America was the first to mature and is still the
leader in ceramic injection molding, but now rapid growth in MIM
is shifting to Asia. North America still is the major consumer of
MIM components. The numbers are the percentages of the total
sales
141
 Table of Contents
Chapter 8: Market Considerations
alloys and sterling silver, as well as tungsten alloys and tool
steels. Technical feasibility is well demonstrated for all of
these materials. Thus, what is probably the largest barrier
would be in the economics. For example, good quality tita-
nium powder for MIM is between $120 and $150 per kg for
spherical powder, while angular hydridedehydride powder
is approximately $80 per kg. Since wrought titanium products
cost $100 per kg, MIM is mostly viable for small and com-
plicated components, but these are sensitive to contamination
and properties. Implants are in production after several years
of seeking approval.
Copper is a relatively new area for MIM. Although MIM
Figure 8.11. Material use based on component sales, showing how
stainless steels dominate the value of MIM products. Stainless
steels also dominate the tonnage data, followed by steels, tungsten
alloys, nickel alloys, and several other materials
Figure 8.12. Example of a novel MIM product, shown here as a
microelectronic package that would be formed from Kovar to
enable glass sealing at the wire holes, while a high-thermal-
conductivity W-Cu is inserted under the semiconductor to dissipate
heat
142
of copper was demonstrated early, the cost of a high-purity
powder delayed applications. Tool steels are similar, in that
early demonstrations were made for applications in woodcut-
ting and metal forming, but powder cost delayed the
applications.
A few electronic alloys are in use via MIM, mainly Kovar
for glassmetal sealing packages, and Invar for low-thermal-
expansion applications.
Composites offer unique growth opportunities, especially
for applications that do not require ductility. The addition of
ceramic particles to a MIM stainless steel feedstock is one
means to form a hard and wear-resistant corrosion-resistant
material, such as desired for shears and hand tools.
Heat dissipation and microwave absorption in microelec-
tronic systems are growing concerns. New microcomputers
contain tens of millions of transistors on a single chip, with
connectors of submicrometer widths. The high computing
power and small chip size generate heat that must be removed
for maximum computing performance. Thermal fatigue aris-
ing from onoff heating (expansion) and cooling (contraction)
is a problem with high-thermal-expansion materials used in
early designs. Accordingly, MIM has emerged as a fabrication
route for the creation of unique designs and materials. For
example, Figure 8.12 illustrates one concept where the outer
package is formed from Kovar for glassmetal sealing at the
feed-through holes, but the embedded heat sink in the middle
of the device is molded from W-Cu. In the optimal applica-
tion, computer-chip mounts have been fabricated with a ther-
mal conductivity over 220 W/(m C) and a thermal-expansion
coefficient of 6 ppm/C. Only a few useful combinations of
thermal expansion and thermal conductivity exist. From a cost
perspective, W-20Cu is probably the most suitable material,
but has a high density. Surprisingly, copper solutions have
been accepted by many firms for the control-system heat dis-
sipaters in hybrid vehicles, including novel heat pipes and
vapor chambers.
Applications
The metal powder injection molding field enjoys applica-
tions in a broad array of fields. But this is also a difficulty,
since MIM as a technology is relatively small and new, so it
is very taxing for the industry to support and market to a wide
variety of uses.
Generally, markets for MIM can be divided into the fol-
lowing main categories:
automotive: engine timing components, interior switches,
sensors, fuel injectors, turbocharger vanes, brake calipers,
heating-system controls, steering components
consumer: watchcases, watch bands, camera switches,
eyeglass frames, eyeglass hinges, luggage buckles, cell-
phone hinges, vibration weights, toothbrush gears
dental: endodontic tips, orthodontic brackets, hand tools,
cleaning tools, implants
 Table of Contents
electronics: disk drive parts, computer latches, electronic
packaging, wire-bonding tools, sensors, controls
firearms: trigger guards, sights, sight mounts, triggers,
handgun bodies
hardware: cutting tools, hand tools, fasteners, hinges,
locks
industrial: valves, fittings, magnetic sensors, connectors,
solenoids, cutters
medical: surgical tools, implants, biopsy tools, retractors,
scalpels
military: gun mounts, bullets, projectiles, automatic
firearms, fuses, timing devices
other: aerospace, business machines, jewelry, chemical
devices, sporting goods
Figure 8.13 is the partition by firm specialization, based on
these areas, showing about half the firms focus on industrial,
medical, or electronic products. Note that in parts of the world
the firms were set up to produce military products, typically
firearms and ammunition.
Based on sales, the three largest segments are products for
industrial, automotive, and consumer applications. These are
followed by medical, dental, electronic, firearms, computer,
hand tools, sporting, watches, and so on. These partitions are
not uniform by region, since Europe is more active in watch
and automotive components, North America leads in medical,
dental, and firearm components, while Asia is larger in com-
puter, cellphone, and communication devices.
Firearms remain one of the stable applications for MIM.
These are small, complex shapes (trigger guards, sights, safety
switches, and triggers) formed from steels, heat treated and
surface oxidized to provide black oxide coatings. Typical pro-
duction rates are up to 100,000 units per design and products
often run for 10 years. Orthodontic brackets are one of the
longest-running applications for MIM, with stainless steels
and nitinol (NiTi) and a few nickel-free (cobaltchromium)
alloys dominating. Medical applications have grown over the
past 15 years, with significant gains in quantities and applica-
Figure 8.13. Relative specialization for MIM firms based on the
applications where they focus their efforts. This plot is based on
the number of firms
Chapter 8: Market Considerations
tions, mostly from stainless steels. Fortunately, the tolerances
for medical devices are not as tight as in automotive applica-
tions, and, similarly to firearms, the product life is several
years.
In recent years there has been an intersection between con-
sumer electronics and telecommunications, where MIM is used
for hinges, swivels, weights, disk drive parts, logos, and similar
components. They must be corrosion resistant, so they are
chromium or nickel-plated steel, or formed from stainless steel,
the geometries are very complex, the tolerances are very tight,
and production volumes are highly variable. Industrial appli-
cations, such as in valves, fittings, and connectors, were first
demonstrated in Japan and now are a routine aspect of MIM.
Most of these are ferrous materials. Hand tools emerged over
the past five years, including many plumbing tools, universal
pliers, and electrical-cable-installation devices. These rely on
low-alloy steels, heat treated and coated with a black oxide.
Table 8.4 provides yet another view of the production
activity in 2009, in this case based on material consumption
by application area and geographic region. In recent years the
major growth has been in automotive, medical, mechanical,
telecommunication devices, and consumer products. For
example, between 2003 and 2006, the production volume
changes in these fields were as follows:
automotive 33%
medical 13%
mechanical and industrial components 23%
information systems and cellphones 25%
TABLE 8.4. PERCENT POWDER USE BY MIM
APPLICATION AREA
North
America Europe Asia
automotive 26 60 13
mechanical 39 3 37
(industrial & hardware)
medical 23 23 2
consumer 4 4 44
other 8 10 4
With respect to global metal powder use, the majority of
powder is now devoted to automotive applications (tur-
bochargers, fuel injectors, valve lifters), followed by 21%
going to electronic, computer, and consumer electronic appli-
cations (laptop latches, cellphone hinges, wrist watchcases),
18% for industrial and hardware (valves, fittings, couplings,
hand tools), 5% for medical (surgical tools), and 5% for dental,
aerospace, sporting, and jewelry applications. The powder use
gives emphasis to high-volume, but not necessarily high-value,
applications. Out of the same powder it is possible to form a
few firearm components or thousands of orthodontic brackets,
but the latter has 100-fold higher valuation.
143
 Table of Contents
Chapter 8: Market Considerations
EMERGING OPPORTUNITIES
Much opportunity exists to expand the applications for
MIM, especially if improved dimensional control is achieved
to eliminate post-sintering secondary coining, grinding, and
machining. Currently, as maturation occurs the rate of sales
growth is slowing in parts of the world. Even so, the rapid
expansion of MIM applications in cellphones, computers, and
other consumer and communication devices leads to signifi-
cant upcoming growth. Figure 8.14 provides a forward pro-
jection based on the growth rates in each region up to 2015.
This plot takes the prior sales data and projects forward with
assumed smooth growth.
In the early years of MIM, growth showed very large per-
centage gain, but the sales base was small. In recent years, the
absolute gains are large, but as a percentage of sales the
growth has slowed. Right now the shift to extensive MIM
growth in Asia is driving much of the overall expansion.
A healthy growth comes from the stable base for MIM
where many of the products have long life cycles. Add to this
an abundance of new projects in expanded markets. One
example is seen in the fuel capsules designed for use in fusion
reactors. When commercialized each fusion reactor will use
over 6 million capsules per year. In the U.S. alone there are
already 120 fission power reactors, so assuming a fusion reac-
tor build to this same capacity gives an eventual MIM market
exceeding $500 million per year for the U.S. Nearer term,
metal powder injection molding is undergoing rapid expan-
sion in consumer and automotive products, and this continues
to add considerably to the existing applications.
As mentioned already, growth is not uniform around the
world. Considerable expansion is taking place in Asia. With a
relatively steady growth rate, MIM should pass $2 billion in
global sales before 2020.
Figure 8.14. Sales curve shown earlier in Figure 8.4, now extended
to 2015 based on estimated product gains. The upper line
corresponds to the total powder injection molding field, while the
lower curve applies to MIM
144
Historically, the metals activity in powder injection mold-
ing is about four times larger than ceramics, while ceramics is
about four times larger than cemented carbides. This ratio of
MIM = 4 x CIM
CIM = 4 x CCIM
gives a partition that has changed some over the years, but is a
good reflection of the overall materials fields. Thus, for a
simple rule the global PIM market tends to run 1.25 times the
annual MIM market. Some special opportunities in new
materials arise at the intersection of metals and ceramics,
where particulate composites arise. This is now about 1% of
sales, mostly in SiC-Al, but also in a few other systems. These
are new compositions, not generally recognized by design
engineers, so the sales growth is slow. However, as a differen-
tiation from traditional metal working, machining, casting,
forging, and such, MIM particulate composites should be a
fruitful target.
Applications
Early interest existed in converting existing machined or
cast components into MIM. To avoid the conversion chal-
lenges, effort is required to demonstrate benefits prior to prod-
uct specification. Shape complexity without resorting to
secondary operations provides the largest gain for metal
powder injection molding. Secondary steps are common in
the technology, yet often are very expensive. Eliminating the
expense provides many new application opportunities.
Historically the user community shifts to MIM when there is
at least a 30% price reduction. To deliver these savings
requires more dimensional precision, automated production,
and faster tool generation. In these areas, technical feasibility
has been demonstrated and will advance as industry focuses
on core issues.
For example, several automotive engine components await
improved dimensional precisionfuel injectors, turbocharg-
ers, rotational sensors, valve lifters, and devices in airbags,
braking systems, and even sparkplugs. A good example is in
the fuel injectors for gasoline and diesel systems. Initial
designs require a high hardness, excellent wear resistance, and
corrosion resistance to fuel attack. Tolerances on the plunger
are in the 20 m range (0.0008 in.). Grinding after sintering
is a costly step. A significant cost reduction comes from sin-
tering to final dimensions.
Figure 8.15 is a comparison of growth prospects for several
application areas. Future applications for MIM will leverage
growing technology acceptance and apply the technology to
more demanding applications. Two near-term options are for
broad use in aerospace and biomedical applications. Although
many demonstrations have been made for aircraft and rocket
components, the technology has only qualified a few compo-
nents for the most demanding applications. Now several firms
are making significant progress in qualifying nickel superal-
loys and titanium alloys through the extensive testing required
 Table of Contents
for acceptance. Likewise, biomedical implants such as knees,
hips, shoulders, and other body parts are formed using powder
injection molding, but most in use are for short-duration
implantation. However, significant effort is taking place to
expand the implants to heart valves and other longer-term
devices. At the same time, new standards are emerging from
bodies such as ASTM for these MIM materials from titanium.
Microminiature components are a new target for metal
powder injection molding. It follows from the base established
by the production of orthodontic brackets. Some of this is
showing up in components for cellphones, cameras, eye-
glasses, jewelry, and watches. Other uses are in digital
systems, biomedical sensors, genomic testing, implanted
transducers, telecommunication devices, optoelectronic
devices, dental implants, notebook computer latches, and
replacement bones for the human ear.
One novel use for powder injection molding is in the for-
mation of small high-feature units, such as arrayed posts of
slots in ceramics. The metalized and assembled units form
ultracapacitors used to store electrical energy, useful in power
tools, automobiles, buses, military vehicles, computers, and
cellphones. As one firm operating in this area claims, future
automotive power will derive from a suitcase-size unit that
will power the vehicle for days at a time. Each suitcase is com-
posed of thousands of stacked injection molded arrays with
microminiature features.
Another large application for MIM is intentionally porous
materials. Beyond filters, several applications arise in biomed-
ical systems. For example, porous titanium scaffolds for tissue
ingrowth are under testing. The elastic modulus of titanium is
much higher than that of human bone, but if the titanium is
fabricated with large pores and a final density near 40%, then
it is nearly a perfect match in strength and elasticity of bone.
Figure 8.15. Relative growth rates for PIM and MIM production
distributed over several of the commonly discussed application
areas
Chapter 8: Market Considerations
Considerable research is going into the production of distended
metals with surface functional ceramic coatings for implants.
The pores are generated by mixing an insoluble plastic particle
into the feedstock. The amount of added plastic determines the
final porosity and the size of the plastic sets the pore size.
New Concepts in MIM
Future opportunities are emerging from ongoing R&D
efforts and although several might not progress to commer-
cialization, it is still curious to see how many new inventions
are arising based on MIM, including the following:
stents that are absorbed into the human body over time,
removing the tendency of current stents (treatments to
stabilize arteries, for example) to form blood clots; read-
sorbable stents are formed using alloys that dissolve into
blood over time and already a magnesium MIM alloy has
been patented for this application
ultra-high-thermal-conductivity composites (for exam-
ple, copperdiamond) for heat sinksdemonstrations
with 580 W/(m C) thermal conductivity have been made
by MIM for use in supercomputers, high-end servers,
phased-array radar, military electronics, hybrid-vehicle
control systems, gaming computers, and other applica-
tions involving high-performance computing
vapor-chamber designs, typically from copper, where a
porous internal chamber act as a heat pipe to evaporate a
working fluid for rapid heat transport from electronics,
with early applications in hybrid-vehicle control systems,
gaming computers, and servers
LED heat sinks where copper arrays are used to mount
the semiconductor, with reports of 100 g arrays with
costs as low as $0.75 per mount
microminiature MIM for medical minimally invasive
surgical tools, involving very small components for end
manipulators, such as cutters, grasps, and drug delivery;
most are from stainless steel and example components
are being sold in the range of $2 to $15 each
electronics and related devices also require microminia-
ture MIM components in cellphones, computers, hand-
held devices
implants for a broad array of biomedical applications
tooth posts, ligament alignment, hearing-canal recon-
struction, drug delivery, and heart valves
microarray devices with hundreds to thousands of pins,
posts, or holes for disposable lab-on-a-chip devices used
in blood testing, disease assessment, analysis of DNA to
predict disease, and protein tests; the biochip market is
targeted to reach $3.8 billion in sales by 2013 and con-
siderable research is taking place on how MIM can sup-
port this effort
titanium biocompatible structures, such as for tissue
affixation, implants, surgical tools, tool implants, and
even sporting devices; several firms have focused on this
145
 Table of Contents
Chapter 8: Market Considerations
possibility, and many are relying on hydroxyapatite
titanium composites
hardware from tool steel, such as threading devices for
cast iron plumbing or water pipes, hand tools, valves and
fittings, handles, forming tools, drills, dies
hermetic packages for microelectronics using Kovar to
enable glass-to-metal sealing; some small example MIM
parts when sealed with leads are selling for $30 each
aerospace applications in smaller superalloy bodies such
as IN 625, 718, 723, or Hastelloy X, where high detail,
good surface finish, and shape complexity are financially
attractive
Industry Changes
The MIM industry has established credibility through the
mass production of several small, complex components. The
period of intense new-company formation has ended and now
growth comes from the existing firms making significant
inroads with new customers and markets. Sales growth is
coming with broad acceptance of metal injection molding for
automotive, medical, consumer, and electronic applications.
These are smaller devices with demanding properties.
It is instructive to review the market forecast made in 1990
that identified the industrys growth barriers as follows:
low powder consumption (with little standardization)
long debinding times
long lead times to production (especially for tooling)
poor vendor knowledge of the industry requirements
lack of customized equipment
high powder cost
lack of industry standards
lack of personnel who understand the process (especially
at the customers)
poor access to capital
Today, those barriers are gone, but the lack of industry
coordination (marketing, materials, processes, properties, and
standards) remains as a difficulty. The formation of an inter-
national MIM alliance is one positive step toward correcting
these problems. Most important, in the past few years Powder
Injection Moulding International has become the forum for
sharing news, statistics, research results, and product
announcements in the community.
Processing-cost reductions via automation continue to be
a major need in order to expand into automotive applications.
Powder injection molding is a route to lower-cost, higher-
quality components, especially when compared with machin-
ing. However, MIM is only now becoming sufficiently
sophisticated to take on the large projects. Further, the toler-
ances need to be improved to the 25 m range and even
tighter for the high-payout areas.
FUTURE TRENDS
In recent industry surveys, about 70% of the firms report
146
planned expansion, but 10% are planning to exit the field. The
industry will probably converge on about 200 actors, 10 of
which will be very large. Projections indicate the sales growth
rate will sustain 8% or more annual increase from year 2010
up to 2017. At that point the estimated annual sales will be
near $2 billion.
Growth will continue in several application areas. Many
pilot projects are showing an expanded range of uses. Much
opportunity still lies in converting investment cast stainless
steels to MIM or die cast zinc to higher-strength MIM steel.
The addition of titanium and tool steels to the list of MIM
materials will accelerate future growth, since both are difficult
to process by other net-shape routes.
A growth barrier is the lack of industry visibility and poor
knowledge of the process by the design community. This need
for improved customer education is not being addressed
directly as yet.
To succeed against competitive technologies, MIM must
meet the engineering requirements at the lowest cost (engi-
neering, tooling, capital equipment, and piece cost). As indi-
cated in Figure 8.16, the metal powder injection molding
technology has high marks with respect to the variety of avail-
able materials, component shape complexity, surface finish,
and final product strength, with a range of sizes (from micro-
miniature to a few hundred mm), and excellent productivity.
On the other hand, it suffers because of the high costs of tool-
ing and raw materials.
The successes for MIM lie in the combined attributes that
include:
overcoming the property limitations inherent to plastics
expanding shape capabilities beyond stamping, forging,
and fine blanking
exceeding the property and shape-range limitations
inherent in presssinter powder metallurgy
Figure 8.16. A graphic comparison of relative technology
strengths to give a contrast between powder injection molding and
several other net-shape metal-forming technologies. The
comparison is based on technical attributes of material strength,
component-size range, productivity, breadth of materials,
component complexity, and surface finish
 Table of Contents
Chapter 8: Market Considerations

providing a lower cost when compared with machining M. Sonderegger, B. Unternahrer and A. Oberli, Application
providing productivity levels not attainable with isostatic of the MIM Technology for SWATCH Irony Watch Cases and
pressing and slip casting Watch Components, Second European Symposium on Pow-
avoiding the defects, surface-finish limits, and tolerance der Injection Molding, 2000, European Powder Metallurgy
limits associated with casting Association, Shrewsbury, UK, pp. 235242.
In short, the best places to apply MIM are where plastic
D. Whittaker, Powder Injection Moulding Looks to Automo-
molding would be capable of forming the shape, but plastics
tive Applications for Growth and Stability, Powder Injection
lack the mechanical, thermal, or other properties.
Moulding International, 2007, vol. 1, no 2, pp. 1422.
RESOURCES International Powder Metallurgy Directory 20102011, 2010,
R.G. Cornwall and R.M. German, An Analysis of the Inovar Communications, Shrewsbury, UK.
Powder Injection Molding Industry Global Market, Advances
R. Zauner, A. Altenburger and R. Nagel, New Application
in Powder Metallurgy and Particulate Materials2001,
for MIM of Tungsten: Divertor Modules for ITER, Powder
Metal Powder Industries Federation, Princeton, NJ, part 4, pp.
Injection Moulding International, 2007, vol. 1, pp. 5052.
1116.
N. De Cristofaro and A. De Nicolo, A Case Study of MIM
Components for Automotive Applications, Second European
Symposium on Powder Injection Molding, 2000, European
Powder Metallurgy Association, Shrewsbury, UK, pp. 215
219.
R.M. German, Medical and Dental Applications for Micro-
miniature Powder Injection MouldingA Roadmap for
Growth, Powder Injection Moulding International, 2009,
vol. 3, no. 2, pp. 2129.
R.M. German, Titanium Powder Injection Moulding:
A Review of the Current Status of Materials, Processing,
Properties, and Applications, Powder Injection Moulding
International, 2009, vol. 3, no. 4, pp. 2137.
D. Gonia, MIM Parts for Automotive Applications Produc-
tion and Economic Aspects, Journal of the Japan Society of
Powder and Powder Metallurgy, 1999, vol. 46, pp. 849852.
J.L. Johnson, Mass Production of Medical Devices by Metal
Injection Molding, Medical Device and Diagnostic Industry,
2002, November, pp. 4853.
M. Kimura, Powder Metallurgical Manufacture of a Rocker
Arm for Variable Valve Timing Mechanism by MIM Process,
Ninth Case Studies on New Product Development, Japan Pow-
der Metallurgy Association, Kyoto, Japan, November 2000,
pp. 3544.
T. Moritz, and R. Lenk, Ceramic Injection Moulding: A
Review of Developments in Production Technology, Materials
and Applications, Powder Injection Moulding International,
2009, vol. 32, no. 3, pp. 2334.
J.A. Sago, H. Chen, M.W. Boradley and J.K. Eckert, Devel-
opment of Cobalt-Chromium MIM Alloys Suitable for Med-
ical Device Applications, Advances in Powder Metallurgy
and Particulate Materials2009, Metal Powder Industries
Federation, Princeton, NJ, part 4, pp. 111117.

147
Table of Contents
 Table of Contents

Chapter 9
SUPPLEMENTAL INFORMATION

Outline Overview
Cost-Calculation Approach
Tool Costs
Calculation Approach
Optimization of Multiple-Cavity Tooling
Other Tool-Cost Factors
Example Tool-Cost Estimation
Summary of Important Tool-Cost Factors
Feedstock Costs
Cost of Raw Materials
Operation Parameters
Mixing Costs
Feedstock for Golf Clubs
Feedstock-Cost Summary
Batch-Size Effects
Component Costing
Operational Characteristics
Molding Costs
Debinding Costs
Sintering Costs
Finishing Costs
Tight-Tolerance Effects
Per-Piece Costing
Additional Costs
Golf Club Example
How Much Will It Cost?
Cost Sensitivity
Typical Component Analysis
Component-Mass Effect
Golf Club Example
Bottom Line on Costing
Definitions
Nomenclature
Units and Conversion Factors
Helpful Relations

149
 Table of Contents
Chapter 9: Supplemental Information

Outline Prefixes
Fundamental Units
continued Derived Units
Important Constants
Conversion Factors
Misconceptions
Standards
Patent Sources
Conferences and Publications
Trade and Professional Organizations
Resources

OVERVIEW
A cost model for MIM is generated where a larger number
of parameters are required, but via this model the important
sensitivities become evident. One of the common puzzles is
why pricing varies so much between vendors. Via this cost
model, it is evident that busy facilities have the lowest costs,
so the incremental addition of more business often comes at a
low price. On the other hand, if the facility is too busy, then
inefficiencies become evident. One of the important points
captured in cost modeling is how incremental changes in pro-
duction cause large cost changes. For example, a second shift
must be added or a new furnace must be purchased. The
model presented here is the first comprehensive compilation
of MIM costing.
Besides costing, other underpinnings to MIM are treated
in this chapter. For example, consistent nomenclature ensures
accurate communication between the parties involved in any
project. This chapter provides definitions, conversion factors,
and several facts on the technology, its documentation, the
trade exhibits and conferences, patent situation, and other de-
tails. As a newer technology MIM has reached out to generate
standards for the materials, testing, properties, and specifica-
tions. Through standardization the user community gains con-
fidence with MIM.
COST-CALCULATION APPROACH
Economic viability is critical to sustained growth for a new
technology. An understanding of engineering economics helps
designers to choose between processes or to modify a design
to reduce its manufacturing cost. Statistics show that the de-
sign stage typically accounts for less than 10% of a projects
costs, yet decisions made in the design stage impact the ma-
jority of the production cost. Metal powder injection molding
lacks a well-established cost history, so designers are often
uncertain about how best to use the technology. This slows
growth, even when MIM might have a significant advantage.
Accordingly, this chapter reviews the engineering economics
of MIM to help best utilize this newer process. When coupled
with component-design guidelines, the combination provides
a potent means to design for improved performance with
lower production costs. Throughout this chapter, a sand wedge
golf club head is used to illustrate the procedures.
Two words are defined for this discussion:
manufacturing cost is the cumulative expense incurred
in the production of a product; it includes the direct pur-
chases of raw materials, the expenditures for energy and
labor, and prorated items such as maintenance, depreci-
ation, insurance, and taxes
sale price reflects the value of the component to the user;
it is the amount of money exchanged for the fabricated
component and normally includes the manufacturing cost
and allocations for a return on investment and risk
Ideally, cost and price should be related, but early in a tech-
nology that is not the case. During the 1980s, MIM products
were selling at a price that sometimes was just 25% of the
manufacturing cost and bankruptcy was common. Today,
metal powder injection molding is profitable, meaning that on
average the cost is below the price.
The information presented in this chapter follows engineer-
ing economic analysis as developed in powder metallurgy, ce-
ramics, cemented carbides, and plastics molding. Powder
injection molding depends on a combination of steps and costs
previously encountered in those fields. For the core manufac-
turing steps, unit operation calculations are developed for each
step to accumulate a total fabrication cost. An outline of the
costing model is repeated from earlier and is given in Figure
9.1, while Table 9.1 provides a rough industry-wide cost sum-
mary breakdown for MIM component production. Although

