Naval Material Science and Engineering USNA PDF

EN380
Naval Material Science and Engineering

Course Notes
With Contributions by:
Thomas Dawson, Victoria Johnson, Paul Miller, Michael Morabito, Michael

Schultz (USNA) and G.W. Swain (Florida Institute of Technology)
United States Naval Academy

Chapter 1: Materials in the Marine 5.4 Pitting Corrosion
Environment 5.5 Other Localized Corrosion
1.1 Characterization of Seawater 5.6 Selective Leaching aka Dealloying
1.2 Organic Matter and Biology 5.7 Erosion-Corrosion
1.3 The Physical Environment 5.8 Stress-Corrosion-Cracking (S.C.C.)
1.4 Classification of Marine Environments 5.9 Intergranular Corrosion
1.5 Corrosion factors for Carbon Steel in Marine 5.10 Fretting
Environments 5.11 Biological corrosion
1.6 Historical Perspective 5.12 Stray current corrosion
Chapter 2: Corrosion and Electromotive Chapter 6: Corrosion Protection and

Potential Monitoring
2.1 Corrosion cell 6.1 Cathodic Protection: Sacrificial Anodes
2.2 Specific Examples of Electrochemical 6.2 Cathodic Protection: Impressed Current
Reactions 6.3 Measures of Corrosion Rate (Weight Loss and
2.3 Measurement of Potential Strength Tests)
2.4 Standard Electromotive Force (Surface 6.4 Corrosion Allowance Calculations
Potential)
2.5 Correction for Solution Concentration: Nernst Chapter 7: Microstructure of Materials
Equation 7.1 Fundamentals
2.6 Reference electrodes 7.2 Bonding
2.7 Galvanic series 7.3 Crystalline Structure
2.8 Passivating films 7.4 Density Calculation
2.9 Concentration Cells 7.5 Direction Indices and Miller Indices
2.10 Effect of Alloying 7.6 Polymorphism
2.11 Pourbaix, or pH Diagrams 7.7 Coordination Number and Packing Factor
7.8 Solidification
Chapter 3: Corrosion Thermodynamics
3.1 Gibb’s Free Energy Chapter 8: Metals and Alloys
3.2 Nernst Equation 8.1 Types of Alloying
3.3 Electrochemical kinetics of corrosion 8.2 Hume-Rothery Rules for Alloying
3.4 Corrosion rates 8.3 Material Systems
8.4 One-component material systems
Chapter 4: Polarization 8.5 Two component material systems
4.1 Activation and Concentration Polarization 8.6 Derivation of Lever Rule
4.2 Scaling of Corrosion Measurements 8.7 Metal Processing
4.3 Determination of Corrosion Current for Single 8.8 Work Hardening and Annealing
Metals 8.9 Precipitation Hardening of Alloys
4.4 Mixed Potential Theory 8.10 Iron-Carbon Systems
4.5 Alternate construction of Polarization 8.11 Eutectoid Point for Iron-Carbon Systems
Diagrams 8.12 Continuous-Cooling Transformation Diagram
for Iron-Carbon Systems
Chapter 5: Corrosion Types 8.13 Heat Treatment of Steel
5.1 Uniform Corrosion 8.14 Specifics of Quenching Process for Iron-
5.2 Galvanic Corrosion Carbon Systems
5.3 Crevice Corrosion 8.15 Case Hardening of Steel
i
12.4 Effect of Mean Stress using Goodman-
Chapter 9: Stress Soderberg Relation
9.1 General Stress State and Principal Stresses 12.5 Stress concentrations
9.2 Determination of Principal Stresses (Two 12.6 Miner's Rule for Fatigue Under Varying
Dimensional) and Mohr’s Circle Stress Levels.
9.3 Determination of Principal Stresses (Three 12.7 Fatigue Under Combined Stress
dimensional) 12.8 Specialized Formulae for Two Dimensions
9.4 Maximum Shear Stress 12.9 Equivalent Stresses in Terms of Maximum
Shear
Chapter 10: Deformation 12.10 Connection Between Fatigue and Failure.
10.1 Stress-Strain Diagrams and Material Behavior
10.2 Material Characteristics Chapter 13: Wood
10.3 Elastic-Plastic Response of Metals 13.1 Directionality (isotropy)
10.4 True stress and strain measures 13.2 Mechanical Properties
10.5 Yielding of a Ductile Metal under a General 13.3 Typical Uses in the Marine Environment
Stress State - Mises Yield Condition. 13.4 Structural Analysis Equations
10.6 Maximum shear stress condition 13.5 Fastening
10.7 Creep 13.6 Control of Degradation
13.7 Plywood
Chapter 11: Fracture
11.1 Brittle vs. Ductile Fracture Chapter 14: Concrete
11.2 Temperature Effects and Fracture Analysis 14.1 What is Concrete?
Diagram 14.2 Variations of Concrete
11.3 Griffith Theory of Brittle Fracture. 14.3 Mechanical Properties of Concrete:
11.4 Fracture mechanics of high-strength materials 14.4 Design Options
11.5 Connection with Fracture Mechanics
11.6 Stress Corrosion Cracking (SCC) Chapter 15: Composites
11.7 Non-Destructive Testing 15.1 Fiber Reinforced Polymers (FRP)
15.2 What are Polymers?
Chapter 12: Fatigue 15.3 FRP Fiber Orientation
12.1 Fatigue Failure of Materials 15.4 FRP Fabrication
12.2 Fatigue Testing Procedures 15.5 Vacuum Bagging
12.3 S-N Fatigue Curves 15.6 Mechanical Properties of FRP
ii
EN380 Naval Materials Science and Engineering Course Notes, U.S. Naval Academy
CHAPTER 1:
MATERIALS THE MARINE ENVIRONMENT

(This chapter is adapted from Swain (1996) and Schultz (1997))
1.1 Characterization of Seawater

1.2 Organic Matter and Biology
1.3 The Physical Environment
1.4 Classification of Marine Environments
1.5 Corrosion factors for Carbon Steel in Marine Environments
1.6 Historical Perspective
Materials used in marine environments are subject to chemical, physical, and biological
deterioration. These factors make materials selection, design, and protection critical to the
effective and safe functioning of a structure, vessel, or component for its design life. An
understanding of the marine environment and its impact on materials is essential to anyone
working in these areas.
1.1 Characterization of Seawater
Seawater Composition
Seawater is a complex solution of inorganic, organic, and biological components. These can
interact with materials to cause corrosion and to degrade their properties.
Inorganic Components
The typical composition of sea water is shown below, but it must be remembered that in tropical
waters higher salinities may be experienced. And, in coastal waters, the complete spectrum from
fresh to sea water is found. One must also account for temperature differences, redox potential,
and physical activity.
CHEMICAL COMPOSITION OF SEAWATER, 19ppt Chlorinity

Salinity (ppt) = 0.03 + 1.805 Cl-
Anions g/kg of water Cation g/kg of water
Chloride 19.35 Sodium 10.76
Sulfate 2.70 Magnesium 1.29
Bicarbonate 0.14 Calcium 0.41
Bromide 0.067 Potassium 0.39
Borate 0.0044 Strontium 0.0079
Fluoride 0.0014
Swain, G.W. (1996) “OCE-4518 Protection of Marine Materials Class Notes”, Florida Institute of Technology.
Schultz, M.P. (1997) “OCE-4518 Protection of Marine Materials Class Notes”, Florida Institute of Technology.
1-1
Specific Conductance of Seawater

The specific conductance of seawater relates to its ability to conduct electricity. This has
implications to corrosion rates and cathodic protection. Specific conductance is a function of
temperature and chlorinity. Resistance is the reciprocal of conductance.
Table of the Specific Conductance of Seawater (-1cm-1)

Chlorinity, ppt Temperature, oC
0 5 10 15 20 30
1 0.001839 0.002134 0.002439 0.002763 0.003091 0.003431
2 0.003556 0.004125 0.004714 0.005338 0.005971 0.006628
3 0.005187 0.006016 0.006872 0.007778 0.008702 0.009658
4 0.006758 0.007845 0.008958 0.010133 0.011337 0.012583
5 0.008327 0.009653 0.011019 0.012459 0.013939 0.015471
6 0.009878 0.011444 0.013063 0.014758 0.016512 0.018324
7 0.011404 0.013203 0.015069 0.017015 0.019035 0.021121
8 0.012905 0.014934 0.017042 0.019235 0.021514 0.023868
9 0.014388 0.016641 0.018986 0.021423 0.023957 0.026573
10 0.015852 0.018329 0.020906 0.023584 0.026367 0.029242
11 0.017304 0.020000 0.022804 0.025722 0.028749 0.031879
12 0.018741 0.021655 0.024684 0.027841 0.031109 0.034489
13 0.020167 0.023297 0.026548 0.029940 0.033447 0.037075
14 0.021585 0.024929 0.028397 0.032024 0.035765 0.039638
15 0.022993 0.026548 0.030231 0.034090 0.038065 0.042180
16 0.024393 0.028156 0.032050 0.036138 0.040345 0.044701
17 0.025783 0.029753 0.033855 0.038168 0.042606 0.047201
18 0.027162 0.031336 0.035644 0.040176 0.044844 0.049677
19 0.028530 0.032903 0.037415 0.042158 0.047058 0.052127
20 0.029885 0.034454 0.039167 0.044114 0.049248 0.054551
21 0.031227 0.035989 0.040900 0.046044 0.051414 0.056949
22 0.032556 0.037508 0.042614 0.047948 0.053556 0.059321
Dissolved Gases
All gases present in the atmosphere are also found in seawater. They are, however, at a lower
partial pressure due to their limited solubility in seawater. In general the only gases normally
considered are nitrogen (because of its overwhelming percentage) and oxygen and carbon dioxide
because of their importance in corrosion. Other gases, such as ammonia, hydrogen sulfide, and
hydrogen also have important implications to materials performance.
1-2
Solubility of Oxygen in Seawater
1-3
Seawater pH
The pH of surface waters typically is in the range 7.8 to 8.4. This value is maintained by the effect
of the carbonate equilibria and the presence of cations; Ca, Mg, Na, & K. The addition of CO 2, a
decrease in temperature or an increase in pressure will cause the pH to fall. The removal of CO 2,
an increase in temperature, or a decrease in pressure will cause the pH to rise. Seawater pH effects
the corrosion rates of metals, and alters the calcareous deposits formed at metals cathodes.
Seawater Temperature
The surface water temperatures varies from about -1.8 C at the poles to 30 C at the equator. The
water in the deep ocean has a more constant temperature of about 4 C except where thermal vents
are active in the seabed. Higher temperatures will be found at heat exchangers and desalination
plants
An increase in temperature increases seawater conductivity and decreases oxygen concentration.

This effects corrosion rates. Increases in temperature also reduces the mechanical properties of
thermoplastics.
A decrease in temperature may cause materials to become brittle and many failures of steel
structures have been attributed to brittle fracture a low temperature.
1-4
Typical Profiles of Oxygen, Temperature, and Salinity in the Ocean.
1-5
1.2 Organic Matter and Biology
Organic matter and marine organisms are generally more abundant in coastal areas. Their impact
on materials may be due to biodegradation, microbial induced corrosion (MIC), wood boring, or
biofouling. Their impact to the design life and operational efficiencies of structures is often
underestimated, and materials selection or biological control may be necessary.
Penetration of Water by Visible Light.

This limits the Depth to which Plant Life can Grow
1-6
The Size and Shape of Typical Fouling Organisms
1-7
1.3 The Physical Environment
Materials used in ocean engineering have to withstand the physical conditions imposed on them
by wind, waves and currents. These cause loading, stress, fatigue, scour, abrasion, and impact
damage.
Beaufort Wind Scale and Sea State
1-8
Wind, Wave and Current Effects on Offshore Structures
1.4 Classification of Marine Environments
For engineering and corrosion control purposes the marine environment may be divided into the
following regions; atmospheric zone, splash zone, tidal zone, submerged zone, and mud or
sediment line. These environments have been described with respect to their effect on steel.
1-9
1 - 10
1 - 11
1.5 Corrosion factors for Carbon Steel in Marine Environments
Atmospheric Corrosion
Atmospheres may broadly be divided into rural, industrial and marine. The most corrosive are the
marine environments and if augmented by industrial pollution (acid rain) then severe corrosion
will occur. A good local example is the problems associated with the HCl produced by the solid
rocket boosters from the shuttle launch at Cape Canaveral. This has played havoc with corrosion
of the steel rocket gantries which traditionally used zinc as a coating to protect the steel from the
marine environment. The zinc is now dissolved by the acid fall-out.
Typical atmospheric corrosion rates for various locations
Splash Zone Corrosion

This is considered to be the most corrosive of all marine environments. Field data provides the
following guidelines for different environments.
Condition Years Exposure Rate, mpy

Quiet 5 20
Moderate 5 37
Rough 4 55
1 - 12
Immersed Corrosion
The factors affecting the corrosion of steel in the immersed condition are shown below
1 - 13
1.6 Historical Perspective
The marine environment has provided a challenge to engineers for thousands of years. The earliest
surving boat is the Khufu Barge, discovered in a pit next to the great pyramid of Egypt, dated
around 2500 BC. This boat was over 143 feet long, demonstrating the skill of Naval Architects
over 4500 years ago. It was constructed out of Cedar, a rot-resistant wood that grew in Lebanon.
The hull timbers were lashed together with rope. Part of this was because of the extreme expense
of metal fasteners.
Khufu barge, Egypt, 2500 BC, Dimensions: 43.6 m x 5.9 m [143 ft. x 19.5 ft.] Material: wood
Bronze and Copper

The bronze age dates from around 2000-500 BC. Copper and Bronze found their way into marine
design in terms of small nails (large nails were made of wood “tree nails” or “trunnels”), and in
specialized castings. For instance, Greek Triremes used bronze rams to sink other ships.
1 - 14
Iron
Historically, Iron was in short supply. A knife found in the tomb of King Tut was determined by
scientists to be of meteorite origin. The ability to process iron ore into tools and weapons was a
major challenge. The “Iron age” began somewhere around 1000 – 500 BC depending on the
locality. Iron weapons were of strategic importance, and there was significant geopolitical change
during this period.
While iron was used in fasteners and reinforcements for ships for a very long time, wood was the
most important building material for ships from pre-history until the 1800’s when iron became
available in large plates, and ships such as the Great Britain (1838 designed by I.K. Brunel) were
constructed. This ship was 322' long, 50' 6" wide and displaced 3,680 LT. She had an iron keel
1" thick and 21" wide and hull plate 3/8 to 3/4" thick and riveted to frames of angle iron. The iron
enabled larger ships to be built and supported the increasingly powerful steam ships.
Great Britian
The British Navy was slow to convert to iron for two reasons. Firstly a cannon ball fired at close
range could easily penetrate 3/4" iron plate and yet 8" of oak would effectively stop it. Secondly,
copper had successfully been used for many years as a means of preventing fouling. The
corresponding loss in performance and the galvanic action between iron and steel, however, caused
the steel to corrode and the copper to lose its antifouling qualities. This situation remained until
satisfactory copper based antifouling coatings were produced.
Steel
Iron was eventually replaced by steel. Prior to the early 1940's most steels used in construction
were low carbon steels (y = 32 ksi). These performed well but there were occasions when the
steel failed by brittle fracture. The most notable of these occurred in welded cargo ships and
tankers built in the USA during the World War II. Of about 5,000 ships built, over 1,000 had
experienced structural failures within three years of construction. About 20 ships actually broke
in two. This lead to the American Bureau of Shipping or ABS (1948) specifying notch toughness
1 - 15
requirements for hull steel and specifying grades and steel making practices. The improvements
to steel were furthered by submarine development which led to the use of high tensile strength
(HTS) steels with yield strengths of 50 ksi, and now HY80, HY100 and HY120 steels are
commonly used for offshore applications.
USS Virginia (SSN 774)
Other Alloys
Other alloys have been used for ship hull construction. Aluminum, discovered by Hans Oersted,
University of Copenhagen, 1825, is commonly applied to marine structures. The first boat built
out of aluminum was in 1890, Zepher a 17' launch. The early alloys used copper as an alloying
agent and as such were unsuitable for marine service. However, the Washington Disarmament
Conference, 1922, limited the total navy displacements and this revived interest in aluminum as a
construction material. By 1940, aluminum was used in about 100 US warships as weight saving
material, mainly for topside applications. For example the USS Independence (aircraft carrier)
contains 2.25 million lbs of aluminum which saved 2 million lbs in weight. In 1951, the USS
United States used 2,000 tons of aluminum, saving 8,000 tons in weight. Today aluminum is being
used in many applications where weight savings are desired, such as the entire hull structure of the
Littoral Combat Ships.
USS Independence (LCS 2)
Accompanying the development of the steam ship came requirements for hardware to support the
engines, propulsion systems, piping, rigging, etc. The hardware had to be manufactured from
1 - 16
materials that could withstand both the mechanical and corrosion abuse imparted by operating in
the marine environment. Many of the parts were manufactured form cast and wrought irons, steels
and aluminum; but others were made from copper based alloys and later stainless steels.
Copper based alloys were some of the first metals to be fabricated by man. About 400 BC Homer
described the Trojan War where 1200 ships, 80' long with 50 oars were equipped with bronze
beaks for ramming. The next copper alloys to find regular use were the brasses. Now there are a
whole family of copper alloys that find regular application including the use of 90:10 copper-
nickel hull plate for small boat construction.
Stainless steels were first developed around 1910 in England and Germany. By the 1920s,
commercial production in the US was begun by companies such as Allegheny, Armco, Carpenter,
Crucible, Firth-Sterling, Jessop, Ludlum, Republic, Rustless and US Steel. Production was,
however, limited partially due to the problems of obtaining raw materials when civil unrest was
active in Africa and Asia, which are traditional sources of nickel. The modern family of stainless
steels show increasing use in marine engineering. They maybe divided into: martensitic, ferritic,
austenitic, duplex, and precipitation-hardening stainless steels.
Continued Use of Wood

With the advent of the screw propeller problems arose with maintaining through hull stern bearings
and glands. This was solved by the use of Lignum vitae (guaniacum tree) a tropical hardwood
from the Caribbean which had excellent self lubrication properties and was also extremely hard.
The wood is still used in the sterntube bearings of many ships currently operating, hence the title
of a recent article, “Lignum Vitae: Wood So Bad-Ass, It's Used to Make Shaft Bearings for
Nuclear Submarines (and More)”
During World War II, plywood was used extensively in high-speed, light weight patrol craft.
Wood was also used in minesweepers because it is non-magnetic. Plywood still continues to be
used today in small boat construction.
Offshore Structures
At the same time as shipping activities were increasing, so were the services required to support
them on land and to ensure safe navigation and passage. Perhaps the earliest example of offshore
structures were the lighthouses. Perhaps the most famous of these structures was the Eddystone
light, built on some rocks about 15 miles south of Plymouth, England. The first structure was built
by Winstanley in 1698 from wood and anchored to the bottom by iron rods. It was replaced in
1706 by a stone structure clad with wood and designed by John Lovett and John Rudyard. This
burnt down in 1759, and John Smeaton designed a stone structure to take its place. In 1852 Trinity
House decided that a taller lighthouse was required and moved Smeaton’s structure to Plymouth
and replaced it with the building that remains there to this day. In any event, the use of stone,
wood, concrete, iron, and steel are commonplace for coastal and offshore facilities.
1 - 17
U.S. Navy R2 Tower Offshore Oil Platform
U.S. Naval Base in Okinawa
New materials and material requirements are continually emerging at what sometimes seems an
accelerating rate. The 1960’s, 70’s, and 80’s saw the development of the offshore oil industry,
which has led the way to improved structural materials and specifications and to the use of metals
such as titanium and high nickel alloys for critical applications. The world’s navies have continued
to use and develop advanced materials for weapons, sonar, nuclear submarines, etc. Perhaps the
greatest impetus for materials development and experimentation has come from racing sail and
power boats. In these sports, the application of high strength to weight ratio composites provides
the cutting edge to evaluate how these materials can withstand the marine environment.
Reference
Swain, G.W. (1996) “OCE 4518 - Protection of Marine Materials Class Notes”, Florida Institute
of Technology, Melbourne, FL.
1 - 18
CHAPTER 2
CORROSION AND ELECTROMOTIVE POTENTIAL
(from Swain (1996), Schultz (1997) and Dawson (2003))
2.1 Corrosion cell

2.2 Specific Examples of Electrochemical Reactions
2.3 Measurement of Potential
2.4 Standard Electromotive Force (Surface Potential)
2.5 Correction for Solution Concentration: Nernst Equation
2.6 Reference electrodes
2.7 Galvanic series
2.8 Passivating films
2.9 Concentration Cells
2.10 Effect of Alloying
2.11 Pourbaix, or pH Diagrams
What is corrosion?
 Webster’s Dictionary - corrode (v.) To eat away or be eaten away gradually, especially by
chemical action.
 NACE Corrosion Basics - corrosion may be defined as the deterioration of a material (usually a
metal) because of a reaction with the environment.
Why do metals corrode?

Most metals are found in nature as ores. The manufacturing process of converting these ores into
metals involves the input of energy. During the corrosion reaction the energy added in
manufacturing is released, and the metal is returned to its oxide state.
Metal Ore reduction
   Metal oxidation
(add electrons)
   Corrosion Products
(strip electrons)
In the marine environment, the corrosion process generally takes place in aqueous solutions and is
therefore electrochemical in nature.
Corrosion consequences
Economic - corrosion results in the loss of $8 - $126 billion annually in the U.S. alone. This impact
is primarily the result of:
1. Downtime
2. Product Loss
3. Efficiency Loss
4. Contamination
5. Overdesign
Safety / Loss of Life

Corrosion can lead to catastrophic system failures which endanger human life and health. Examples
include a 1967 bridge collapse in West Virginia which killed 46. The collapse was attributed to
stress corrosion cracking (SCC). In another example, the fuselage of an airliner in Hawaii ripped
open due to the combined action of stress and atmospheric corrosion.
2-1
2.1 Corrosion cell
Corrosion occurs due to the formation of electrochemical cells. In order for the corrosion reaction
to occur five things are necessary. If any of these factors are eliminated, galvanic corrosion will
not occur. THIS IS THE KEY TO CORROSION CONTROL! The necessary factors for corrosion
to proceed are:
1. ANODE - the metal or site on the metal where oxidation occurs (loss of electrons). The anode
has a more negative potential with respect to (wrt) the cathode and is termed less noble wrt the
cathode.
2. CATHODE - the metal or site on the metal where reduction occurs (gain of electrons). The
cathode has a more positive potential wrt the anode and is termed more noble wrt the anode.
3. ELECTROLYTE - the electrically conductive medium in which the anode and cathode reside.
4. ELECTRICAL CONNECTION - the anode and the cathode must be electrically connected.
5. POTENTIAL DIFFERENCE - a voltage difference must exist between the anode and the
cathode.
Schematic of the corrosion cell
The electrochemical cell is driven by the potential difference between the anode and the cathode.
This causes a current to flow, the magnitude of which will be determined by the resistance of the
electrochemical circuit (i.e. Ohm’s Law, I=V/R). The three main types of electrochemical cells
are:
 Concentration Cells - this is where the anode and the cathode are the same material, but
concentrations of reactants and therefore potential differ at the electrodes. These can be oxygen
concentration or metal ion concentration cells.
 Bimetallic Cells - where the anode and cathode are different materials.
 Thermo-galvanic Cells - where the anode and the cathode are of the same material and the
composition of the electrolyte is the same but the temperature at the electrodes are different.
In the corrosion cell, metal ions formed from metal oxidation (cations) migrate from the anode to
the cathode through the electrolyte. The electrons given off by this oxidation reaction move from
the anode to the cathode through the electrical connection. Current flows from cathode to the anode
through the electrical connection and from the anode to the cathode in the electrolyte.
2-2
Anode half-cell reaction

Oxidation of the metal at the anode may be expressed by the following half-cell reaction:
M  M n   ne 
Cathode half-cell reaction

The reduction half-cell reaction at the cathode depends mainly on environmental conditions. The
following six reactions represent common cathodic reactions along with the conditions in which
they generally occur:
1. O2  2 H 2 O  4e   4OH  aerated neutral to alkaline water

 
2. O 2  4 H  4e  2 H 2 O aerated acidic solutions
3. 2 H   2e   H 2  hydrogen evolution (in acids)
4. M n   ne   M metal deposition
5. M n   e   M (n 1) metal reduction
Example: Magnesium is submerged into a bath of HCl. What would the predominate anodic and
cathodic reactions be?
Solution:
Anodic Mg  Mg 2   2e 
Cathodic 2 H   2e   H 2 
Overall reaction Mg  2 HCl  MgCl 2  H 2 
2.2 Specific Examples of Electrochemical Reactions
Keeping in mind the definition of corrosion as a destructive electrochemical process, resulting in

material wastage, we examine here the basic concept of a corrosion cell and its associated reactions.
Iron Bar in Hydrochloric Acid

To begin, we consider a simple idealized case where a perfectly homogeneous bar of iron (every
part the same as every other part) is placed in a container of hydrochloric acid (Hcl). On placing the
bar in the acid, some of the iron will initially tend to go into solution as ferrous iron (Fe ++) as a
result of attractive forces exerted by the ionized acid solution. Such a situation will, however, leave
the departure points on the bar with electrons, causing negative charge so the tendency for ion
formation is thus counteracted by local electrostatic attraction from these fixed electrons, i.e. the
reaction stops. If the bar is perfectly homogeneous, as assumed, there will accordingly be no loss
of metal to the solution and therefore no corrosion of the bar.
Iron and Copper in Hydrochloric Acid

Suppose, however, that we now place a second metal bar, say a copper, in the acid and connect a
wire between them as indicated in the figure below:
2-3
Iron and Copper in Hydrochloric Acid
In this case, we have what is called a corrosion, or galvanic cell. Again, there is a tendency for the
iron to be drawn into solution as ferrous ions and leave behind electrons. But now, these electrons
will no longer be fixed in place at the departure sites of the ions. They will, in fact, be drawn through
the iron bar and connecting wire by the natural potential difference existing between the two metals.
In this way, the iron ions are thus fully drawn into solution because the freed electrons are drained
away rather than remain at the departure sites to attract the ions back to the metal. Of course, when
the electrons reach the copper bar, they must be used in some way if the process is to continue. This
happens because the electrons can combine with the hydrogen (H +) ions in solution to form
hydrogen (H2) gas on the surface of the bar which then bubbles off. In addition, the Fe ++ ions can
combine with the Cl- ion in solution to form ferrous chloride. Thus, as long as the above conditions
exist, corrosion of the iron bar will continue.
In the above example, oxidation occurs at the anode according to the reaction:
Fe  Fe++ + 2e-
Reduction occurs at the cathode according to the reaction:
2H+ + 2e-  H2(gas)
A compound also forms as described by:
Fe++ + 2Cl-  FeCl2
Which is ferrous chloride. (Note how the reaction is balanced, so remember your chemistry).
Iron and Copper in Water

Suppose now that we place the iron and copper bars in water rather than HCl. In this case, the anode
reaction is still:
Fe  Fe++ + 2e-
2-4
The cathode reaction is, however, now mainly
O2 + 2H2O + 4e-  4(OH)-
that is the production of hydroxide ions. The additional reaction in the electrolyte is
Fe++ + 2(OH)-  Fe(OH)2
which is ferrous hydroxide. This compound in the presence of oxygen is converted to ferric
hydroxide according to the reaction
4Fe(OH)2 + O2 + 2H2O  4 Fe(OH)3
Ferric hydroxide is orange to red-brown in color and compromises most of ordinary rust.
Non-Perfect Bar
The above discussion assumed a homogeneous iron bar such that if placed alone in an electrolyte,
no corrosion would occur. In reality, though, no bar is perfectly homogeneous, and when placed
alone in an electrolyte, a part of it will act as an anode and a part as a cathode, thus giving rise to a
local corrosion cell as indicated in the figure below
Local Anodes and Cathodes
As corrosion proceeds, the location of the anode and cathode regions will shift around on the bar
and the bar will corrode more or less uniformly. This is the basis for the uniform corrosion (rusting)
of an iron bar immersed in water. At the anode region, the reaction as before is
Fe  Fe++ + 2e-
At the cathode region
O2 + 2H2O + 4e-  4(OH)-
2-5
As discussed above, ferric hydroxide is formed from the Fe++ and OH- ions and this compound (rust)
precipitates out of solution. Of course, if a non-perfect bar of iron is connected to a copper bar, as
described earlier, the cathodic behavior of the copper bar will generally override that of any regions
of the iron bar itself, and the entire iron bar will act as an anode.
When two bars are placed in an electrolyte and connected, the question of which one acts like the
anode and which like the cathode is determined by the potential difference existing between them.
The same is also true regarding regions of a non-perfect bar.
2.3 Measurement of Potential
Consider again the iron and copper bars discussed earlier. To fix ideas, we now suppose the iron to
be immersed in a solution of iron ions (Fe++) with a concentration of 1 mole/liter. Similarly, we
suppose the copper is immersed in a solution of copper ions (Cu ++) with a concentration of 1
mole/liter. We also suppose a voltmeter connected between the two bars as indicated in the figure
below:
Iron and Copper Bars with a Voltmeter Connection
If we measure the potential difference when no current exists, we find with the voltmeter
hook-up indicated that
ΔV = VFe - VCu = -0.789 volts
Since this shows the potential of the copper to be greater (i.e. less negative), the direction of the
current, externally will be from the copper to the iron when the (open circuit) voltmeter is replaced
by a connecting wire. By convention, the current direction is opposite to the flow of electrons. The
electrons would therefore flow from the iron to the copper through the connecting wire and the iron
would thus act as the anode and corrode and the copper would act as the cathode and would not
corrode. This, of course, is consistent with what we assumed earlier when discussing the iron and
copper bars.
2-6
Standard Hydrogen Electrode

It is useful to use a standard hydrogen electrode to measure the potentials of the various metals for
comparative purposes. The hydrogen electrode consists of a platinum tube immersed in a 1
mole/liter solution of H+ ions through which hydrogen gas is bubbled, as shown below:
Hydrogen Half-Cell Electrode, As connected to Measure Potential
Note: With the wire in place, it is equivalent to not having a barrier between the 1 mole/liter ion
solutions. The potential reaction of the hydrogen half-cell is
1/2 H2  H+ + e-
the potential reaction being to the right when the half-cell is anodic to the sample and to the left
when it is cathodic.
2.4 Standard Electromotive Force (Surface Potential)
The surface potential of a metal is a measure of its activity. When a metal is immersed in an aqueous
environment, both oxidation and reduction reactions occur until some equilibrium is reached. These
reactions tend to create an electrical double layer at the surface which establish an electrical
potential. The more positive metals are said to be more noble and less reactive, while the more
negative metals are called base metals and are highly reactive. The standard potential of metals are
given in the following table termed the standard electromotive force series. It should be made clear
however that a metal’s actual potential can be greatly altered by its environment.
2-7
Standard EMF Series Table [from Jones (1996)]

(Note, these potentials are in a solution of 1 mole/liter of their own ions, measured against a
standard hydrogen electrode)
2-8
Consider the potential difference between iron and copper. We have
ΔV = VFe - VCu = -0.447 - 0.342 = - 0.789 volts
as before. The iron would thus act as an anode in a corrosion hook-up and corrode.
Next consider potential difference between iron and zinc.
ΔV = VFe - VZn = -0.447 - (-0.762) = +0.315 volts
Hence, in this case the zinc would act as the anode and corrode while the iron would act as the
cathode in a corrosion hookup. If we put the (+) probe on the iron and the (-) probe on the zinc, we
would read +0.323 volts on the voltmeter. This means that the current would be from the iron to
the zinc and hence that the electrons would flow from the zinc to the iron in accordance with our
previous discussion.
It should be noted that any metal listed in the electromotive series table will act as an anode when
connected to one above it in the series and hence will corrode.
For example, if tin (Sn) and zinc (Zn) are connected in a corrosive cell, the tin will act as the cathode
and be protected and the zinc will act as the anode and corrode. Similarly, if iron is placed in a
solution of nickel ions, the iron will corrode and the nickel will plate out at the local cathode regions.
This is known as a DALIC process. The nickel plating of steel is a common process for the
restoration of dimensions on shafts and the mating surfaces of press fittings. It is also important to
note that given time, as ΔV increases, the rate of reaction also increases.
2.5 Correction for Solution Concentration: Nernst Equation
The potential values listed in the electromotive series table are for a unit concentration (1 mole/liter)
of the metal ions in solution. For other concentrations, we may calculate the potential using the
Nernst equation, which simplifies to:
E = Eo + (0.0592/Z) log10(c)
where Eo is the potential at unit concentration, Z is the valence of the ion (Z = 2 for iron, 3 for
aluminum, etc.) and c is the concentration (moles/liter).
Example: As an example, we may calculate the electrode potential of iron in contact with a solution
of 10-3 moles/liter of Fe++ ions. We have from Table 1, Eo = -0.440 V and
E = -0.440 + (0.0592/2)log10(10-3)
or
E = -0.529 V
2-9
2.6 Reference electrodes
Half-cell reference electrodes are used in corrosion measurements to determine the potentials of
specific metals in selected aqueous environments. The hydrogen half-cell provides the basic
standard, but in practice is awkward to use. For this reason, several other types of half-cells have
been developed. Some of these reference cells are listed in the following table along with their
potential with respect to the hydrogen half-cell and location of use.
Commonly used half-cells [from Swain Classnotes (1996)].