150
 Table of Contents
Chapter 9: Supplemental Information

Figure 9.1. Flowchart showing the basic MIM component-costing Figure 9.2. Detailed outline of the MIM tool-costing model in
model where each manufacturing step is analyzed for the cost flowchart form
accumulation
the time to machine the features for a single cavity, and then
TABLE 9.1. RANGES OF TYPICAL MIM PRODUCTION scales that cost to multiple cavities. Tool cost is calculated
COSTS from the construction time multiplied by a shop hourly rate,
cost item range with additions for profit (before tax, dividend, and interest)
direct labor 10 to 30% and risk aversion. Such an approach does not allow for rework.
feedstock cost 15 to 50% In reality, tool shops are averse to risk when fabricating a tool
tooling and molding 8 to 20% that is different from their prior experience base; thus, a risk-
debinding and sintering 3 to 15% aversion factor is introduced to allow for possible errors. Like-
interest and depreciation 5 to 9%
inspection, secondary operations 10 to 50% wise, business levels impact tool quotations, as evident by the
general, administrative, sales expenses 8 to 20% mapping of cost vs. delivery; contrary to MIM production,
busy tool shops are often more costly and have a longer deliv-
ery time.
there are recognized differences in all of the factors, including
labor, interestingly labor is not a dominant factor; MIM suc- Calculation Approach
ceeds in both expensive and low-labor-cost regions. Figure 9.2 shows the tool-costing flowchart, adapted here
based on values taken from the MIM industry. The first step
TOOL COSTS is to calculate the mold-base cost,
Tool-cost estimates for MIM are obtained using principles
encountered in plastic injection molding. Although many mod- CB = 2 (BT + 0.45 Ad Dd0.4) (9.1)
els exist, the following approach, adapted from Boothroyd,
Dewhurst, and Knight, has proven most accurate for MIM CB is the cost of manufactured mold base including cooling
tools. This model assumes a standard mold base and calculates and ejection parts (units of $)

151
 Table of Contents
Chapter 9: Supplemental Information
BT is the cost of a manufactured standard tool-steel base (units
of $)
Ad is die-plate area (units of cm2), which depends on the part
and molding machine
Dd is the combined thickness of die plates (units of cm), which
depends on part and machine
The factor of 2 In Equation 9.1 accounts for drilling cooling
channels as well as machining inserts, runners, sprues, and
other features that scale with the tool size. For hot-runner sys-
tems, major cost additions are required which are not included
here.
Calculation of the time to construct the tool cavity requires
an estimate for each of the major tool actions:
tS is the estimated hours to form each side pull (units of h)
tL is the estimated hours to form each internal lifter
(units of h)
tU is the estimated hours to form each unscrewing device
(units of h)
nS is the number of side pulls (integer)
nL is the number of internal lifters (integer)
nU is the number of unscrewing devices (integer)
where the cumulative time to form the major tool actions is
given as
tC = nS tS + nL tL + nU tU (9.2)
tC is the cumulative time for forming the major tool actions
(units of h)
Usually, the time to machine an unscrewing device is
roughly twice that for machining a lifter (say 250 h vs. 150
h), and in turn the lifter takes roughly twice the time to
machine a side pull (65 h).
A separate calculation is used for the time to form the cav-
ity and any cores in the base. The first of these is the time to
machine the projected area and the associated ejector pins,
tA = 5 + 2.5 AP0.5 + 0.085 AP1.2 (9.3)
tA is the machining time associated with cutting the projected
area (units of h)
AP is the projected area of part perpendicular to the direction
of mold opening (units of cm2)
The square-root dependence on part area relates to the ejector
pin machining and the 1.2 power dependence relates to the
empirical scaling law observed in machining the projected
cavity.
Next is estimation of the machining time associated with
the geometric complexity, which increases nonlinearly with
the number of features as follows:
tG = 5.83 (CI + CO - 0.1 NR0.7)1.27 (9.4)
152
tG is the manufacturing time to cut the geometric features of
the part (units of h)
CI is the inner complexity of the part (dimensionless)
CO is the outer complexity of the part (dimensionless)
NR is the number or repeating features, such as holes of the
same diameter (integer)
The two complexity parameters are determined as follows:
CI or CO = 0.1 (NS + NH) (9.5)
NS is the number of surface protrusions, bumps, or patches
(integer)
NH is the number of holes or depressions (integer)
Beyond the machined features in the tooling, additional
time is required to machine the parting plane depending on its
curvature or planarity,
tP = FP AP0.5 (9.6)
tP is the time to manufacture the parting plane (units of h)
AP is the part projected area perpendicular to the direction of
mold opening (units of cm2)
FP is a factor depending on the type of curvature (dimension-
less), estimated as follows:
FP = [3NL + (0.45 + NN2)] (9.7)
NL is the number of radial curvatures on the parting line
(integer)
NN is the number of steps on the parting line (integer)
A flat parting line is most typical and least expensive, while
complex parting planes, described by curvatures and steps,
add to the expense.
Tooling tolerances are more stringent than part tolerances,
and in many estimates the tool tolerance is less than 20% of
the component allowance. Accordingly, considerable time is
used to fabricate a highly toleranced tool. The estimate of ad-
ditional machining time attributed to tight tolerances is given
as follows:
tT = m FT tG (9.8)
tT is the manufacturing time associated with the required tol-
erances (units of h)
tG is the manufacturing time associated with the geometric
complexity (units of h)
FT is a dimensionless tolerance factor given as follows:
FT = 0.000167/TR (9.9)
 Table of Contents
Chapter 9: Supplemental Information

where
TR = ( Ti)1/m
the single-cavity cost by a procedure described as follows:
(9.10)
CT = CS NCK (9.14)
TR is the root-mean of the tolerances (units of mm)
Ti is the tolerance on the ith dimension (units of mm)
Ti indicates the product of the tolerances (multiply all m
CT is the cumulative tool-set cost for a multiple-cavity mold
(units of $)
tolerances times each other)
CS is the cost of a single-cavity mold (units of $)
m is the number of specified tolerances (integer)
NC is the number of cavities (integer)
K is a scaling factor associated with second and subsequent
Complexity concepts are a means to approach the cost of the
cavities (dimensionless)
tolerance factor as captured in Equation 9.8. Namely, cost is
directly proportional to the number of tolerances m. Generally
Recent studies on the scaling factor K suggest it is in the range
tighter tolerances increase the tool cost in a logarithmic man-
from 0.70 to 0.74. For most of the MIM situations, a value of
ner, which is approximated by the inverse factor in Equation
0.72 is typical.
9.9. Finally, as the overall geometric complexity increases,
the cost of tighter tolerances adds to the geometric factor cap-
Optimization of Multiple-Cavity Tooling
tured by Equation 9.4, which carries over into Equation 9.8.
It is possible to optimize the economic benefit from the use
Next is an appreciation that the specified tool-surface finish
of multiple-cavity tooling based on these equations. Letting
has an added expense. The time to achieve the specified tool-
NC represent the number of cavities, it is possible to consider
surface finish is estimated as
the tradeoff between more cavities (with an increase in ma-
chining cost) and the reduced molding time. The calculation
tF = FF (tA + tG)(1 + 0.05 YT) (9.11)
of the molding rate per hour MH is covered later, but assume
such a value is known (say in the $30/h range). Then the ben-
tF is the time associated with surface finishing (units of h)
efit from shorter molding time with multiple cavities is cal-
tA is the time associated with the projected area as calculated
culated with respect to the cost of machining the extra cavities,
above (units of h)
giving
tG is the time associated with the geometric complexity as cal-
culated above (units of h)
CM = CS NCK + (NP /NC)(MH / MR) (9.15)
YT is a factor (0 for no and 1 for yes) to indicate texturing, let-
tering, or insignias
CM is the total cost of molding (units of $)
FF is the surface-finish factor (dimensionless), given as
CS is the cost of a single-cavity tool (units of $)
K is the tool cost scaling factor (dimensionless)
FF = 0.125 + 0.06/RS (9.12)
NC is the number of tool cavities (integer)
NP is the production number or number of parts in a
RS is the surface roughness specified on the tooling (units of
production batch per year (integer)
m)
MH is the molding machine cost per hour (units of $/h)
MR is the average molding rate given as the cycles per hour
The factor YT notes whether lettering, insignias, numbering,
(units of 1/h)
or special textures are added, as these add to the overall cavity
and geometric construction times.
A minimum in the combined molding and tooling cost occurs
So far we have focused on the cost of the mold base and
when the first derivative of Equation 9.15, total molding cost
then estimated the time needed to create the required cavity
CM with respect to the number of cavities NC, equals zero. Set-
and detail in that base. To convert to a cumulative machining
ting the derivative to zero and rearranging the terms gives the
cost requires summation of the individual machining times.
optimal number of mold cavities NCo as follows:
The cost of a single-cavity tool is calculated from the con-
stituent times as follows:
NCo = [NP MH / (MR K CS )]1/(1+K) (9.16)
CS = CB + RH ( tC + tA + tT + tG + tF + tP) (9.13)
Normally, Equation 9.16 is rounded down and held to an even
number, so beyond 1 cavity, the normal progression is 2, 4,
RH = is the toolmakers hourly rate for mold fabrication (units
and on up to 32 cavities.
of $/h)
Two fundamental barriers show up in this optimization.
The first is from the molding machine: either the clamp force
The cost of a multiple-cavity tool depends on the dilation of
or the shot size might not be able to handle the optimized

153
 Table of Contents
Chapter 9: Supplemental Information

number of cavities. For example, the peak molding pressure
multiplied by the projected area of each part multiplied by the
number of cavities must be well below the machine clamping
force. The second barrier is associated with the space required
for mold actions, with fewer cavities being possible as the
number of unscrewing devices, lifters, or side pulls increases
per cavity. Two cavities might be possible with two motions,
but only one is possible with three or more motions.
Other Tool-Cost Factors
Finally, there must be consideration of profit, tax, and in-
terest. For shops that have not previously fabricated a similar
tool, there is a risk aversion which necessitates a large safety
margin in the tool cost. For the MIM customer, the final tool
cost is the only element of concern, but for the tool fabrication
shop, underpricing errors can be fatal. Alternatively, conser-
vative costing will lead to high prices and an idle business.
Risk aversion then describes the pricing policy (beyond profit,
taxes, and interest) beyond the cost calculations. For example,
if the tool shop has a goal of 15% profit (before tax and inter-
est), and estimates a 10% possible error in a quote for a new
tool, aversion to risk might lead them to include a 25% margin
in the quotation. Thus, the final mold cost for the customer is
given as follows:
CP = CT (1 + CA) / [(1 - P) (1 - RA)] (9.17)
says that a four-cavity tool would be the lowest project price
for the intended production quantity.
From the above model, the calculated four-cavity tool price
gives a tool cost starting near $23,000 for an hourly rate of
$25 per hour and 20% combined overhead and profit. It in-
creases to a high of $38,000 for $60 per hour and 35% com-
bined overhead and profit. The median quote was $34,870
which corresponds to a 35% overhead plus profit and $48 per
hour rate. The low bid was $27,800, which corresponds to
20% overhead plus profit and $41 per hour. Other combina-
tions would be possible as well. There is evidence that the
high bidder was too busy and would have had to pay overtime
to meet the production schedule, since they had the longest
delivery time. Also, one of the higher bidders had previously
taken on a MIM job and had a bad experience, so they in-
cluded a high risk-aversion factor in their quotation. In other
words, all of the quotations could be explained using reason-
able model assumptions.
Although this tool-costing model cannot predict actual
quotation prices, due to shop-specific cost factors, it still pro-
vides a guide to the cost implications from various decisions
on part features, number of cavities, tolerances, surface finish,
and other factors that might be adjustable during the design
stage of a project.
As a final illustration of tool-cost calculations, assume a
new golf club head is to be fabricated using MIM. The club

CP is the price of the tool set to the customer (units of $)

CT is the cumulative tool-set cost for a multiple-cavity mold TABLE 9.2. COMPARISON OF TOOL PRODUCTION
(units of $) BIDS
CA is a fixed administrative cost factor; 0.15 is a common delivery 7 to 12 weeks
value (dimensionless) mean price $43,134
RA is the risk-aversion factor reflecting uncertainty from lack median price $34,870
standard deviation $18,040
of experience (dimensionless) high/low ratio 2.8
P is the target profit prior to tax and interest per job, usually
a factor between 0.05 and 0.3; industry average profit is
0.07 (7%) although planned to be larger (dimensionless)

Example Tool-Cost Estimation

Using this model, the mathematics of mold-price compu-
tation would seem straightforward, but determining the proper
factors to use in the formulation is a problem. Accordingly,
quoted mold costs are highly variable between tool shops. As Figure 9.3.
an illustration, in one recent round of quotations, the price for A stainless steel
a tool varied as outlined in Table 9.2, which gives the mean, sand wedge golf
club design consid-
median, and standard deviation prices, and the high/low ratio, ered for production
and notes the range of promised delivery times. The 2.8 ratio by MIM. This wedge
between the high and low quotes indicates substantial differ- is approximately 10
ences between vendors who are looking at the same engineer- cm deep, 10 cm tall,
ing drawings. For reference, this component required one side and 5 cm across at
the bottom. Table 9.3
pull and had four surface bumps, four holes, and 47 dimen- gives the data used
sioned features, with a typical tolerance of 0.02 mm, and 0.1 for projecting the
m surface finish. The optimization based on Equation 9.16 MIM tool cost

154
 Table of Contents
TABLE 9.3. SAND WEDGE TOOL QUOTATION INPUT AND CALCULATED PARAMETERS
input parameters
Ad plate area (units of cm2): 729
AP projected area of part perpendicular to mold opening
(units of cm2): 60
BT cost of a standard base (units of $): 1,000
CA fixed administrative-cost factor (dimensionless): 0.15
Dd thickness of die plates (units of cm): 20
K scaling factor associated with multiple cavities
(dimensionless): 0.7
m number of specified tolerances (integer): 60
MH molding machine cost per hour (units of $/h): 30
MR average molding cycles (units of 1/h): 125
NC number of cavities (integer): 1
NH number of holes or depressions (integer): 6
NL number of radial curvatures on the parting line (integer): 0
NN number of steps on the parting line (integer): 0
NP number of curvatures on the parting plane of the mold
(integer): 0
NP production number of parts (integer): 60,000
NR number of repeating features (integer): 0
nS number of side pulls (integer): 2
NS number of surface protrusions, bumps, or patches (integer): 8
P target profit prior to tax and interest (dimensionless): 0.20
RA risk-aversion factor (dimensionless): 0.00
RH toolmakers hourly rate (units of $/h): 55
RS surface roughness specified on the tooling (units of m): 0.2
tL estimated hours to form each internal lifter (units of h): 130
head is a 56 sand wedge, with a MIM mass of 200 g. The
design is shown in Figure 9.3. Sand wedges generally range
near 300 g, but this club head uses a 100 g tungsten heavy
alloy insert to reach 300 g with a lower center of gravity (not
included in the calculations outlined here). Production for
such a design would typically be in the 50,000 to 100,000 per
year range, so only a single-cavity tool is justified. The cal-
culation input and output parameters are given in Table 9.3.
Using a standard mold base, typical tolerances (0.1 mm), 60
specified dimensions, and a 0.2 m tool-surface finish, the
predicted tool price would be $22,525 with no risk factor, 15%
overhead, and 20% pre-tax profit. An engineering charge
might be added along with a risk factor, so the final price
might be quoted closer to $25,000.
If we wanted to significantly reduce this cost, then attention
to the side pulls, depressions, holes, and number of toleranced
features would be the way to generate the most savings. With
judicious reengineering, realizing that most of the dimensions
are not critical and do not jeopardize the function, it is possible
to cut this tool cost by 30%.
Summary of Important Tool-Cost Factors
This introduction to MIM tooling shows that a few features
quickly increase the tool cost. A first point is to avoid speci-
fying multiple cavities unless there is an economic benefit as-
sociated with the total part-production quantity. Next, for
Chapter 9: Supplemental Information
TR root-mean of the tolerances (units of mm): 0.08
tS estimated hours to form each side pull (units of h): 65
tT manufacturing time associated with the tolerances
(units of h): 1
YT indicates texturing, lettering, or other engraved features
(units of 1 or 0): 0
calculated parameters
CB mold base cost (units of $): 4,175
CI inner complexity of the part (dimensionless): 0.6
CM total cost of molding (units of $): 20,050
CO outer complexity of the part (dimensionless): 0.8
CP price of the tool set (units of $): 22,525
CS cost of a single cavity mold (units of $): 15,670
CT cumulative cost for a multiple cavity tool set (units of $): 15,670
FF surface finish factor (dimensionless): 3.125
FP factor for parting plane curvature (dimensionless): 0.45
FT tolerance factor (dimensionless): 0.21
NCo optimal number of cavities (integer): 1
tA time associated with cutting the projected area (units of h): 43
tC cumulative time to form the major tool actions (units of h): 130
tF time associated with surface finishing (units of h): 22
tG time associated with the geometric complexity (units of h): 9
tM total time for machining (units of h): 209
tP time associated with the parting plane (units of h): 4
tS time associated with side pull action (units of h): 65
tT time associated with tolerances (units of h): 1
lower production volumes it might be less expensive to per-
form secondary operations rather than create complicated,
multiple-action toolssuch as adding threads after sintering
instead of using an unscrewing tool component. Depending
on the cost of cutting the threads, as a secondary operation, it
is possible the unscrewing device is not cost justified. Other
ways that may help lower the tool-machining cost is to reduce
the part size and complexity by a reduction in the number of
dimensions, toleranced features, holes, and depressions. Fur-
ther, relaxed tolerances help reduce tool cost. In the case of
the golf club, only two of the features require tight control, so
less stringent tolerancing helps reduce cost. Finally, a low
tool-shop hourly rate greatly lowers the final price.
FEEDSTOCK COSTS
A few materials with the powder characteristics desired for
MIM are widely available, so there is a potential cost advan-
tage if the designer can use the more common materials in
new designs. An understanding of raw-material costs helps
the design community appreciate the economic implications
associated with various materials. A titanium golf club head
is used to illustrate the options in powder types and feedstock
formulations.
Although the MIM technology was developed in the years
between 1940 and 1980, and has been in continuous use since
the 1970s, still there is little quantitative information available
155
 Table of Contents
Chapter 9: Supplemental Information

or formal benchmarking of costs. In this analysis, the costs of spherical prealloyed titanium powders sell for $150/kg. In
powders and powderbinder feedstocks are discussed. Usually some cases the demand cycles have significant impact on both
these are purchased constituents for all variants of the MIM price and availability, shifting raw-material cost up and down
process. A comparison is given for the cost of powders, poly- over time.
meric binders, and mixing for comparison with the price of Further, there has been a progressive price reduction for
premixed feedstock. The comparison shows self-mixing is a the raw materials as overall demand increases. For example,
natural transition to maintain lower costs as production vol- the average sale price for stainless steel powder has declined
umes grow. 30% in the past 10 years. This is the largest-use segment in
Costs in MIM often trace to differing availabilities of small MIM, accounting for 50% of the metallic component sales,
powders. In some compositions, namely, cemented carbides so it provides the greatest price competition.
and ceramics, the powders were available long before injec- Table 9.4 provides powder and feedstock prices for some
tion molding started. On the other hand, some powder com- popular chemistries. This table is a sampling and is not in-
positions are custom fabricated for MIM. This leads to a wide tended to be comprehensive and will change over time. For
material-cost range when MIM is compared to competitive each composition there might be six or more offerings, but
technologies. Thus, certain formulations have almost no ma- only one was selected to provide an overview of relative
terial-cost penalty associated with the MIM route, while other prices. For example, ten different firms offer the more popular
materials have a high cost when prepared in the size and shape stainless steel powders and at least ten firms sell stainless steel
characteristics needed for MIM. Appreciation of this differ- feedstock. For a 316L stainless steel (Fe-19Cr-9Ni-2Mo), the
ence helps guide the designer and producer to fruitful devel- raw-ingredients cost varies with the metal market, but in late
opment projects. 2010 was about $5.00/kg, melting energy is about $0.10/kg,
The current analysis relies on supplier cost data and reports and atomization costs about $1.50/kg, mostly for the inert gas.
on MIM costing for the raw powders and feedstock. The treat- The difference between the direct powder-production cost
ment is extended to include the cost of mixing. The latter ($6.60/kg) and purchase price ($15.00/kg) reflects a combi-
comes from research on MIM feedstock formulation in con-
tinuous mixers. Since mixers as well as powders and binders
all affect the cost, a generalized model is provided and applied TABLE 9.4. APPROXIMATE POWDER AND FEEDSTOCK
COSTS (various global suppliers, actual price depends
to several formulations. on supplier, quantity, particle characteristics, and raw
material costing)
Cost of Raw Materials composition powder powder feedstock
The first step is to select the engineering material, which $/kg $/kg
has a cost determined by the supply chain. In many applica-
aluminum gas atomized 75 103
tions the powder or feedstock cost is small compared to the bronze water atomized 11 18
component fabrication cost, since the components tend to be cobalt reduced oxide 45 58
small. Even so, powder or feedstock cost can determine the copper water atomized 12 19
viability of MIM when compared to competitive technologies. Invar (Fe-42Ni) gas atomized 21 31
Prior publications have detailed the size, shape, purity, pack- iron carbonyl 8 16
iron-nickel (Fe-8Ni) reduced oxides 20 28
ing, and other characteristics favorable for MIM. A small pow- Kovar water atomized 25 38
der size is required to allow sufficient interparticle strength nickel carbonyl 94 112
during debinding to retain the molded shape, and a rounded stainless 17-4 PH water atomized 15 28
powder with a high packing density ensures easy filling out stainless 304L water atomized 14 23
of the mold cavity. stainless 316L gas atomized 20 29
stainless 316L water atomized 15 24
Understanding the cost penalty or advantage associated stainless 430L water atomized 15 25
with different compositions is an important aspect of the de- stainless 440C water atomized 13 22
sign process. Compositions that are popular tend to have steel 4130 water atomized 15 25
lower powder prices. As with most manufacturing, price de- steel 8630 water atomized 16 27
creases as consumption increases. For example, stainless steel superalloy Hastelloy X gas atomized 26 41
superalloy IN 625 gas atomized 30 46
powder with a median particle size of 15 m produced by superalloy IN 718 gas atomized 25 40
water atomization decreases in price from $27/kg at a volume titanium hydridedehydride 88 150
of 100 kg to $11/kg at 10,000 kg, and falls to under $8/kg in titanium 6-4 gas atomized 125 180
long-term contracts. tool steel M2 gas atomized 22 57
Likewise, for titanium, at 30 kg the price is $85/kg for an- tungsten reduced oxide 33 68
tungsten heavy alloy mixed powders 45 67
gular powder and $120/kg for coarse spherical powder, but at tungstencopper milled oxides 75 104
120 kg quantities the price falls to $100/kg, while smaller

156
 Table of Contents
nation of factors that includes atomization yield in the desired
particle-size range, operational overhead, equipment depreci-
ation, sales expenses, and profit.
Binders for MIM are also highly variable in cost depending
on the formulation. The most common formulations are based
on mixtures of waxes and polymers. Market prices for large
quantities of commodity polypropylene or polyethylene are
as low as $2.40/kg. Special waxes with prescribed levels of
molecular weight and crystallinity can range up to $5.00/kg.
Water-soluble thermoplastic polymers are priced as high as
$10.00/kg. Polymers with surfactants cleaved onto the back-
bone molecule are also in this price range. Surfactants, plas-
ticizers, and lubricants are usually added at low concentrations
(about 5% to 15%) to the waxpolymer mixture to tailor prop-
erties. For many of the binder formulations these agents are
the same as those encountered in plastic injection molding, so
they are commonly available stearates or similar species with
modest costs in the $2.00/kg range. So, the common thermo-
plastic binder system has a direct cost in the $2.00/kg to
$10.00/kg window, depending on the formulation. For this
analysis, the typical waxpolymer system gives a binder cost
of $4.50/kg. Such a binder is used in concentrations from 3%
to 15% to form the feedstock: lower concentrations with high-
density metals such as tungsten and lower concentrations with
low-density ceramics. Since the polymer is generally lower in
cost than the powder, the net effect is to lower the feedstock
cost per unit mass when compared to the powder cost. In other
words, on a raw-material basis, feedstock is lower in intrinsic
cost than the powder from which it was fabricated.
Finally, raw materials must be levied for purchasing, trans-
portation, inventory, and insurance costs. These tend to range
near 10% to 15% of the material cost; these calculations use
12% of the material cost. For smaller operations, the protocol
is to purchase feedstock, but for larger operations the cost of
feedstock is better controlled by internal mixing. This calcu-
lation will analyze the latter to calculate the annual material
use corresponding to a shift from purchased feedstock to in-
ternal mixing.
Operation Parameters
One major impact on feedstock production comes from the
operational context, which goes beyond the raw-material cost.
Included in this are interest rates, building-rental rates, em-
ployee benefits, capital-equipment depreciation schedules, ad-
ministrative expenses, and financial policies. These factors
affect the mixing costs. For example, assume the equipment
depreciation time is YD (years). Then a mixer can be assessed
an hourly cost based on its use rate and occupied space. For
the feedstock-mixing operation, this means that a straight
hourly-use rate can be computed as follows:
HD = CE / (tU YD) (9.18)
Chapter 9: Supplemental Information
HD is the hourly rate for capital equipment depreciation ($/h)
CE is the equipment capital cost ($)
tU is the hours of time the equipment is used per year (h)
YD is the capital equipment depreciation time (y)
The use time depends on the number of hours the device is
operating each day, number of days per week, and weeks per
year; it typically ranges from 800 hours per year (16 hours per
week) to 7,000 hours per year for full-scale operations. These
values roughly correspond to outputs ranging from 3 to 30
tons/month. Normally, the use factor needs to include every-
thing from lunch breaks to holidays, vacations, shutdowns,
and idle weekends. Most facilities have excess capacity in
mixing, so the mixer-utilization rate is important in determin-
ing the cost of mixing. It is well recognized that MIM must
strive toward continuous, full-time operations in order to be
cost effective. Unfortunately, much of the MIM industry uses
their mixers at a rate of less than 60% of full capacity.
The cost for rental of the facility RT ($/m2) is likewise pro-
rated over the equipment by first determining the calculated
cost per unit time per unit area for the space,
CR = RT A F/ tU (9.19)
CR is the rental cost per unit time ($/h)
RT is the facility-rental rate per unit area per year ($/(m2/y))
A is the approximate area required for each device (m2)
F is the ratio of total space to manufacturing space (dimen-
sionless)
tU is the use time per year (h/y)
The factor F reflects the need to allocate the rental cost of idle
space against the productive activities. For many MIM oper-
ations this factor is between 1.2 and 1.5. Thus, idle space in
the facility creates a burden on each unit operation, since the
rental must be allocated to each productive device. Alterna-
tives include a burden based on percent time use of a facility
or just a multiplier on all costs.
Next are the cost allocations for utilities, maintenance, and
labor. These are calculated based on electricity, gas, com-
pressed air, water, and other resource-use rates gathered on an
hourly basis,
CU = Ci (9.20)
CU is the total hourly burden for each operating device (units
$/h)
Ci is the individual contribution to the burden, such as elec-
tricity use per hour ($/h)
i is a counter for the several unit contributions
For example, if the material is abrasive, then mixer mainte-
nance will be high and that cost needs to be captured on a
157
 Table of Contents
Chapter 9: Supplemental Information
prorated-use basis. Annual maintenance in industries involved
in sintered materials tends to be 5% of the equipment cost.
Labor costs differ with the level of automation. As automa-
tion is added the capital-equipment-depreciation expense in-
creases about as fast as the labor expense decreases. However,
in most parts of the world the workforce can be easily adjusted
with the workload, while capital-equipment depreciation must
be recaptured independently of the annual use. The ability to
keep costs variable with manual labor often leads to the lower-
risk solution to use manual labor in low-labor-rate regions. As
a result there is a progressive shift in MIM manufacturing to-
ward low-labor-rate geographic regions with less automation.
Mixing Costs
Wear-resistant twin-cam or twin-screw mixers with the
high shear required for mixing MIM feedstock can produce
from 15 to 50 kg/h, depending on the material, particle size,
and operating characteristics. The capital cost for new mixers
is from $140,000 to $250,000, and in a few cases reaches
$450,000, with additional expenses for feeders, loaders, dry-
ers, and other auxiliary devices for granulation, pre-mixing,
and forming pellets.
Mixing with a high-shear MIM compounder has a direct
energy content estimated at 1 MJ/kg (and can be higher for
smaller powders). This is a direct charge of about $0.02/kg
for energy or about $1.00/h in most locations. Purging gas,
cooling water, and compressed air combine with the electricity
to create a total variable cost in the $2.60/h range, which in-
cludes $1.35/h for maintenance. The labor cost based on U.S.
averages in plastics is $14.66/h including benefits. Thus, the
variable cost for mixing is $17.26/h. However, space rental
and equipment depreciation expenses are accumulated even
if the mixer is idle. For a production mixer capable of 50 kg/h,
this amounts to approximately $51,000/y in fixed expenses
independent of the output.
Assume a production compounder mixing about 50 kg/h
of feedstock. If the mixer is busy most of the time (30 tons/
month), then as noted above the mixing charge is small and
amounts to just $0.53/kg ($17.26/h variable expense and
$9.27/h prorated fixed expense based on 22 h per day, 5 days
per week, 50 weeks per year, and $51,000 annual fixed ex-
pense). For lower-use rates, say 5 tons/month, the processing
cost is more like $1.20/kg or nearly $60/h. For comparison,
toll mixing services in the plastic field range from $0.12/kg
(40 tons/day) to $5.50/kg (300 kg minimum order). On the
other hand, a premium of from $4 to $20/kg over powder cost
is associated with the purchase of pre-compounded feedstock.
One of the burdens is the capital-equipment-depreciation
charge. In operations with low mixer-use rates it is not fruitful
to purchase a mixer and it is more prudent to pay the premium
for pre-compounded feedstock to avoid the capital expense.
As an example, consider a facility that uses 1,000 kg/month
of 17-4 PH stainless (AISI 630). The raw-powder cost is
158
$9,350/month for 1 Mg (1,000 kg), plus $292/month of
binder. There is an additional 12% cost for purchasing, trans-
portation, insurance, and inventory or $1,157/month. The cap-
ital investment of $195,000 provides a depreciation expense
of $3,250/month (five-year depreciation). Allocated space cost
would be approximately $125/month. For mixing one 1 Mg/
month, the utility and labor costs amount to $345/month, and
an additional charge of $100/month is included for cleaning
the mixer. This gives a total cost for 1,000 kg of AISI 630 of
$14,807/month. Purchased feedstock in low quantities is
quoted at $26.62/kg, and 1,078 kg would be required to de-
liver 1,000 kg of product. With a 12% purchasing, transporta-
tion, insurance, and inventory cost, the pre-compounded
feedstock would cost $32,139/month. The saving by self-mix-
ing, when compared to purchased or premixed feedstock, is
$17,332/month. If the operation sets a financial goal of 30%
return on assets, then this annual ratio of 3.6 savings-to-asset
cost per month makes a clear case for self-mixing. Hence, a
financially astute MIM operation would probably shift from
purchased feedstock to self-mixing when consumption rates
approached approximately 1 Mg/month. In fact, firms with
consumption of just 150 kg/month have elected to mix their
own special formulations.
Accordingly, in most full-time MIM operations, mixing
costs probably range near $30/h or $0.60/kg, and are higher
for lower-density ceramics, especially with a submicrometer
particle size. Eventually, in large tonnage situations, mixing
can be scaled up to 4 to 5 tons/h with an anticipated asymp-
totic cost near $0.20/kg. Unfortunately, MIM has not achieved
this scale, resulting in a premium price for feedstock. Most
mixers can generate 50 kg/h of homogeneous feedstock, yet
most of the industry is consuming feedstock at a lower rate.
Indeed the industry average is 9,000 kg of feedstock con-
sumed per molding machine per year (which can be estimated
at 180 h of mixing per molder per year) some of which is in
the form of recycled sprues and runners. Since the larger op-
erations have 30 to 60 molding machines, this works out to a
requirement for one or two mixers. Consequently, there is ex-
cess mixing capacity in the MIM industry. The largest opera-
tions consume 2,000 kg of feedstock per day. As the
particle-size decreases, more work of mixing is required, so
the cost is higher and output is lower, but these estimates still
show an imbalance.
Feedstock for Golf Clubs
To finish this phase of the costing discussion, attention is
given to feedstock costs for production of the sand wedge in-
troduced the tool-costing section. Once the tool cost is esti-
mated, the next phase in the economic analysis is associated
with the feedstock cost. This exercise is needed during the de-
sign stage to ensure overall viability of the concept. Produc-
tion is targeted for 100,000 clubs over a year, with normal
operating schedules (90% effective use of a facility that is run-
 Table of Contents
ning 83% of the year). Overhead is set to 18% and minimum
return on assets is set to 30% per year with an alternative profit
goal of 15% before taxes and interest.
If stainless steel AISI 630 grade (17-4 PH) is selected as
the MIM material, then the powder price is $9.35/kg (discount
price for orders over 5 tons). A total of 20 tons is needed, since
each head weighs 200 g. If the feedstock is self-mixed, it will
take about three weeks using a single mixer (50 kg/h). Assume
the operation is fully engaged, so there is no need to prorate
idle time, then the feedstock cost will be $10.80/kg with a
95% process yield, and the piece cost for feedstock will be
$2.45 each. However, if a pre-compounded feedstock is used,
then the piece cost increases. For example, one of the com-
mercial offerings for AISI 630 is priced at $26.62/kg; then the
feedstock cost increases to $6.77 for each club. These calcu-
lations assume the runner and sprue are reused. An alternative
way to lower the cost would be to use mixtures of elemental
and master alloy powders that interdiffuse during sintering to
produce the desired stainless steel alloy. This could be self-
mixed or purchased as premixed feedstock.
Over the years, there has been much discussion and eval-
uation of the options for combining MIM with titanium to fab-
ricate golf clubs. Taking the same operating parameters as
used for the AISI 630 version above, it is possible to substitute
titanium. The lower titanium density reduces the mass to 115
g. Two versions are possible:
1) self-mixing which gives a feedstock cost of $7.40 each
(assuming the volume of the head remains constant)
2) purchased pre-compounded feedstock which gives a
feedstock cost of $27.58 each
As will be detailed in the subsequent part of this series, this
latter value is over twice the sale price of the MIM stainless
steel golf club while the former is near the cost of the invest-
ment cast head. It would be difficult to convince the customer
that MIM titanium is justified with this differential just in
feedstock cost, especially since the titanium club will have a
lower mass and may not be as successful in biting into the
sand.
Feedstock-Cost Summary
There are many different designs of mixers, many suppliers
of powders, and some significant company operating differ-
ences that affect mixing and feedstock costs. These example
calculations are used to illustrate the costs of feedstock and
the barrier such costs represents for MIM applications. Such
cost barriers are larger for metals, which must compete against
casting, forging, machining, stamping, fine blanking, and
other conventional fabrication routes with lower raw-material
costs. In ceramics and cemented carbides, the MIM route has
less of a raw-material-pricing difference when compared to
extrusion, slip casting, die pressing, and cold isostatic press-
ing; thus, large and more massive MIM structures are more
common in these materials. Contrary to popular myth, the bar-
Chapter 9: Supplemental Information
rier to large part production via MIM is economic, not tech-
nical, since large MIM parts (several kg) are in production via
ceramics.
Powder, binder, and mixing costs are illustrated here for a
busy MIM operation. In such a case the typical cost of feed-
stock is dominated by the powder cost. The incremental cost
for mixing is about $0.60/kg and the polymer cost is usually
less than the powder and lower in density, so it dilutes the cost
per kg for the feedstock. Currently the MIM industry is in a
mode of excess mixing capacity. Typical statistics are cited
for the industry and forward projections suggest that mixing
could approach $0.20/kg as an asymptotic limit based on cur-
rent costs. As such, powder costs are the main component in
the feedstock cost for internally mixed feedstock. For smaller
operations, the premium price paid for pre-compounded feed-
stock is probably justified. These calculations show that con-
sumption levels of just 1,000 kg/month provide justifications
for internal mixing.
BATCH-SIZE EFFECTS
This analysis of MIM cost is based on direct allocation of
charges against a project, assuming an abundance of projects,
so only that proportion of the facility used for any given pro-
duction run is assigned to the cost calculation. In reality, idle
shops are more expensive while shops running at capacity en-
counter high overtime or outsourcing expenses. Ideally, MIM
should be like a hotel: the charges in a hotel depend on the
number of nights you stay, not on the hotels occupancy rate.
However, like airlines, even hotels reprice to maximize profits
as demand increases. So far the opposite philosophy seems to
be at work in MIM. If the MIM shop is not very busy, then it
allocates the cost for idle time (for example, equipment depre-
ciation and rent) to fewer active jobs, leading to increasing
prices as business declines. One attribute of this approach is
that shops that are busier than planned produce higher profits.
Batch-size effects are significant. For example, consider
the cost of powder. Popular compositions have lower powder
prices. As with most manufacturing, price decreases as con-
sumption increases. For example, 15 m stainless steel pow-
der produced by water atomization is priced as follows:
$27/kg at 100 kg
$11/kg at 10,000 kg
$8/kg in long-term contracts
As another example of batch-size effects, consider the in-
jection cost depends on the cycle time and number of cavities.
There is a set-up cost to install tooling, but once it is installed
the internal operating cost per hour for a molder is a standard
ratesay, $25/h, while toll molding is more like $45/h. This
cost represents a combination of factorsdepreciation, rental,
electricity, labor, overhead, and profit. It is desirable to use
multiple cavities to improve productivity at the expense of
machining the tooling.
Multiple-cavity tooling implies a tradeoff between up-front
159
 Table of Contents
Chapter 9: Supplemental Information

tool cost and productivity in manufacturing. As the number

of cavities increases, the MIM project cost decreases while
the tooling cost increases, as shown in Figure 9.4 for a cell-
phone stainless steel part. Since the multiple-cavity tooling is
amortized over a large number of shots, the higher productiv-
ity dominates. This plot shows the tool cost and total project
cost vs. the number of tool cavities for a small stainless steel
cellphone component. For this case, the lowest total cost to
the customer comes with an eight-cavity mold.
To lower the molding cost, faster cycles via improved cool-
ing are required. Faster cooling allows for a shorter molding
cycle or more parts per hour, so there is less molding cost per
part. Because of heat transfer, the cooling time scales with the
square of the section thickness; hence, thin sections are attrac-
tive. For one component with a 125 mm (5 in.) wall thickness, Figure 9.4. Tooling cost (right) and total project cost (left) in
the cooling time was 5 min. This slow molding cycle accu- 1,000s of dollars, plotted vs. number of tool cavities for a stainless
mulates a cost of $2/part and is one reason why thick parts are steel cellphone component produced by MIM
not often fabricated using MIM.
Powder injection molding could fabricate a single compo-
nent. However, the cost would be exceptionally high. In gen-
eral, unit costs tend to stabilize when the production quantity
exceeds 300,000 per year. Note that this is for a single order.
If this is broken into 12 orders of 25,000/month, then the piece
cost is higher. The difference reflects the need to set up and
run for, say, one week out of every month. For example, a tool
change in a molding machine requires a few hours. This cost
must be amortized over the production run. Smaller batch
sizes have a larger burden per part; alternatively a larger lot
size has a smaller set-up burden per piece. Hence, prices are
sensitive to the batch size, not just the total order size.
For smaller components there is a greater sensitivity to
batch size. The approximate point of stabilization is termed
Figure 9.5. Comparison of the production economic-batch-size
the economic batch size. Each shaping technology has a char- impact on MIM cost, showing how the MIM situation continues to
acteristic number where piece price approaches an asymptotic improve at higher volumes vs. machining and die casting. This plot
value with increasing quantity. The concept is illustrated in illustrates how MIM thrives on larger production orders
Figure 9.5 using a MIM disk-drive component, where MIM
is contrasted with screw machining and investment casting
followed by machining. A plateau is evident for each technol-
ogy where periodic expenses arise to offset further gains from
amortized set-up costs. So the order size impacts unit cost
through several aspects, including repair, maintenance, set-
up, design, and tool wear.
Set-up of a molding machine requires a few hours. This
cost which might typically range near $400 is amortized over
the production run. If it is a single component, then that com-
ponent has a $400 burden. However, if the lot size is 1,000
pieces, then the burden is $0.40 each, while if the tool set suc-
cessfully runs for 10,000 pieces, then the set-up burden is just
$0.04 each. With higher-volume products, tools successfully
fabricated up to 2 million parts, lowering the tool set-up cost
Figure 9.6. These three curves illustrate how relative cost depends on
per part to just $0.0002 each. production volume for stainless steel components that are 5 g, 20 g, or
Other factors related to batch size impact the cost. 100 g in mass. These data show at larger MIM production volumes
Schematic curves showing relative cost vs. production volume the smaller components tend to give the greatest relative cost reduction

160
 Table of Contents
Chapter 9: Supplemental Information

are given in Figure 9.6 using a loglog scale for three stainless nificant cost reductions as the economic batch size reaches
steel components at 100 g, 20 g, and 5 g. At a volume of 1 one million units, but the cost of producing larger components
million each, the actual cost is $3.25, $0.55, and $0.30 each, (in the 100 g range) is insensitive to the batch size once
although final values depend on the design details. However, 100,000 units are fabricated.
assuming nominally the same complexity, this plot shows the
cost reduction with larger orders. Note that the smaller com- COMPONENT COSTING
ponent enjoys a much steeper and longer price decline, since MIM component-fabrication costs show scattered behav-
the material cost is small, while for the larger component cost ior, reflecting differences between sites in productivity, capac-
is dominated by raw materials and there is relatively less price ity, equipment, technology, labor rates, taxes, depreciation,
decrease with increasing volume. As another example of the and materials consumption. Many smaller MIM operations do
volume impact on cost, 85 g diesel-engine turbochargers cost not have the experience needed for accurate costing. This is
$20 each to produce at a quantity of 250,000/year, but the cost evident by companies where sales are only 50% of operating
declines to $4 each at 3 million/year. costs. In the long term only profitable strategies are realistic.
Metal powder injection molding excels at components For some short-term customers of MIM, working with an un-
below 10 g. Only single-cavity tooling is justified for most profitable operation is not a concern and they take advantage
components at batch sizes below 100,000, and at batch sizes of the pricing errors. However, for longer-term production, re-
of one million the move would be to two- or four-cavity tool- liance on inherently unstable MIM operations is not viable,
ing. The higher the component mass, the lower the number of so firms with large orders insist on accurate costing to ensure
cavities. Smaller components (in the 520 g range) see sig- supplier stability.
Pricing variability is common in business transactions and
especially true in MIM. As an example, a medical-device quo-
TABLE 9.5. COMPARISON OF BIDS FOR MIM tation for 20,000 parts per year gave the results summarized
MEDICAL DEVICE
in Table 9.5. The ratio between the highest and lowest price
metric part cost mold cost total project is more than a factor of 4. This raises the question as to pos-
mean $2.95 $16,392 $75,368 sible quotation errors of differences in business activity at the
median $2.04 $17,027 $56,743 various sites.
deviation $2.02 $7,408 $45,821 For accurate costing calculations, considerable information
lowest bid $1.55 $7,600 $42,600
highest bid $8.00 $31,000 $191,000
is required, including details that are elusive without experi-
high/low ratio 5.2 4.1 4.5 ence. Here we rely on a unit-cost calculation, as diagramed in
Figure 9.7. It assumes a yield for each step (molding, debind-
ing, and sintering), and allocates the labor, energy, consum-
ables, rent, depreciation, and other charges against the
product, largely based on the active time spent in each manu-
facturing operation. Considerable sophistication can go into
such calculations to account for differences in handling, cycle
time, automation, and factors such as vacation policies, benefit
packages, and depreciation schedules.
The current engineering economic analysis is based on a
direct allocation of charges against a project, assuming an
abundance of projects so only that proportion of the facility
used carries into the cost calculation. In MIM, however, it
seems that operations losing money more typically bid low,
even below cost, leading to a further spiral toward bankruptcy.
By the current model, if a shop is not very busy, then it must
allocate the cost for idle time to fewer jobs, leading to higher
quotations. As a benchmark, in the summer of 2003, globally
there were 258 active MIM companies running at slightly over
70% capacity.
The overall costing calculation is composed of stages that
Figure 9.7. Involved outline of the unit-costing concept as applied consider the tool cost, material cost, batch size, and individ-
to each of the MIM manufacturing steps. This approach accounts
for each component and its mass, energy, labor, and other costs as
ual-component cost. This installment deals with the latter por-
accumulated at each manufacturing step. Obviously it is beneficial tion, the unit operations and determination of cost allocation
to reduce the number of manufacturing steps per piece at each processing step. A quotation would assemble

161
 Table of Contents
Chapter 9: Supplemental Information
all of these costs into a final bid. Additionally, risk and pro-
duction-balancing factors might be considered in forming a
final quotation.
Operational Characteristics
A major impact on component-production cost comes from
the operational context. Included in this are interest rates,
building rental rates, employee benefits, depreciation sched-
ules, and such. For example, the equipment-depreciation time
YD (years) can be different between companies or different
locations. Each production device is assessed an annual de-
preciation charge. When the fabrication of a component uses
that device for a portion of the year, then it is assessed for its
relative use. Such calculations would take place for each step
giving a rate HD computed as follows:
HD = CE / (tU YD) (9.21)
HD is the hourly depreciation rate for capital equipment (units
of $/h)
CE is the equipment capital cost (in $)
tU is the hours of time the equipment is used per year
The use time depends on the number of hours the device is
operating each day, number of days per week, and weeks per
year; it typically ranges from 4,000 to 7,000 hours per year
for full-scale operations. Such a factor includes everything
from lunch breaks to holidays, vacations, and weekends. Note
the sensitivity to the utilization rate, implying low-use device
time has a higher rate, to the point where if it is only used once
a year there is a singular yet high cost.
The cost for rental of the facility is likewise prorated over
the equipment by first determining the calculated cost per unit
time per unit area of space as follows:
CR = RT A / (tU F) (9.22)
CR is the rental cost per unit area per unit time (units of
$/(h/m2))
RT is the facility-rental rate per year (units of $/m2)
A is the area allocated to each device (units of m2)
tU is the use time (units of h)
F is the ratio of manufacturing space to total space
(dimensionless)
Idle space creates a burden since the facility-rental cost is
spread over just the manufacturing space and each devices
relative use time. This calculation is repeated for each unit op-
eration, and if the equipment occupies a large area then costs
grow. Likewise, plant expansion without direct contributions
to increased output results in an increased unit cost.
Next are the required allocations for utilities, maintenance,
and labor. These are calculated based on electricity, gas, com-
162
pressed air, water, and other resource rates gathered on an
hourly basis as follows:
CU = Ci (9.23)
CU is the total hourly burden for each operating device (units
$/h)
Ci is the individual contributions to the hourly burden, such
as electricity use per hour (units $/h)
For example, if the material is abrasive, then maintenance will
be high and that cost needs to be captured on a prorated use
basis. Depending on automation, the labor allocations to each
unit process will differ. As automation occurs there is an added
depreciation expense. In North American MIM the reductions
in labor expenses with automation are about the same as the
increases in capital equipment depreciation expenses. How-
ever, since labor can be varied with the workload, while cap-
ital equipment depreciation must be captured independently
of the annual use, this leads to the natural decision for MIM
to use manual labor in low-labor-rate regions with low levels
of automation.
Molding Costs
Because of the large plastic injection molding industry,
molding costs are well established. Molding averages $18/h
in Asia and ranges up to $40/h in the U.S. and even higher in
Germany. If toll molding is used, then the cost includes over-
head, profit, and other factors, giving rates in the $37$45/h
range in the U.S. for the size molders used for MIM. Most
components are generated in tool sets with one to eight cavi-
ties. As the number of cavities increases, the molding cost per
part decreases, but the tool cost increases, as illustrated earlier.
The cost-justified transition from single-cavity to two-cavity
tooling is usually between 100,000 to 300,000 pieces per year.
Four cavities are not justified until annual volumes exceed 1
million parts per year. Currently 7 million parts per month are
produced on a single machine with a 32-cavity mold operating
24 hours per day. Recent demonstrations have gone to 336-
cavity tooling and 12 s cycle time for a simple MIM shape.
As a benchmark, the MIM industry averages slightly more
than $1 million in annual sales per molding machine, about
twice that of plastic molding.
For cost estimation purposes, section-thickness effect on
cycle time must be estimated, although actual times cannot be
determined until the mold is constructed. Accordingly, cycle
time is estimated using a parametric equation that is sensitive
to the feedstock composition:
tC = CM + CE + CT WE2 (9.24)
tC is the molding cycle time (units of seconds)
CM is a parameter reflecting the empty-molding-machine
 Table of Contents
cycle time (units of seconds)
CE is the extra time associated with various tool options (units
of seconds)
CT is a feedstock parameter associated with cycle times to
mold temperature, feedstock temperature, and feedstock
thermal conductivity (units of s/cm2)
WE is the effective or root mean square wall thickness (units
of cm)
For a typical steel or stainless steel feedstock with a wax
polymer binder, the parameter CM is 7 s and CT is 30 s/cm2.
The extra-time parameter CE comes from added steps that
slow the molding operation. For example, an unscrewing
operation associated with an internal thread might add up to
10 s, a degating operation might add 3 s, and use of a robot
can add 5 s.
Cooling time usually dominates the total cycle time. Cal-
culation of the effective wall thickness relies on the outer em-
bracing dimensional envelope of the component using a
modified empirical formulation found in plastics and invest-
ment casting to calculate an effective-thickness parameter for
heat transfer:
WE = [3 (L1f1 L2f2 L3f3)2 / [(L1f1 L2f2)2 + (9.25)
(L1f1 L3f3)2 + (L3f3 L2f2)2]1/2
M is the final component mass (units of g)
T is the theoretical density of the materials (units of g/cm3)
L1 is the maximum length dimension of the component (units
of cm)
f1 is the effective density of the component along the length
direction (dimensionless)
L2 and L3 are width and thickness dimensions of the compo-
nent (units of cm)
f2 and f3 are the corresponding effective densities for width
and thickness (dimensionless)
The maximum dimensions describe the outer embracing box
around the component, while the effective densities indicate
the holes, undercuts, and other features that provide cooling
contact area, so the cooling time inherently links to the effec-
tive density of the component.
Debinding Costs
Debinding costs are highly variable, ranging from options
such as immersion in chlorinated hydrocarbon solvents to
freeze drying. The nonuniform cost traces to the capital-equip-
ment cost and dwell time in the equipment. Among debinding
technologies, capital-equipment costs vary by a factor of 10.
For example, if aggressive solvents are used, then the capital
expense for an environmentally approved low-emission
system is in excess of $450,000, but debinding is fastwith
penetration rates averaging 2.5 mm/h. On the other hand, if
Chapter 9: Supplemental Information
debinding relies on a water-soluble binder, then the equipment
cost is dramatically lower, near $20,000. However, penetra-
tion rates are slower, so the cost difference is not as large since
more units are needed and the parts dwell longer in each unit.
Retort furnaces for thermal debinding capable of 100 kg/day
can be purchased for as little as $7,000. For small MIM com-
ponents, production debinding costs usually range near
$2.5/kg for a cycle that takes about 8 hours. As a first guess,
most debinding operations average a cost from $25 to $100
per hour, depending on the cycles, equipment type and size,
labor content, and utility rates. The load and cycle time deter-
mine how this cost is partitioned into a per-piece debinding
cost. With some components the cycle time is long and the
packing in the retort is poor, leading to a high per-piece cost.
For illustration, assume thermal debinding in a batch retort
with a capital cost of $80,000 (retort and handling equipment).
Based on a 5,889 hour use per year and five-year depreciation,
the depreciation expense is $2.38/h of use. Based on rental
rates and space, the unit might accrue a rental change of
$0.09/h, utility charge of $9.08/h (electricity, gas, water),
maintenance charge of $1.36/h, and average labor charge of
$9.44/h (monitoring, loading, and unloading). This totals to
$27.48/h in direct costs, but because of loading and unloading,
a higher cost of $33.43/h is used. The component thickness
now enters into the determination of the debinding cycle time:
tD = CS + CX WE2 (9.26)
tD is the debinding time (units of h)
CS is the stabilization time for the reactor, independent of sec-
tion thickness (units of h)
CX is the reaction coefficient for the debinding process (units
of h/cm2)
WE is the effective or root mean square wall thickness (units
of cm)
This empirical route is for thermal debinding, yet it is fairly
reflective of the wall thickness effect. Most debinding
processes have a time dependence on the section thickness
squared. For example, in batch catalytic debinding several
hours are required to establish the nitrogen purge, heat the re-
action vessel, and cool after debinding and CS is about 4 h and
CX is about 10 h/cm2. In thermal debinding, CS reflects the
time to reach the burnout temperature and CX reflects the pen-
etration rate; for one binder these values are 6 h and 70 h/cm2.
Obviously these parameters depend on the binder and debind-
ing technology, but clearly thicker sections require longer
times and by this model accumulate higher costs.
Sintering Costs
Sintering costs in MIM range from $30$300/h. The cost
per part depends on the load in the furnace, and cycle time
(usually from 630 h). This translates to nominal costs in the
163
 Table of Contents
Chapter 9: Supplemental Information