Half-Cell Potential Ref. Environment
SHE
(v)
Copper : Copper Sulfate +0.3160 Soil
Tenth Normal Calomel +0.3337 Laboratory
Normal Calomel +0.2800 Laboratory
Saturated Calomel +0.2415 Laboratory
Silver : Silver Chloride (0.1M KCl) +0.2880 Seawater
Silver : Silver Chloride (Seawater) +0.2222 Seawater
Silver : Silver Chloride (3.8M KCl) +0.1990 Seawater
Hydrogen 0 Laboratory
Zinc -0.7600 Seawater
The following sub-sections present more details on the construction of Saturated Calomel, Saturated
Copper-Copper Sulfate and Silver-Silver Chloride half cells.
Saturated Calomel Reference Electrode

In establishing the galvanic series of metals and alloys in seawater, use is generally made of the
saturated calomel half-cell, which derives its name of calomel from one of its ingredients, mercurous
chloride (Hg2Cl2). The potential reaction is
2Hg + 2Cl-  Hg2Cl2 + 2e-
The potential of the saturated calomel half-cell relative to the hydrogen half-cell is + 0.246V.
The cell is illustrated below.
2 - 10
Figure: Saturated Calomel Reference Electrode
Saturated Copper-Copper Sulfate Reference Electrode

The saturated copper-copper sulfate half-cell is an equally useful cell for corrosion measurements.
A sketch of the cell is illustrated below. When used, it is immersed in seawater where the porous
plug makes the connection to the cell. The half-cell potential reaction is Cu  Cu++ + 2e- and
its reverse. The potential of the copper-copper sulfate half-cell, relative to the hydrogen electrode
is +0.316 V
Silver-Silver Chloride Half-Cell

This electrode consists of silver chloride covering a silver plated platinum wire sealed in a glass
tube, as indicated below. When the electrode is immersed in a chloride solution (such as seawater),
the potential reaction is:
Ag + Cl-  AgCl + e-
The potential of the silver-silver chloride electrode relative to the hydrogen half-cell is +0.266V.
2 - 11
Figure: Silver-Chloride Half Cell
2.7 Galvanic series
In marine applications, we are obviously not so much interested in the potential of a metal immersed
in a solution of its own ions as we are in the relative potentials of metals placed in seawater. Such
a ranking is known as the galvanic series.
The electrochemical series presented earlier can only be applied to oxide free surfaces at ion
concentrations for which the standard potentials are valid. When metals are exposed in a more
complex electrolyte such as seawater, the galvanic series may be used to help predict if corrosion of
a metal is possible. The galvanic series for many commonly used metals is given in the table on the
next page. It should be noted that for some metals, such as the stainless steels, there are significant
differences in the potential they are likely to exhibit. These differences are generally owed to the
condition of the metal surface. For example, 316 stainless steel has a potential of about -0.1v ref
saturated calomel when it is passive (protected by a thin oxide film). If the oxide layer is
compromised, the potential may shift to -0.4v and corrode. In service, severe localized attack may
occur at active sites.
2 - 12
Galvanic Series [from Fontana (1986)].
2 - 13
2.8 Passivating films
Passivity can be defined as the loss of chemical reactivity exhibited by certain metals under specific
environmental conditions. In some cases, oxide films which form on the surface of a metal exposed
in an electrolyte can have a marked effect on its corrosion behavior. This is notable with metals
near the top of the electromotive series. Aluminum, for example, would corrode rapidly in seawater
if it were not protected by a thin oxide surface film.
Alloys containing chromium, iron, nickel, and titanium can become “passive” in strong oxidizing
solutions when they form protective surface films. In this state, they may have a resistance to
corrosion which is orders of magnitude greater than the unfilmed or “active” metal surface. The
stainless steels can often exhibit this behavior. The passivating film is usually of the order of only
3 nm in thickness, however. This means that the film can be quite delicate. The ability of the surface
film to adhere or self heal if compromised by turbulence or mechanical effects can be the
determining factor in their corrosion rate. It is in cases which the surface film does not remain intact
that stainless steels can undergo severe localized attack.
Nonpassivating (left)/ passivating metal (right) [from Corrosion Basics (1984)].
2 - 14
2.9 Concentration Cells
As stated previously, the electrochemical cell may be the result of differing concentrations of
reactants on a metal surface. This can be in the form of either oxygen or metal ion concentrations.
In both the case of the oxygen concentration cell and the metal ion concentration cell, the surface
potential is lowered at the site of lower concentration. Our earlier example of the rotating copper
disk was a metal ion concentration cell. In this case the higher velocity on the periphery of the disk
led to a turbulent boundary layer. This is much more diffusive than its laminar counterpart. The
metal ions produced at the surface move more rapidly from the surface on the periphery. A lower
ion concentration at the metal surface is established, and corrosion occurs on the outer part of the
disk. Copper and its alloys are the most susceptible to setting up a metal ion concentration cell.
If we this time look at a rotating iron disk placed in aerated seawater, it is observed that the corrosion
attack occurs near the center of the disk. Why is this? In this case, an oxygen concentration cell is
established on the iron surface. The oxygen on the metal surface is able to reach a higher
concentration at the periphery of the disk, where turbulent flow allows higher diffusion. At the
center, oxygen concentrations at the metal surface are lower. This causes attack at the center of the
iron disk. Iron and its alloys are the most susceptible to forming an oxygen concentration cell.
Both the metal ion and oxygen concentration cell can also be established in metal crevices. In the
case of the metal ion cell, corrosion occurs just outside the crevice. In the oxygen concentration
cell, attack occurs in the crevice. In all these cases, if the concentrations are known, the driving
potential can be predicted with the Nernst equation.
2.10 Effect of Alloying
The graphs depicting the change in corrosion rate with increasing corrosion potential show how the
amount of oxidizing agent in solution can have a great effect on the corrosion rate of a metal.
Choosing proper alloying metals for a given metal can have a similar effect. Not only can alloying
reduce the amount of oxidizer needed to make the metal go “passive”, but it can also reduce the
corrosion rate of the metal while it is in the passive state. A good example would be adding 18%
chromium to iron. This produces what is termed a ferritic stainless steel. The corrosion rates are
shown in the following graph.
2 - 15
Effect of alloying on corrosion rate [from Corrosion Basics (1984)].
It can be seen that the stainless steel reaches passivity at a much lower corrosion potential and its
corrosion rate in its passive state is much lower than that of iron.
2.11 Pourbaix, or pH Diagrams

A plot for a given metal which expresses its potential versus pH, is termed a Pourbaix diagram. The
diagrams are generally constructed using equilibrium constants, solubility data, and a form of the
Nernst equation which includes a pH term. These diagrams are some very important uses. These
include:
 Predicting if corrosion will occur.
 Finding what the corrosion products may be.
 Forecasting what effect environmental changes may have on corrosion.
The following is a Pourbaix diagram for iron immersed in water.
2 - 16
Pourbaix diagram for iron [from Jones (1996)].
Some interesting information can be gleaned from this diagram. The most striking thing is that if
iron is kept at a potential of less than -1.2 v with respect to a hydrogen half cell, it should not corrode
in a solution of any pH. To keep the iron protected, however, and external voltage would be
required. This is the basis of cathodic protection, which we will discuss later in more detail. It can
also be observed that in a potential range of about 0.7 v and -0.6 v and pH below 9, the corrosion
product will be the ferrous ion. At more positive potentials, the ferric ion would be produced. In
other cases, ferric or ferrous hydroxide and complex iron ions may be formed.
References
Corrosion Basics: An Introduction (1984) National Association of Corrosion Engineers, Houston.
Fontana, M.G. (1986) Corrosion Engineering, 3rd Edition, McGraw Hill, New York.
Jones, D.A. (1996) Principles and Prevention of Corrosion, 2nd Edition, Prentice Hall, Upper Saddle
River, NJ.
2 - 17
Schultz, M.P. (1997) “OCE-4518 Protection of Marine Materials Class Notes”, Florida Institute of
Technology.
Swain, G.W. (1996) “OCE-4518 Protection of Marine Materials Class Notes”, Florida Institute of
Technology.
Dawson, T. (2003) “EN380 Course Notes” United States Naval Academy.
2 - 18
CHAPTER 3
CORROSION THERMODYNAMICS
3.1 Gibb’s Free Energy
3.2 Nernst Equation
3.3 Electrochemical kinetics of corrosion
3.4 Corrosion rates
As we have observed, corrosion reactions inevitably involve electron transfer. For this reason, the
reactions may be considered electrochemical in nature. Thermodynamics can provide a basis for
the understanding of the energy changes associated with the corrosion reaction. It can, in general,
predict when corrosion is possible. Thermodynamics cannot predict corrosion rates. The rate at
which the reaction proceeds is governed by kinetics.
3.1 Gibb’s Free Energy
The Gibb’s free energy, given by the following equation, provides us a tool with which to predict
if a corrosion reaction is thermodynamically possible:
G  -nFE
where : G  Gibb' s free energy (Joules)
n  electrons transfered in oxidation reaction (mol e - )
F  Faraday' s constant (96,500 J/v - mol e - )
E  Standard emf potential = E ox  E red (volts)
0 o
E red  Standard potential for cathode half cell (volts)

o
E ox  Standard potential for anode half cell (volts)

0
If G is positive, the reaction will not proceed. If G is negative, the reaction is possible.
Example: Steel is placed in aerated seawater with a neutral pH. Is corrosion of the steel possible,
why? (Assume valence of 2)
Solution:
Anodic Fe  Fe 2  2e  Eoxo = 0.447 v
 
Cathodic O 2  2 H 2 O  4e  4OH Eredo = 0.820 v
Overall reaction 2 Fe  O2  2 H 2 O  2 Fe OH 2
E = Eoxo + Eredo
E = 1.267 v
G  -nFE (see variable definitions above)
G  -(2 mol e - )(96,500 J/v - mol e - )(1.267 v)
G  - 244,531 J
negative sign indicates corrosion reaction, as written, is possible
3-1
3.2 Nernst Equation
As mentioned earlier, the potential values of a metal are modified by the environment.
Concentrations of anodic and cathodic reactants will alter the balance between the oxidation and
reduction reactions. The Nernst equation allows us to calculate metal potentials under differing
metal ion or oxidation/reduction conditions. The Nernst equation may be stated as follows:
RT
E cell  E o  2.3 log 10
oxid .
nF red .
where : E cell  Cell potential under environmental conditons (volts)
E o  Standard reduction potential @ 25 0 C and unit activity (volts)
R  Universal gas constant (8.3143 J/mol o K)
T  Absolute temperature (degrees Kelvin)
n  electrons transfered in the reaction (mol e - )
F  Faraday' s constant (96,500 coulombs/mol e - )
oxid .  activity of oxidized species (M)
red .  activity of reduced species (M)
At standard temperature and pressure (25oC and 760mm Hg) this may be simplified to the
following (shown earlier):
E cell  E o 
0.059
log10
oxid .
n red .
The Nernst equation can also be written for each half cell as is shown in the following metal ion
concentration cell example.
Example: A circular copper coupon is rotated in seawater. A gradient in the metal ion
concentration is set up on the disk surface. On periphery of the disk copper ion concentration is
0.001 M. Near the center of the disk the copper ion concentration is 10 M. What are the potentials
of anodic and cathodic sites on copper? Where will the metal loss occur? Assume STP.
Solution: In this case we find the standard EMF for the following equation:
Cu 2  2e   Cu
Eo = 0.342 volts
E outside  E o 
0.059
log10 
 
 10 3

n  1 
E center  Eo 
0.059
log10 
 
 101 

n  1 
3-2
E outside  0.342v 
0.059
-
 
log10 10 -3  0.2535v
2 mol e
E center  0.342v 
0.059
-
 
log10 101  0.3715v
2 mol e
Metal will be lost from the outside of the disk (it is more negative)
E overall  E center  E outside  0.118v
3.3 Electrochemical kinetics of corrosion

We now have a tool to predict if the corrosion reaction is possible, but it would also be handy to
predict how fast the reaction will proceed. In theory, Faraday’s law can be used do this. Faraday’s
law may be stated as follows.
m  Izt
where : m  mass of metal lost to corrosion (grams)

I = corrosion current (amps)
a
z = electrochemical equivalent = (g/A - s)
nF
a  atomic weight of corroding metal (grams)
n  electrons transfered in oxidation reaction (mol e - )
F  Faraday's constant (96,500 A - s/mol e - )
t  time of reaction (seconds)
3-3
Example: A steel coupon with an anode surface area of 1000 cm2 is placed in an electrolyte. The
corrosion current is measured to be 1 mA. What mass of steel will be lost in 6 hours? What is the
corrosion rate in g/cm2/day? In mpy? Assume valence of 2.
Solution:
m  Izt
I = 0.001A
a 55.847g
z= = - -
= 2.89x10 -4 g/A - s
nF (2 mol e )(96,500 A - s/mol e )
 60min  60s 
t = 6 hr    = 21,600 s
 1hr  1min 
m = (0.001A)(2.89x10 -4 g/A - s)(21,600s)

m = 6.255x10 -3 g
To find the corrosion rate in g/cm2/day, first divide by anode area and time.
m 6.255x10-3 g  10 6 g  3600s  24hr 

rate =       25.02g/cm 2 /day
At (1000cm2 )(21,600s)  1g  1hr  1day 
To find the corrosion rate in mpy, divide by the metal density.
25.02g/cm 2 /day 25.02g/cm 2 /day  365days  1mil 

rate =  
3 
 
 7.20x10 g/cm  yr  2.54x10 cm 
6 3
rate  0.499 mpy
3.4 Corrosion rates
The rate of corrosion (or current density) is determined by the potential difference beween the
anode and the cathode and the resistance of the corrosion cell. The corrosion current is therefore:
V
I
R
The resistance of the cell may be as a result of electrical resistance or electrode polarization. The
greater the resistance the lower the corrosion current and from Faraday’s law the lower the mass
loss. A high resistance within the corrosion cell is beneficial for control. This resistance may
result from one or more of the following factors:
 Resistance of the electrical connection between anode and cathode.
 Resistance of the electrolyte.
3-4
 High concentration of anode metal ions in solution.

 Reactant build-up at the cathode.
 Lack of reactants at the cathode.
The action of these resistances may be expressed in a polarization diagram. These diagrams plot
potential difference versus current (or log current). The slope of the curve represents the
resistance.
The following table gives density, atomic mass, valence, and corrosion rate for various metals.
Corrosion data for various metals [from Swain Classnotes (1996)].

Element Atomic Mass Valence Electrochemical Corr. rate Density
Equivalent equivalent to 1
A/cm2
(g/mole) (g/coulomb) (mm/yr) (g/cm3)
Magnesium 24.31 2 1.26E-04 0.023 1.74
Zinc 65.38 2 3.39E-04 0.015 7.13
Aluminum 26.98 3 9.30E-05 0.011 2.72
Iron 55.85 2 2.89E-04 0.013 7.20
Iron 55.85 3 1.93E-04 0.087 7.20
Copper 63.54 1 6.58E-04 0.023 8.94
Copper 63.54 2 3.29E-04 0.012 8.94
Nickel 58.71 2 3.04E-04 0.011 8.89
3-5
CHAPTER 4
POLARIZATION
4.1 Activation and Concentration Polarization

4.2 Scaling of Corrosion Measurements
4.3 Determination of Corrosion Current for Single Metals
4.4 Mixed Potential Theory
4.5 Alternate construction of Polarization Diagrams
Previously, our consideration of potentials has dealt with equilibrium considerations when no
current actually existed between the cathode and the anode. When a net current to or from an
electrode exists, the electrode is no longer in equilibrium and the measured potential is altered.
The alteration being such as to oppose the current. The extent of the potential change caused by
the current is known as polarization. Such change is caused by various physical and chemical
factors at the electrodes.
Polarization may be defined the shift in electrode potential which results from the effects of current
flow w.r.t. the zero current flow potential. All corrosion reactions involve current flow and will
alter the potential of the metal surfaces involved. The degree of polarization will be determined
by the resistance of the corrosion cell. The higher the cell resistance, the larger the shift in
potential.
V  IR
Anodic and cathodic polarization can be measured using the cell illustrated below
Laboratory Polarization Cell
As the current is allowed to develop through the variable resistor, the potential difference between
the two metals will decrease as illustrated in the graph below. This decrease in potential difference
is due to the polarization of the two electrodes. Note that the potential is plotted on a linear scale
while the current is plotted on a logarithmic scale, plotted on semi-log paper. This causes the data
to plot approximately as straight lines.
4-1
Polarization Graph for Iron-Copper Corrosion
At II in the figure above, the polarization of the cathode (copper) and the anode (iron) is as indicated
and the potential difference (ΔV) between the two is reduced from its open-circuit value (V Cu -
VFe). Finally, when the resistor is completely shorted out of the circuit, Icor is the maximum
corrosion current and Vcor is the corresponding voltage. Note that the potential difference will not
be zero even when the resistor is shorted since there is internal resistance in the electrolyte.
When the potential change caused by current (polarization) occurs mostly at the anode, the
corrosion is said to be anodically controlled. When polarization occurs mostly at the cathode, it is
said to be cathodically controlled. These conditions are illustrated in the figure below.
System Corroding under (a) anodic and (b) cathodic control
4.1 Activation and Concentration Polarization
There are two different types of polarization, that is, two ways that electrochemical reactions of
corrosion are retarded. These are referred to as activation and concentration polarizations.
Activation polarization is used to indicate retarding factors inherent in the reaction itself. Thus,
for example, with a metal corroding in acid, a finite time is required to form hydrogen gas at the
cathode regions even when an ample supply of electrons exist.
4-2
In contrast, concentration polarization refers to the retardation of an electrochemical reaction of