$6/kg range, but the actual value varies from a low near $1/kg
to a high of $30/kg. For example, a small continuous sintering
furnace costs approximately $30/h to operate, but one with a
six-fold higher production rate costs $120/h (4-fold cost in-
crease gives 6-fold output gain). However, if there is insuffi-
cient work to keep the larger furnace busy, then it is a burden.
Cost reductions drive MIM to larger equipment, but if there
is not adequate business to fill the equipment, then per-part
costs skyrocket. Larger and busy operations inherently have
lower costs.
The furnace design depends on the intended production
quantity, materials to be sintered, type of atmosphere, and
post-sintering cooling rate. Production MIM batch sintering
furnaces exist for loads ranging from less than 1 kg to over
300 kg (2 lb. to 660 lb.). Continuous sintering can reach levels
up to 100 kg/h. Larger batch furnaces typically operate on cy-
cles lasting from 12 to 36 h. On the other hand, large pusher
furnaces operate on entry-to-exit times ranging up to 24 h.
Continuous furnaces tend to have lower operating costs, but
are less flexible. Thus, they are used only for sintering similar
components. Large components require slower heating and
larger furnaces, thereby incurring even higher sintering ex-
penses. Contrary to popular myths, large component produc-
tion by MIM is not paced by technology; rather, the rapid
escalation of costs with size is the major barrier.
As with molding and debinding, there is a sintering-cycle
time related to section thickness; however, this is mostly re-
lated to heating rates and soak times needed to obtain uniform
shrinkage. Smaller components can be heated quickly, with
demonstrations in continuous sintering of times near 2 h for 3
g components. Thicker and more-massive components are
heated slowly to avoid distortion due to thermal gradients.
Hence, the sintering-cycle time scales with the component
thickness as follows:
supports are necessary. Depending on the numbers of setter
trays, part size, and furnace size, the setter costs can be sig-
nificant. Most of these setters need repair and replacement
after, say, 100 cycles, so a replacement cost is also a part of
the sintering calculation.
As an example of sintering costs, a 6 g tool-steel piece that
must be individually staged and vacuum sintered has a sinter-
ing cost of $0.10 each, but an 8 g stainless steel component
that packs tightly in the furnace is vacuum sintered for $0.05
each. As an example of a low-cost sintering, a simple 25 g
steel component is sintered in a continuous-belt furnace in a
nitrogen-based atmosphere for just $0.02 each.
Finishing Costs
Finally, there are many finishing steps required prior to
shipping. These include secondary treatments such as machin-
ing, heat treatment, or electroplating. Inspection and packag-
ing are other final steps that add expense.
Each of the finishing steps typically has a charge based on
mass, time, or area. Because of the many factors involved,
these are not single-valued parameters. However, a listing of
some of the common steps and approximate cost factors is
provided in Table 9.6. The bottom line is that each unit oper-
ation has a calculated part or hourly rate. Rarely is a value
analysis performed, but in situations where constraints exist
there are some interesting pricing options.
Tight-Tolerance Effects
One of the significant cost issues in MIM comes from tight
tolerances. Tight tolerances either reduce yield or increase the
secondary operations such as machining or grinding. Process
yields fall quickly as tolerances increase. The industry average
coefficient of variation is 0.22% (dimensional standard devi-

tF = CH + CP WE2 (9.27) TABLE 9.6. EXAMPLE FINISHING COSTS FOR MIM

operation costing basis cost $
tF is the furnace cycle time (units of h)
CH is a furnace parameter associated with heating and cool- anneal per kg 1.54
cadmium plate per m2 0.54
ing (units of h)
chromium plate per m2 1.55
CP is a component parameter reflecting cycle times for coining per h 25.00
more-massive components (units of h/cm2) copper plate per m2 0.70
WE is the effective or root mean square wall thickness (units machining per h 24.30
of cm) nickel plate per m2 0.78
packaging per part 0.02
precipitation harden per kg 2.31
For batch sintering in a vacuum furnace, the heating and cool- prime and paint per m2 1.40
ing rates are often 10C/min (18F/min), and for sintering quench and temper per kg 2.31
stainless steel at 1,365C (2,490F) this gives 6 h for CH with surface carburize per kg 1.54
no load. As the section thickness increases, the empirically surface finishing per h 24.30
surface nitride per kg 2.31
determined factor for CP is 120 h/cm2, reflecting the need to
surface steam oxidation per kg 4.41
place holds in the cycle for outgassing and soaking for larger tumbling per kg 0.22
bodies. zinc plate per m2 0.39
There is a setter cost for sintering, since customized part

164
 Table of Contents
ation divided by the mean size); thus, specifications tighter
than about 0.5% (2.25 standard deviations) impact yield and
increase cost. Likewise, the more dimensions that are held to
tight tolerance the larger the probability the part will be re-
jected. Accordingly, sorting is required to separate those in
specification from those not acceptablethis adds a costly
extra step. More of an impact is from the expense associated
with production of the rejected components, an expense that
must be captured in the final shipped material. Thus, error
budget concepts need to be employed to balance cost and tol-
erancing. Use of secondary operations to hold tolerances
(coining or machining) can have a dramatic impact on costs.
Thus, tight tolerances are possible, but usually lower
process yield and add expensive secondary operations. As the
number of tight tolerances increases the economics often be-
come less attractive; hence, MIM defaults to machining di-
rectly from wrought materials in cases were several features
are held in close tolerances.
Per-Piece Costing
This cost model is an accounting schedule that has unit costs
per time, and cumulative process costs are combined with feed-
stock and subcontracted secondary costs to form a production
cost estimate. Now the attention turns to the specific compo-
nent and its production quantity and production rate. High-
intensity projects require multiple molders and multiple tool
sets. Basically, in the unit operation approach, multiple oper-
ating units do not change the cost, since attention is on how
many parts can pass through each step or machine per hour.
The unit operating cost is then allocated to each part as
follows:
PCi = CTi / U (9.28)
U is the average number of parts produced per hour (units of
1/h)
PCi is the cost per piece (units of $)
CTi is the unit operating cost per time (units of $/h)
These calculations are repeated for each unit operation and
the values are summed to estimate the piece price, along with
feedstock cost and yield factors.
A penalty should be assessed when a component creates
an imbalance in the production operation. Molders, debinding
devices, and sintering furnaces come in integers, so facility
sizing might rely on a formula such as one mixer, six molders,
eight debinding retorts, and two furnaces or something of that
sort to make up a manufacturing complex. Depending on the
cycle times for each step, it is possible for production to be
out of balance, so some of the equipment sits idle. If there is
sufficient business, then the spare devices are shifted to other
products (possibly overloading downstream equipment). For
example, a spindly component with thin sections will mold
Chapter 9: Supplemental Information
quickly, but give a low part density in the debinding and sin-
tering devices. Consequently, molding will be out of balance
from the other operations. If the molder sits idle, then there is
a cost for idle time that needs to be accounted for in the pric-
ing, otherwise overall financial performance sill suffer. This
capacity-balance cost needs to be calculated for the idle time
based on the characteristics of the part and production facility.
Some facilities are better suited for certain sizes or shapes;
thus, the capacity balance cost will be highly variable with
equipment and its operation. Typically this cost is distributed
over the quoted production quantity, and might amount to a
few percent of the cost in some cases.
Additional Costs
Other factors impact the final price. These are largely as-
sociated with overheadpurchasing, finance, sales, market-
ing, research, management, and other business expenses that
must be covered, but are only indirectly associated with a par-
ticular component. Further, there are often one-time (nonre-
curring) expenses associated with fixtures, inspection tools,
handling devices, and other custom-designed items. These can
easily exceed $20,000 or more and must be amortized over
the initial production order. Finally, profit is required to invest
in the future; across the MIM industry this is highly variable
and is as low as 50% loss to 20% profit.
Insurance is a means to reduce risk. Payments are made
periodically and only occasionally are claims made, but these
tend to be much larger than the payments. Risk analysis is as-
sociated with determining the exposure to loss. A modified
form of risk analysis is seen in MIM pricing. When a project
forces a molder into a new realm of production experience,
often the prudent manager includes a contingency or fudge
factor in anticipation of surprises. Some of the problems might
be slow cycle times, more tool wear, or problems with defor-
mation in sintering. Aversion to risk leads to an extra cushion
in the price quotation to accumulate reserves in case a problem
arises after production starts.
Generally, the risk aversion or contingency in pricing in-
creases as the quote strays from familiar ground. This is an-
other reason for variability in price quotes, since different
vendors will have different experience levels. For example,
on a high-volume project, over eight firms participated in the
early quotation. The highest quote was from a smaller firm
with less experience in automation and high-volume produc-
tion, while the lowest quote was from a firm with extensive
prior production experience in high-volume components of
similar size and shape. In other words, larger firms with broad
production experiences will have more accurate quotes, but
not necessarily the lower quotes. Smaller firms with less ex-
perience often are the low bidders, since they do not recognize
some of the problems. Even a small loss per part times a large
number of parts can quickly bankrupt a firm.
165
 Table of Contents
Chapter 9: Supplemental Information
Golf Club Example
To demonstrate the approach, we examine the costs for pro-
ducing a sand wedge via MIM. The component was intro-
duced in the tooling calculation and treated in the feedstock
section as well. A single-cavity tool is estimated to cost
$22,525 and with an engineering charge the final mold price
was estimated at $25,000. Production is scheduled for 100,000
club heads over a year, with normal operating schedules (90%
effective operation for a facility that operates 83% of the year).
Overhead is set to 18% and profit is set to 15% before taxes
and interest.
Stainless steel grade 17-4 PH is selected as the baseline
MIM material, at a bulk cost of $9.35/kg. The outer dimen-
sions of the 200 g club head are 17 by 12 by 2 cm and the
mixed feedstock cost was estimated at $2.45 each. Sintering
and debinding are both sized for 20% packing of the parts
(80% of the furnace volume is space between the parts or set-
ters). Critical dimensions are coined after sintering, the exte-
rior is bead blasted, and polishing is performed on the stem.
The project will consume 22 days of mixing, 90 days of mold-
ing (58 s cycle time in single-cavity tool), 214 days of debind-
ing, 176 days of sintering, and 147 days of finishing.
Obviously, multiple debinding and sintering furnaces can be
used to better balance steps, for example, by using two de-
binding retorts and sintering furnaces to balance with the
molding. In each operation, time is added for loading and
unloading, giving an effective equipment use near 62% of full
capacity.
The projected cost is $6.67 each. The feedstock (self-
mixed) is 37% of the cost, molding is 5%, debinding is 15%,
sintering is 37%, and finishing is 6%. Normal overhead and
profit then make the price $9.96 each. Further, charges for risk
aversion and capacity imbalance (the parts consume much sin-
tering time and little molding time) could dilate the bid price
to $10.75 each. For 100,000 club heads, the tool cost would
be 2.5% of the project price.
If the sand wedge is made from titanium, then the mass de-
creases to 117 g for the same volume (not realistic, since mass
is a dominant factor in this design). Even with the lower mass
the costs are much higher. Using titanium powder at $53/kg
gives a final part price of $17.91 each, but if $220/kg titanium
powder is used the part price jumps to $53.23 each. What
would be the component cost if feedstock were free? For the
sand wedge, this works out to $6.45. If the titanium powder
were priced like stainless steel powder, then the part price be-
comes competitive with investment casting. Scenarios with
the model show that the fabrication cost is almost a linear
function of mass, material cost is a linear function of mass,
and yield declines with mass, so the overall price is essentially
a linear function of mass. At a low mass, fabrication costs
dominate. Hence, the key to large components by MIM is two-
fold: reduce powder costs and reduce cycle times for larger
sections.
166
HOW MUCH WILL IT COST?
So what do some MIM components cost to produce? A few
examples are listed in Table 9.7. The sale prices are typically
much higher than the model predicted cost, since operation
variables such as administrative, sales, interest, and research
expenses enter. This model was applied to 16 commercial
components where purchased price was disclosed along with
sufficient detail to apply the model. These parts ranged from
$0.17 to $16.32 each. The correlation coefficient between pur-
chase price and computed cost was 0.97 (very significant),
with an offset of $0.15 and slope of 1.1 (if the model sug-
gested $1.00 cost, then the price would be $1.25). The error
in this estimate was 33%, so the actual purchase price is highly
variable even when the cost factors are known. However, the
model shows nearly a linear dependence on component mass.
Surveys of the MIM industry show a wide distribution of
products in production. For example, in a survey of 220 MIM
parts selected at random, the median was just over an inch
(26 mm) in largest dimension and 8 g in median mass. In this
sample set, the largest component was 193 mm (7.5 in.) in
length with a 43 g mass. On the other hand, the largest mass
was 1,097 g (2.4 lb.) with a maximum dimension of 137 mm
(5.4 in.). These findings show that MIM has found most of its
economic successes in lower mass, and slender or spindly
geometries.
Likewise surveys of the average MIM component sale
price show $0.92/part is most typical. Prices vary consider-
ably, yet this suggests that a typical MIM part is selling for
about $115/kg. Although 55% of the industry sales are stain-
less steels, this is not reflective of the considerable variation
in powder costs.
Another way to analyze cost is to appreciate that on aver-
age in MIM about 32% 17% (one standard deviation) of the
manufacturing cost is attributable to the cost of powder (dom-
inated by stainless steel). Often the component sale price is at
least twice the manufacturing cost. Since the average MIM
grade stainless steel powder is $13.27/kg, this translates into
an average component sale price near $80/kg (low of $53/kg
and high of $177/kg). A few aerospace, medical, superalloy,
titanium, and microminiature components are much higher. A
survey of MIM industry sales independently estimates this at
$132/kg in North America. The error with this approximation
is large, but such numbers provide a guide to what MIM costs.
Clearly MIM is most appropriate for smaller, higher-value,
and complicated components, and struggles with cost issues
as the size and section thickness increase. In a few instances,
pricing for molded parts is as much as $14,000/kg, but those
tend to be small components. For example, some orthodontic
parts at 0.2 g each are priced at $0.35 each, giving the equiv-
alent of $1,750/kg. For comparison, the plastic injection mold-
ing industry averages 33 g per part which sells for an average
of $7.59/kg, or $0.25/part using feedstock that is typically
priced at $2 to $3/kg.
 Table of Contents
Chapter 9: Supplemental Information

TABLE 9.7. EXAMPLE PRODUCTION COSTS FOR MIM COMPONENTS

(Ignores Sales, Administration, Taxes, Profit, Interest, and General Overhead and Markup Items)
quantity,
component name material million per year mass, g unit cost, $
biopsy blade 17-4 PH stainless 0.02 0.01 2.70
blade 410 stainless 0.03 2.50
clock mount 316L 0.1 14 1.87
connector 316L stainless 0.2 40 1.35
disk drive latch 17-4 PH stainless 12 2.4 0.17
electronic package Kovar 0.01 10 4.25
gear set low-alloy steel 3 90 2.20
hermetic package Kovar 0.3 10 2.70
knee implant 316L stainless 0.05 200 400
laparoscopic jaw 17-4 PH 0.1 5 0.60
magnetic latch Fe-50Ni 1.2 5 0.18
orthodontic bracket 316L stainless 6 0.03 0.08
projectile W-1.5Ni-0.5Fe 1 130 5.00
root-canal tip 17-4 PH 0.05 0.9 16.67
self-aligning hook 17-4 PH 60 0.1 0.60
solenoid housing Fe-Ni alloy 0.1 180 35
sleeve guide 316L stainless 1.3 30 4.35
toothbrush gear 316L stainless 10 0.34 0.35
turbocharger Ni superalloy 0.3 100 9.00
union connector copper 0.4 6 0.35
valve plate Ni superalloy 0.005 36 100
watch case 316L stainless 8 31 1.00

COST SENSITIVITY tC is the molding cycle time (units of s)

Several variants exist in MIM production equipment. Low-
pressure molders have lower capital costs, but mold with
slower cycles when compared to high-pressure molders. Some
materials have high-heat capacities or low thermal conductiv-
ities, so heating and cooling cycles vary between materials.
With respect to the component design, thick sections cool
slower in the mold and heat slower in the furnace and cost
more to produce. If a process is well understood, then time
estimates are possible for each production step, allowing cal-
culation of the total production cost based on material, process
details, and equipment characteristics.
A first task is to derive equations showing the cost. Price
is calculated from the cost by adding additional factors such
as risk, taxes, insurance, and profits. Pie charts generated at
this point show where cost reduction efforts might be most
fruitful. Further, sensitivity analysis is possible based on a first
derivative of those equations.
Price analysis provides a quantitative means to evaluate
possible design or process changes. This is illustrated by the
molding cycle. The time to form a component is often domi-
nated by the cooling time, which depends on the section thick-
ness. To lower cost it is possible to reduce the maximum
section thickness without compromising function, say by
10%. The cycle time in molding is calculated as follows:
tC = tM + tE + tW W2 (9.29)
tM is the empty-molding-machine cycle time (units of s)
tE is the extra time associated with various tool options (units
of s)
tW is the cooling-time factor, which depends on mold temper-
ature, feedstock temperature, and feedstock thermal con-
ductivity (units of s/cm2)
W is the wall thickness (units of cm)
For typical feedstock, the parameter tM is 7 s and tT is 30
s/cm2; for a simple mold the extra-time parameter tE is 3 s. Of
the 18 s cycle time, 10 s is independent of the section thick-
ness, but the remaining 8 s would benefit from a thinner wall.
The 8 s cooling time corresponds to a wall of 5 mm. A 10%
reduction to 4.5 mm reduces the molding cycle time from 18
s to 16 s. At a cost of $20/h, this reduces the molding cost per
part by more than 10%. Unfortunately, molding often is one
of the lower-cost steps, so this design change cuts the molding
cost by 10%, yet may have no meaningful impact on the final
price. It is an example of solving the wrong problem.
Figure 9.8 shows a 2.2 g cellphone component from stain-
less steel as an illustration. When production reached millions
per year, the part price dropped to $0.27 each using water-
atomized 316L powder and batch sintering.
Figure 9.9 is a pie chart showing the various cost contri-
butions; price is dominated by finishing, profit, and overhead.
Curiously, these three factors are not MIM costs. With respect
to MIM, sintering and feedstock costs are the most important

167
 Table of Contents
Chapter 9: Supplemental Information

factors. A redesign to reduce the wall thickness from 2.1 mm
to 1.8 mm with the same mass lowers the price per part by
3%. However, if the wall thickness and mass are decreased
by 10%, then the price drops by 8%. Continuous sintering
gives a larger savings, and if continuous thermal debinding
and sintering are combined, a total 15% price reduction oc-
curs. Hence, nearly 23% price reduction is possible via section
thickness and process changes. Finally, if the profit is cut in
half, then the price drops to less than $0.18 each, yet is still
profitable.
This case illustrates that some costs are not important. To
attend to the important factors shows that a MIM calculation
is needed to first identify the controlling parameters. The first
step is to apply the model, determine the sensitivities over the
range of practical options, then implement strategies to meet
the design objectives, yet lower the price.
An important approach comes from sensitivity analysis.
Each of the cost contributors has a different amplification
factor. Small changes to dominant factors provide large gains.
For example, sintering might have ten times the impact on
final cost when compared to molding. Thus, changes in
section size to increase molding productivity might be less
important when compared to efforts to reduce sintering time.
The sensitivity factor F provides a rationalization of various
influences:
we will use a 316L stainless steel component for the analysis.
Figure 9.10 is a pie chart of price contributions for this com-
ponent using 4-cavity tooling at one million parts per year.
The calculations assume 15% general overhead, 10% profit,
and 95% process yield for a factory that is typical as detailed
in Table 9.8.
The sensitivity analysis considered several factors to de-
termine the price changes with each. The sensitivities are
ranked in Figure 9.11. Other factors considered, but not listed,
had even smaller impact in this case. The decision on pur-
chased feedstock vs. self-mixing shows up as the most signif-
icant parameter. Efforts to lower the mass, improve process
yield, use lower-cost raw materials, and improve sintering all
promise to improve efficiency.
As demonstrated in Table 9.9, a few factors have signifi-

F = (X/C) (C/X) (9.30)

where X is an operating parameter such as labor cost, and C

is the component price, so this is the normalized change in
price (C/C) divided by a normalized change in one of the op-
erating parameters (X/X). This approximates the partial deriv-
ative in the price vs. parameter space of the cost model. Since
many factors are involved, and each component is different,
the sensitivity factors change with each component, operation, Figure 9.8. 2.2 g stainless steel cellphone knuckle shown next to
the assembly; this component is used for illustrating the detailed
material, and design. MIM cost analysis

Typical Component Analysis

The typical MIM component tends to cluster into a design
window that can be defined by the following parameters:
mass80% range between 0.3 g and 60 g
projected area80% range between 200 mm2 and
2,500 mm2
thickness80% range between 1 mm and 14 mm
maximum dimension80% range between 8 mm and
80 mm
complexity80% range between 20 and 160 features
materialmajority contain iron

The median component has 60 features, 500 mm2 projected

area, 4 mm thickness, a maximum length of 25 mm, and mass Figure 9.9. Chart giving the relative cost contributions to the
of 8 g. All of these do not exist at the same time, yet these sta- fabrication of the 2.2 g stainless steel knuckle component formed
tistics help define where MIM is successful. For this analysis, by MIM (part shown in Figure 9.8)

168
 Table of Contents
Chapter 9: Supplemental Information

cant impact on the final price; shown here are the benefits
from self-mixing, reducing mass 10%, improving process
yield to 98%, shifting to water-atomized powder, improving
furnace loading, and shortening sintering time. The price
drops to almost half that initially calculated. These changes
are realistic, since coring or section thickness changes can be
used to reduce the mass, process yields of 98% are realized
on larger orders, water-atomized stainless steel powders are
widely used in MIM, changes in the setters and packing of
parts can improve furnace loading, and thin substrates, heat
exchangers, and convective cooling can reduce the sintering-
cycle time. Curiously, labor rates, capital-equipment cost, and
facility-rental rates were less important factors.
Figure 9.10. Further illustration of cost contributions, showing
relative contributions to the price for a typical 8 g stainless steel Component-Mass Effect
MIM component
The above example was for the median 8 g stainless steel
component. At the other extremes, 10% of the components
TABLE 9.8. TYPICAL MIM PARAMETERS USED FOR are 0.3 g or less and 90% are 60 g or less. Table 9.10 compares
COST CALCULATIONS the upper and lower component attributes with the typical
part = 8 g, 316L stainless steel, 7.9 g/cm3, 1 million per
year, 25 mm long, 60 features
tolerance = 0.1 mm
tool base cost = $1,000
number of side pulls = 1
projected area of part = 500 mm2
number of surface bumps and depressions = 10
surface finish on tooling = 0.2 m
lettering on tooling = yes
toolmaker hourly rate = $50/h
number of cavities = 4
material = 316L stainless steel
powder = -20 m water atomized
powder cost = $13.06/kg
binder cost = $4/kg
facility-rental rate = $60/m2 per year
production floor area = 75% of factory
factory labor rate = $8.50/h
benefit rate = 15%
setup-engineering cost = $16,000
depreciation schedule = 10 years
maintenance provision = 5%
mixing-facility cost new = $195,000 Figure 9.11. Price-sensitivity factors for typical MIM component,
mixing rate = 50 kg/h showing the relative impact of various process or design decisions
molding-facility cost new = $90,000 with respect to final component price. Note that self-mixing of the
molding-cycle time = 30 s feedstock and reduction of the component mass are the two largest
debinding-facility cost new = $90,000 factors
debinding-furnace volume = 0.25 m3
debinding-furnace loading factor = 10%
debinding-cycle time = 18 h TABLE 9.9. PRICE REDUCTIONS POSSIBLE FOR A
sintering-facility cost new =$590,000
TYPICAL MIM COMPONENT
sintering-furnace volume = 0.42 m3
sintering-furnace loading factor = 10% parameter initial value improved value
sintering-cycle time = 24 h
electric rate = $0.10/kW-h feedstock purchased self-mixed
cooling-water rate = $2.00/h part mass, g 8 7.2
gas rate = $0.60/m3 process yield, % 95 98
process yield = 95% powder type gas-atomized water-atomized
operational hours per year = 6,000 h furnace loading, % 10 25
overhead = 15% sintering-cycle time, h 24 18
profit = 10% unit cost, $ 0.66 0.36

169
 Table of Contents
Chapter 9: Supplemental Information
(8 g) component. The larger part would be characteristic of a
turbine impeller, pump housing, or solenoid housing, while
the smaller part would be characteristic of a fiberoptic con-
nector, orthodontic bracket, or surgical biopsy tool.
These two cases benchmark the range of MIM products.
Figure 9.12 plots the price contributions from several factors
for the three component sizes. The 8 g component was most
sensitive to feedstock (mass), yield, and sintering cost reduc-
tions. For the larger (60 g) component, the comparative plot
shows that feedstock cost is even more important, followed
by sintering. On the other hand, for the small (0.3 g) compo-
nent, the price is dominated by finishing and tooling costs.
From the sensitivity analysis, cost-reduction efforts are most
fruitful if applied to the major cost contributors.
Golf Club Example
The MIM technology has repeatedly been applied to the
fabrication of golf clubs. This discussion has examined the
economic feasibility for producing a sand wedge. The com-
ponent was used in the tooling, feedstock analysis, and com-
ponent cost treatments. A single-cavity tool was estimated at
$25,000. Production was set at 100,000 club heads over a year.
The estimated price using self-mixed water-atomized 17-4 PH
stainless steel feedstock was $9.96 each, assuming the cus-
tomer purchased the tooling. When titanium was considered,
TABLE 9.10. PARAMETER SPACE FOR MIM
COMPONENTS
cumulative population, % 10 50 90
mass, g 0.3 8 60
number of features 20 60 160
projected area, mm2 200 500 2,500
thickness, mm 1 4 14
maximum dimension, mm 4 25 80
Figure 9.12. Comparison of price contributions for small (0.3 g),
medium (8 g), and large (60 g) MIM components. The small
component is most sensitive to tooling and finishing costs while the
large component is most sensitive to feedstock and sintering costs
170
the mass decreased, but the price increased to $17.91 each
using hydridedehydride titanium powder, and increased to
$53.23 each using spherical atomized titanium powder.
If we examine the price factors for the stainless steel ver-
sion, then several small changes are identified to lower the
price. For example, custom fixtures would increase furnace
loading in debinding and sintering to 20%, while convective-
gas cooling would reduce sintering time to 18 h. Other realis-
tic changes, such as improving process yield to 98%, would
bring the price to an asymptotic level near $6.12 each. This
approaches the price currently paid for investment cast golf
club heads, so a viable project emerges, but with great effort.
For titanium there is little hope for MIM to compete against
investment casting, even with low-cost powder.
Bottom Line on Costing
Once a cost model is created, then it is possible to deter-
mine what factors are dominant. Pie charts are given for small,
typical, and large components. Further, the discussion on golf
club production helps capture the reality of MIM for some tar-
get applications. Sensitivity analysis provides data on where
attention will have the greatest payout. It is demonstrated that
a few changes can reduce component price by 30% to 50%.
In general, the most fruitful changes are to perform self-mix-
ing (except for very small operations), use coring and other
design changes to reduce mass, maximize process yield, rely
on lower-cost powders, and push sintering technologies for
shorter times and higher loading. For smaller components,
price is dominated by tooling and finishing costs, while for
larger components the price is most sensitive to feedstock and
sintering factors.
DEFINITIONS
The following definitions first give the word or phrase
commonly encountered in MIM follows by a brief sentence
or two on how this is used in MIM.
Acicular powderneedle-shaped particles good for
debinding strength but generally detrimental to the MIM
forming process since they resist flow in molding.
Adiabatic formingthe use of a frozen powderwater feed-
stock that melts and flows under pressure, but refreezes
when pressure is released.
Admixed powdera small, discrete powder mixed with an-
other powder for lubrication, bonding, or alloying. One
common means of forming MIM alloys or composites is
via admixed powders.
Agglomerationa tendency for small MIM particles to stick
together and appear as larger particles. It is a common
problem with nonconducting powders such as ceramics,
especially as the particle size decreases.
Air classificationa most common means to separate
specific size classes of powders by differential settling in
 Table of Contents
a high-velocity air stream or cyclone.
Alloy powdera powder in which each particle is composed
of the same mixture of two or more constituents, especially
common in gas- or water-atomized powders such as stain-
less steels.
Aluminathe compound of aluminum and oxygen Al2O3
that is the most common material used in ceramics, often
used as a furnace lining, hardware, or setter support for
sintering.
Angle of reposethe angle from the horizontal plane that a
pile of loose powder will assume when freely poured
through an orifice, providing a measure of the interparticle
friction and ease of shape retention during debinding.
Apparent densitythe mass of a unit volume of powder in
the loose condition, usually expressed in g/cm3. For MIM
powders it is often 50% of the theoretical density for the
material.
Aspect ratiothe ratio of the maximum particle dimension
to the minimum dimension.
Atomizationthe dispersion of molten material into droplets
by a rapidly moving stream of gas or liquid (usually water),
or by centrifugal force. The droplets solidify into particles.
Atomized powderalloy or metal powder produced by the
disintegration and subsequent solidification of a molten
metal stream.
Attritiona mechanical milling or grinding process that typ-
ically employs a stirred or tumbled container filled with
small balls that act as a crushing medium. The tumbling
action causes repeated collisions against particles mixed
with the balls. The process is widely used for deagglomer-
ation, partial alloying (W-Cu, WC-Co), and particle-size
reduction in ceramics.
Austenitethe face-centered cubic crystal structure of pure
iron that is not stable below 910C. However, with alloying
this nonmagnetic form of iron can be stabilized at room
temperature and even cryogenic temperatures; austenitic
stainless steels are a common example where sufficient
nickel is employed to stabilize the nonmagnetic phase at
room temperature.
Barrelthe heated portion of the feedstock-flow path before
the nozzle. The barrel holds the feedstock under pressure
while providing heat to melt the binder system.
BET surface areathe specific surface area as measured by
gas adsorption according to the Brunauer, Emmett, and
Teller theory. Widely used in characterization of ceramic
MIM powders and expressed as square meters per gram,
m2/g.
Bimodala particle-size distribution exhibiting two mode
(mode is the most common or most frequent) sizes.
Although such a powder can result from mixing two dif-
ferent particle-size distributions, some powder-production
techniques naturally produce this distribution. Its main ad-
vantage is in producing a higher solids loading in MIM
Chapter 9: Supplemental Information
feedstock. It is common to use trimodal mixtures in ce-
ramic MIM.
Binderthe polymer mixture that provides lubrication and
strength to the powder. It is critical to the feedstock fluidity
during molding and to the strength of the molded compact.
The binder is evaporated during debinding and sintering.
Binder formulations generally fall into a few classes, but
exact formulations are usually proprietary. The binder
chemistry impacts the debinding process.
Bingham flowviscous flow of a feedstock with initial yield
strength, meaning the stress must exceed the yield strength
prior to initiation of viscous flow.
Blendingthe thorough intermingling of powders of the
same nominal composition, for example, the combination
of water-atomized and gas-atomized stainless steels.
Brown statea term originally used to describe a ferrous
MIM component following debinding by slow heating in
air to remove the polymer, where oxygen caused the iron
to oxidize into a rust-colored component.
Burnoffthe removal of the polymer binder via preheating
prior to sintering.
Capillary rheometera device for measuring viscosity by
applying pressure on molten feedstock and determining the
flow-rate dependence on applied stress as the feedstock is
pushed through a small capillary tube. It is useful in deter-
mining feedstock flow and homogeneity.
Carbo-nitridea heat treatment in a nitrogen and methane
atmosphere that causes both carbon and nitrogen to diffuse
into the steel surface for strengthening and hardening.
Carbon controla measure of the ability to remove the MIM
polymer without contamination of the powder. Carbon con-
trol is quantified by the final carbon level and the unifor-
mity of that level between parts, or over subsequent
operations, fabrication days, or feedstock batches. For
some materials the desire is no final carbon (austenitic
stainless steel, alumina, titanium), while others require a
precise final carbon level, and in the extreme cemented car-
bides and silicon carbides require very high carbon levels.
Carbonyl powderpredominantly an iron powder, but may
also be nickel. These powders are prepared by the thermal
decomposition of a metal carbonyl molecule. The resulting
particle size is typically in the size range from 1 to 10 m.
The particle shape can be spherical, agglomerated, or
highly angular.
Carburizationdecarburizationtwo events critically re-
lated to carbon control, since the addition of carbon to a
material occurs by carburization from binder or atmosphere
sources, while decarburization occurs by reacting carbon
in the material with hydrogen, oxygen, or carbon monox-
ide.
Case carburizea post-sintering heat treatment aimed at dif-
fusion of carbon into the surface of a ferrous MIM com-
pact. The carbon increases strength and hardness.
171
 Table of Contents
Chapter 9: Supplemental Information
Catalytic debindingthe partial removal of the polymer via
heating in an atmosphere containing an agent that induces
depolymerization. The classic process involves extraction
of polyacetal using a nitrogen atmosphere doped with nitric
acid.
CCIMcemented carbide powder injection molding, where
the inorganic phase in the polymer binders is predomi-
nantly a mixture of cementing phase (cobalt is typical) and
a carbide (tungsten carbide is typical); a subset of powder
injection molding (MIM).
Cemented carbidea solid composite consisting of a metal
carbide and a binder phase, usually cobalt or nickel alu-
minide. The composite is formed by liquid-phase sintering
a mixture of the carbide and binder metal powders. The
WC composites are known as hardmetals in Europe. They
are produced by MIM and used for wear, tooling, cutting,
or drilling applications.
Centrifugal atomizationthe formation of spherical parti-
cles by combining a melt with a centrifugal force such that
the melt is disintegrated into high-velocity droplets that
spheroidize prior to solidification.
Cermeta composite body consisting of ceramic particles
bonded with a metal. Cemented carbides are the most com-
mon cermets in production.
CIMceramic powder injection molding, where the inor-
ganic phase in the polymer binders is predominantly a ce-
ramic or mixture of ceramics; a subset of powder injection
molding.
Clamping forceone of several measures of the capabilities
of a molding machine. In this case the available force for
holding the mold together while pressurized feedstock is
filling the cavity. If the applied pressure times the projected
part area exceeds the clamping force, then the cavity will
open or flash during molding.
Classificationseparation of a powder into fractions accord-
ing to particle size.
Closed-loop feedback controla molding concept for pre-
cise dimensional control where the pressure inside the
molding cavity is monitored during filling and used to ad-
just the molding machine operation to ensure repeatable
filling, weight, and final dimensions. The weight and di-
mensional scatter in final parts is greatly reduced by using
this control logic as compared to open-loop or adaptive
controls.
Closed porean isolated pore not linked to the external sur-
face, usually formed in sintering after a component reaches
a density of approximately 92% of theoretical, resulting in
capture of the process atmosphere in the pores.
Coarseningthe progressive enlargement of the grain size
or pore size during sintering due to diffusion, coalescence,
or solutionreprecipitation processes. With respect to prop-
erty control, microstructure coarsening is very important.
Coefficient of variationthe standard deviation divided by
172
the mean value, giving a nondimensional measure of uni-
formity such as in dimensions, weight, or properties.
Coiningthe final pressing of a sintered compact to obtain a
definite surface configuration, flatness, or surface finish.
Co-moldingalso known as two-color molding because of
the use in plastics. Two materials are shot into the same
cavity from different injector units to form layers, inter-
connections, or other differences in materials with position
in the cavity.
Compositea mixture of two or more powders that form a
multiple phase structure, typically designed to deliver prop-
erties that are a hybrid of the constituent properties.
Compressionthe removal of air from melting feedstock by
applying heat and pressure. In a reciprocating screw mold-
ing operation this is achieved by tapering the screw to re-
duce the space for the feedstock during metering toward
the screw tip.
Computer-aided designthe use of a computer program
geared to perform stress, fit, dimensional, and other calcu-
lations and to allow visualization prior to fabrication.
Computer-aided manufacturingcomputer controlled ma-
chines used to ensure proper fabrication and resource uti-
lization; included in the broad category of computer
integrated manufacturing and computer-aided manufactur-
ing activities might be inventory control, maintenance
schedules, production scheduling, tool-path analysis, and
cost analysis.
Critical loadingthe maximum volume fraction of solid par-
ticles that can be incorporated in a polymer binder without
forming pores while still allowing flow in normal injection
molding situations.
Cross linkingthe formation of bonds between polymer
chains to give rigidity and strength to the polymer. Ther-
mosetting polymers that harden on first heating are exam-
ples of cross-linked polymers.
Cycle timea critical measure of molding-equipment pro-
ductivity, it is the time for completion of one molding
cycle. It can be measured from the time to start filling to
the start of the next fill. Cycle times from a few seconds to
5 minutes are encountered in MIM production.
Debindinga step between molding and sintering where the
majority of the binder used in molding is extracted by heat,
solvent, catalysis, or other techniques. The debinding tech-
niques are highly variable between production sites. Ther-
mal debinding is most common and is the oldest version,
but several operations rely on alternatives or combinations
of methods.
Delaminationthe cracking of a molded compact, often lead-
ing to a hairline crack oriented perpendicular to the ejection
direction that is hidden until debinding or sintering.
Densificationthe change in porosity divided by the initial
porosity due to pressing or sintering. A term loosely asso-
ciated with property gains in sintering.
 Table of Contents
Densitythe mass divided by the volume, usually expressed
in g/cm3 (equivalent to Mg/m3) or sometimes given as a
ratio to pycnometer or theoretical density.
Dew pointa measure of atmosphere purity based on water
content, it is the temperature where moisture condenses out
of a process atmosphere. Generally it is assumed a low dew
point (low condensation temperature) corresponds to a
cleaner atmosphere.
Differential scanning calorimetrya means to determine
the heat flow into or out of a MIM sample. Usually it is ap-
plied at lower temperatures where the polymer melts, crys-
tallizes, or evaporates.
Differential thermal analysisthe careful measure of tem-
perature and temperature difference during heating for a
MIM material and a reference. Whenever the polymer
melts or a phase transformation occurs the test sample will
lag behind the reference, while if there is a reaction the
sample will heat faster than the reference.
Dihedral anglea microstructure feature associated with
phase boundaries in a material. It is evident when a grain
boundary intersects with a pore or liquid phase, resulting
in a groove in the grain boundary at the point of intersec-
tion. Low dihedral angles are associated with high grain-
boundary energies and unstable microstructures.
Dilatant flowviscosity that changes with flow conditions
where the mixture actually dilates under stress.
Dilatometrymeasurement of dimensional change during
thermal processing to determine the sources of sintering
densification, phase transformation, or other causes of
dimensional-control problems. Usually dilatometry is con-
ducted with a laser or a pushrod that makes contact with
the specimen in a furnace. Sintering experiments intended
to find the densification temperature for a powder are con-
ducted using constant heating-rate dilatometry.
Dimensional controlthe repeatability of final dimensions
in a MIM operation as measured by part-to-part, day-to-
day, and batch-to-batch scatter. Usually quantified by the
standard deviation (or a multiple of standard deviations,
perhaps as large as six) observed in a dimension as nor-
malized by that dimension, expressed as a percent, for
example, a coefficient of variation of 0.1%.
Ductilitya measure of the permanent stretching or defor-
mation a material can take prior to failure. Ceramics have
no ductility, while materials such as stainless steel and alu-
minum exhibit large ductilities, often measured at 30% to
60% stretch or elongation prior to failure.
Ejectionthe final stage of molding where the powder
binder compact is forced out of the die.
Ejector pinsmechanical pins that insert into a die cavity to
act as a means to push the molded component out of the
tooling.
Elastic modulusthe material property linking stress and
strain. Also known as Youngs modulus or stiffness. In
Chapter 9: Supplemental Information
MIM feedstock it also is important because it determines
residual stress and the stress relaxation time in cooling.
Elasticitythe spring-back of feedstock after ejection from
a tool set. Formally, the elastic modulus is the material pa-
rameter that links stress to strain, but in MIM it is largely
related to die sticking and dimensional-control problems
since ejection and debinding induce stresses that might
cause warpage or loss of component precision.
Elemental powderpowder of a single chemical species like
iron, nickel, titanium, copper, or cobalt, with no alloying
ingredients.
Elongation to fracturea measure of ductility, since this is
the amount of permanent plastic stretch a material under-
goes prior to failure in a tension test. The most common
measure used for ductility.
Equiaxed powderparticles with approximately the same
size in all three (perpendicular) dimensions. A sphere
formed by gas atomization is the classic example of this
desired particle shape.
Feedstockthe mixture of powder and binder used in injec-
tion molding. Its formulation involves decisions on the
powder composition, particle characteristics, binder for-
mulation, mixing practice, and ratio of powder to binder.
Ferritevarious compositions based on iron oxide used to
fabricate common magnets, usually containing other metal
oxides or ceramics, including zinc oxide. It is also the name
given to the body-centered cubic form of pure iron.
Ferritic steelan alloy based on iron that consists of a body-
centered cubic-crystal structure. It is the most common
form of low-alloy steel when slow cooled, but is usually
heat treated to form martensite.
Fillingthe first phase of the molding cycle where feedstock
is flowing into the mold cavity under pressure from the
screw.
Finishing operationsthe steps applied to a MIM compo-
nent after sintering to tailor the dimensions, properties, or
attributes to the application; examples include machining,
polishing, heat treatment, straightening, and electroplating
steps.
Flashinga lip of extruded feedstock that penetrates along
the parting line of a die cavity due to excessive pressure,
poor tool tolerances, or binder separation from the powder.
Flow analysiscomputer-simulated molding to assess the lo-
cation of the gate, runner, vent, and cooling passages and
other important aspects of tool design and molding to min-
imize errors in production.
Flow timethe time required for a powder sample to flow
through an orifice in a standardized test. The flow time
gives a measure of the interparticle friction. Most powders
used in MIM are not free flowing due to the typical small
particle size.
Fracture toughnessa measure of the resistance to crack
propagation in a material, related to the applied stress when
173
 Table of Contents
Chapter 9: Supplemental Information
the crack starts moving and the crack size. High-fracture-
toughness materials, such as stainless steels, provide more
safety when compared with low-fracture-toughness mate-
rials such as glass or ceramics.
Freeze firinga concept developed in the 1960s using water
as the binder and freezing in the tool cavity. The frozen
compact is ejected and subjected to sublimation to extract
the water without melting the ice.
Gas-atomized powdera rounded or spherical powder
formed by the disintegration of a melt stream by a high-
pressure-gas expansion nozzle. The particles solidify dur-
ing free flight after atomization.
Gatethe constricted opening into the flow path at the entry
to the die cavity in the injection molding tool set. It should
be the first portion of the flow path to solidify after filling
the mold.
Gelationa binder-setting process where a macromolecule
grows in a binder solution to form a highly interlinked
structure with most of the binder water trapped in cells
formed by the long-range molecule.
Gelcastinga variant of low-pressure injection molding
where the binder consists of a monomer that polymerizes
in the die cavity, forming a rigid polymer to hold the par-
ticles in place, often supported by a catalyst addition just
as the feedstock is molded.
Granulationa term describing the agglomeration of pow-
der or the breaking apart of lumps, runners, sprues, or parts
that are reformed into pellets for reloading into the molding
machine.
Green statethe condition of the molded component prior
to debinding or firing. The term comes from the ceramic
concept of greenware in reference to formed bodies that
are not sintered.
Green strengththe strength of the as-molded component
at room temperature.
Guided wavean in situ inspection technique involving
wave propagation through a MIM compact while still in
the mold. Mathematical transform of the scattering behav-
ior detects defects, allowing for sorting on ejection.
Hardmateriala group of compounds that are typically
metal carbides, borides, oxides, or nitrides that exhibit a
high hardness. To fabricate components, various combina-
tions of the hardmaterials are combined with cementing
metallic phases during liquid-phase sinteringfor exam-
ple, TiC is sintered with Ni or Fe to form a cermet.
Hardmetallike hardmaterials, but usually reserved for the
WC-Co family of hardmaterials or cermets. This name for
the cemented carbides is more popular in Europe but in
general it reflects the high hardness after sintering.
Hardnessa formal test of resistance to penetration by a
pointed or rounded indenter under a given load. There are
many useful hardness tests where the depth or width of the
penetration is measured. Often this relates to other material
174
properties such as strength.
Heavy alloya class of high-density alloys based on tungsten
with small concentrations of alloying additions such as
nickel, iron, or copper. These alloys are liquid-phase sin-
tered from mixed elemental powders to create a composite
material. Most of the applications are for weights, radiation
shields, thermal management heat sinks, or projectiles.
Hot isostatic pressinga process combining temperature and
high-pressure gas to fully densify a sintered MIM structure.
Used only for very-high-performance structures.
Hot runnera tool cavity where the flow path is kept hot
between shots to eliminate recycle of runners and sprues.
The built-in heaters and valves are coordinated with the
molding machine to ensure no freezing in the flow path.
Impregnationliquid polymer filling of open pores from an
external surface, used to seal open pores after sintering for
improved lubrication, corrosion resistance, or machining.
Inconela variety of nickel-base alloys invented by Interna-
tional Nickel Company designed for high-temperature ap-
plications, ranging from furnace components to jet engines.
Infiltrationthe process of filling the pores of a compact
with a lower-melting-temperature metal or alloy. It is one
means of forming low-cost tooling or making dense struc-
tures via molding a porous preform and filling the pores
with liquid metal, such as Al or Cu.
Injection moldinga hydrostatic forming technique for
shaping powders using plastic binders and relatively low
temperatures and pressures.
Interparticle frictionthe friction between powders which
limits sliding, packing, and densification.
Invarlow-thermal-expansion alloys of ironnickelcobalt
where a martensitic phase transformation is balanced
against the thermal expansion coefficient to give a near-
zero thermal expansion coefficient over a range of temper-
atures.
Irregular powdera powder which lacks shape symmetry
in the individual particles.
Jettinga condition that arises with the rapid filling of an in-
jection mold where the feedstock shoots across the mold
and fills back toward the gate. Generally this is unaccept-
able for quality components.
Knit linethe same as a weld line. A linear defect occurring
where feedstock streams merge in the cavity because of
two gates or flow around a core or other solid portion of
the die.
Kovara glass-to-metal sealing alloy used for microelec-
tronic packaging and other situations where a matched
thermal expansion or graded structure is required.
Laminationa layered structure or cracking in the pressed
compact resulting from ejection stresses exceeding the
green strength.
Liquid-phase sinteringsintering at a temperature where a
liquid and solid coexist due to chemical reactions, partial
 Table of Contents
melting, or eutectic liquid formation. It is most useful for
stainless steels, tungsten, cemented carbides, cermets, and
ferrous alloys containing phosphorus, boron, copper, or sil-
icon.
Low-pressure moldingthe use of lower pressures and low-
viscosity binders (largely water-based or wax-based) to fill
out a complex tool cavity without packing the shape. This
route is successful for components where internal flaws are
not a concern, such as nozzles, spray tips, or other geome-
tries where external geometry is the key concern.
Lubricantan organic additive which is mixed into the feed-
stock to minimize die wear and aid in ejection after com-
paction.
Maraging steela high-strength, high-toughness class of
ironnickelmolybdenum alloys that lack carbon, yet form
their strength by an age-hardening process in the marten-
sitic phase. Strength levels of 4 GPa or 600 ksi are possible
with these alloys.
Martensitethe distorted ferrite crystal structure (the com-
mon variant is a distorted body-centered cubic phase that
forms a body-centered tetragonal structure) due to carbon
supersaturation in the rapidly cooled ferrite. Martensite is
hard and brittle, hence it is usually tempered to partially
relax the hard phase for improved toughness.
Mean sizethe average value from the particle-size distribu-
tion.
Mechanical alloyingthe formation of an alloy powder by
milling elemental powders for a prolonged time; frequently
used to create amorphous or dispersion-strengthened alloy
powders via attritor milling.
Median sizethe centroid of the particle-size distribution,
where half of the particles are larger and half are smaller;
not necessarily the mean, but easily identified as the 50%
value. Also known as the D50 particle size.
Melt indexa measure of flow at low-shear-strain rates,
where a capillary tube is used to extrude molten feedstock
under a dead load. The melt index depends on the capillary
tube diameter and applied load, but is always reported as
the grams of feedstock collected from the tube in 10 min-
utes.
Meteringcontrolled forward extrusion of molten feedstock
past the screw tip and check ring to ensure the proper shot
volume is ready for the next filling event.
Microminiaturecomponent dimensions measured in the
micrometer size range, requiring microscopes for evalua-
tion. These are typically formed using special machines
with smaller shot size to enable better precision for small
bodies.
Micromoldinga new class of MIM technologies geared to
the production of components in the millimeter and mi-
crometer size range, requiring new technologies in tool fab-
rication, nanoscale powders, and molding machine
operation and construction. Most of the early applications
Chapter 9: Supplemental Information
are in microelectronics, medicine, sensors, and optical
communication systems.
Microstructurethe detailed information on the microscopic
phases, pores, grains, defects, heterogeneities, and other
property-controlling features.
MIMmetal powder injection molding, where the inorganic
phase suspended in the binder is predominantly a metal or
alloy powder. Common engineering alloys are possible by
mixing elemental powders and forming the alloy during
sintering (homogenization) or by use of a prealloyed pow-
der where each particle contains all of the elements. MIM
is a subset of powder injection molding (PIM).
Mixingthe thorough intermingling of powders of two or
more different compositions. It is also used to describe the
compounding of feedstock by the thorough distribution of
binders between particles.
Mold-flow simulationthe use of computer simulations for
analysis of the flow, packing, venting, sizing, cooling, and
other events and tool or machine parameters associated
with MIM forming operations.
Mold releasea spray or coating that reduces component
sticking to the die cavity, aiding ejection without defects.
Moldabilitya relative measure of the ease of filling out a
tool cavity during injection molding. It can be determined
by the length of filling for a long, narrow passage. The cur-
rent test of plastic moldability is to measure the filling of
a spiral, which has been altered in MIM to a zig-zag fill
test to induce powderbinder separation along the flow
path.
Multimodala powder-size distribution which exhibits
several modes, possibly generated by blending several
monosized powders.
Nanoscalepowders or microstructures with sizes that can
be measured in nanometers. Typically the powders are less
than 100 nm in size, or less than 0.1 m.
Near-net shapea discrete component without a need for
final machining which many production technologies, in-
cluding MIM, attempt to form. Together these technologies
are considered net-shape approaches, but when critical di-
mensions cannot be held without a final machining step
they are termed near-net-shape technologies.
Net shapea compact manufactured to final density and
dimensions without the need for machining. MIM is a net-
shape process.
Newtonian flowan idealized viscosity situation applicable
to a few fluids where there is no sensitivity to the shear-
strain rate, only temperature. The stress is proportional to
the shear-strain rate.
Nodular powderirregular particles with knotted, rounded
shapes. This is characteristic of water-atomized powders.
Open porea pore completely through a compact from one
surface to another. During thermal debinding the pores must
be open to allow the escape of evolving vapors. Open pores
175
 Table of Contents
Chapter 9: Supplemental Information
close in sintering at a density near 92% of theoretical.
Oxidationreductiona combination of atmospherepow-
der reactions that can extract oxygen (reduction) or deposit
oxygen (oxidation) in a powder compact during heating,
especially during debinding and sintering. Reduction con-
ditions are usually required for sintered metallic materials.
Packing pressurethe peak pressure encountered in the
molding operation once the die is filled, prior to freezing
of the gate. It is precisely controlled for weight uniformity
and optimized final dimensional control.
Particle sizethe controlling linear dimension of an individ-
ual particle, as determined by analysis with screens, lasers,
or other sensing techniques.
Particle-size analyzeran automated device for determina-
tion of the particle-size distribution. These are widely used
in the research, product-development, and quality-control
functions as part of MIM to ensure repeatable powders for
the process.
Parting linethe linear mark on a compact where two
separate tool or die pieces mated during shaping. In injec-
tion molding it is where the two halves of the die joined
together.
Pelleting or pelletizingthe formation of discrete chunks of
feedstock with repeated sizes and shapes that allow easy
flow and filling of the molding machine.
PIMpowder injection molding is the comprehensive term
for forming inorganic engineered components in a plastic
molding machine using thermoplastic binders and inert
powders, with subsets including metal powder injection
molding (MIM), ceramic powder injection molding (CIM),
and cemented carbide powder injection molding (CCIM).
Planetary mixerusually these have two offset mixing
blades that rotate around individual shafts and the two
blades further rotate around a center axis. The net effect is
intermixing and stirring and shear to produce MIM feed-
stock, usually in a heated vessel.
Plasma debindinga new technology similar to catalytic
debinding it involves a low-pressure plasma for depoly-
merization of the binder from the sample surface inward
to avoid heating and softening for distortion control.
Plunger moldinga hydraulic plunger is used to push
molten MIM feedstock into a die cavity, where motion and
pressure are controlled by the applied hydraulic pressure.
Pneumatic moldinguse of an air-pressure head over
molten feedstock to push it into a die cavity. This is the
lowest-cost and least-precise form of MIM, yet is widely
employed in the fabrication of large ceramic structures.
Polydisperseimplies a broad powder-size distribution, cov-
ering a wide range of particle sizes with no clear mode size.
Pore sizethe size of the holes or voids between powder par-
ticles, often measured by microstructure quantification,
mercury porosimetry (open pores only), or other tests that
include gas flow, gas condensation, and polymer intrusion.
176
Porosimetera device for measuring the size of the open
pores using high pressures and mercury-intrusion tech-
niques.
Porositythe amount of void space in a powder compact.
Most MIM materials have less than 5% porosity after
sintering.
Powderparticles of solid matter characterized by a small
size, less than 1 mm in size. Most MIM powders are below
20 micrometers in maximum size.
Prealloyed powdereach particle contains an intimate mix-
ture of two or more elements in a prescribed ratio to form
an alloy; examples include brass, bronze, steel, and stain-
less steel.
Presinteringthe heating of a compact to a temperature
lower than the normal sintering temperature to gain
strength for subsequent handling, including machining.
This is often performed by heating above the highest tem-
perature required to thermally decompose any residual
binder polymer.
Pressure-assisted sinteringsintering with the application
of an external pressure. It is often performed by initially
sintering in vacuum and subsequently pressurizing the fur-
nace to densify any remaining closed pores. Best applied
to high-performance alloys or difficult-to-sinter alloys and
is commonly employed for nitrides that decompose during
sintering (AlN and Si3N4) and brittle materials that require
full density (WC-Co for example).
Pressure controlthe final phase of molding where the gate
is not frozen and the quantity of feedstock in the die cavity
is controlled by the pressure, thereby ensuring uniform
weight and dimensions in the final MIM component.
Proper control is achieved by monitoring pressure sensors
in the die-cavity wall.
Pseudoplastic flowa form of viscous flow where there is a
strain-rate sensitivity to the viscosity; generally it is the op-
posite from dilatant flow, since the viscosity decreases with
higher strain rates.
Pycnometera device for measuring the theoretical density
of a loose powder or preform. Helium and granular fluids
are commonly employed to encase the powder or compo-
nent for determination of true volume, and when coupled
with an independent mass determination leads to a density.
PZTpiezoelectric ceramic based on oxides of lead (Pb), zir-
conium (Zr), and titanium (Ti) used for ultrasonics, sen-
sors, sonar, and various signal-propagation or collection
applications, based on shapes fabricated by MIM.
Rapid prototypeseveral techniques emerging that bypass
a need for tooling by using computer-controlled freeform
fabrication to generate a first MIM green body, typically
via building the feedstock into a three-dimensional object
from a stack of closely spaced laminates.
Reciprocating screwinjection molding with a screw lo-
cated in a heated barrel. During compression, metering,
 Table of Contents
melting, and forward advance of the feedstock, the screw
is turning, while during mold filling it becomes a plunger
to quickly fill the cavity.
Reduced powdermetal powder produced by the chemical
reduction of a compound, most typically a metal oxide.
Refractorya metal or ceramic having a high melting tem-
perature, usually over 1,700C. Example refractory metals
are tungsten, molybdenum, rhenium, and zirconium, while
example refractory ceramics are alumina, zirconia, yttria,
and chromia.
Rheologythe study of strength, and elastic, plastic, and vis-
cous flow of polymers and feedstocks used in MIM.
Runnera portion of the feed path for filling an injection
molding die; the runner is between the sprue and gate, the
latter being the inlet to the actual die cavity.
Screwthe key portion of an injection molding machine for
metering and filling the die cavity. It has a taper along the
length to compress the feedstock as it is metered for re-
moval of trapped air and a check ring at the tip to allow
forward plunging motion during mold filling.
Screw moldinga reciprocating screw is used to mix, pres-
surize, de-air, and convey molten feedstock to finally be
injected into a tool cavity. This is the most widely em-
ployed means of injection molding precise components.
Secondary operationsthose activities performed to
adjust dimensions or properties of a component after sin-
tering. Examples include heat treatment, coining, machin-
ing, electroplating, impregnation, and shot peening.
Segregationnonuniform distribution of ingredients, such as
powder separation by size, shape, or density, or chemical
separation in the microstructure of a solidified material.
Sendustalloys of ironaluminumsilicon that have impor-
tant combinations of hardness and magnetic properties suit-
able for magnetic card readers and other high-use magnetic
applications. The alloy hardness is such that MIM is the
most credible manufacturing route.
Setterthe tray or shaped substrate for support of MIM com-
pacts during debinding and sintering.
Shear ratea measure of the rate feedstock is deformed and
forced to flow into a die cavity. It has units of inverse sec-
onds. Actually it is the shear-strain rate, representing the
sheared change in length divided by the original length in
unit time. For MIM, shear rates of several thousand per
second are normal, meaning the slug of feedstock is
quickly deformed to fill out the cavity.
Shot sizea historical basis for sizing molding machines.
This is usually given in terms of the amount of polystyrene
that can be moved through the nozzle in a single forward
motion of the screw or plunger.
Shrinkagea decrease in dimensions of a compact which
occurs during sintering, usually 15% or up to 25% in a few
cases and as low as 2% with multimodal powders.
SiAlONa group of commercial alloys based on silicon
Chapter 9: Supplemental Information
nitride, with liquid-phase formers such as alumina and yt-
tria. These are very strong and lightweight materials with
exceptional strength and reasonably high-temperature
properties (up to 1,200C) useful in high-temperature aero-
space, diesel engine, and military systems.
Sigma mixera closed mixer consisting of two blades that
look like the Greek capital letter sigma. These blades lift
and separate feedstock to form a mixture of moderate ho-
mogeneity. A major advantage for MIM is the small batch
size, but with low homogeneity due to the poor shearing.
Silicon carbidethe man-made compound SiC that is used
for applications ranging from sandpaper to high-tempera-
ture heating elements. It has excellent high-temperature
strength and oxidation resistance and is formed by MIM
when complex shapes are required without post-sintering
machining. Most systems rely on boron and carbon addi-
tions to improve sintering.
Silicon nitridecompositions based on the stoichiometric
compound Si3N4; in MIM this compound is used with var-
ious liquid-phase-sintering additives to form SiAlON com-
positions.
Sink marka shallow surface cavity that forms during cool-
ing. It is usually indicative of underpacking prior to gate
freezing and is corrected by a higher molding pressure.
Sinter hardencontrolled cooling from the sintering tem-
perature designed to induce desirable transformations and
microstructures, to avoid post-sintering heat treatments.
Usually it is applied to ferrous alloys to control martensite
formation on cooling.
Sinteringthe thermal process which bonds and densifies
the molded powders. It increases the compact strength via
diffusion or related atomic-level events. Most of the me-
chanical, magnetic, or other properties of a MIM part are
developed in the sintering cycle.
Sintering diagrama process map showing the interaction
of the key variables with respect to the densification of a
powder. The variables include particle size, grain size, tem-
perature, time, and applied pressure.
Solids loadingthe relative powder volume in feedstock de-
signed for binder-assisted shaping. Common injection
molding feedstocks have solids loadings near 0.6 or 60 v/o
powder.
Solvent debindingthe extraction of the binder or some
portion of the binder by leaching into a solvent. This is per-
formed by immersion or exposure to solvent vapors. The
solvent might be flammable, toxic, or simply water or
ethanol.
Speed controlthat initial portion of the molding stroke
where feedstock is flowing to fill the mold cavity and no
feedback signals are available; thus the molding machine
is simply controlled by the position of the screw during its
forward motion.
Spherical powderpowder with a uniform spherical shape
177
 Table of Contents
Chapter 9: Supplemental Information
and a size that can be characterized by a diameter. Gas-
atomized powders are often spherical.
Spruethe initial inlet into the die set for injection molding
feedstock. The sprue is usually filled by the nozzle and is
tapered for easy extraction after mold filling. It feeds the
runner system, which in turn feeds the gate.
Stainless steela wide range of alloys based on iron and
chromium that give corrosion resistance in most common
corrosive environments. The most popular MIM alloys are
300 series (austenitic) that contain high nickel levels, 400
series (mostly ferritic or heat treatable into martensitic) that
have little nickel, and precipitation-hardened alloys (mostly
heat treated into martensitic) such as 17-4 PH.
Stereolithographya group of rapid-prototype techniques
that construct a green MIM component from a computer
design by building thin sequential layers that match cross-
sections through the design, typically using laser, ink jet,
cutting, or micro-nozzle concepts.
Stress relaxationthe removal of residual stresses in a MIM
component by controlled cooling or reheating to relax the
polymer entanglement or other sources of stress.
Superalloythe highest-performance alloys used for the
most demanding applications. Typically these are formu-
lated with a wide range of components, based on nickel,
with tailored properties that allow use in moving compo-
nents in jet turbines.
Supercritical extractionthe use of high pressure and mod-
erately high temperatures to heat a solvent over the critical
point for binder removal. The most common supercritical
fluid used in MIM is carbon dioxide. Over the critical point
the gas is compressed to a density equal to that of the liq-
uid, so there is no volume change on binder removal.
Supersolidus sinteringa liquid-phase-sintering process ap-
plied to prealloyed powders where sintering occurs over
the solidus temperature, thereby nucleating liquid within
the particles to enable rapid densification of larger parti-
cles. It is used for high-alloy systems, such as tool steels,
stainless steels, cobaltchromium alloys, and superalloys.
Surfactanta surface active agent added to the binder sys-
tem in MIM feedstock to induce binder wetting of the pow-
der surface. There are many trade secrets, but most
powders prove responsive to simple soap compositions.
Tap densitythe density of a powder obtained when it is vi-
brated for a prolonged period. The tap density represents
the highest packing density possible for a powder without
the application of pressure.
Tensile strengththe maximum strength attainable prior to
failure during uniaxial-tension testing; properly known as
the ultimate tensile strength, but often given as the tensile
strength.
Thermal debindingextraction of the polymer in debinding
by the application of heat. The classic processes relied
on slow heating in air to evaporate the binder over a period
178
of days.
Thermogravimetric analysisa term that means measuring
the sample weight change vs. temperature during heating.
It is used for determination of polymer evaporation tem-
peratures by performing constant heating-rate experiments
and recording the sample weight vs. temperature.
Thermoplasticbinders or polymers (like wax) that soften
on heating, but stiffen on cooling, and can repeat the heat-
ing and cooling process without undergoing degradation.
Thermosettingbinder or polymer hardening due to heating
to a temperature where crosslinks form between the poly-
mer chains. Unlike a thermoplastic that can be softened by
reheating, a thermosetting binder becomes hard on the first
heating cycle and cannot be subsequently softened.
Thixomoldinga direct injection molding technique that is
a hybrid between powder injection molding and die cast-
ing. The alloy is partially melted to form a solidliquid
mixture that is initially stiff, but with continued stirring be-
comes low in viscosity and easily molded. Many MIM op-
erations had hoped for an aligned benefit for thixomolding
lower-temperature alloys and lighter-weight alloys. Cur-
rently only magnesium alloys are commercial, but alu-
minum and titanium variants have been demonstrated.
Thixotropica material-flow behavior where there is a yield
strength that must be overcome to initiate flow and with
shear the viscosity decreases; however, during aging weak
bonds form to increase viscosity.
Torque rheometera computerized mixer that allows inter-
active experiments to measure the torque for mixing as var-
ious changes are made in surfactants, powders, binders,
temperature, or shear rate. The output from the torque
rheometry provides an effective basis for formulation of
MIM feedstocks.
Toughnessliterally the energy to cause failure or fracture
per unit of test material cross-sectional area. It can be esti-
mated by the product of strength and ductility, but is more
typically measured using a swinging pendulum that frac-
tures a standard test sample (Charpy test). The kinetic en-
ergy loss by the pendulum is the toughness. In design, the
most important parameter is the fracture toughness, which
measures the stress required to cause a crack to propagate.
Transverse rupture strengththree-point fracture test ap-
plied to brittle materials or green compacts to assess relative
strength. The three-point bend test configuration is most
common for these materials with a four-point test some-
times used to measure the strength of sintered ceramics.
Twin-screw processora mixer or extruder consisting of
two intermeshing screws which convey, shear, and mix a
powder or powderbinder mixture during transport through
the mixer.
Ultimate tensile strengththe highest engineering stress en-
countered in the tensile testing of an engineering material.
Ultrasonic inspectionthe use of high-frequency sound
 Table of Contents
waves for detection of cracks or other defects. The most
common procedure uses through or back transmission
waves out of the die cavity, but new technology allows
guided wave inspection in the die cavity during molding.
Vacuum debindinga variant of thermal debinding that
gives extraction of binder by progressive heating in a vac-
uum to distill the molten binder out of the powder compact.
Usually vacuum debinding is performed with a gas inlet to
drag escaping binder to the pumps.
Ventthe thin channel in a tool set that allows air to escape
while molten feedstock enters through the gate. It is located
opposite from the gate and should not be sealed until late
in the molding cycle to avoid trapped air pockets in the
component.
Viscositythe resistance to flow of a feedstock or polymer.
Formally it is the proportionality between the stress and
applied shear-strain rate. A high viscosity makes molding
difficult. A low viscosity is easier to mold, but there are
often difficulties with flashing.
Water-atomized powderirregular or ligamental particles
formed by impacting a molten metal stream with high-
pressure water.
Water-based bindermostly gelation binders that become
rigid after molding by the formation of a long-range mol-
ecule with water in the structure. Other forms rely on water
saturated with silicates that form glasses on freezing.
Water-soluble bindera class of binders that rely on poly-
mers that are water soluble so solvent debinding takes
place in water. Most common are polyethylene oxide,
polyvinyl alcohol, and polyethylene glycol.
Waxpolymera general term for polymers used as binders
that consist of low-molecular-weight polymers and wax
fillers. The polymers are usually simple molecules such as
polyethylene, polypropylene, or co-polymers such as eth-
ylene vinyl acetate. These are the classic binders used in
MIM processing.
Weld linea linear defect or mark on a compact surface
where two portions were bonded, either in molding or sub-
sequently, such as in sinter bonding.
Weibull modulusa statistical parameter associated with the
spread in fracture strengths for ceramics. A high Weibull
modulus indicates a narrow spread in strength and few
manufacturing defects.
Wicking debindingthe most popular means of binder ex-
traction, especially for wax-based binders, where the com-
pact is packed in wicking powder and heated to a
temperature where the wax melts. The molten wax flows
by capillary action into the surrounding packing powder
and leaves the compact supported as polymer is extracted.
Yield strengththe end of the linear elastic region for most
metals. The yield strength is the highest stress a structure
can endure without permanent deformation and should be
the maximum stress used in engineering designs.
Chapter 9: Supplemental Information
Youngs modulusthe elastic modulus expresses the propor-
tionality between stress and strain during elastic deforma-
tion prior to permanent deformation; it is calculated
assuming linear behavior by dividing the applied stress by
the strain and for metals such as steel can range near 200
GPa (30,000,000 psi).
YTZyttriatoughened zirconia, where a few percent of
Y2O3 is added to ZrO2 to help in phase stabilization such
that a phase transformation occurs in the zirconia under se-
vere stress, such as at a propagating crack tip, to block frac-
ture.
Zeta potentialan electrochemical measure of the surface
charge or surface chemistry of a fine powder in a suspen-
sion.
Zirconiaa stoichiometric compound of zirconium and oxy-
gen in the ratio ZrO2. It is often phase stabilized by adding
other ceramics. It is widely employed as a MIM material
because of its exceptional wear resistance and difficulty in
grinding, and is colored to form a wide variety of semi-
precious jewelry items.
NOMENCLATURE
As with words, and consistent definitions, it is also impor-
tant to establish consistent symbols, especially since engineers
often speak in mathematical terms. It is frustrating to have
symbolic information provided without a guide as to how the
symbols are defined. It is unclear why the physics community
uses the theta symbol for time when the rest of us use t. Any-
how, here is my pass at a consistent nomenclature for MIM
giving commonly used mathematical symbols:
= numerical constant in calculations, dimensionless
= solid cone angle, degree
= strain or fractional porosity (sometimes given as %)
f = elongation at fracture, a strain, usually given as %
= viscosity, Pa-s
= surface energy, J/m2
= thermal conductivity, W/(m K)
= mean free separation or mean free path, m
= density, g/cm3 or kg/m3
= stress or strength, Pa; typical units here are MPa or
106 Pa
= shear stress, Pa; typical units here are MPa or 106 Pa
= friction coefficient, dimensionless
max = magnetic permeability, dimensionless
= angle, degree
= angle, degree
= solids loading, dimensionless or often given as
volume percentage
= 3.14159
= adhesion force, frequency, or other atomic-level
parameter
= atomic volume, m3/atom or m3/mol
A = area, m2
179
 Table of Contents
Chapter 9: Supplemental Information
B = magnetic response, T
c = fractional molar solubility, dimensionless
C = microstructure parameters such as contiguity,
connectivity
CP = constant-pressure heat capacity, J/kg or J/mol
dP = pore diameter, m; usually given as m
D = particle size or particle diameter, m; usually given
as m
if subscript is added such as 50 this refers to the
50% smaller size or median
DL = solid-phase diffusivity in liquid phase, m2/s
e = strain, dimensionless
E = elastic modulus or elasticity, N/m2; typical units are
GPa or 109 Pa
f = fraction, solids fraction or fractional solids loading
F = fractional attribute, fractional solid, fractional
liquid, and so on, dimensionless
F = force, N
g = gravitational constant, m/s2
G = grain size, m; typical units are m
h = height of component, m
H = enthalpy or heat, J/kg
Hc = coercive force, A/m
I = impact energy, toughness, J/cm2
k = Boltzmanns constant, 1.3810-23 J/(Katom) constants, basic measuring units, and conversion factors that
KIc = fracture toughness, MPa(m)2
l and L = characteristic sample dimension, m
M = mass, g
N = work-hardening exponent, dimensionless
P = pressure, N/m2
PG = gas pressure, N/m2
r = component radius, m
R = universal gas constant, 8.32 J/(Kmol)
Ra = average surface roughness, m or typically given
as m
RA = tensile test reduction in area at fracture,
dimensionless, often given as %
S = fractional saturation, dimensionless
t = time, s
T = temperature, K; more typically given as C and at
times F
V = volume fraction, dimensionless
W = weight fraction, dimensionless
w = width of component, m
x = coordinate, m
X = sinter bond neck size, m; often given in m
y = coordinates, m
Y = shrinkage, dimensionless; often given as %
z = coordinate, m
Z = tooling expansion factor, dimensionless; often
given as %
subscript A = additive
180
subscript B = base
subscript f = fracture or failure
subscript g = per grain
subscript i ...iv... ix = counting subscript
subscript L = liquid
subscript max = maximum value
subscript min = minimum value
subscript o = original or wrought
subscript P = pore (or vapor phase)
subscript S = solid
subscript LV = liquidvapor
subscript SS = solid-state
subscript SL = solidliquid
subscript SV = solidvapor
subscript U = ultimate
subscript V = vapor
subscript Y = yield
UNITS AND CONVERSION FACTORS
The primary units for international trade and commerce are
the SI units (Le systme international dunits), sometimes
called the metric units. This book follows international stan-
dards, and where possible secondary English units are given
parenthetically. In the SI units, there are fundamental defined
allow for easy manipulation between related topics, such as
energy, heat flow, and heat capacity.
Helpful Relations
Here are a few simple relations that help introduce the
basic conversions:
a meter is just a little longer than a yard (3 feet)
25 millimeters (25 mm) is about one inch
a common U.S. dime (10 cents) is about one millimeter
thick
a kilogram is a little more than two pounds
one gram (0.001 kilogram) is a little heavier than a paper
clip
one liter is a little larger than a quart
water boils at 212F (degree Fahrenheit) which equals
100C (degree Celsius)
water freezes at 32F which equals 0C
the 180F gap between freezing and boiling water is
100C
4,000 joules is about one kilocalorie (what the food
industry calls a calorie)
usual daily caloric intake for an adult human is 8,000,000
joules (8 MJ)
humans would need just over 2 kWh per day if they
could run on electricity
one hectare (10,000 square meters) is about two and one-
half acres
 Table of Contents
Chapter 9: Supplemental Information