corrosion resulting from concentration changes in the solution next to the metal surface. A good
example is provided by the cathode region of a metal corroding in seawater, that is,
O2 + 2H2O + 4e-  4(OH)-
which is known as the hydroxyl reaction. Because of its low solubility in seawater, the amount of
oxygen in contact with the cathode areas is small and readily consumed by the reaction. For the
reaction to continue, additional oxygen must therefore diffuse to these areas. The diffusion of
oxygen is a relatively slow process and accordingly gives rise to concentration polarization of the
cathode.
With metals corroding in seawater, the main retarding effect on the corrosion is generally the
concentration polarization resulting from the slow diffusion of oxygen to the cathode areas. The
"bottleneck" in the corrosion process is therefore the cathode reaction, and any situation which
would increase the rate of the oxygen diffusion would increase the rate of corrosion. Thus, for
example, water flowing past a steel pipe is more corrosive than still water because of the increased
oxygen supply. Conversely, de-oxygenated water is far less corrosive than ordinary. This is the
prime reason that steam plant feedwater and reactor coolant water is "deaerated" prior to use.
Because corrosion of metals in seawater is cathodically controlled, the relative size of the anode
and cathode has an important effect on the corrosion. The larger the cathode, relative to the anode,
the faster the corrosion since more surface area on the cathode is in contact with the seawater and
thus in contact with the oxygen. In seawater, small steel (or iron) rivets in a large copper plate
would fail quickly by accelerated corrosion. If the reverse was used, ie steel plates riveted together
with copper rivets, the corrosion would be greatly reduced. These effects can be illustrated by a
polarization diagram.
Effect of Area on Polarization
With the larger cathode area, A2 the polarization (potential change caused by current) of the
cathode is reduced and the corrosion current is increased, thereby accelerating the corrosion of the
anode. These polarization diagrams may also be plotted as a function of current density, below:
4-3
Polarization diagram [from Jones (1996)].
4.2 Scaling of Corrosion Measurements
We may use laboratory scale polarization cells such as illustrated earlier to determine the corrosion
current for two small metal samples. Under appropriate conditions (such as flow of the electrolyte
or bio fouling), we may then scale this measurement to predict the corrosion current for the same
metals when attached to a full scale field set-up.
The procedure is based on the fact that the corrosion current per unit of (wetted) anode area is a
function of the types of metals, the ratio of the area of anode to cathode, and the type and condition
of the electrolyte. Thus, we have
I AA
= F( ,electrolyte,metals)
AA AC
Now, if we fix all the dependent variables in this equation, the independent, I/A A, will be fixed.
Thus, if we keep the same anode and cathode metals, duplicate the electrolyte, and require
AA AA
( )LAB = ( )FIELD
AC AC
I I I I
Then ( )LAB = ( )FIELD??( )LAB = ( )FIELD
AA AA AC AC
4-4
Example: We wish to find the corrosion current existing in a field (ocean) structure when 50 ft 2
of steel (anode) is attached to 25 ft2 of copper (cathode). Laboratory data for a steel bar of 5 in 2
and a copper bar of 2.5 in2 give corrosion current (in seawater) of 5 ma (5 x10 -3 amps).
Solution: First, since the ratio of anode to cathode area is equal to 2, we can directly scale the lab
data.
Therefore,
AA AA
( )LAB=2,( )FIELD=2
AC AC
(I)LAB (I)FIELD
=
(AA)LAB (AA)FIELD
(AA)FIELD 50 x 144
(I)FIELD= (I)LAB or IFIELD= x (5 x 10_3)= 7.2 amps
(AA)LAB 5
Thus, so long as the conditions of the seawater in the lab approximate these in the ocean, the
expected current in the field is 7.2 amps.
4.3 Determination of Corrosion Current for Single Metals
When two metals are in electrical contact in an electrolyte such as seawater, we can determine the
corrosion current and potential using the polarization cell illustrated in Figure 8. When a single
bar or object is placed in seawater, however, the anodes and cathodes will be located at small,
continually changing areas of the bar. As discussed earlier, an indirect method for determining
the corrosion current and potential must be used. Consider the set-up shown below consisting of
a specimen.
Cell for study of polarization of a single metal
If the specimen is attached to the negative side of the power supply, electrons will be drawn from
4-5
the auxiliary electrode and supplied to it. Thus the specimen will be made to act as a cathode even
if it is naturally anodic to the auxiliary electrode. Similarly, if the specimen is attached to the
positive side of the power supply, it can be made to act as an anode, with electrons being drawn
from it and supplied to the auxiliary, regardless of its natural tendency toward the auxiliary
electrode.
Using this cell, we thus first make the specimen act as an anode and determine current and potential
data for a number of power supply settings. Next we repeat the process when the specimen is
made to act as a cathode. On dividing the current measurements by the specimen area in contact
with the seawater, we obtain current density measurements which are independent of the particular
size of the specimen used in the test. In this way we can then construct the polarization diagram
shown below.
Polarization Diagram for a Single Specimen in Seawater
When the potential approaches the corrosion potential V cor, certain areas of the specimen will act
as anodes and cathodes regardless of the polarity setting of the power supply. Thus, the
polarization curve will no longer reflect total anodic or cathodic behavior, as indicated by the
portion of the curves in the figure above near V cor.
By extrapolating the pure anodic and cathodic curves, we can estimate the corrosion potential and
the corresponding net corrosion current density existing for the specimen. With the corrosion
current density known we can then calculate the corrosion current for a given surface area in
contact with the seawater simply as the product of the current density and the wetted area.
Example: Listed below are potential - current density polarization data for a metal A in seawater.
Use this data to determine the corrosion current existing for 20 ft 2 of metal A in contact with
seawater.
4-6
Metal A
2
Pot (v) i (ma/in ) Pot (v) i (ma/in2)
-0.70 0.100 -0.52 0.087
-0.74 0.158 -0.40 0.110
-0.80 0.251 -0.32 0.138
-0.85 0.400 -0.20 0.190
4-7
Solution: The data is plotted as shown with log10(i) as the horizontal axis. The corrosion current
density is read as log10(i) = -1.22 or i - 0.06. The corrosion current for the 20 ft 2 area is I = 0.06
x 20 x 144 = 173 ma.
4.4 Mixed Potential Theory
Polarization diagrams such as shown above “Polarization Diagram for a Single Specimen in
Seawater“, if determined for two dissimilar metals, can be used to estimate the corrosion current
and mixed corrosion potential when the two are connected together in a galvanic cell. The method
is based on the requirement that the magnitude of the current must be the same for the anode and
the cathode. Thus, the current densities in the polarization diagrams for each metal must be
converted into actual currents for the given area of metal in contact with the electrolyte (the product
of the current density and respective areas). When plotted together, the intersection of the two
polarization (potential vs current) curves will indicate where the current is the same for both
metals, and thus, the corrosion current. This method is illustrated in the figure below. In this figure,
the corrosion current Icor and mixed potential Vcor are determined as shown. For the two metals, it
is also seen that metal B acts as the anode and metal A as the cathode.
4-8
Example: Listed below are potential-current density data for metals A and B in seawater.
Metal A Metal B
Pot (v) i (ma/in2) Pot (v) i (ma/in2)

-0.66 0.075 -0.54 0.300
-0.74 0.090 -0.70 0.215
-0.86 0.120 -0.86 0.150
Use this data to estimate the corrosion current existing for 20 ft 2 of metal A in contact with 12 ft2
of metal B in a Galvanic cell.
Solution: The current densities of A are multiplied by the area, 20 x 144 = 2880 in 2, to get the
total current. Potential vs log10(I) is then plotted as indicated. Similarly, the current densities of
B is multiplied by its area, 12 x 144 = 1728 in 2 and its potential-current relationship plotted. B is
seen to act as the anode and metal A as the cathode. The corrosion current is read from the
intersection of the two lines as log10(I) = 2.47, or I = 295 ma.
Example: Polarization diagrams for two metal specimens A and B are shown below. Using these,
determine the corrosion current (in amps) for a submerged structure having 45 ft 2 of metal A in
contact with 90 ft2 of metal B.
4-9
Solution: This diagram is for 20 in2 of metal B in contact with 10 in2 of metal A. Since the ratio
of these areas (20/10 = 2) is the same as that of interest (90/45 = 2), we may find the corrosion
current by directly scaling the value found from the data. The corrosion current from the data is
I = 1.9 ma. Hence since the ratio of I/AA is the same for the lab and the field under the conditions
of direct scaling,
(I)LAB (I)OCEAN (AA)OCEAN
= or (I)OCEAN = ILAB
(AA)LAB (AA)OCEAN (AA)LAB
we have
I = 1.9 x 45 x 144/10 = 1231 ma (1.23 amps)
4.5 Alternate construction of Polarization Diagrams
The polarization diagrams discussed above refer to the case where the anode reaction is that of a
metal changing to its ion and the release of electrons, and the cathode reaction is that typical of
seawater where the electrons are used to make hydroxyl ions, as indicated the figure “Polarization
Diagram for a Single Specimen in Seawater” a few pages back. An alternate diagram may be
determined for which the anode reaction is still the formation of the metal ion, but the cathode
reaction is the plating out of the ion from an ion-rich solution. (Such as in an acid bath). The
measurement thus involved is an experimental polarization cell, except that the seawater is
replaced with an electrolyte rich in the metal ion. The figure below illustrates the polarization
diagram obtained by this procedure.
4 - 10
Polarization diagram for case where cathode reaction is the plating out of the metal ion.
An interesting interpretation of this diagram is that the intersection of the anode and cathode curves
gives, not the corrosion current, but instead the equilibrium exchange current where as many ions
are going into solution as are coming out. If we superimpose on the polarization diagram of Fig.
15, the corresponding diagram for the reaction O2 + H2O + e-  OH- we then have the diagram
shown below.
Superimposed Polarization Diagrams for metal/seawater reaction
The solid line in the figure above is seen to correspond to the polarization diagram shown earlier.
The intersection represents the corrosion potential V cor and the free corrosion current density icor
of the metal in seawater. This alternate method for polarization diagrams is thus seen to be one-
step more detailed than that of our earlier discussion.
4 - 11
CHAPTER 5
CORROSION TYPES
5.1 Uniform Corrosion
5.2 Galvanic Corrosion
5.3 Crevice Corrosion
5.4 Pitting Corrosion
5.5 Other Localized Corrosion
5.6 Selective Leaching aka Dealloying
5.7 Erosion-Corrosion
5.8 Stress-Corrosion-Cracking (S.C.C.)
5.9 Intergranular Corrosion
5.10 Fretting
5.11 Biological corrosion
5.12 Stray current corrosion
The diagram on the following page illustrates the main types of corrosion which will be discussed
in this chapter.
5-1
Corrosion types [from Jones (1996)].
5-2
5.1 Uniform Corrosion
This corrosion results from the continual shifting of anode and cathode regions of the surface of a
metal in contact with the electrolyte and leads to a nearly uniform corrosive attack on the entire
surface. An example of such corrosion is the rusting of steel plate in seawater.
If the rate of metal loss is known, allowances can be made in design and maintenance to
accommodate the corrosion. Although it is termed uniform corrosion, it is characterized by the
average surface loss.
Uniform corrosion of storage tanks [from Jones (1996)].
5.2 Galvanic Corrosion

When two different metals are exposed to a corrosive environment, an electrical potential
difference will exist. If the two metals are electrically connected, the more active metal will
become the anode in the resulting galvanic cell and its corrosion will be increased. An example
of such a corrosion cell is the use of steel bolts to hold copper plates together
Not all galvanic corrosion is detrimental. Zinc coated steel, or galvanizing, is used to protect steel,
not because the steel is resistant to corrosion, but because the zine, being anodic to the steel,
corrodes preferntially. Hence, the steel is protected cathodically by making any exposed areas of
steel into cathodes.
5-3
It is generally good practice not to use dissimilar metals unless it is necessary, but if it is, the
following precautions should be used:
 Attempt to electrically isolate the metals.

 Use protective coatings on the metal surface(s), generally the cathode
 Cathodically protect the less noble metal.
 Put corrosion inhibitors into the system.
 Use design in which anodic part may be replaced easily.
 Keep out moisture.
 Use metals that are close to one another in the galvanic series.
 Design so that the anode/cathode area ratio is high.
 Use design allowances to account for the corrosion.
5.3 Crevice Corrosion
Crevice corrosion is a localized attack which occurs when crevices, formed by lapped joints, or
areas of partial shielding, are exposed to corrosive environments. Such resulting cells are referred
to as concentration cells. Two common cases are oxygen cells and metal-ion cells.
Oxygen concentration cells occur when the shielded area becomes depleted in oxygen and the area
acts as an anode relative to the oxide region. As illustrated in Figure 24, the corrosion becomes
quite rapid because of the small shielded area as compared to the unshielded area. Do not get
confused with the concentration polarization that we previously discussed. In the case of an
oxygen cell, we have an oxygen "gradient" that forces the formation of the anode and cathodes
with respect to the oxygen levels.
5-4
Crevice corrosion of stainless steel [from Jones (1996)].
The initial driving force of such corrosion is the oxygen cell. The continued growth is fostered by
the accumulation (often caused by the same factors that produced the low oxygen level) of acidic,
hydrolyzed salts within the crevice. Alloys, such as 18-8 stainless steels, are subject to oxygen
cell crevice corrosion.
Metal-ion cells are formed mainly with copper alloys. The shielded area accumulates corrosion
products and becomes cathodic to the regions outside of the crevice where corrosion products are
kept washed away. The figure below illustrates this type of concentration cell.
Another example of metal-ion cell corrosion occurs when relative speeds of electrolyte over the
metal surface are greater at one point than at another, thus resulting in metal-ion crevice corrosion.
A good example is where a disc of metal is rotating at high speed in seawater. Corrosion occurs
near the edge where linear velocities (v=ωr) are the highest and the metal-ion concentration is low
(since the ions are repeatedly swept away). The high velocity, higher than in regions closer to the
hub of the disc, sweeps away the metal-ions, thus forming anode regions. At the center of the disc,
where velocities are lower, the metal acts as a cathode and is protected (Do not confuse this with
the discussion on "immunity" from corrosion where the low ion concentration was one of the
entering arguments. That was an equilibrium concentration. In this case, metal ions continue to
form because we can't reach an equilibrium concentration).
However, the two concentration cells corrode at different regions of the crevice. The oxygen cell
corrodes under the shielded area while the metal-ion cell corrodes outside of the area. As stated
before, the initial driving force behind the corrosion is either the oxygen or the metal-ion cell. Its
continued growth is governed by the accumulation of corrosion products, clacareous deposits, and
salts within the crevice.
5-5
5.4 Pitting Corrosion
Pitting is an extremely localized attack that eventually results in holes in the metal. It is one of the
most destructive and insidious forms of corrosion. Basically, the alloys subject to pitting are those
that rely on an oxide film for protection, such as stainless steels. The initiation of a pit can be the
result of any of the following:
a) Chemical attack, such as ferrous chloride or aerated seawater on stainless steel.
b) Mechanical attack such as an impact or scratching that removes small areas of the
protective film.
c) Crevice corrosion resulting from tiny deposits on the surface, especially in stagnant
seawater.
Some theories state that pitting is just a special case of crevice corrosion.
Localized pitting of stainless steel [from Jones (1996)].
5.5 Other Localized Corrosion
In addition to the pitting and the crevice corrosion, the following are also localized corrosion:
 Poultice corrosion - attack which occurs at the edge of a damp fabric.

 Deposition corrosion - corrosion process in which a more noble metal deposits on a less noble
metal and causes attack due to a bimetallic cell. It is common in copper/aluminum systems
 Filiform corrosion - localized attack of a metal surface beneath a coating due to oxygen
concentration cells.
5-6
5.6 Selective Leaching aka Dealloying
Selective Leaching corrosion results from areas of a metal surface being different metallurgically
from other, adjacent areas. Brass, for example, is an alloy with zinc and copper in a "solid
solution". It can corrode with the zinc being selectively removed from the alloy, leaving behind
the copper. It makes the alloy porous and compromises its mechanical properties. In brass it may
be identified when its yellow natural color turns reddish or coppery in appearance. It is helpful to
add a small amount of tin to the alloy to prevent dealloying.
Such selective leaching is known as "dezincification". Cast irons can corrode in such a matter that
the iron is selectively corroded away, leaving behind a soft graphite layer. This is referred to as
"graphitization". Other examples are referred to as dealuminification, denickelification,
decobaltification, etc. where the terms refer to the metallic element that is selectively corroded
away.
The mechanism of selective leaching has been explained as follows for a brass alloy:
A. the brass corrodes
B. the zinc ions stay in solution
C. the copper plates back on as a solid layer
The problem with this theory is that the corrosion occurs even under high electrolyte flow
velocities when one would surmise that the copper ion would be swept away before they could
plate out.
A second theory, again for brass, is that the zinc corrodes preferentially, leaving behind copper in
a lattice structure. A corrosion process in which the less noble metal in an alloy is attacked
preferentially and replaced in the matrix by cathodic products. The most common example of this
occurs with brass and is termed dezincification. In the dezincification of brass, the zinc in the
alloy’s matrix is attacked and copper remains.
Dealloying of brass [from Jones (1996)].
5-7
5.7 Erosion-Corrosion
Erosion-Corrosion results from a high velocity electrolyte flow whose abrasive action accelerates
the corrosion. This corrosion is especially severe when the electrolyte contains solids in
suspension. The effect is to remove a protective oxide from the film surface, thus exposing fresh
alloy to corrode. Erosion-Corrosion could be thought of as pitting on a much larger scale.
There is, in fact, a limit to what electrolyte velocities can be tolerated by specific metals. Copper-
nickel alloys are selected for seawater service based on their resistance to erosion-corrosion
(amongst other requirements). This table illustrates these limitations.
Recommended Maximum Velocity to Reduce

Impingement/Velocity Effects in Seawater
Alloys Maximum velocity, fps
Copper 3
90 Cu/10 Ni (with 1.25% Fe) 12
70 Cu/30 Ni (with .5% Fe) 15
85 Cu/15 Ni (with .5% Cr) >15
Titanium >15
In addition to Erosion, other forms of attack related to velocity effects are:

 Cavitation - the deterioration of a surface caused by the sudden formation and collapse of
bubbles and voids due to the turbulence in the liquid. It is generally marked by a pitted or
rough metal surface.
 Impingement attack - localized corrosion caused by turbulence or impinging flow. Generally
there is a critical velocity below which no impingement occurs and above which attack
increases rapidly.
It should be stated that in many cases these three corrosion processes occur simultaneously.
5-8
Erosion corrosion [from Jones (1996)].
5.8 Stress-Corrosion-Cracking (S.C.C.)
Stress-corrosion-cracking occurs with specific alloys under the following threshold conditions:
a) Specific corrosive environment solution composition
b) Minimum tensile stress levels
c) Temperature
d) Metal composition
e) Metal structure
Some examples of scc are the brass and stainless steel alloys. Specific brass alloys will crack in
ammonia containing environments when a minimum threshold tensile stress is reached. Stainless
steel alloys do not crack in ammonia environments, but will crack in chloride solutions.
The interplay of the conditions leading to scc is not well understood. It is believed that the
corrosion causes a pit or surface discontinuity to form on the metal which then functions to act as
a stress concentrator. The presence of a minimum threshold tensile stress, coupled with the
corrosion, causes the crack to propagate. Additionally, during the initial corrosion, the tensile
stresses could cause the protective films on the surface to rupture, thereby exposing the metal to
the corrosive environment.
This particularly dangerous corrosion type can be the result of environmental factors or cyclic
stresses. The following are the major types of cracking attack:
 Corrosion fatigue - the accelerated failure of a metal which undergoes cyclic loading due to its
presence in a corrosive environment.
 Stress corrosion cracking (SCC) - the corrosion induced cracking which occurs in alloys under
high tensile stress. The cracks start on the surface and go inward. It should be noted that the
stress can be the result of cold working, forming, or external loading.
5-9
 Hydrogen embrittlement - the loss in ductility of a metal due to the saturation of atomic
hydrogen in the grain boundaries. It occurs at local cathodic sites and is aggravated by stress
and compounds such a hydrogen sulfide.
 Liquid metal cracking - metals subjected to simultaneous tensile stress and certain molten
metals can undergo this type of cracking. It is most common in mercury/copper alloy systems.
Stress corrosion cracking [from Jones (1996)].
5.9 Intergranular Corrosion
On a microscopic level, metals and there alloys have small, distinguishable regions called grains.
Within an individual grain the orientation of the atomic arrangement (called a lattice) is the same.
Individual grains have different orientations and the boundary between the grains is called the
grain boundary. Normally, grain boundaries are no more reactive in corrosion than the grain itself.
Under certain conditions, however, the grain boundaries are altered from the grain itself by
impurities and/or enrichment (or depletion) of one of the alloying elements.
Heat treatment and welding can lead to changes metal composition which may incite intergranular
corrosion. In severe cases, intergranular corrosion can lead to a marked decrease in mechanical
properties and can, in extreme cases, turn the metal into a pile of individual grains.
One of the most common examples of intergranular corrosion occurs in stainless steels. During
welding of the alloy, or heating in the temperature range of 950 F to 1450F, the alloy becomes
sensitized or susceptible to intergranular corrosion as illustrated in Figure 26. The chromium
carbide (Cr23C6) is not soluble in this temperature range and preciptates out of the grain into the
grain boundary. As a result, the area of the grain adjacent to the grain boundary is depleted of the
chromium and becomes anodic to the rest of the grain and to the grain boundary. The corrosion
of this depleted grain boundary area is very severe and occurs in environments and acids where
the alloy would not normally corrode. The simplest solution to the stainless steel intergranular
corrosion problem is to cast alloys with carbon contents below 0.03%. (This prevents the formation
of the chromium carbide and the chromium stays in solution).
5 - 10
Intergranular attack [from Jones (1996)].
5.10 Fretting
A rapid localized attack which occurs on mated surfaces under load when a small amount
of slip is allowed to occur. It is often observed on bearings, shafts, and gears in mounted
in vibrating machinery. Not only is mechanical damage of the surface possible, but the
protective surface film of the metal is also removed. This in turn hastens electrochemical
corrosion processes.
5.11 Biological corrosion

Biological organisms can play a major role in metal attack. This attack is usually
categorized in the following two headings:
 Microbially induced corrosion (MIC) - aerobic and anaerobic bacteria and other
microorganisms contain enzymes and can produce metabolites which accelerate
corrosion. This can manifest itself in pitting type corrosion of the metal surface.
 Macrofouling effects - barnacles, oysters, and other macrofoulers produce by-products
that are often acidic and can accelerate corrosion. These organisms also create crevices
at their attachment points that can lead to crevice corrosion.
5 - 11
5.12 Stray current corrosion
Corrosion can be accelerated by the action of electrical currents entering a metal from some
external source such as a generator or a battery and leaving the metal to continue its flow
in whole or part through the seawater electrolyte. As an example of stray current corrosion,
consider the bilge pump illustrated below. The electrical connections were made such that
the current found a path from the pump and its through-hull connection to the seawater and
back through the propeller shaft and engine to the battery.
To prevent such corrosion, all metal surfaces exposed to bilge water or projecting through
the hull should be electrically connected (bonded) to the negative side of the battery as
shown in the following figure:
Another example of stray current corrosion damage to a ship is through the use of an
onshore welding generator below. If the return leads to the onshore generator do not have
ample current carrying capacity, some of the current will find its way back through
discontinuities in the paint on the submerged hull, resulting in corrosion of the hull. To
avoid such corrosion, the generator should have been placed on the ship's deck and its AC
lines could have been run from shore. Any AC stray current to shore would have caused
little damage.
5 - 12
Stray current corrosion can also occur when the current from a cathodic protection system
passes from the anode to a metal object which was not designed to be protected before
going to the object which was intended to be protected. The result can be severe local
attack on the unprotected metal where the current exits. It can also attenuate the current to
a point that the object to be protected is underprotected.
Stray current corrosion [from Jones (1996)].
References
Jones, D.A. (1996) Principles and Prevention of Corrosion, 2nd Edition, Prentice Hall,
Upper Saddle River, NJ.
Schultz, M.P. (1997) “OCE-4518 Protection of Marine Materials Class Notes”, Florida
Institute of Technology.
Swain, G.W. (1996) “OCE-4518 Protection of Marine Materials Class Notes”, Florida
Institute of Technology.
Dawson, T. (1993) “EN380 Engineering Aspects of Materials Class Notes”, United States
Naval Academy
5 - 13
CHAPTER 6
CORROSION PROTECTION AND MONITORING
6.1 Cathodic Protection: Sacrificial Anodes

6.2 Cathodic Protection: Impressed Current
6.3 Measures of Corrosion Rate (Weight Loss and Strength Tests)
6.4 Corrosion Allowance Calculations
As inferred during the previous discussion of the Pourbaix diagram, corrosion may be prevented
by introducing electrical currents from external sources to counteract the normal electrochemical
reaction. There are two basic types of so-called cathodic protection systems, namely, sacrificial
anodes and impressed current. Sacrificial anodes consist of expendable electrodes, usually zinc,
aluminum or magnesium, or their alloys, whose natural potentials provide adequate protective
current to a structure. Impressed current systems are those that employ a direct current (DC)
power source to provide the necessary current.
6.1 Cathodic Protection: Sacrificial Anodes
When iron and copper are attached in a galvanic cell, we have seen that the result is the corrosion
of the iron, with electrons flowing out of the iron (the anode) and into the copper (the cathode).
Suppose that we now place a zinc bar in the cell as illustrated below:
Sacrificial Zinc Anode
In this case, we now find, that because of the potentials involved, that electrons flow out of the
zinc and into both the iron and copper. Thus the zinc corrodes and the iron and copper, which
now both function as cathodes, do not. This is the basis for sacrifical anode protection, namely,
the use of an expendable anode to protect another metal from corrosion. Two examples of this
type of corrosion protection are shown in Figure 28. The first is the protection of an ordinary
hot-water heater with a magnesium anode and the second is the protection of a ship's hull in the
vicinity of the bronze propeller by zinc anodes. Without the protection, the steel hull of the ship
is, of course, subject to galvanic corrosion because of the presence of the bronze material used
for the propeller.
6-1
Example Sacrificial Anodes
To determine the amount of material needed for sacrificial anode protection of a structure, we
use Faraday's Law together with an estimate of the current density needed for protection
(typically in the range of 10-15 ma/ft2).
Example: If an aluminum alloy is 100% efficient as a sacrificial anode (i.e. it follows Faraday's
Law exactly) determine the number of pounds of aluminum needed to protect a steel structure
with 80 ft2 of surface area in seawater for 18 months. Assume a current density of 10 ma/ft 2 is
needed for current protection.
Solution
W=_CI
From Faraday' Law we have
27
3
C= = 9.33 x 10-5 grams/amp_sec
96,500
where
w = -7.45 x 10-5 grams/sec

Hence
w=(7.35 x 10-5)(2.205)(10-3)(3600)(24)(365)=5.19lb/year
in lbs/yr
For total consumption in 18 months, we must have (for 100% efficiency)
w = (5.19)(1.5) = 7.7 lbs
6-2
Example: A chemical company advertises aluminum-alloy sacrificial anodes with an average

current output of 1176 amp-hours/lb. Calculate the efficiency.
Solution: From Faraday's Law we have the consumption given as

27
3
C= = 9.33 x 10-5 grams/amp-sec
96,500
hence, the current output is

C-1 = 1.07 x 104 amp-sec/gm
or
C-1 = 1350 amp-hours/lb
The efficiency of the advertised anodes is therefore 1176/1350 = 0.87 or 87%.
Example: An offshore structure has 10,000 ft 2 of submerged area to be protected for 20 years
with aluminum-alloy anodes (85% efficient) weighing 150 lbs each. Determine the number of
anodes required if 15/ma ft2 is needed for protection.
Solution:
Again, from Faraday's Law we have
27
3
C= = 9.33 x 10-5 grams/amp_sec
96,500
and
I = 150 amp
so that
w = -0.014 gm/sec = 973 lbs/year
Hence, w = 973 x 20 = 19463 lbs if the anodes were 100% efficient. Since they are only 85%
efficient, 19463/.85 = 22,900 lbs are required and 22,900/150 = 152.6 or 153 anodes.
6.2 Cathodic Protection: Impressed Current
In an impressed current system of cathodic protection, the current can be supplied from such
sources as storage batteries, rectifiers, or generators depending on convenience and the amount
of current required. The anodes used in an impressed current systems can be expendable, being
made of ordinary steel. Such steel anodes would require periodic replacement, since they are
destroyed at a rate of about 20 pounds/ampere-year by the passage of the protective current. It is
common to use permanent impressed current anodes which are not destroyed or are destroyed
very slowly by the passage of the protective current. These anodes are platinum; platinum
sheathed titanium, tantalum, or niobium. Steel structures exposed to seawater are normally
6-3
protected if they are polarized to a potential of -0.85 volts versus a silver/silver-chloride or a

copper/copper-sulfate reference electrode.
A description of such a system follows. Consider again the case of iron and copper attached
together in a galvanic cell, with the iron corroding and the copper protected. Instead of using a
sacrifical anode to supply electrons to the steel and protect it, we use a DC power supply. In this
case, we have the cell described in the figure below.
Impressed Current Cathodic System
By proper choice of the power supply voltage, we can over-ride the potential difference between
the iron and copper and make each act as a cathode (receive electrons) and be protected. In
selecting the proper potential, it must be the same as the open-circuit potential of the local anode.
This means that the voltage is chosen such that current is neither entering or leaving the iron, nor
is it flowing between the local anode and cathode regions on the iron. By use of a standard half-
cell, as in Figure 29, a polarization diagram can also be produced. The open circuit potential and
applied current can be determined in this manner. Thus, the rule for selecting the proper voltage
setting of the power supply is to adjust the power supply until the potential of the iron (and
copper) relative to a standard half-cell is equal to the open circuit potential of the local anode
regions, which was determined experimentally. Some examples of impressed current protection
are shown below .
6-4
Example: Use the Nernst equation to establish the potential of steel (iron) relative to the
copper/copper-sulfate half-cell needed for protection from corrosion.
Solution: We assume a concentration of 10-3 ions/liter in the electrolyte adjacent to the steel to
indicate essentially negligible corrosion activity. From the Nernst equation, we then have
relative to the hydrogen half-cell,
E = -0.44 +(0.0592/2)Log1010-3
or
E = -0.53 v
The potential of the copper/copper-sulfate half-cell relative to the hydrogen is +0.32V. Hence
E = VFe - VH2 = VFe - VCu + VCu - VH2

or
-0.53 = VFe - VCu + 0.32

or
VFe - VCu = -0.85 v
6.3 Measures of Corrosion Rate
The Galvanic Series gives us information on which metals will corrode when connected with
others in seawater, but it does not tell us anything about the rate of corrosion. Polarization
experiments can give us information on the rate of corrosion of metals connected together or
freely corroding. When time permits, field tests can provide the simplest and most reliable way
to obtain corrosion-rate data.
6-5
Weight-Loss Tests. These experiments are conveniently made using flat plates, as shown in the
figure below.
Sample for Weight Loss Test
The weight of the plate is given in terms of geometry of the figure above and the specific weight
γ of the material as
2γabh = W
and the rate of weight loss is given by differentiation as
2γ(abh + abh + abh) = W

or with
a b
h = 2 = 2
denoting the corrosion rate, as
2h 2h
2γabh( + + 1) = W
a b
For thin plates, this expression yields
W
h =
2γab
o
Writing A = 2ab = total surface area of plate and W = Δw/Δt, where Δw denotes weight loss in
time Δt, we have
Δw
h =
γAΔt
The corrosion rate may vary with time under various circumstances, but for purposes of
calculation we may generally assume it to be constant. In examining test data, it is usual to
replace Δw and Δt by w and T where w denotes the total weight loss in time T, so that the
equation becomes
W
h =
γAT
These two equations are of course equivalent only if the rate of weight loss is constant.
6-6
Example. A steel plate of dimensions 3 x 6 x (1/8) inches is known to suffer a weight loss of
0.025 lb when exposed in seawater for 6 months. Determine the corrosion rate.
Solution: We have A = 2(3 x 6) = 36 in2, γ = 0.28 lb/in3, T = 0.5 years and W = 0.025 lbs.
Hence the equation gives
0.025
h = (0.28)(36)(0.5) = 0.005 inchesyear = 5 MPY
If, in place of the thin plate, we have a specimen of arbitrary shape, we may construct a weight-
loss formula similar to that for the flat plate. Consider, in particular, the effect of corrosion on
the cross section shown in the figure below.
During a time Δt, the surface of the solid will move inward an amount _hΔt. If S denotes the
lateral surface area of the solid, the volume loss will be _hΔtS and the weight loss Δw = γ_hΔtS,
where γ denotes the specific weight. Hence, solving for h we have
Δw
h =
γSΔt
If we take S to be the initial lateral surface area of the solid and write W = Δw, T = Δt, we have
W
h =
γST
Because the lateral surface area changes with time, this equation will be only valid only as long
as the lateral area change is small.
6-7
Example. A cylindrical steel test rod of 0.50 inches in diameter and 2 inches in length is exposed
in seawater for 3 years. The measured weight loss is 0.0165 lbs. Find the corrosion rate.
Solution:
We have γ = 0.28 lb/in3, T = 3 years

and
S = (π x .5 x 2) + (2 x π x 0.252) = 3.53 in2
0.0165
h = (0.28)(3.53)(3) = 0.0055 inchesyear = 5.5 MPY
Note: This is not a particularly good example of our assumption of the total area, S, staying
constant. If one recalls that corrosion occurs through the wetted surface area, it can be shown
that since 12% of the area is at the ends of the cylinder, 12% of the corrosion must have occured
there. The remaining 88% of the corrosion occured through the cylinder walls.
Strength Reduction Test. Instead of using the weight loss test to determine the corrosion rate, we
may use a strength- reduction test. Suppose we have a thin plate with a failure (breaking) force F
applied over area 2ha, as shown in the figure below:
We assume the failure stress Y is the same, regardless of the amount of corrosion so that
F
Y A
= =1
Yo Fo
Ao
where the subscript o denotes initial value. Solving for A/A o, we have
A F
Ao = Fo
now the area A is given by A = 2ah where (neglecting corrosion of the edges as is permissible
6-8
for thin plates)

h = ho-ht
hence we have
A 2ah h
= =1- t
Ao 2aho ho
so that _h may be determined as
F
(1- )h
Fo o ΔFho
h= =
T F oT
where T denotes the duration of exposure.
Example: A thin-plate of 1/8 inch thickness is exposed for 6 months in seawater. The initial
yield strength of the plate was 28,000 lbs. After exposure the strength is 25,000 lbs. Determine
the corrosion rate _h.
ΔFho
h =
FoT
(3000) (0.0625inches)
hdot = (28000) (0.5years) = 0.0134inchesyear = 13.4 MPY
Solution:
Suppose that a standard cylindrical tensile specimen is used in place of the flat-plate specimens
in the above scheme. In this case, we have as before
A F
Ao = Fo
But now,
A R hdott
= ( )2 = (1 - )2
Ao Ro Ro
so that
F
1-
Fo
h=( )Ro
T
6-9
Example: A standard 0.5 inch diameter tensile specimen is found to suffer a 15% loss in
strength when exposed in seawater for 2 years. Determine the rate of corrosion.
F Fo _ F F
Δ = =1- = 0.15
Fo Fo Fo
F
= 0.85
Fo
0.25
h = (1 - .85)( )
2
h = 0.0099inches/year = 9.9MPY
6.4 Corrosion Allowance Calculations
The following examples illustrate engineering calculations associated with corrosion

allowances.
Example: A submerged pipeline is corroding on the outside with a corrosion rate of 15 MPY
and on the inside with a rate of 5 MPY. Find the design wall thickness of the pipe if the design
life is 30 years.
Solution: The inside radius is denoted by r1, and its initial value by r1o. If the corrosion
o
rate is h 1, then
r1 = r1o + h1T
where T = life of the pipe. Similarly, with r2 denoting the outside radius and r2o its initial value.
If the corrosion rate is _h2, then
r2 = r2o - h2T
From a corrosion standpoint we require that r 1 = r2 at T = 30 years. Hence the initial wall
thickness required is
r2o - r1o = (h1 + h2)T
r2o - r1o = (.20)(30) = 0.6inches
From a true design standpoint, we would also need to account for the fact that this pipeline is
going be carrying a fluid at a design flow rate and that it is going to be exposed to both internal
and external pressures. At T = 30 years there will some required wall thickness to withstand this
pressure, p = absolute value of (pexternal - pinternal) and at T = 0, there will be a minimum r1 to allow
for the rated flow at the design pumping pressure. Since we know which material is going to be
used, we also know its shear strength τ. We can now solve for the required wall thickness and
subsequently the outside radius.
pr1
r2 - r1 = + (h1 + h2)T
σ
6 - 10
Example: A steel cable is to be used as an underwater guide wire in an off shore structure. In
the absence of corrosion, a section area of 1 in 2 is recommended. If tests show the rate of
corrosion at the site and for this material to be 25 MPY, what initial area of cable should be
used for a design life of 20 years.
Solution: With Ro denoting the initial cable radius and R its radius after T years, we have
where _h is the corrosion rate. Now πR2 = 1 and Ro = 0.564 inches.