Prefixes V (volt) = J/C

Associated with the various measures are prefixes that help (ohm) = 1 V/A
convey the powers of 10 associated with the unit, so a meter S (siemens) = 1 A/V
(m) when divided into 100 is known as a centimeter (cm), and T (tesla) = Vs/m2
when divided into 1,000 is known as a millimeter (mm), and H (henry) = Vs/A
so on down to one part in a million which is a micrometer F (farad) = As/V
(m), and one part in a billion is a nanometer (nm). These pre- Wb (weber) = Vs
fixes are standardized independent of the unit as follows: lm (lumen) = cdsr
T (tetra) = 1012
G (giga) = 109 Important Constants
M (mega) = 106 A large body of factors is created from defined constants.
k (kilo) = 103 For example, the speed of light is a combination of length
c (centi) = 10-2 and time. Below are important constants encountered in
m (milli) = 10-3 engineering:
(micro) = 10-6 acceleration of gravity = 9.8 m/s2
n (nano) = 10-9 atomic mass unit = 1.66110-24 g
p (pico) = 10-12 Avogadros number = 6.0221023 molecules
Bohr magneton = 9.2710-24 Am2
Fundamental Units Boltzmanns constant = 1.38110-23 J/K
As a basic definition, the SI units assume certain funda- electric permittivity of vacuum = 8.85410-12 C/V
mental units for time, mass, length, current, temperature, and elementary charge = 1.60210-19 C
such. These are designated as follows: Faradays constant = 9.65104 C/mol
length = m (meter) gas constant = 8.314 J/(molK)
time = s (second) gas volume at standard temperature and pressure =
mass = kg (kilogram) 0.0224 m3
amount = mol (mole) electron mass = 9.1110-28 g
current = A (ampere) permeability of vacuum = 7.958105 H/m
temperature = K (kelvin) Plancks constant = 6.62610-34 Js
plane angle = rd (radian) speed of light = 2.998108 m/s
solid angle = sr (steradian) StefanBoltzmann constant = 5.6710-8 J/(m2sK4)
luminous intensity = cd (candela)
Note that we tend to use the degree Celsius for temperature, Conversion Factors
which is based on the triple point of water (this is the only The shift from one measurement system to another in-
temperature where solid, liquid, and vapor water coexist and volves a conversion factor. The listing below allows conver-
is designated as 0C). This occurs at slightly more than sions. The metric unit is shown in terms of equivalent units in
273 K. other systems. For example, one meter is 39.4 inches. The re-
verse conversion is based on dividing. To go from inch to
Derived Units meter, divide by 39.4, given that one inch is 0.0254 meter (or
Besides the defined units, some common terms and units one inch equals 25.4 millimeter). There are several popup or
are widely used in the SI system and are simply derived from Internet routines that handle most of these conversions.
the fundamental units listed above:
Hz (hertz) = 1/s length conversions
h (hour) = 3,600 s 1 m = 39.4 in. (inch)
min (minute) = 60 s 1 m = 3.28 ft. (foot)
L (liter) = 10-3 m3 1 m = 1.09 yd. (yard)
t (metric or long ton) = 1,000 kg 1 cm = 0.394 in. (inch)
C (Celsius) = K 273 1 mm = 0.0394 in. (inch)
N (newton) = kgm/s2 1 m = 39.4 in. (microinch)
Pa (pascal) = N/m2 1 nm = 10 (angstrom)
bar (bar) = 0.1 MPa
J (joule) = Nm area and volume conversions
W (watt) = J/s 1 cm2 = 0.155 in.2 (square inch)
C (coulomb) = As 1 m2 = 1550 in.2 (square inch)