0.564 = Ro - (0.025)(20)
or
Ro = 1.06 inches, and the required initial area is 3.53 in 2.
6 - 11
CHAPTER 7
MICROSTRUCTURE OF MATERIALS
7.1 Fundamentals
7.2 Bonding
7.3 Crystalline Structure
7.4 Density Calculation
7.5 Direction Indices and Miller Indices
7.6 Polymorphism
7.7 Coordination Number and Packing Factor
7.8 Solidification
7.1 Fundamentals
Materials consist of atoms or molecules (two or more atoms bonded together). In gases, the atoms
or molecules are in constant translational motion; in liquids, this motion is greatly reduced; and in
solids, there is essentially no translational motion.
A. Atomic Scale:
Arrangement of atoms
Fundamental processes (i.e. molecular bonding and forces, etc.)
B. Microscopic Scale:
Properties observable with a microscope (i.e. phases, grains, etc.)
C. Macroscopic Scale:
Typical scales of measurement (i.e. cm, inches, yield point, etc.)
Figure 1: Scales of Investigation of Materials
7-1
7.2 Bonding
In the solid state, the atoms are held together by ionic, covalent, or metallic bonding.
Ionic Bonding: Bonding force results from an electrical attraction between ions of opposite charge.
Most commonly occurs in materials with loosely held electrons or nearly filled outside shells.
Covalent Bonding: Bonding occurs when atoms share electrons, usually to complete the outside
shell of one or both of the atoms.
Metallic Bonding: Bonding occurs when a few electrons from an incomplete outer shell break
free, forming a "cloud" of free electrons which move about the metal structure. The positively
charged "core" maintains an attraction to the negatively charged "clouds," and other similar "cores"
are likewise attracted to the "cloud," thus creating a bond between the various atoms within the
metallic structure.
Material Symbol Type of Bond Bond Energy, eV (per atom)

Cesium Chloride CsCl Ionic 3.4
Sodium Chloride NaCl Ionic 4.0
Aluminum Oxide Al2O3 Ionic 31.4
Magnesium Oxide MgO Ionic 20.4
Silicon Si Covalent 4.7
Diamond C Covalent 7.4
Silicon Carbide SiC Covalent 6.4
Mercury Hg Metallic 0.7
Aluminum Al Metallic 3.4
Iron Fe Metallic 4.3
Tungsten W Metallic 8.8
7.3 Crystalline Structure
Solids formed by these bonds may be either crystalline or non-crystalline in nature. Crystalline
materials have a periodic repeating of the atoms or groups of atoms throughout the material. Non-
crystalline materials have a random arrangement of the atoms.
Crystalline solids possess symmetry of atomic arrangement. This symmetry involves a space
lattice and a basis.
Space Lattice: Involves the concept of points in space with equal surroundings. Consider the two-
dimensional representation of points shown in Figure 2. This represents a simple lattice, since
each interior point has identical surroundings. The lattice has associated with it a lattice cell which,
in this two dimensional case, is simply a square cell with all sides of equal length "a". Obviously
a lattice could be constructed with a rectangular cell shape.
7-2
(a) (b)
Figure 2: (a) Lattice and (b) Basis
Basis: Suppose we have two atoms, A and B, arranged as shown in figure 3. If we place atom A
of the set on each lattice point in Figure 2, we then have a crystalline structure. (A crystalline
structure has a set of atoms repeating itself periodically and having identical surroundings as
shown in Figure 3 ). The set of atoms that repeats itself, i.e. A-B, A-B, is known as the basis. The
basis may consist of a single atom or a complex group of atoms.
Figure 3: Crystal structure in terms of lattice and basis
The above discussions of the lattice and basis are for the illustrated two-dimensional case, but can
easily be expanded into the real world three-dimensional case. In three-dimensions, there are only
fourteen possible lattice cells that can exist. Some of the more common of these are illustrated in
figure 4.
7-3
Figure 4: Common Lattice Structures
For metals, the basis often consists of a single atom such that the corresponding crystalline
structure consists of the appropriate space lattice with atoms located at its lattice points. For
example, iron has a crystalline structure consisting of single atoms located at the points of the
body-centered cubic lattice. Aluminum has a structure consisting of single atoms located at the
points of a face-centered cubic structure.
An exception to this is the hexagonal close-packed crystalline structure of metals such a zinc and
titanium. Here the lattice is hexagonal and its basis consists of two atoms such that there is an
intermediate row of atoms between the top and bottom of the hexagonal cell
(Figure 5). This gives rise to the so-called ABAB stacking of atoms. Note that the lattice cells are
described by parameters “a” and “c”.
Figure 5: Hexagonal Close Packed Lattice
Hard-Ball Model. If we consider atoms to be hard spheres of definite radius in contact with other
atoms in the directions of closest approach, we can relate the atomic radius to lattice parameter.
For the face-centered cubic (FCC), the spheres are in contact along the face diagonals. For a body-
centered cubic (BCC) the spheres must be in contact along the cube diagonal.
7-4
√3𝑎
4𝑟 = √3𝑎 → 𝑟 =
4
Figure 6: Body Centered Cubic
√2𝑎
4𝑟 = √2𝑎 → 𝑟 =
4
Figure 7: Face Centered Cubic
Figure 8: Hexagonal Close Packed
7-5
7.4 Density Calculation
The density of the material may also be calculated from a knowledge of the lattice parameters and
the atomic weight (A.W. the weight in grams of Avogadro's Number (6.024 x 10 23) of atoms) of
the material. If n = number of atoms in a lattice cell, then
Or, using the weight of one atom,

A.W
n( )
6.024 x 1023
ρ=
Lattice volume of cell
For metals with cubic arrangement of atoms, the volume of the lattice cell is simply the cube of
the lattice parameter (defining the length of the sides of the cells. Noting that lattice cells repeat,
and can be considered in contact with one another, it can be seen that an atom at a corner of a cell
is actually shared with 7 other cells, so only 1/8 atom belongs to any one cell. The 8 corners of a
cell thus provide only one net atom to the cell.
Also, for a FCC cell, the atoms on each face are shared with an adjacent cell, so that the 6 atoms
on the faces of any cell provide a net of 3 atoms per cell. The corner and face atoms together thus
provide a net number of 4 atoms per cell.
For a BCC cell, the 8 corner atoms contribute the same, namely one atom, while the atom at the
center is not shared by any other cell, thus giving a net number of 2 atoms per cell.
Example: Aluminum (A.W. = 26.98 grams)has a FCC lattice cell, with a measured lattice
parameter (from x-ray diffraction studies) of a = 4.05 x 10 -8 cm. Determine its mass density.
26.98
4( )
6.024 x 1023
ρ=
4.05 x 10-8
gm lb
ρ = 2.70 = 0.10
cm3 inch3
7.5 Direction Indices and Miller Indices
Direction indices and Miller indices may be used to describe the geometry of a particular lattice
structure. Direction indices refer to a specific direction in a cubic unit crystal; whereas Miller
indices refer to planes.
7-6
Direction indices refer to a specific direction in the cubic unit crystal. By standard convention, the
notation is:
[ℎ, 𝑙, 𝑘] = ℎ𝑥 + 𝑙𝑦 + 𝑘𝑧̂
Where, ℎ, 𝑙, 𝑘 are vector components of the directions 𝑥, 𝑦, 𝑧̂ reduced to the smallest integer.
Negatives are denoted with an overbar, so ℎ = −ℎ
The method to find the direction indices is as follows:

1. Locate origin – at the tail of the vector
2. Find the vector projections in 𝑥, 𝑦, 𝑧̂
3. if not all integers, multiply by least common denominator
4. Simplify if needed
5. Place in brackets
Example: Find the direction indices for the following vector.
Solution
1. Origin at (0,0,0)
2. Projections
x y z
1 0 0
3. Integers – Yes!
4. Simplified- Yes!
5. Answer [1 0 0]
Example: Find the direction indices for the following vector.
7-7
Solution
1. Origin at (1,0,0)
2. Projections
x y z
-1/3 1/2 1
3. Integers – No, but multiply by 6 will do
x y z
-2 3 6
4. Simplified- Yes!
5. Answer [2 3 6]
Miller indices refer to a plane in the cubic unit cells.
Method:
1. Choose origin so plane does not pass through it
2. Find axis intercepts with plane in 𝑥 , 𝑦, 𝑧̂
3. Take the reciprocal of the intercepts
4. Form integers by multiplying by the least common denominator
5. Simplify by dividing by the least common denominator
6. Place in parentheses
Example: Find the Miller indices for the plane shown below
Solution:
1. Find Origin (0,0,0)
2. Projections
x y z
+1 infinity infinity
3. Reciprocals
+1 0 0
4. Integers – OK
5. Simplified- OK
6. (1 0 0) ANS
7-8
Example: Find the Miller indices for the plane shown below
Solution:
1. Origin (0,0,0)
2. Projections
x y z
1 1 infinity
3. Reciprocals
1 1 0
4. Integers -OK
5. Simplified – OK
6. (1 1 0)
7.6 Polymorphism.
The ability of a material to have more than one crystalline structure is known as polymorphism.
The transition from one structure to another at a specific temperature and pressure is known as a
polymorphic change.
A common and important example is iron, which has a BCC cell structure (known as
α-Fe) for temperatures up to 910o C. Above this temperature, the structure is FCC (known as
γ-Fe) until a temperature of 1400o C is reached. It then reverts back to an BCC structure
(known as δ-Fe).
Example: Calculate the change in density of an iron specimen as it changes from BCC to FCC at
910o C. The lattice parameter of iron at 20o C is 2.87 Å and the coefficient of thermal expansion
is 15 x 10-6/o C. The lattice parameter of iron at 910o C is 3.65 Å.
Solution:
Mass = (55.85/6.024 x 1023) = 9.273 x 10-23 gm

n=2
BCC lattice parameter at 910o C,
a = 2.87 + 2.87(15 x 10-6)(910-20) = 2.91 Å

7-9
7 - 10
(2)(9.273 x 10-23)
ρBCC =
(2.91 x 10-8)3
gm
ρBCC = 7.53
cm3
(4)(9.273 x 10-23)
ρFCC =
(3.65 x 10-8)
gm
ρFCC = 7.63
cm3
hence
gm
ρFCC - ρBCC = 0.10
cm3
ρFCC - ρBCC
= 1.33%
ρBCC
Example: Using the results above, calculate the change in volume of the iron specimen as it
changes from BCC to FCC at 910o C
Solution: Let M = total mass of the specimen and V be its volume. The density ρ is given
ρ = M/V so ρ/V = M (constant)
ρBCC / ρFCC = VFCC / VBCC = 7.53/7.63 = 0.987
(VFCC - VBCC) / VBCC = (1 - 0.987)/(1/0.987) = - 0.0133 = -1.33%
7.7 Coordination Number and Packing Factor
Additional terms of interest include the coordination number and the packing factor. The
coordination number is the number of nearest neighbors of an individual atom in a lattice structure.
An FCC lattice has a coordination number of 12, while a BCC structure has a coordination number
of 8. The packing factor (p.f.) is the ratio of the volume of atoms (using the hard ball model) to
the total volume of the lattice cell structure.
p.f. = Volume of atoms/volume of lattice cell
For a cubic structure with hard-ball radius r
p.f. = n(4/3)(πr3)/a3
where a = lattice parameter.
7 - 11
Example 1: Find the packing factor for aluminum.
Solution: Aluminum is an FCC structure. As shown previously, the lattice parameter was related
to the radius r by
( 2)
r= a
4
n=4
thus
4 2
(4)( )(π)( )
3 4
p.f. =
a3
p.f. = 0.74
Example 2: Find the packing factor for NaCl.
Solution: NaCl forms an FCC lattice, but now the basis consists of a combination of Na and Cl
atoms. As a result the volumes in the equation are different than those in the previous example.
The volume of the atoms VA is now
VA = (4/3)π(rNa3 + rCl3) = (4/3)π(.953 + 1.813)
The volume of the lattice, VL, has also changed and is now
VL = (2rNa + 2rCl)3 = (2(0.95) + 2(1.81))3
Thus
p.f. = 4(VA)/(VL) = 0.67
The packing factor is independent of the lattice parameter and the atomic radius provided that only
7 - 12
one atom as present at a lattice site (as is the case for most commom metals of interest in this
course). Similar calculations for a BCC structure show p.f. = 0.68, HCP = 0.74, and 0.52 for a
simple cubic.
7.8 Solidification
When cooling, metal often solidifies as shown in Figure 9. The stages of solidification include
a. Crystal Nucleation
b. and c. Crystal Growth
d. Solid Metal with Grain boundaries
Figure 9: Metal Solidification

Polycrystalline Structures
Because of the solidification process shown above, metals generally consist of regions called
grains, where the crystalline arrangement of atoms exists. These regions differ from one another
only in that the spatial orientation (see figure 10); that is, the orientation of the unit cells describing
the crystalline symetry differ from one grain to another. These grains typically have dimensions
on the order of 10-2 inches or less. They can easily be seen with light microscopes, provided the
specimen is smooth enough, since light is reflected differently from the various grains.
Photographs showing this "microstructure" are known as photo-micrographs.
In many cases, the grains are randomly oriented relative to one another, so that a specimen with
characteristic dimensions of 1 inch will contain many grains. The material will appear isotropic,
(properties the same in all directions) even though the grains are anisotropic (different properties
in different directions).
Single Crystal
It is possible to control the formation of crystals such that a piece of material has a single crystal
structure, rather than the grain boundaries. This is used for solid state devices, and turbine blades.
7 - 13
Figure 10: Polycrystalline Structures
7 - 14
CHAPTER 8
METALS AND ALLOYS
8.1 Types of Alloying

8.2 Hume-Rothery Rules for Alloying
8.3 Material Systems
8.5 Two component material systems
8.6 Derivation of Lever Rule
8.7 Metal Processing
8.8 Work Hardening and Annealing
8.9 Precipitation Hardening of Alloys
8.10 Iron-Carbon Systems
8.11 Eutectoid Point for Iron-Carbon Systems
8.12 Continuous-Cooling Transformation Diagram for Iron-Carbon Systems
8.13 Heat Treatment of Steel
8.14 Specifics of Quenching Process for Iron-Carbon Systems
8.15 Case Hardening of Steel
Alloys consist of a combination of two or more metal elements in the solid state. In engineering
work, alloys, rather than pure metals, are generally used because of their superior properties.
Ordinary steels, for example, are alloys of iron and carbon and are preferred to pure iron because
of strength and other considerations. Important mechanical properties of alloys are related to the
solid phases, or homogenous parts, that form from the combination of elements. Such phases are
observable on the microscopic scale, and their study is important in understanding alloy properties.
8.1 Types of Alloying
Interstitial is where the alloy elements are located in spaces between atoms in the unit cell. For
instance, when carbon (Atomic radius 0.129 nm) is added to iron (0.075nm), the carbon fits into
the gap between the Fe atoms (interstitial site). The solubility depends on the size of these gaps
and the crystal structure.
Substitutional is where the alloying elements are located in vacancies in the unit cell. An atom of
one element substitutes for another in the crystal structure.
8-1
Fig. 1
8.2 Hume-Rothery Rules for Alloying
It is not possible for every combination of materials to form an alloy. The following rules are for
maximum solubility in substitutional alloying:
1. Same Crystal Structure (BCC, FCC, HCP for instance)

2. Same Valency Factor
3. Similar Electronegativity – the tendancy for the atom to attract a shared pair of electrons.
This depends on the atomic number and the distance at which the valence electrons are
from the nucleus.
4. Difference in atomic radius <= 15%
In interstitial alloying, the solute atoms should be <= 59% of the atomic radius of the solvent
atoms.
8.3 Material Systems

A material system refers to a definite amount of material or materials.
The components in a system refer to the smallest number or individual substances that must be
listed to describe the chemical composition of the system. These may be elements (Pb, Sn, Fe, C,
etc.) or compounds (H2O, NaCl, etc.)
The phases in a system denote the homogenous parts of the system that are made up of its
components.
Phase diagrams are graphs showing the phases present under selected conditions of temperature,
pressure, and composition when the system is in thermodynamic equilibrium.
A material system may be specified as in the following example:
8-2
Example: Suppose we have 8 gms of copper and 2 gms of nickel. This yields a system having 80
w/o copper and 20 w/o nickel (w/o= weight percent). This can also be expressed as mole percent
(m/o) or atomic percent (a/o).
8 gm (Cu): (8 gm)/(63.54 gm/mole) = 0.1259 moles

2 gm (Ni): (2 gm)/(58.71 gm/mole) = 0.0341 moles
Since the number of atoms in a mole is constant, the atomic percent is:
8 gm (Cu): 0.1259/(0.1259 + 0.0341) = 78.7 m/o = 78.7 a/o