181
 Table of Contents
Chapter 9: Supplemental Information

1 cm3 = 0.061 in.3 (cubic inch) energy conversions

1 m3 = 35.3 ft.3 (cubic foot) 1 J = 9.4810-4 btu (British thermal unit)
1 m3 = 1,000 L 1 J = 0.737 ft.lb. (foot pound)
1 L = 1,000 cm3 (cubic centimeter) 1 J = 0.239 cal (calorie)
1 L = 0.264 gal (gallons) 1 J = 107 erg
1 L = 1.06 qt. (quart) 1 J = 2.810-7 kWh (kilowatt hour)
1 L = 0.035 ft.3 (cubic foot) 1 J = 6.241018 eV (electron volt)
1 J = 4.83 hph (horsepower hour)
amount of substance conversion 1 J = 1 Ws (watt second)
1 mol = 6.0221023 molecules 1 J = 1 VC (volt coulomb)
1 kJ = 0.239 kcal (kilocalorie)
density conversions
1 Mg/m3 = 1 g/cm3 power conversions
1 g/cm3 = 0.0361 lb./in.3 (pound per cubic inch) 1 W = 0.737 ft.lb./s (foot pound per second)
1 kg/m3 = 10-3 g/cm3 1 W = 1.3410-3 hp (horsepower)
1 kW = 1.34 hp (horsepower)
temperature conversion
to convert K to F (Fahrenheit), multiply by 1.8 then thermal conversions
subtract 459.4F 1 J/(kgK) = 2.3910-4 btu/(lb.F) (British thermal unit
to convert K to C, subtract 273 per pound per degree Fahrenheit)
to convert C to F (Fahrenheit), multiply by 1.8 then 1 J/(kgK) = 2.3910-4 cal/(gC) (calorie per gram per
add 32F degree Celsius)
1 W/(mK) = 0.578 btu/(ft.hF) (British thermal unit
heating and cooling-rate conversions per foot per hour per degree Fahrenheit)
1 K/s = 1C/s = 1.8F/s 1 W/(mK) = 2.3910-3 cal/(cmsC) (calorie per
1 K/min = 1.8F/min centimeter per second per degree Celsius)

mass conversions viscosity conversions

1 g = 0.035 oz. (avoirdupois ounce) 1 Pas = 1 kg/(ms)
1 kg = 2.2 lb. (pound) 1 Pas = 10 P (poise)
1 Mg = 1.1 short ton (2,000 pounds) 1 Pas = 103 cP (centipoise)

force conversions stress-intensity conversion

1 N = 0.102 kg force
1 N = 105 dyne
1 N = 0.225 lb. force (pound force)
pressure, stress, strength conversions
1 Pa = 0.0075 torr (millimeter of mercury)
1 Pa = 10 dyne/cm2 (dyne per centimeter square)
1 kPa = 0.145 psi (pounds per square inch)
1 MPa = 9.87 bar (atmosphere)
1 MPa = 145 psi (pounds per square inch)
1 MPa = 0.145 kpsi (thousand pounds per square inch)
1 GPa = 145 kpsi (thousand pounds per square inch)
1 kg/cm2 = 14.2 psi (pounds per square inch)
velocity conversions
1 km/h = 0.62 mph (miles per hour)
1 km/h = 54.7 ft./min (foot per minute) = 0.91 ft./s
(foot per second)
1 mm/s = 2.36 in./min (inch per second)
1 MPam = 0.91 kpsiin. (kilopounds per square inch
times square root inch)
magnetic conversions
1 T = 104 G (gauss)
1 A/m = 1.25710-2 Oe (oersted)
1 Wb = 108 Maxwell
MISCONCEPTIONS
Many engineers think of MIM as a new field. In many re-
gards it is new, but various forms of the technology have been
in commercial use since 1975. Over the 36 years MIM has
been in practice, some folklore has arisen, leading to some
broadly held misconceptions. This section gives a few and
helps clarify some of the common myths.
Special molding machines are required.
Certain molding-machine characteristics are better suited
to MIM, especially for higher levels of precision, but many
firms are successful with a variety of molding machines. The

182
 Table of Contents
evidence suggests that there is no single manufacturer or
design that is necessarily best.
There is easy money to be made in MIM.
With its maturation, MIM has largely remained a healthy
technology that continues to grow by satisfying more sophis-
ticated customers. But most of the firms with favorable finan-
cial positions tend to have much practical experience, and
almost all of the leaders are well over 10 years old. These
leaders have established manufacturing systems which are
benchmarks for all net-shape manufacturing. To assume that
a late entry will jump into this class is not realistic. Yes, there
is success if you work long and hard, but MIM is not a route
to quick riches.
Only a few firms understand the technology.
There are more than 350 firms, several university pro-
grams, and a dozen research institutions supporting the tech-
nology developments in MIM. None of these has a lock on
the technology. The basic concepts are well documented in
books, and many of the suppliers will gladly share technology
to help create new customers. As is often known as Prados
Principle, 20% of the firms have 80% of the sales, but it is not
wise to assume any single firm is smarter or more knowledge-
able than another. The field has many successes, failures, and
much cross-fertilization, so there is no single pathway or tech-
nology that will give success.
A license is required to practice MIM.
Some of the first commercial developments in powder in-
jection molding date from the 1930s and 40s, so long ago we
have trouble reconstructing the history. In the 1960s, Corning
used ceramic molding to form tableware, a technology that
survives in the ceramic casting core business. In MIM, the
original patents were issued in the 1970s. Thus, with the tech-
nology aging this much, many of the suppliers provide an ex-
cellent support base for start-ups. The generic technology
works well and most of the ingredients and equipment can be
ordered with help from the equipment manufacturers. So, if a
company wants to have its hand held during the start-up phase
using a license, that is fine, but to think there is a proprietary
or special technology that requires a license does not reflect
reality.
You need a materials scientist to succeed.
This is a myth that is close to reality in some cases. Several
plastic injection molders have successfully entered MIM, but
at times they struggle with controlling impurities, optimizing
properties, selecting proper heat treatments, and other basic
aspects of materials engineering. These can be mastered with-
out full-time staff. One start-up used a skilled metallurgist who
was staying home with young children on an as-needed basis.
Another plastic firm initially relied on feedstock and furnace
Chapter 9: Supplemental Information
vendors for support. A plastic molder in Canada used a net-
work of small contracts to seed critical developments during
start-up. However, to become established at more than, say,
$5 million in annual sales requires mastery of three technolo-
giesmaterials engineering, manufacturing engineering, and
plastic molding, but you can start with only one or two.
Powder costs will continue to fall.
As MIM grows, some of the powders (especially preal-
loyed stainless steel) have declined in cost as consumption
has increased. But in all cases, consumption has increased
faster than the cost reduction. In other words, if powder sales
(dollars) have increased 20% per year then powder shipments
(tons) went up 40% per year. Along with improved process
yields, the increasing tonnage generated a price decrease. For
some alloy chemistries this sort of volume growth is not going
to be sustained.
Component producers that purchase the same feedstock
allow for movement of an order between sites.
Clearly, a trend has been to rely on purchased feedstock,
but that does not cure the need for similar molding practices,
processing equipment, and secondary operations. In general,
tool design is highly variable and customized to the shop and
molding machine. Hence, moving an order to a second site
might encounter problems beyond the feedstock, so there is
no assurance of success along these lines.
Newer binders enable cost-effective production of large
components.
The binder is not a barrier to molding large components.
Large stainless steel components were formed from in the
1980s using MIM, ranging up to 12 kg (26 lb.), using a tradi-
tional waxpolymer binder. Today large ceramic components
are in production at several sites. There are problems with
large components, but more significant are the economic bar-
riers. In metals, as the mass increases the raw material cost
difference (powder vs. casting cost) leaves plenty of margin
for machining after casting. Thus, changing the binder will
not solve the critical powder-cost issue. Indeed, several eco-
nomic problems arise with large MIM components; besides
powder cost there are penalties because the molding machines
are larger, molding cycles are slower, and debinding times are
longer. Accordingly, MIM is less cost effective as size in-
creases, independent of the binder.
Low-pressure molding is less costly.
Low-pressure MIM machines are used to reduce machine
cost and tool wear. They are used by about 15% of the com-
bined metals and ceramics industry. The manufacturing cost
is now lower with low-pressure molding, since these options
lack automation and usually only fill a single cavity. So with
a lower purchase price (capital cost) there is a burden of a
183
 Table of Contents
Chapter 9: Supplemental Information
higher operating cost. For example, in a comparison across
the industry, low-pressure molding shops show lower sales
per employee. More important, without a high molding pres-
sure the components tend to have more internal defects. For
surface features (such as in sandblast nozzles and watchcases)
there is no problem.
Continuous sintering in hydrogen is a MIM evolution.
The use of hydrogen sintering in a pusher furnace was ap-
plied to refractory metals and stainless steel from prior to the
1940s. It was widely used for that application at many sites
prior to the first commercial MIM use in 1985. Today, con-
tinuous furnaces constitute about a quarter of the installed sin-
tering capacity in MIM and prove most productive for
long-running components. Thus, they are more common in
large Asian shops fabricating consumer products, cellphone
parts, and watchcases.
The MIM technology is only capable of holding
tolerances at 0.5%.
High tolerances are always possible using post-sintering
machining. The real challenge is in forming tight tolerances
without coining or machining. The as-sintered tolerances de-
pend on the age of the technology, recognizing newer opera-
tions employing better binders, molders, and process control
during debinding and sintering. Essentially all of the industry
can hold dimensional scatter to within 0.3% (one standard
deviation), but a few have reached a coefficient of variations
as small as 0.02% with newer controls, such as closed-loop-
feedback control during molding. A few products are in pro-
duction with absolute tolerances of 2 to 5 m.
Metal powder injection molding is just another form of
powder metallurgy.
Although a few companies practice both MIM and tradi-
tional die compaction and sintering (what is generally implied
by the term powder metallurgy), largely the two have little in
common other than relying on metal powders. MIM powders
are much smaller and round if not spherical, and sintering tem-
peratures are higher such that the final density and performance
are higher.
STANDARDS
The standards used in MIM usually start at the national
level, say, in Japan or the U.S., and then become adopted in
the powder community, and spread to national bodies, even-
tually becoming part of the ISO standards. For North America,
the example would be the Metal Powder Industries Federation
(MPIF) as the regional trade organization, which then feeds
its standards into the American Society for Testing Materials
(ASTM), which participates in the International Standards
Organization (ISO) global standards. The standards are
reviewed and updated periodically, and expanded to handle
184
new alloys, processing options, or test parameters. Listed
below are some of the ones seen in MIM.
Testing Powders
Analysis by Microscopical Methods for Particle Size
Distribution of Particulate Substances of Subsieve Size
ASTM E20
Apparent Density of Non-Free Flowing Metal Powders Using
the Hall ApparatusMPIF 04, ASTM B212, ISO 3923/1
Apparent Density of Non-Free Flowing Metal Powders Using
the Carney ApparatusMPIF 28, ASTM B417, ISO
3923/1
Apparent Density of Metal Powders Using the Arnold
MeterMPIF 48, ASTM B703
Apparent Density of Refractory Metals and Compounds by
Scott VolumeterASTM B329
Average Particle Size of Metal Powders Using the Fisher
Subsieve SizerMPIF 32, ASTM B330, ASTM C72
(see also ISO 10070)
Flow Rate of Free-Flowing Metal Powders Using the Hall
ApparatusMPIF 03, ASTM B213, ISO 4490
Oil Absorption of Pigments by Spatula Rub-outASTM
D281
Particle Size Distribution of Metal Powders and Related
Compounds by Light ScatteringASTM B822
Particle Size Distribution of Refractory Metals and
Compounds by X-ray Monitoring of Gravity Sedimenta-
tionASTM B761
Particle Size Distribution of Tungsten Metal Powder by
TurbidimetryASTM B430
Sampling Finished Lots of Metal PowdersMPIF 01, ASTM
B215, ISO 3954
Tap Density of Metal PowdersMPIF 46, ASTM B527, ISO
3953
Testing of Sintered Products
Carburized Case Hardness and Case Depth of Sintered
PartsMPIF 37, ISO 4507
Density Determination of MIM Components (Gas Pycnome-
try)MPIF 63
Determination of Charpy Impact Energy of Unnotched Metal
Injection Molded Test SpecimensMPIF 59
Determination of Corrosion Resistance of MIM Grades of
Stainless Steel Immersed in 2% Sulfuric Acid Solution
MPIF 62
Green Strength of Compacted Metal Powder Specimens
MPIF 15, ASTM B312, ISO 3995
Impact Strength of Sintered Metal Powder SpecimensMPIF
40, ISO 5754
Material Standards for Metal Injection Molded PartsMPIF
35
Nickel-Iron Powder Metallurgy Soft Magnetic Alloys
ASTM A904
 Table of Contents
Nickel-Silver Sintered Metal Powder Structural Parts
ASTM B458
Sintered Austenitic Stainless Steel Structural PartsASTM
B525
Sintered Metal Materials and Hardmetals Determination of
Young's ModulusASTM E111, ISO 3312
Sintered Metal Materials, Excluding Hardmetals, Determina-
tion of Transverse Rupture StrengthISO 3928
Soft Magnetic Iron Fabricated by Powder Metallurgy
ASTM A811
Standard Specification for Metal Injection Molding (MIM)
Ferrous PartsASTM 883
Standard Test Method for Linear Thermal Expansion of Solid
Materials with a Push-Rod DilatometerASTM E228
Transverse Rupture Strength of Sintered Metal Powder Test
SpecimensMPIF 41, ASTM B528, ISO 3325
PATENT SOURCES
There are a large number of patents in MIM, dating from
about the mid-1970s. The easy means to sort through the U.S.
Patents is via the Web site www.uspto.gov by entering the
quick search for patents zone (PTO = patent and trademark
office). You can search by title, keyword, inventor, year, or
other features. Be careful in relying on MIM since this has
many other definitions besides metal powder injection
molding.
Once a patent is identified, a good copy is available as a pdf
file at www.pat2pdf.org as a free download. Happy hunting.
CONFERENCES AND PUBLICATIONS
Every two years there is a World Congress of Powder Met-
allurgy, rotated between Asia, Europe, and North America.
This is the typical forum for showing the latest developments
in MIM. Additionally, each regional trade organization (listed
below) has an annual conference and MIM is usually well rep-
resented at these as well. Quite often, to foster more attention
on MIM, a short course and tabletop exhibition, standards
committee meeting, technical presentations, and company ca-
pability briefing is organized, usually in a desirable location,
such as Orlando or Los Angeles. While the World Congress
will often draw 1,500 participants, the conferences are about
half that, and the MIM topical meetings attract typically 120
participants.
TRADE AND PROFESSIONAL ORGANIZATIONS
There is no trade association specifically for MIM. In the
various geographic regions, various powder metallurgy mar-
keting and trade associations operate and they embrace MIM
as part of their activities, including coverage at annual confer-
ences, articles in the magazines, and training courses. The trade
associations focused on powder metallurgy are as follows:
Chapter 9: Supplemental Information
European Powder Metallurgy Association, Shrewsbury, UK
www.epma.com
Japan Powder Metallurgy Association, Tokyo, Japan
www.jpma.gr.jp
Metal Powder Industries Federation, Princeton, NJ, USA
www.mpif.org
In addition, a new Asian Powder Metallurgy Association
is forming to combine the separate organizations in Japan, Tai-
wan, South Korea, and China. Until that is fully commis-
sioned, each country has its individual trade organization.
Likewise, many other countries have small efforts, for exam-
ple, in Turkey, South Africa, Russia, and Thailand.
RESOURCES
International Comparison of Manufacturing Productivity and
Unit Labor Cost Trends, 2001, Report USDL 01-543, Bureau
of Labor Statistics, United States Department of Labor, Wash-
ington, DC, 26 September 2002.
M. Ashby and D. Cebon, Estimating Process Cost: Cost
Modelling for Selection, Version 4, Cambridge Engineering
Selector, Granta Design, Cambridge, UK, 2002.
G. Boothroyd, P. Dewhurst, W. Knight, Product Design for
Manufacture and Assembly, second edition, 2002, Marcel
Dekker, New York, NY, pp. 359-379.
R.M. German, Engineering Economics of Powder Injection
Molding Component Production: Part 1, Tool Costing, P/M
Science and Technology Briefs, 2003, vol. 5, no. 2, pp. 511.
R.M. German, Engineering Economics of Powder Injection
Molding Component Production: Part II, Feedstock Costs,
P/M Science and Technology Briefs, 2003, vol. 5, no. 3, pp.
1116.
R.M. German, Engineering Economics of Powder Injection
Molding Component Production: Part III, Production Costs,
P/M Science and Technology Briefs, 2003, vol. 5, no. 4, pp.
1422.
R.M. German, Engineering Economics of Powder Injection
Molding Component Production: Part IV, Price Sensitivity,
P/M Science and Technology Briefs, 2004, vol. 6, no. 1, pp.
510.
R.M. German and D. Blaine, Production Cost Sensitivity
Analysis for Metal Powder Injection Molding, Advances in
Powder Metallurgy and Particulate Materials2004, Metal
Powder Industries Federation, Princeton, NJ, part 4, pp. 110.
R.M. German and A. Bose, Injection Molding of Metals and
Ceramics, 1997, Metal Powder Industries Federation, Prince-
ton, NJ.
185
 Table of Contents
Chapter 9: Supplemental Information

R.M. German and S.J. Park, Handbook of Mathematical

Relations in Particulate Materials Processing, 2008, Wiley,
Hoboken, NJ.
D. Gonia, MIM Parts for Automotive Applications
Production and Economic Aspects, Journal of the Japan So-
ciety of Powder and Powder Metallurgy, 1999, vol. 46, pp.
849852.
U. Haupt and H. Walcher, Powder Injection Moulding Con-
cepts for Presses and Tooling, Powder Injection Moulding,
Proceedings of the First European Symposium on Powder In-
jection Moulding, European Powder Metallurgy Association,
Shrewsbury, UK, 1997, pp. 7079.
D.P. Hoult and C.L. Meador, Manufacturing Cost Estimat-
ing, Materials Selection and Design, G. E. Dieter (ed.), 1997,
ASM International, Materials Park, OH, pp. 716722.
H. Rees, Mold Engineering, 1995, Hanser/Gardner Publishers,
Cincinnati, OH.
R.D. Stewart, Detailed Cost Estimating, Mechanical
Engineers Handbook, second edition, M. Kutz (ed.), 1998,
Wiley Interscience, New York, NY, pp. 2,1172,142.
W.J. Tobin, Is There a Best Practice for Costing Parts?,
Injection Molding, 2001, September, pp. 4548.
I.R. Van Gestel, A New Production Method for High Value
Parts, Powder Injection Molding Technologies, R.M. Ger-
man, H. Wiesner, and R.G. Cornwall (eds.), 1998, Innovative
Material Solutions, State College, PA, pp. 3342.

186
 Table of Contents

INDEX
A bronze statue, 23, 33
acceptance criteria, 67 brown, 9
accuracy, 60 burner cover, 109
adjustable vane, 109 burnout cycle, 19
adjustment ring, 109 business-machine applications, 111
aerospace applications, 106107, 146 butterfly-valve plate, 107
aesthetics, 125
air conditioner fittings, 122 C
airbag pivot, 108 cantilever, 46
alumina, 72, 76, 83, 85, 171 capital-equipment investment, 138
aluminum nitride, 87 captive molders, 136
aluminum, 73, 74, 80 carbon control, 171
annual sales history, 138139 carbo-nitrides, 24
apparent density, 12, 73, 171 carbonyl powder, 12, 171
applications, 9, 25, 105, 139, 142, 145 carburization, 27
arrow head, 130 casting comparison, 102
arthroscopic connector, 125 catalytic debinding, 19, 172
arthroscopic implant, 127 cavity design, 51, 57
articulator gear, 126 cavity transducer, 56, 64
artistic items, 131132 cellphone applications, 1, 111113
assembly, 42 cellphone cover, 112
atmosphere sintering, 20 cellphone knuckle, 101, 168
atmosphere-stabilized compositions, 90 cellulose, 13
audio jack mount, 112 cemented carbide, 60, 47, 72, 73, 79, 83, 88, 114, 172
automation, 146 centrifuge component, 37
automotive applications, 108111, 142 ceramics, 49, 64, 8287
automotive magnetic sensor, 56 cermet, 47, 89, 172
automotive ratchet, 108 Charpy testing, 8384
check ring, 17
B chemistry specification, 74
backbone polymer, 13, 18 chess set, 132
balanced runner, 53 chopstick rest, 132
barrel, 15, 171 chromium evaporation, 79
batch mixer, 13 clamping force, 15, 172
batch sintering, 21 closed features, 40
batch size effects, 159161 closed pore, 172
bi-material, 22 closed-loop feedback control, 74, 160
binder, 9, 13, 171 coarsening, 172
binder contraction, 57 cobaltchromium, 79, 80
biocompatibility, 75, 77, 90 coefficient of variation, 60, 172
biopsy tip, 23, 126 coining, 18, 27, 172
birdshot, 120 cold deformation, 27
blemishes, 42, 50 cold plate, 113
bonded materials, 91 color, 8586
brake fitting, 110 comparison to powder metallurgy, 3
brazing, 27 competitive technologies, 104, 139140
bronze filter, 39 complexity, 32, 42, 64, 99, 140, 152

187
 Table of Contents
Index

component cost analysis, 100, 161 design, 31, 32, 49

composite, 23, 83, 88, 9091, 172 design effects on costs, 102
compressed air nozzle, 120 design window, 170
compression, 172 diamond, 91, 114115
compressor disk, 39 die-cavity pressure, 57
computer applications, 113 dimensional control, 173
computer latch, 8 dimensional scatter, 4, 61, 60
computer modeling, 28, 64, 65 disk drive component, 7, 32, 113
conductivity, 76 drill tip, 42, 115
conferences, 185 ductility, 173. See elongation
conformal cooling, 55
conformal setter, 39, 64 E
constants, 181 eccentric weight, 130
consumer applications, 123124, 142 economic batch size, 140, 160
continuous mixer, 13 effective density, 34
continuous sintering, 21 ejection, 15, 54, 55, 173
contrast with other technologies, 8 ejector pins, 173
controlled-expansion alloy, 68 elastic modulus, 23, 79, 87, 173
controlled porosity, 22, 25, 91 electric-toothbrush gear, 123
conversion factors, 180182 electrical applications, 118
convertible-roof clip, 109 electrical properties, 76
cooling channel, 15, 55, 66 electronic applications, 118, 130, 143
cooling stage, 28 electronic package, 42, 43, 118, 128
cooling time, 55 electroplating, 27
copper, 73, 74, 76, 79, 80 elemental powder, 173
copper feedstock, 14 elevated-temperature properties, 85
copper powder, 12 elongation, 47, 8081, 89, 173
copperberyllium, 55 embossing, 41
coring, 44, 50, 55 emerging opportunities, 144
corner radius, 55, 62 end mill, 114
corrosion resistance, 29, 68, 75, 77, 90, endodontic tips, 117
corrosion testing, 93 engineering specifications, 33
cost calculation, 97, 150, 161, 167168 environmental properties, 75
cost reduction, 102 equipment costs, 99100
cost sensitivity, 167168 error budget, 64
counterbalance weights, 79 ethylene vinyl acetate, 13
critical loading, 172 evaluation mode, 137
custom molders, 136 evaluation overview, 105
cutting edge, 63 expansion factor, 52
cutting-tool applications, 113114 external threads, 40, 46
cycle time, 172
F
D F15 alloy. See Kovar
debinding, 9, 18, 19, 172 F75. See cobaltchromium
debinding costs, 163 fan gate, 54
decorative handles, 132 fatigue, 23, 58, 85, 89
defects, 65, 68 feature combinations, 39
defense applications, 114115 feedstock, 2, 9, 12, 61, 173
definitions, 170179 feedstock cost, 155159
density, 79, 88, 172 feedstock-flow path, 52
density gradients, 28 feedstock suppliers, 73
dental applications, 116118, 142 ferric chloride testing, 78
dental implant, 45 ferrous alloys, 47, 74, 76, 77, 80, 94. See also steel and

188
 Table of Contents
Index

stainless steel handcuffs, 122

fiber optic transceiver, 128 handgun body, 39
figurine, 36 handgun components, 119
filled polymers, 91 handtool applications, 120
filling, 66, 173 hanger, 112
filter, 22, 91 hard material, 174. See cemented carbide
financial performance, 140 hardness, 47, 81, 83, 88, 89, 174
finishing costs, 164165 hardware applications, 120122, 143, 146
finishing operations, 173 Hastelloy X, 79, 80
finite element analysis, 66 hearing aid receiver, 118
firearm applications, 6, 118120, 143 heart-surgery component, 126
firing initiator, 108 heat capacity, 87
first article, 66 heat dissipation, 26, 87
fishing weights, 129 heat sinks, 145
flashing, 53, 173 heat treatment, 27
flat surfaces, 39, 50 helical threads, 40
flip slider, 112 helium pycnometry, 68
flow body, 106 hermetic package, 118
foam, 91 heterogeneous microstructure, 23, 25, 89
fracture elongation, 8081 high-temperature properties, 76, 85
fracture strength, 49 hinge barrel, 112
fracture toughness, 83, 173 historical shifts, 6, 136
fragmentation tips, 116 holes, 40, 41, 50
frictional drag, 20 hot deformation, 27
front grip, 119 hot isostatic pressing, 48, 80, 81, 86, 174
fuel injectors, 110 hot-runner system, 53, 174
full-scale production mode, 137 household applications, 123124
future trends, 146 housing block, 119
hydraulic pressure, 57
G hydride formation, 90
gas-atomized powder, 12, 174 hydrochloric acid testing, 78
gas-assisted molding, 23 hydroxyapatite, 90
gate, 174
gate blemish, 54 I
gate size, 15, 54 identification number, 40
gear, 65 impact toughness, 8384
geographic breakdown, 141 implants, 90, 126127, 145
geometric details, 34, 38 imprints, 45
golf club, 101, 129, 155, 166, 170 impurity, 79, 93
granulation, 14, 174 Inconel, 174
graphite furnace, 20 Inconel 718, 47, 80, 85, 106
gravity, 65 industrial applications, 7, 40, 120122, 143
green-density distribution, 66 industry changes, 146
green joining, 22 industry statistics, 136
green machining, 22, 59 industry structure, 135136
green state, 9, 174 inertial properties, 76, 79
green strength, 174 infiltration, 90
growth barriers, 146 inhomogeneities, 13
gun sight, 33, 42 injection molding, 9, 12, 174
insignias, 25
H inspection 66, 68
hair trimmer, 44 instrumentation applications, 124
handbook properties, 49 internal cavity, 40

189
 Table of Contents
Index

internal threads, 40 median particle size, 13, 73

Invar, 76, 79, 81, 80, 90, 94, 174 median size, 175
iron aluminide, 85 medical applications, 125127, 143
iron, 73, 76, 77, 80 medical hand tool, 37
iron powder, 12 metal powder injection molding, 9
ironneodymiumboron, 91 metering, 175
ironnickel. See steel 2200 and steel 2700 microarray devices, 23, 41, 145
microelectronic package, 53, 127129, 142, 146
J microminiature structures, 23, 41, 61, 126, 145, 175
jetting, 174 micromolding, 175
jewelry applications, 124125 microstructure, 88, 175
joining, 27 microwave absorption, 26
military applications, 119, 143
K minimally invasive manipulator, 126
key terms, 9 misconceptions, 182
knife, 130 mixed powder, 47, 72
knife blade, 41 mixing, 13, 175
knock-out feature, 64 mixing costs, 158
knuckles, 112 mold design, 51
knurled surface, 40 mold-flow simulation, 175
Kovar, 74, 76, 77, 79, 94, 118, 174 mold life, 50
mold refurbishing, 15
L molding, 2, 15, 61
labor rates, 99 molding costs, 162163
large structure, 23, 39 molding cycle, 57
linkage, 33 molding machine, 15
liquid-phase sintering, 174 molding pressure, 18
lock components, 121 molybdenum 72, 73
logos, 26, 31 molybdenum powder, 12
low-pressure molding, 23, 174 mounting bracket, 56
low-alloy steel, 68 multiple cavities, 18, 53, 153154, 160
multiple gates, 54
M multiple-purpose tool, 120
machining, 27
machining comparison, 102 N
magnetic properties, 76, 77 near-net shape, 175
magnets, 91 net shape, 9, 102, 175
manufacturability, 31 new concepts, 145
manufacturing cell, 99 nickel, 73, 77
maraging steel, 8, 9, 84, 135147 nickel superalloy, 47, 72
market segments, 139 nickeliron. See steel 2200, steel 2700
market specialization, 143 niobium, 73, 80
mass range, 37, 40 nitrogen alloying, 24, 48, 90
mass variation, 57, 62 nomenclature, 179180
master alloy, 47, 72 nominal chemistry, 73
material selection, 47, 71, 75 nondestructive techniques, 69
material standards, 93 novel options, 22, 58
material use, 141 nozzle, 42, 52
maturity levels, 137138 number of cavities, 53
maximum dimension, 36
mean, 69 O
mechanical latch, 8 open pore, 175
mechanical properties, 29, 47, 48, 68, 76, 8085 operational characteristics, 141, 157, 162

190
 Table of Contents
Index

optical connector, 64 price, 97

optical housing, 34, 131 price reduction, 169
optical properties, 76, 85 price sensitivity, 169
optoelectronics applications, 127129 printer head, 111
orthodontic bracket, 37, 116117 printer yoke, 65
over-molding, 40 process attributes, 1, 2, 6, 11, 25
oxidation resistance, 75, 79 process capability, 60, 61, 69
oxide ceramics, 47 process simulation, 28
oxide-reduced powder, 12 process variation, 62
oxygen contamination, 79 production costs, 98, 167
production levels, 137
P production quantity, 32, 34, 140, 160
P20 steel, 58 professional organizations, 185
packing pressure, 176 projectile, 18, 43
packing, 28, 57 property scatter, 82
paper hole punch, 123 property specification, 94
paraffin wax, 13 protrusions, 40, 45
particle, 9 publications, 185
particle packing, 65 pump housing, 21
particle size, 176 pusher plate furnace, 21
parting line 15, 53, 59, 99, 176 putter, 129
patent sources, 185 pycnometry, 68
peanut oil, 13
performance, 32, 34 Q
phased-array radar, 145 qualitative design criteria, 31
pilot-facility costs, 100 quality plan, 66, 6768
pilot mode, 137 quantitative design criteria, 32
pivot, 112
plastication, 56 R
polishing, 27, 36 radiation shield, 127
polypropylene, 13 rapid prototyping, 15, 55, 176
porcelain, 83 reciprocating screw, 176
pore former, 25, 92 redesigns, 44
porosity, 22, 91, 176 reduced powder, 177
porous dental implant, 117 refractory metal furnace, 20
postage meter, 111 research and development, 138
post-sintering treatment, 48, 62 residual porosity, 4
powder characteristics, 72, 73 rheology, 13, 177
powder cost, 74, 75, 156 ribs, 40, 45
powder metallurgy comparison, 102 rifle sight, 42
powder production, 73 robot system, 15, 57
powder testing, 184 rocker arms, 110
powder use, 75, 143 rocket nozzle, 106
powderbinder separation, 18 rotational sensor, 108
prealloyed powder, 47, 72, 176 rubber mold, 59
precious metals, 124 runner, 15, 177
precipitation harden, 27 rupture modulus, 47, 8384
precision, 60
premixed feedstock, 14 S
presintering, 20 safe-and-arm rotor, 116
presssinter powder metallurgy, 3 safety switches, 121
pressure control, 176 sales growth rate, 137139
pressure-assisted sintering, 176 sales projections, 144

191
 Table of Contents
Index

salt-spray testing, 78 stainless 17-4 PH, 20, 27, 48, 74, 76, 77, 78, 79, 80, 82, 84,
sand wedge, 101, 155 85, 88, 94
scissors, 37 stainless 304L, 80
screw, 15, 177 stainless 316L, 61, 66, 74, 77, 79, 80, 82, 84, 85, 94
screw position, 18 stainless 318, 80
screwdriver holder, 7 stainless 410, 80, 84
sculpture, 32 stainless 414, 80
seal ring, 106 stainless 420, 58, 80, 84, 94
secondary operations, 27, 177 stainless 430L, 79, 80, 84, 94
segregation, 177 stainless 440C, 58, 80, 88
self-mixing, 14 stainless 630, 48. See stainless 17-4 PH
sensor applications, 124 stainless PANACEA, 80
serrations, 6, 45 stainless steel, 68, 73, 74, 7788, 178
service failure, 58 stainless steel microstructure, 4
setter, 63, 171 stainless steel powder, 12
shape complexity, 33 standard deviation, 69
sharp corners, 41 standard process, 29
shear-strain rate, 14 standardized chemistries, 73
shear thinning, 14 statistical process control, 68
short shot, 18 stearic acid, 13
shot size, 177 steel, 73, 74, 76, 77, 80, 94
shower safety valve, 121 steel 2200, 74, 77, 79, 80, 84, 85, 94
shrinkage, 177 steel 2700, 74, 79, 80, 84, 85, 94
shrinkage factor, 15 steel 4140, 74, 80, 84, 94
silica, 76 steel 4340, 74, 80, 85
silicon carbide, 72, 85, 90, 177 steel 4605, 74, 80, 84, 94
silicon nitride, 72, 83, 85, 177 steel 4620, 74
sink marks, 17, 28, 54, 177 steel 4640, 80, 84
sintered density, 52 steel 4650. See steel 4605
sintering, 9, 20, 177 steel properties, 4
sintering costs, 163164 steering clevis, 110
sintering shrinkage, 3, 14, 20, 52 stents, 145
sintering simulation, 28, 65 stepped mold closure, 54
sintering substrate, 20 stereolithography, 59
size range, 36 sticking, 54
slenderness, 36 stiffeners, 39
slides, 15, 55 strength, 5, 47, 8081, 8384, 87, 89
slots, 41 structural properties, 8085
small details, 41 summary characteristics, 38
small-scale production mode, 137 superalloy, 85, 106, 178
soft magnetic alloys, 68, 124 superalloy Hastelloy X, 79
soft tooling, 33, 58 superalloy Inconel 718, 47, 80, 85, 106
software packages, 65 superalloy Udimet, 81
solenoid valve body, 124 surface carburization, 27
solids loading, 13, 52, 177 surface finish, 26, 35, 42, 85, 99
solvent debinding, 19, 177 surface roughness, 62, 63
soup spoon, 35 surface treatments, 27
splines, 39 surgical stainless steel. See stainless 17-4 PH
sporting applications, 129130 swirler, 106
spray nozzles, 122
sprue bushing, 56 T
sprue, 178 tableware, 35
stabilizer fin, 115 tap density, 178

192
 Table of Contents
Index

taper, 15, 40 turbocharger, 16, 39, 65, 109

technology status, 8, 146 twin-screw mixer, 13
telecommunications applications, 130131 two-material molding, 22, 26, 40, 44
temperature-control passages, 55
tensile properties, 47, 48, 68, 8083, 87, 89 U
tensile strength, 178 Udimet 700, 81
tensile testing, 9293 ultimate tensile strength, 47, 8081, 178
testing procedures, 68, 92, 184185 undercut, 3, 40, 41
texture 42, 45 uniaxial die compaction, 3
thermal conductivity, 26, 8788 unique aspects, 6
thermal debinding, 18, 178 unit operations, 98, 161
thermal expansion, 26, 8788 units of measurement, 180182
thermal fatigue, 87, 142
thermal management, 87 V
thermal processing, 2, 12 vacuum debinding, 179
thermal properties, 26, 76, 8788 vacuum sintering, 20
thermal softening, 85 valve body, 27
threads, 46 vane-lever block, 109
tissue attachment structure, 24, 117118 vapor chambers, 145
titanium, 7, 24, 45, 72, 73, 74, 7980, 85, 89 vapor pressure, 49
titanium 6-4, 74, 79, 80 vendor differences, 29, 61
titanium 6-7, 80 Venn diagram, 1
titanium powder, 12 vent, 18, 54, 179
tolerances, 62 vibration weights, 79
tool cavity, 14, 15, 52 viscosity, 14, 179
tool-cost estimation, 154 void, 16, 54
tool costs, 98, 140, 151
tool design, 14, 51 W
tool life, 99 waffle pattern, 40, 45
tool materials, 15, 58 wall thickness, 36, 39, 40, 103
tool motions, 5 watchcase, 18, 34, 59, 86
tool options, 58 watches, 124125
tool-steel powder, 12 water-atomized powder, 12, 179
tool steel, 21, 58, 73, 81, 88 water debinding, 19
tooling, 17 water-assisted molding, 23
tooling tolerance, 64 wear failure, 58
toothbrush gear, 123 wear properties, 23, 68, 76, 88
toughness, 178 wear tips, 114
toy applications, 123124 Weibull modulus, 83
trade organizations, 185 weld line, 53, 54, 179
train wheels, 123 welding, 27
translucency, 86 wine stopper, 124
transverse rupture strength, 47, 8384, 178 wire-bonding tool, 114
trigger guard, 118 wire-saw bead, 114115
tripod mount, 123 wood, 91
tungsten, 72, 73 woodcutting tool, 114
tungsten alloys, 26, 47, 76, 79, 81, 85
tungsten heavy alloys, 47, 8384 Y
tungsten medallion, 14 yield, 64
tungsten powder, 12 yield strength, 47, 49, 8081, 179
tungsten projectile, 43
tungstencopper, 26, 76, 81 Z
turbine disk, 107 zirconia, 37, 83, 179

193
 Table of Contents
Index

194
cover_Layout 1 4/7/2011 2:05 PM Page 1

Metal
Injection
Molding

A Comprehensive MIM Design Guide

Metal Injection Molding
A Comprehensive MIM Design Guide
Randall M. German

GERMAN

ISBN No. 978-0-9819496-6-6

Metal
Injection
Molding
A Comprehensive MIM Design Guide
Randall M. German
 Table of Contents

How to Use This Electronic Version of the Book

Metal Injection Molding

A Comprehensive MIM Design Guide
All parts of this book are accessible via links from the Table of Contents; individual chapter
subsections are also accessible via links from the outline found on the first page of each chapter.
Simply click on any link (highlighted in blue) and youll instantly move to the destination you
selected. Every page in this electronic version offers a link to the table of contents, so you can
always find what youre looking for with one click.
 Table of Contents

Metal Injection Molding

Table of Contents
 Table of Contents

Metal Injection Molding

A Comprehensive MIM Design Guide

Randall M. German

Metal Powder Industries Federation

Princeton, New Jersey 08540
 Table of Contents

Copyright 2011 Metal Powder Industries Federation

All rights reserved. Published 2011

ISBN No. 978-0-9819496-6-6

Cover photo: Young-Sam Kwon

No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including
photocopying, recording, scanning, or by any information storage and retrieval system, without the written permission of the
Metal Powder Industries Federation, 105 College Road East, Princeton, N.J. 08540. The information in this book and the
identification of design decision processes, property mapping, and material selection techniques, as well as cost and management
metrics, is proprietary intellectual property of the author and is not for distribution, reproduction, or use beyond that of a single
purchase.
Table of Contents

To Savannah
Table of Contents

PREFACE...........................................................................ix
ACKNOWLEDGMENTS.................................................xi
AUTHOR BACKGROUND ...........................................xiii
Chapter 1. INTRODUCTION ...........................................1
OverviewWhat is MIM? ...................................................1
Process Essentials .................................................................2
Where MIM Differs from PM ..............................................3
Unique Aspects of MIM .......................................................6
Historical Evolution..............................................................6
Market and Technology Status..............................................8
Contrast with Other Technologies ........................................8
Key Terms.............................................................................9
Resources ..............................................................................9
Chapter 2. PROCESS OPTIONS ....................................11
Process Overview................................................................11
Background ....................................................................11
FeedstockPowder Plus Binder ...................................12
Tooling and Materials ....................................................14
Molding..........................................................................15
Debinding.......................................................................19
Sintering .........................................................................20
Novel Options .....................................................................22
Attributes and Advantages ..................................................25
Secondary Operations .........................................................27
Process Simulations ............................................................28
Vendor Differences .............................................................28
Resources ............................................................................29
Chapter 3. DESIGN PRINCIPLES.................................31
Design Cycles .....................................................................31
Recognizing Good Candidates............................................31
Qualitative Criteria.........................................................31
Quantitative Criteria.......................................................32
Geometric Details ...............................................................34
Effective Density Concept .............................................34
Surface Finish ................................................................35
Size Range .....................................................................36
Mass Range ....................................................................37
Geometric Summary ......................................................38
Features...............................................................................39
Flat Face.........................................................................39
Closed Features ..............................................................40
Small Details ..................................................................41
Holes, Slots, and Undercuts ...........................................41
Imprints, Protrusions, Ribs ............................................45
Threads...........................................................................46
Materials .............................................................................47
Properties ............................................................................47
TABLE OF CONTENTS
Design Support ...................................................................49
Resources ............................................................................50
Chapter 4. MANUFACTURING CONSIDERATIONS ...51
Mold Attributes and Limitations.........................................51
Cavity Design.................................................................51
Parting Line....................................................................53
Sprue, Runner, and Gate ................................................53
Vent ...............................................................................54
Temperature Control Passages .......................................55
Ejector ............................................................................55
Slides and Cores.............................................................55
Tool Motions.......................................................................56
Tool-Cavity Design.............................................................57
Novel Solutions...................................................................58
Blemishes from Tooling......................................................59
Geometric Feature Combinations .......................................60
Dimensional Tolerances......................................................60
Tolerance Approaches.........................................................62
Process Yield.......................................................................64
Computer Simulation for Defect Avoidance.......................65
Quality Plans.......................................................................66
Inspection............................................................................68
Quality Tests...................................................................68
Quantitative Techniques.................................................69
Nondestructive Techniques ............................................69
Resources ............................................................................69
Chapter 5. MATERIAL PROPERTIES .........................71
Available Materials .............................................................71
Powders..........................................................................73
Feedstock .......................................................................73
Standardized Chemistries...............................................73
Cost Attributes ....................................................................74
Materials Selection .............................................................75
Key Classifications ............................................................76
Electrical and Magnetic .................................................76
Corrosion and Biocompatibility.....................................77
Inertial Properties...........................................................79
Structural Properties.......................................................81
Tensile Properties of Metallic Materials ...................81
Rupture Strength of Ceramic Materials ....................82
Impact, Fracture, and Fatigue Properties ..................83
Elevated-Temperature Properties ..............................85
Optical Properties...........................................................85
Thermal Properties.........................................................87
Wear ...............................................................................88
Sintered Microstructures.....................................................88
Composite Materials ...........................................................89
High-Elastic-Modulus Cermets .....................................89
Heterogeneous Microstructures .....................................89
Atmosphere-Stabilized Compositions ...........................90
vii
Table of Contents
Biocompatible Materials ................................................90
Infiltrated Preforms ........................................................90
Bonded Materials ...........................................................91
Controlled-Porosity Structures.......................................91
Testing Standards ................................................................92
Material Standards ..............................................................93
Resources ............................................................................94
Chapter 6. COSTS ............................................................97
Typical Project Costs ..........................................................97
Tooling Cost........................................................................98
Tool Life .............................................................................99
Equipment Costs .................................................................99
Component Costing ..........................................................100
Competitive Forces ...........................................................102
Design Effects on Processing Costs..................................102
Resources ..........................................................................103
Chapter 7. APPLICATIONS..........................................105
Overview...........................................................................105
Aerospace..........................................................................106
Automotive .......................................................................108
Business Machines ............................................................111
Cellphones.........................................................................111
Computer Devices.............................................................113
Cutting Tools.....................................................................113
Defense .............................................................................114
Dental................................................................................116
Electrical and Electronic Components..............................118
Firearms ............................................................................118
Hand Tools ........................................................................120
Hardware and Industrial....................................................120
Household and Consumer.................................................123
Instrumentation and Sensors.............................................124
Jewelry and Watches.........................................................124
Medical .............................................................................125
Microelectronics and Optoelectronics ..............................127
Sporting.............................................................................129
Telecommunication...........................................................130
Summary...........................................................................131
Resources ..........................................................................132
Chapter 8. MARKET CONSIDERATIONS ................135
Industry Structure .............................................................135
Overview......................................................................135
Historical Shifts ...........................................................136
Production Levels ........................................................137
Sales Breakdown..........................................................137
Competitive Technologies ................................................139
Financial Performance ......................................................140
Geography....................................................................141
Materials ......................................................................141
Applications .................................................................142
Emerging Opportunities....................................................144
Applications .................................................................144
viii
New Concepts in MIM.................................................145
Industry Changes .........................................................146
Future Trends ....................................................................146
Resources ..........................................................................147
Chapter 9. SUPPLEMENTAL DATA ...........................149
Overview...........................................................................150
Cost-Calculation Approach...............................................150
Tool Costs .........................................................................151
Calculation Approach...................................................151
Optimization of Multiple-Cavity Tooling ....................153
Other Tool-Cost Factors...............................................154
Example Tool-Cost Estimation ....................................154
Summary of Important Tool-Cost Factors ...................155
Feedstock Costs ................................................................155
Cost of Raw Materials .................................................156
Operation Parameters...................................................157
Mixing Costs ................................................................158
Feedstock for Golf Clubs .............................................158
Feedstock Cost Summary ............................................159
Batch-Size Effects.............................................................159
Component Costing ..........................................................161
Operational Characteristics ..........................................162
Molding Costs ..............................................................162
Debinding Costs...........................................................163
Sintering Costs .............................................................163
Finishing Costs.............................................................164
Tight-Tolerance Effects................................................164
Per-Piece Costing .........................................................165
Additional Costs...........................................................165
Golf Club Example ......................................................166
How Much Will it Cost? ...................................................166
Cost Sensitivity .................................................................167
Typical Component Analysis .......................................168
Component-Mass Effect ..............................................169
Golf Club Example ......................................................170
Bottom Line on Costing...............................................170
Definitions ........................................................................170
Nomenclature....................................................................179
Units and Conversion Factors...........................................180
Helpful Relations .........................................................180
Prefixes ........................................................................181
Fundamental Units .......................................................181
Derived Units ...............................................................181
Important Constants .....................................................181
Conversion Factors ......................................................181
Misconceptions .................................................................182
Standards...........................................................................184
Patent Sources...................................................................185
Conferences and Publications...........................................185
Trade and Professional Organizations ..............................185
Resources ..........................................................................185
INDEX..............................................................................187
 Table of Contents

PREFACE
The combination of powder metallurgy and plastic molding results in an exciting fabri-
cation technology, termed metal powder injection molding (MIM). It changes the design
rules, allowing more complex components with more features in a high-production-volume
technology.
This design guide is constructed to help the reader take advantage of MIM. The metal
powder injection molding process has a related ceramic cousin CIM. Together the variants
of MIM, CIM, and similar technologies for carbides and composites, are known as PIM or
powder injection molding. These forming routes allow designers to combine the cost-
effective shaping attributes of plastic injection molding with the performance and engineer-
ing properties associated with durable and known engineering materials. About 80% of
powder injection molding is in the form of metals, so generally MIM and PIM are used
nearly interchangeably.
This guide organizes the facts on the process options, materials, properties, design
features, manufacturing tolerances, applications, markets, and costs to enable the design
community to best utilize this net-shape production route.
Consideration of MIM during the design process leverages its unique attributes to ease
production, lower costs, and increase process yield. This is because small design changes
can greatly enhance process yield and avoid costly secondary operations.
Today, MIM involves a large cadre of highly skilled and competitive manufacturing
firms. Although MIM is not new, exceeding production rates over a billion of parts per year,
guides on how to best use the technology are not well distributed. This design guide helps
rectify the situation, by identifying good design candidates to guide the user to best practices
in materials, properties, features, sizes, shapes, tolerances, and costs.

Randall M. German

ix
Table of Contents
 Table of Contents
ACKNOWLEDGMENTS
A few individuals helped tremendously with the background
to this book. Robert Cornwall provided insight on how to
collect market statistics and assess business trends. Animesh
Bose helped organize application and process details
fundamental to this book. Kay Leong Kim, Lye King Tan, and
Chee Tian Yeo provided insight to the production environ-
ment. Karl Hens made sure I was fully grounded in molding
technologies. Seong Jin Park likewise guided the computer
simulation gains.
Several former students, sponsors, and industry friends
guided my efforts by providing pictures, samples, data, and
comments. Besides the individuals acknowledged above,
notice is given to the special help provided by the following
individuals:
James Adams (Metal Powder Industries Federation,
Princeton, New Jersey, USA),
Sundar Atre (Oregon State University, Corvallis,
Oregon, USA),
Eric Baril (Industrial Materials Institute, National
Research Council, Boucherville, Quebec, Canada),
Paul Busch (Injectamax, Escondido, California, USA),
Qiaoling Chen (Hunan Injection High Technology,
Changsha, Hunan, China),
Zhahan Chen (Beijing Jarain Powder Injection
Technology, Beijing, China),
Santosh Das (Polymer Technologies, Clifton,
New Jersey, USA),
Kim Gon (PIM Korea, Gyoungasn-city, Gyoungbuk,
Korea),
Uwe Haupt (Arburg, Lossburg, Germany),
Kuen Shyang Hwang (National Taiwan University,
Taipei, Taiwan),
John Johnson (Engineered Products, Huntsville, Ala-
bama, USA),
Stefan Joens (Elnik Systems, Cedar Grove, New Jersey,
USA),
Yoshiyuki Kato (Epson-Atmix, Aomori-ken, Japan),
Chhattar Kucheria (Ortho Organizers, Carlsbad,
California, USA),
Young-Sam Kwon (CetaTech, Sacheon, Gyeongam,
Korea),
Jimmy Lu (Honeywell, Tempe, Arizona, USA),
Seong Jin Park (Pohang University of Science and
Technology, Pohang, Korea),
Tracy Potter (Advanced Powder Products, Philipsburg,
Pennsylvania, USA),
Julian Thomas (Springfield Munitions, Kersey,
Pennsylvania, USA),
Nick Williams (Inovar Communications, Shrewsbury,
UK),
Zulkefli Bin Zainal (Advanced Materials Technologies,
Singapore).
My son Eric German (Mitchell Silberberg & Knupp, Santa
Monica, California, USA) provided legal guidance.
I am especially thankful to Julian Thomas and to Rick
Sharbaugh (Spring Mills Studio, Spring Mills, Pennsylvania,
USA) for their help in preparing many of the images.
Besides the individuals mentioned above, thanks go to the
following individuals for their help in providing input data
relevant to the text: Antonyraj Arockiasamy (Mississippi State
University), Abdulhakim Belhadjhamida (Deloro Stellite),
Anthony Fanelli (Honeywell), Anthony Griffo (Smith Inter-
national), Anish Upadhyaya (Indian Institute of Technology
Kanpur), Barbara Lograsso (Michigan Technological Univer-
sity), Barry Rabin (Sage Technology and Development),
Bo Li (Yuelong Supefine Metal Materials), Jim Neill
(CM Furnaces), Debby Blaine (Stellenbosch University),
Thomas Gladden (SA Fabriques d'Ebauches), Haorong Zhang
(NetShape Technologies), Johnny Bruhn (SECO Tool),
Jonathan Wroe (European Powder Metallurgy Association),
Jonathon Fischer (MoldMaster), Joseph Rose (Penn State
University), Louis Campbell (Eaton), Mark Janney (Porvair
Advanced Materials), Maryann Wright (Remington Arms),
Matthew Bulger (NetShape Technologies), Mu-Jen Yang (Air
Force Research Laboratory), Neal Myers (Kennametal),
Pavan Suri (Heraeus), Rajiv Tandon (Magnesium Elektron),
Rudolf Zauner (Verbund-Austrian Renewable Power), Shin
Tian Lin (National Taiwan University of Science and Tech-
nology), Stanley Zalkind (TekForm Management), and
Yunxin Wu (Tsinghua University).
Draft chapters were reviewed by several volunteers who
suggested means to improve the presentation. My special
thanks go to Aditya Bothate, Wei Li, Kyle Katzmiller, and
Ridvan Yamanoglu for taking the time to read the manuscript,
making helpful suggestions, and supporting the timely com-
pletion of the project.
xi
Table of Contents
 Table of Contents

AUTHOR BACKGROUND
Rand German is Professor and Associate Dean of Engineering
at San Diego State University. Previously he was a chaired pro-
fessor at three universities where he conducted academic re-
search on automotive, powder metallurgy, and materials
engineering problems; Rand is an emeritus professor at the
Pennsylvania State University. He has been involved in the start-
up of a dozen firms, some of which have successfully commer-
cialized core MIM technologies. He obtained his doctoral
degree in Materials Science at the University of California
Davis, and obtained his MS and BS degrees from The Ohio
State University and San Jose State University. In addition, he
has taken intensive management training at the Hartford Grad-
uate Center and Harvard University.
He is a Fellow of ASM International and Fellow of APMI International, and has awards
that include the Nanyang Professorship, Tesla Medal, Japan Institute for Materials Research
Lectureship, Penn State Engineering Society Premiere Research Award, University of Cal-
ifornia at Davis Distinguished Engineering Alumni Award, San Jose State University Award
of Distinction, Ohio State Distinguished Alumni Award, and Technical Development Award
from the Japan Institute of Metals.
Rand has more than 930 published articles, 15 books, 19 edited books, and 24 patents.
He chairs several annual short courses, workshops, and symposia; he started the annual
Powder Injection Molding Symposium in 1990. He is technical advisor or member of the
board of directors for several organizations ranging from start-up companies and venture
capital boards to research institutes and private companies. He is involved in a wide range
of research projects that include studies on spark sintering, hard materials consolidation, ti-
tanium by MIM, microminiature components, developments in superfine powders, and
tungsten processing. He and his wife Carol live in Del Mar, California, and they have two
sons and three grandchildren who all live in California.

xiii
Table of Contents