2 gm (Ni): 0.0341/(0.1259 + 0.0341) = 21.3 m/o = 21.3 a/o
The phases possible in a one component system are limited to liquid, solid, and gas states (Fig 2).
A phase diagram consists simply of a graph of pressure vs. temperature, with the phase regions
indicated.
Fig. 2 Phase diagram for a one-component system (water)
8.5 Two component material systems
Materials with two components are called binary systems. They are commonly encountered in
engineering materials (brass consisting of copper and zinc, carbon steel consisting of iron and
carbon, etc.) The phase diagram of a binary material is customarily displayed in a temperature vs.
composition format, with the pressure held fixed at atmospheric.
In the phase diagrams,

L denotes liquid
α, β, γ are various solids
α+ β or L+ γ indicate states where both are present
8-3
Two-Components with Complete Solubility: For a system having components completely soluble
in the solid state, the phase diagram is especially simple, since only one phase exists in the solid
state. See Figure 3 for the phase diagram of the material.
Two-Components with Partial Solubility (Eutectic): For a binary system having components only
partially soluble in the solid state, two solid phases will exist, and the phase diagram is much more
complex than Figure 3. A special case is an eutectic diagram shown in Figure 4. “Eutectic” means
that when the two components are mixed, the melting point is lower than the melting temperatures
of the two components separately. At point d, the eutectic reaction Liquid, L  α + β occurs.
Two-Components with Partial Solubility (Eutectoid): When the liquid region is replaced by a third
solid phase γ, we have the eutectoid reaction γ  α + β and associated phase diagram. Figure 5
shows such as phase diagram with both a Eutectic and Eutectoid point.
Two-Components with Partial Solubility (Peritectic): For a binary system having components only
partially soluble in the solid state, we may also have a phase diagram having a peritectic point P,
where the reaction L + β  α occurs (Figure 6)
Two-Components with Partial Solubility (Peritectoid): If the liquid in the peritectic reaction is
replaced by a third solid phase, γ, we have a peritectoid reaction α + γ  β. This is illustrated by
Figure 7.
Figure 3: Components with Complete Solubility (Copper-Nickel System)

Melting temperature depends on percentage of nickel. There is a solid phase, then a mixture of
liquid and solid (L  α) and then a complete liquid phase L
8-4
Figure 4: Two Components with Partial Solubility (Silver Alloy Copper System)
At the Eutectic point (d) the liquid changes to a solid L  α + β
(This is at a lower temperature than when the separate components become solid)
Figure 5: Case 2 - Two Components with Partial Solubility (Iron Carbon System)
At Eutectic point, the liquid changes to a solid L  α + β
At the Eutectiod point γ  α + β
(This is at a lower temperature than when the change occurs for the individual components)
8-5
Figure 6: Two Components with Partial Solubility - Peritectic

At the Peritectic point, the reaction L + β  α occurs
Figure 7: Two components with Partial Solubility- Peritectoid Reaction -

peritectoid reaction α + γ  β
8-6
8.6 Derivation of Lever Rule
Consider a temperature-composition point in a phase diagram where two phases exist, as shown
in Figure 8 for temperature T1 and composition X2. The chemical composition of the two phases
are:
α - phase: X1 w/o B, 100 - X1 w/o A:
β - phase: X3 w/o B, 100 - X3 w/o A.
Figure 8: Phase Diagram for Derivation of Lever Rule
This may be seen by imagining that we start with pure A material and add increasing amounts of
B. When we reach the composition X1 w/o B, the α phase has become saturated and can accept
no additional amount of B. As we add still more B, the second β phase must appear. The β phase
will be saturated, and its composition will be X3 w/o B, as read from the right side of the diagram.
Now we would like to determine how much α phase and β phase material will exist at a temperature
T1, and the composition X2 w/o B. To arrive at this, we use the lever rule.
Let A1 be the amount of α phase present

A3 be the amount of β phase present
A2 be the total amount of material
8-7
The balance of material requires that
A1 + A3 = A2
And the balance of the B material requires that
X1A1 + X3A3 = X2A2
Combining these two equations yields
A1 (X3 - X2)
=
A2 (X3 - X1)
which is the fractional weight of the α phase, and
A3 (X2 _ X1)
=
A2 (X3 _ X1)
which is the fractional weight of the β phase.
Example: A Cu-Ni alloy (Fig 9) contains 73% cu and 27% Ni at 1200C. Find (1) Weight % Cu
in solid and liquid phases and (2) % of alloy in solid % liquid phases.
8-8
Solution:
Step 1: Draw horizontal tie line at 1200C
Step 2: Project down for Cu
Liquid, 𝑊 = 80%Cu
Solid, 𝑊 = 70%Cu
Step 3: For 𝑊 = 27% Nickel, project up to tie line
𝑊 −𝑊 30% 𝑁𝑖 − 27%𝑁𝑖 3
𝑋 = = = = 0.3
𝑊 −𝑊 30%𝑁𝑖 − 20%𝑁𝑖 10
Is the fraction of alloy liquid.
𝑊 −𝑊 27% 𝑁𝑖 − 20%𝑁𝑖 7
𝑋 = = = = 0.7
𝑊 −𝑊 30%𝑁𝑖 − 20%𝑁𝑖 10
Is fraction of the alloy that is solid
𝑋 + 𝑋 = 1 OK!
8-9
Example: Consider the lead-tin (Pb-Sn) diagram shown in Figure 20. For an alloy consisting of
40 w/o Sn determine the composition and amount of the α and β phases present at 100 o C. Also,
determine the relative amount of material formed from the eutectic reaction.
Figure 20: Lead-Tin Phase Diagram
Solution: At T1 (100o C), the composition of the α phase is

5 w/o Sn, 95 w/o Pb
The amount of the α phase present is

(A1/A2) = (99 - 40)/(99 - 5) = 62.8% by weight
The composition of the β phase is

99 w/o Sn, 1 w/o Pb
The amount of the β phase present is

(A3/A2) = (40 - 5)/(99 - 5) = 37.2 % by weight
To determine the amount of material that experienced the eutectic reaction, we note that just above
the eutectic temperature of 183o C, the remaining liquid had the eutectic composition 62 w/o Sn
and 38 w/o Pb. The amount of remaining liquid was (40 - 19)/(62 - 19) = 48.8%. On cooling
below the eutectic temperature, this liquid experienced the eutectic reaction l  α + β. The
amount of eutectic material present at 100o C is 48.8 % (by weight). The amount of α and β phase
present in the eutectic material at 100o C is (99 - 62)/(99 - 5) = 39.4 % α, 60.6 % β.
8 - 10
Example : Shown in Figure 21 is the phase diagram for the H 2O - NaCl system. Knowing that
seawater is 3.5 w/o NaCl, determine:
(a) the amount of NaCl is 100 lbs of seawater,
(b) the relative amount of ice and brine at 10o F, and
(c) the composition of the ice and brine at 10o F.
Figure 21: Phase Diagram for Seawater
Solution: (a) the amount of NaCl in 100 lbs of seawater is

(100)(.035) = 3.5 lbs
(b) the amount of ice and brine at 10o F is

(16.5 - 3.5)/(16.5) = 78.8% ice
3.5/16.5 = 21.2% brine
(c) the compositions are

0% NaCl in ice and 16.5 % NaCl in the brine.
8 - 11
8.7 Metal Processing
The manufacturing processes can alter crystal microstructure and thus influence the strength,
ductility, and impact toughness of a material. While a variety of finishing processes are suited to
specific applications, we will focus on casting, hot and cold rolling, extrusion, forging, and
drawing.
Casting:
Molten metals can be cast into large ingots in a direct chill casting unit. These large ingots are
typically rolled into sheets or plates, or else extruded into structural shapes like I-Beams. Metals
can also be cast into smaller, complex shapes using metal or sand molds.
Fig 10 Casting
Hot and Cold Rolling:

To make large sheets or plates, large ingots are often hot-rolled initially (to allow for a larger
thickness change). A schematic is shown below. The hot rolling process continues until the plate
reaches its desired thickness or until it cools such that rolling becomes impossible. Plates will be
reheated and rolled until they reach the desired thickness.
Fig. 11 Hot Rolling
After hot rolling, plates are usually cold rolled. Cold rolling hardens and strengthens the metal
by introducing internal stresses (dislocations) in the material microstructure. To re-soften the
material, remove internal stresses, and improve toughness, the material may be reheated in a
process called annealing.
The amount of cold working is often expressed as a percentage related to the change in a material’s
thickness, i.e.:
8 - 12
%𝐶𝑊 = × 100%
Where %CW indicates the percent cold work done on a material, tho and thf indicate the initial and
final thickness of the material, respectively.
Extrusion:
Extrusion is the forming of a material through plastic deformation by forcing it though a die under
high pressure. Cylindrical rods and hollow tubes of most metals are fabricated this way. This
process is often used to form a wide variety of cross sections for aluminum, copper, and their
alloys.
Fig 12 Extrusion
Forging:
Forging is the process of a material being pressed or formed into its desired shape, which can be
irregular (unlike rolling or extruding). The temperature of forging is important for the final
strength, hardness, and ductility of the material. Typically forged metals are tougher and more
durable than cast metals. Forging involves hot and cold working by nature, and can act to reduced
voids in the metal’s microstructure.
Fig 13 Forging
Drawing:
Drawing is a cold working process typically used to make wire, in which a metal rod (such as
copper) is pulled through a carbide nib. Deep drawing is a form of drawing in which a punch and
die are used to form cup-shaped objects from sheet metal. Aluminum cans are formed by deep
drawing.
Drawing is a form of cold work, and the percentage of cold work can be expressed in a 2-D analogy
to the equation for rolling:
, ,
%𝐶𝑊 = × 100%
,
8 - 13
Where Axc,o and Axc,f indicate the initial and final cross-sectional areas of the metal rod,
respectively.
Fig 15 Drawing
8.8 Work Hardening and Annealing
When a material is plastically strained the yield stress is increased. In many of the manufacturing
techniques listed earlier, the material is plastically deformed in order to fabricate the desired shape.
As Figure 16 shows, although this may significantly increase the yield strength of the material, it
also makes it more brittle. This change is a result of the dislocation density, or imperfections in
the lattice, increasing as a result of the deformation.
Figure 16: Work Hardening
If we take a work-hardened material and subject it to a sufficiently high temperature for a specified
time, we can reduce the dislocation density and the yield stress will return to its initial value. This
is known as annealing the material, which is a form of heat treatment. Due to other aspects, such
as recrystallization, subjecting the work-hardened material to a high temperature for too long a
8 - 14
time, may result in a material whose yield stress is less than it was before the material was work-
hardened.
8.9 Precipitation Hardening of Alloys
Consider the phase diagram below in Figure 17. Suppose we have material held at temperature T 2
and then suddenly quench it to room temperature, T r. There is insufficient time for the β phase to
form and we therefore have an unstable α phase existing at room temperature. Because it is
thermodynamically unstable, the β phase would eventually form at room temperature, although it
would take many years. If we now heat the material to a temperature T 1, the β will begin to form.
As the β phase begins to form, it transforms the unstable α phase into a stable α phase forming a
matrix surrounding numerous extremely small β grains. After some time, the maximum hardness
is reached and the material can again be quenched to room temperature, thus yielding a
precipitation hardened material (Figure 18).
If the material is held at the treatment temperature for too long, the material will begin to soften.
The most important reason for this is that the numerous small β grains which produced the
maximum hardness continue to grow. Some of the β grains grow into larger β, obviously
decreasing the number of total β grains. Other factors including the strain-energy present and grain
boundary depletion are also at work here.
Figure 17: Phase Diagram of Mg-Sn System
8 - 15
Figure 18: Alloy Hardness vs. Precipitation Time at Elevated Temperatures
8.10 Iron-Carbon Systems
Alloys of iron and carbon for the system for common steels and cast irons. Generally, steels have
a carbon content between 0 and 2 % by weight, while cast irons have a carbon content between 2
and 5 % by weight. The iron-carbon phase diagram is shown below in Figure 19.
The phases are defined as follows:

(a) Ferrite α phase: BCC crystalline phase
(b) Austenite γ: FCC crystalline phase
(c) δ phase: high temperature BCC crystalline phase
(d) Cementite or carbide Fe3C: mixture
(e) Pearlite: layered two phase mixture generated by transforming austenite to ferrite and
carbide by the eutectoid reaction γ  α + Fe3C
8 - 16
(f) Martensite: a non-equilibrium phase in steels formed by rapid cooling of austenite. It

arises because, with extremely rapid cooling, there is insufficient time for the carbon atoms
to realign themselves from their locations within the FCC structure of austenite to their
locations in the BCC structure of ferrite. As a result, the structure is "trapped" between
FCC and BCC, and actually results in a body centered tetragonal lattice. Martensite is very
hard and brittle, and by itself, has very few practical applications. Its value lies in
combinations of ferrite and martensite obtainable by various heat treatments.
(g) Bainite: mixture of ferrite and cementite formed by isothermal transformation of
austenite under selected conditions.
Figure 19: Iron-Carbon Diagram
8 - 17
Martensite
Coarse Pearlite Fine Pearlite
Upper Bainite Lower Bainite

Fig 20
8.11 Eutectoid Point for Iron-Carbon Systems
The eutectic and eutectoid points on the phase diagram are important for heat treatment. The
eutectoid point for Iron-Carbon is at a temperature of 727 C and 0.77% carbon by weight. See the
figure below (fig 21) for an enlarged view of this region of the phase diagram.
8 - 18
 Eutectoid steel has exactly 0.77% carbon. When cooling below the eutectoid temperature
(727 C) it becomes 100% Pearlite. The resulting steel is high strength and wear resistant,
used for things like music wire or railway track.
 Hyper-Eutectoid has more than 0.77% carbon content. When cooling below the eutectoid
temperature, it forms a combination of cementite and pearlite. This is more brittle.
 Hypo-Eutectoid has less than 0.77% carbon content. When cooled below the eutectoid
temperature, it forms a combination of pearlite and ferrite. This is more ductile.
The lever rule, discussed earlier, can be used to determine the proportions.
Fig 21
8 - 19
8.12 Time-Temperature-Transformation (TTT) Diagrams for Iron-Carbon Systems
Suppose we rapidly cool steel of eutectoid composition (0.8 % C) from the austenitic range down
to some temperature where α and Fe3C are the stable phases. If we could observe the resulting
transformation process, we would find that, initially, the material is 100% unstable austenite. After
some time, the ferrite and cementite would begin to form (either as pearlite or bainite) and after
sufficient time, the transformation of the austenite would be complete. This process is represented
by the so-called Time-Temperature- Transformation (TTT) diagram shown in Figure 22.
Notice that, if we suddenly cool the austenite to a temperature M s (martensite start) or less, some
martensite will form. The remaining austenite will then transform to bainite. If we suddenly cool
the material to temperature Mf (martensite finish), we will have 100% martensite and no further
transformation will occur.
Figure 22: TTT Diagram for Steel
8 - 20
8.12 Continuous-Cooling Transformation Diagram for Iron-Carbon Systems
The TTT diagram refers strictly to the cases where the steel is suddenly cooled from the austenite
(γ) range down to some new temperature. In most instances, the cooling is not instantaneous, but
rather occurs at a finite rate. As an approximation, the TTT diagram may also be used in this case,
but a more accurate Continuous-Cooling Transformation (CCT) Diagram should be used when
available. This diagram is shown in Figure 23 for a eutectoid steel.
Figure 23: Continuous Cooling Transformation for Steel
If we suddenly cool the material at such a rapid rate, that when plotted on the TTT diagram the
time vs temperature line does not intersect the γ + α + carbide phase (also known as "missing the
knee" but instead goes straight from the γstable through the γunstable region to a temperature Mf
(martensite finish), we will have martensite with no additional phase. When the cooling rate is
such that both the beginning and end curves are reached before reaching the martensitic zone, no
(or very little) martensite will form.
8 - 21
8.13 Heat Treatment of Steel
By varying the amount of martensite present in the material, the mechanical properties of the steel
can be varied. Martensite alone is very hard and strong but also is extremely brittle. A compromise
between strength and ductility can be reached by varying the amount of martensite present in
equilibrium with ferrite and carbide.
Fig. 24 Heat Treatment of Eutectic Steel
The procedure for forming this mixed material is called "heat treatment." First, the material is
rapidly cooled (quenched) so as to form nearly 100% martensite. At room temperature, this non-
equilibrium phase is, for all practical purposes, stable and will not further transform to ferrite and
carbide. If we then take the 100% martensite and heat it to an elevated temperature (say 700 o C)
and hold it there for some length of time (tempering), the martensite can be softened (stress
relieved) and decompose. The changes that occur are temperature dependent and range from
merely a reduction in strain-energy and dislocation density to microstructural changes within the
martensite itself. The end result is a material that retains some of the high strength and hardness
characteristics of pure martensite while restoring some ductility.
As shown in Figure 25, the hardness obtainable with martensite is dependent on the carbon content.
Since it is the "trapping" of the carbon atoms which forms martensite, and thus hardens the metal,
pure iron can not be hardened by quenching
8 - 22
Figure 25: Hardness
8.14 Specifics of Quenching Process for Iron-Carbon Systems
The rate of cooling affects the resulting grain structure and thus the hardness, ductility, and strength
of the steel specimen. Quenching is typically achieved in one of the following mediums. From
fastest to slowest, steel can be quenched in brine (fastest), water (second fastest), air, or molten
salt (slowest). We often use Time-Temperature-Transformation curves (TTT Curves), described
earlier, to predict a steel’s properties after heat treatment. The following discussion uses figure
26, below.
The first step in any heat treatment process is to heat the material above 727.⁰C. This step is called
austenitizing.
The dark blue curved lines on the diagram indicate the start and end of the transformation from
austenite to pearlite. Each of the lines labeled (a)—(g) indicate idealized cooling curves (remember
that the rate of cooling affects the transformation rate):
Curve (a) shows a steel sample rapidly quenched in brine or water. The idealized cooling
curve never approaches the transformation to pearlite, and therefore all austenite (ϒ) is
converted to Martensite. Martensite is an extremely hard and strong crystal structure with
8 - 23
a very low ductility. It is rarely used structurally without tempering. The temperature at
which austenite will transform to martensite is around 250⁰C
Curve (b) shows a steel that is heated to 727 ⁰C, hot quenched (in molten salt) to ~695 ⁰C,
and held at that temperature for over 2 hours to form 100% coarse pearlite. 100% coarse
pearlite is very ductile as hot quenching removes strain-induced stresses, but it is also soft
and weak.
Curve (c) shows a steel that is heated to 727 ⁰C, hot quenched to 610 ⁰C, and held at that
temperature for over 3 minutes to form 100% fine pearlite. 100% fine pearlite has a
uniform structure and higher strength than coarse pearlite. The cutoff between coarse and
fine pearlite varies but is around 650 ⁰C.
Curve (d) shows a steel that is heated to 727 ⁰C, hot quenched to 580 ⁰C, held for ~5
seconds then water quenched to form a structure of 50% fine pearlite and 50%
martensite.
Curve (e) shows a steel that is heated to 727 ⁰C, quenched to ~475⁰C, and held for over 3
minutes to form upper bainite. Bainite is an intermediate structure between pearlite and
martensite. It forms between 250 - 550 ⁰C. Upper Bainite is distinguished by its coarser
cementite particles. IT forms between 330 – 550 ⁰C.
Curve (f) shows a steel that is heated to 727 ⁰C, quenched to ~295⁰C, and held for over 2
hours to form lower bainite. Lower bainite is characterized by its finer cementite
particles.
Curve (g) shows a steel that is heated to 727 ⁰C, cooled to ~300 ⁰C, and held for 30 minutes
then rapid quenched (in water) to form 50% lower bainite and 50% martensite.
8 - 24
Fig 26: TTT Diagram of Eutectoid Steel
8.15 Case Hardening of Steel
Case Hardening is when metal is placed inside a carbon-rich environment. It is heated to very
high temperatures to introduce carbon to the outside of the structure. This results in a hard outer
surface, but a tough and strong inner surface. The process involves diffusion, following Fick’s
Second Law.
Fick’s second law tells us that when case hardening steel, the carbon concentration in air at the
surface, Cs, the carbon concentration at some distance x into the surface, Cx, and the initial carbon
concentration of the steel, Co, can be related according to:
𝐶 −𝐶 𝑥
= 𝑒𝑟𝑓 ( )
𝐶 −𝐶 2√𝐷𝑡
where D is the diffusivity constant in m2/s and t is time.
A plot of the error function (erf) is shown in the Figure 27 below. Typically, this is tabulated for
increasing values of z as shown in the following table.
8 - 25
Figure 27: erf(x) vs. x
Tabulated values for erf(z)
8 - 26
CHAPTER 9
STRESS
9.1 General Stress State and Principal Stresses

9.2 Determination of Principal Stresses (Two Dimensional) and Mohr’s Circle
9.3 Determination of Principal Stresses (Three dimensional)
9.4 Maximum Shear Stress
9.1 General Stress State and Principal Stresses
The mechanical properties of materials are best described in terms of the state of stress which the
material experiences. For example, the "break-down" in the elastic behavior of a bar under simple
tension loading is described by the condition that the break-down will occur when the stress in the
bar equals the yield stress of the material. For general loading, rather than simple tension, a similar
but more complex law applies involving the magnitudes of the various stresses existing.
To describe mechanical properties, we thus first need to consider the general concept of stress at a
point in a material.
A general stress state of a point in a solid consist of three normal stresses σ x, σy, σz and six shearing
stresses τxy, τyx, τxz, τzx, τyz, and τzy as shown in figure 1.
Figure 1: General Stress State
Each of the stresses (or stress components) represents a force per unit area acting on the small cube
of material. By equilibrium, it is easily seen that τ xy = τyx, τxz = τzx, τyz = τzy
(if τxy =/ τyx etc, the block would try to rotate) so that three normal and three shearing stresses are
needed to describe a general stress state.
9-1
Principal Stresses: Planes on which only normal stresses act are referred to as principal planes and
the normal stresses as principal stresses. For any stress state, we may always find principal planes
and principal stresses, as described below.
9.2 Determination of Principal Stresses (Two Dimensional) and Mohr’s Circle
First, consider the two-dimensional plane stress case. At any point the stresses will be as shown
in figure 2.
Figure 2: 2-D Principal Stresses
For principal stresses, we wish to find surfaces on which only normal stresses act. Considering
the element of figure 3, which is part of figure 2, let σ be the principal stress acting on the plane
δs. Equilibrium requires that:
σnx = σxnx + τxyny

σny = τxynx + σyny
where nx = cosθ and ny = sinθ
rewriting the equations as
(σx - σ)nx + τxyny = 0

τxynx + (σy - σ)ny = 0
it is seen that for a non-trivial solution for nx, ny, we must require that
9-2
The directions of the normals to the planes having these principal stresses can be found since tanθ
= ny/nx.
σ - σx
tanθ =
τxy
Example. Consider the two-dimensional stress state shown below in figure 4 and determine the
principal stresses and the planes on which the act.
Figure 4
Solution. From eq. (3), we find that σ1 = 13 ksi, σ2 = 3 ksi. Using eq. (4), we then find that for σ1
= 13, tan θ1 = (13-11)/4 = 0.5 and θ1 = 26.6o. Similarly, for σ2 = 3, tan θ2 = -2 and θ2 = -63.4o.
These planes are shown in figure 5.
9-3
Fig. 5
Notice that the planes are perpendicular to one another. This can be shown to be true in general,
and we thus see that if we had chosen the element with the orientation of figure 5 initially, the
shear stresses would have been zero and only normal principal stresses would be involved.
Example: Consider the previous example using Mohr’s circle
Solution: The above equations are easily seen to yield the Mohr's circle method for principal stress
determination. Taking horizontal and vertical axes with normal stress plotted along the horizontal
and a shear stress along the vertical, we have (adopting the convention that normal stress is plotted
positive on the σ-axes and shear stress, tending to rotate the element clockwise, is plotted on the
positive (τ axes) on the sketch shown in figure 6.
σx + σ y
c= 2
2
r= (σx - c)2 + τ
xy
τxy
tan2θ1 = σ - c
x
σ1 = c + r , σ 2 = c - r
Figure 6: Mohr’s Circle
9-4
9.3 Determination of Principal Stresses (Three dimensional)
To find the principal stresses for the general three-dimensional case, we define direction cosines
nx, ny, and nz of the normal n to the principal plane as shown in figure 7.
nx = cos θx
ny = cos θy
nz = cos θz
Figure 7
The equilibrium equations are
(σx - σ)nx + τxyny + τxznz = 0

τxynx + (σy - σ)ny + τyznz = 0
τxznx + τyzny + (σz - σ)nz = 0
For a non-trivial solution for nx, ny, and nz, the determinate of coefficients must vanish. Thus
(σx-σ) τxy τxz 

 τxy (σy-σ) τyz  = 0
 τxz τyz (σz-σ,)
Expanding this determinant yields a cubic equation whose roots are the principal stresses.
9-5
Where
The roots of the equation, usually determined numerically, are the principal stresses. To find the
direction cosines, nx, ny, and nz of the plane in which, say σ1 acts, we may use the above equations
with σ = σ1, together with the geometric condition
nx2 + ny2 + nz2 = 1

Similar procedures apply for σ2 and σ3.
9.4 Maximum Shear Stress
It is not difficult to show, using methods of analysis similar to those applied above, that the
maximum shear stress existing at a point is given by the greatest of the quantities
σ1 - σ2 σ2 - σ3 σ3 - σ1

 , , 
 2   2   2 
if we label the principal stresses such that σ1 > σ2 > σ3, then
θ1 - θ3
τmax =  
 2 
This greatest shear stress acts on a plane perpendicular to the principal plane associated with σ 2
and at 45o to those associated with σ1 and σ3.
9-6
Example. Consider the stress state defined by
σx = 2 ksi, σy = σz = 0
τxy = τyz = 0, τxz = 1 ksi
Solution. We find the three principal stresses as
σ1 = 2.414 ksi, σ2 = 0, σ3 = -0.414 ksi
To find the directions associated with each, use the first two equilibrium equations and the relation
nx2 + ny2 + nz2 = 1.
For σ = σ1
-0.414 nx + nz = 0
-2.414 ny = 0
nx2 + ny2 + nz2 = 1
thus we find
nx = ± 0.9237
ny = 0
nz = ± 0.3826
and
θ1x = 22.5o or 157.5o
θ1y = 90o
θ1z = 67.5o or 112.5o
Similarly, for σ2 we find nx = nz = 0, ny = ± 1, so that
θ2x = 90o
θ2y = 0o or 180o
θ2z = 90o
And for σ3 we find nx = ± 0.3826, nz = ± 0.9237
θ3x = 112.5o or 67.5o
θ3y = 90o
θ3z = 22.5o or 157.5o
9-7
CHAPTER 10
DEFORMATION

10.2 Material Characteristics
10.3 Elastic-Plastic Response of Metals
10.4 True stress and strain measures
10.5 Yielding of a Ductile Metal under a General Stress State - Mises Yield Condition.
10.6 Maximum shear stress condition
10.7 Creep
Consider the bar in figure 1 subjected to a simple tension loading F.
Figure 1: Bar in Tension
Engineering Stress () is the quotient of load (F) and area (A). The units of stress are normally
pounds per square inch (psi).
 = F
A
where:  is the stress (psi)

F is the force that is loading the object (lb)
A is the cross sectional area of the object (in2)
When stress is applied to a material, the material will deform. Elongation is defined as the
difference between loaded and unloaded length
∆𝑙 = L - Lo
where: ∆𝑙 is the elongation (ft)

L is the loaded length of the cable (ft)
Lo is the unloaded (original) length of the cable (ft)
10-1
Strain is the concept used to compare the elongation of a material to its original, undeformed
length. Strain () is the quotient of elongation (e) and original length (L 0). Engineering Strain has
no units but is often given the units of in/in or ft/ft.
∆𝑙
𝜀=
𝐿
where:  is the strain in the cable (ft/ft)

∆𝑙 is the elongation (ft)
Lo is the unloaded (original) length of the cable (ft)
Example Find the strain in a 75 foot cable experiencing an elongation of one inch.
Solution:
e( ft ) 1 in (1 ft / 12in )
Strain ( )    1.11x10 3 ft / ft
Lo ( ft ) 75 ft
Stress and strain are calculated from easily measurable quantities (normal load, diameter,
elongation, original length) and can be plotted against one another as in Figure 2. Such Stress-
Strain diagrams are used to study the behavior of a material from the point it is loaded until it
breaks. Each material produces a different stress-strain diagram.
Point 1 on the diagram represents the original undeformed, unloaded condition of the material. As
the material is loaded, both stress and strain increase, and the plot proceeds from Point 1 to Point
2. If the material is unloaded before Point 2 is reached, then the plot would proceed back down the
same line to Point 1.
If the material is unloaded anywhere between Points 1 and 2, then it will return to its original
shape, like a rubber band. This type of behavior is termed Elastic and the region between Points
1 and 2 is the Elastic Region.
The Stress-Strain curve also appears linear between Points 1 and 2. In this region stress and strain
are proportional. The constant of proportionality is called the Elastic Modulus or Young's Modulus
(E). The relationship between stress and strain in this region is given by Equation 5-4.

E or   E

where:  is the stress (psi)

E is the Elastic Modulus (psi)
 is the strain (in/in)
10-2
The Elastic Modulus is also the slope of the curve in this region, solved by taking the slope between
data points (0,0) and (y, y).
UTS
. Slope = E 3
Strain Hardening 5
 Fracture
y
2
S
t
r
e
s Plastic Region
s
Elastic Region
1 4
Strain
Figure 2 Stress/Strain Diagram
Point 2 is called the Yield Strength (y). If it is passed, the material will no longer return to its
original length. It will have some permanent deformation. This area beyond Point 2 is the Plastic
Region. Consider, for example, what happens if we continue along the curve from Point 2 to Point
3, the stress required to continue deformation increases with increasing strain. If the material is
unloaded the curve will proceed from Point 3 to Point 4. The slope (Elastic Modulus) will be the
same as the slope between Points 1 and 2. The difference between Points 1 and 4 represents the
permanent strain of the material.
If the material is loaded again, the curve will proceed from Point 4 to Point 3 with the same Elastic
Modulus (slope). The Elastic Modulus will be unchanged, but the Yield Strength will be increased.
Permanently straining the material in order to increase the Yield Strength is called Strain
Hardening.
If the material is strained beyond Point 3 stress decreases as non-uniform deformation and necking
occur. The sample will eventually reach Point 5 at which it fractures.
10-3
The largest value of stress on the diagram is called the Tensile Strength (TS) or Ultimate Tensile
Strength (UTS). This is the most stress the material can support without breaking.
Example: (This is from the FE Exam) A Steel rod (E=200 GPa) has a circular cross section and
is 10m long. Determine the minimum diameter if the rod must hold a 30 kN tensile force without
deforming more than 5mm. Assume the steel stays in the elastic region. Note, 1 GPa = 10^9 Pa.
Solution:
Knowing the initial length and the change in length permits the calculation of strain.
1𝑚
𝛥𝑙 5𝑚𝑚(1000𝑚𝑚 )
𝜀= = = 0.0005
𝑙 20𝑚
In the elastic region, the stress 𝜎 is directly proportional to the strain 𝜀, by the Modulus of
Elasticity, E
𝐹
= 𝜎 = 𝐸𝜀
𝐴
Rearranging, substituting values and converting units,
𝜎 = 𝐸𝜀 = (200 𝐺𝑃𝑎)0.0005 = 0.1 𝐺𝑃𝑎 = 0.1 × 10 𝑃𝑎 = 0.1 × 10 𝑁/𝑚
The definition of stress 𝜎 = can be used to find the required cross section area.
1000𝑁
𝐹 30𝑘𝑁 𝑘𝑁 1000𝑚𝑚 1000𝑚𝑚
𝐴 = = = 0.0003𝑚 ( )( ) = 300𝑚𝑚
𝜎 0.1 × 10 𝑁/𝑚 𝑚 𝑚
The diameter, 𝑑 is solved from the area of a circle
𝜋𝑑
𝐴 =
4
𝐴 4
𝑑 =
𝜋
𝐴 4 300𝑚𝑚 ∗ 4
𝑑 = = = 19.5𝑚𝑚
𝜋 3.14
10-4
10.2 Material Characteristics
There are five material properties that do a good job at describing the characteristics of a material.
They are strength, hardness, brittleness, toughness, and ductility. Each of these will be discussed
in more detail in the upcoming chapters.
Strength is measure of the materials ability to resist deformation and to maintain its shape.
Strength can be quantified in terms of yield stress or ultimate tensile strength. Both yield stress
and ultimate tensile strength can be determined from tensile test data by plotting a stress strain
curve.
Hardness is a measure of the materials ability to resist indentation, abrasion and wear. Hardness
is quantified by arbitrary hardness scales such as the Rockwell hardness scale or the Brinell
hardness scale. These measurements are obtained by a special apparatus that uses an indenter that
is loaded with standard weights. The indenter can have various shapes such as a pyramid or a
sphere and is pressed into the specimen. Either the depth of penetration or the diameter of the
indentation made is measured to quantify material hardness. Hardness and strength correlate well
because both properties are related to inter-molecular bonding. Figure 3 illustrates the test and
gives conversion scales for relating values to tensile strength.
Ductility is a measure of materials ability to deform before failure. Ductility can be quantified by
reading the value of strain at the fracture point on the stress strain curve or by doing a percent
reduction in area calculation. Low carbon steels, pure aluminum, copper, and brass are examples
of ductile materials.
Brittleness is a measure of a material inability to deform before failure. Brittleness is the opposite
of ductility. Brittleness is not quantified since it is the inability to deform. However, ductility is
quantified as discussed above. Examples of brittle materials include glass, cast iron, high carbon
steels, and many ceramic materials. Figure 4 shows the difference between ductile and brittle
behavior on a stress-strain diagram. This will be discussed in more detail in the chapter on fracture.
Toughness is a measure of a materials ability to absorb energy. There are two measures of
toughness. Material Toughness and Impact Toughness. Material Toughness can be measured by
calculating the area under the stress strain curve from a tensile test (Fig 5). The units on this
measure of toughness are in-lb/in3. These are units of energy per volume. Material Toughness
equates to a slow absorption of energy by the material. Impact Toughness is measured by doing a
Charpy V-notch Test, discussed further in the chapter on Fracture.
The material toughness can be expressed as the work W required to strain a bar to engineering
strain ε.
l ε
W =  Fdl = Aoloσdε
lo 0
or if w denotes the work per unit volume of material W/A olo, then
10-5
w = σdε
0
Thus, the area under the engineering stress-strain curve is a direct measure of the amount of work
per unit volume of the material needed to effect a given engineering strain ε.
Figure 3: Hardness Tests
10-6
Brittle

Ductile
S
t
r
e
s
s
Strain
Figure 4. Ductile and Brittle Behavior
Fig. 5
10.3 Elastic-Plastic Response of Metals
When an ordinary piece of metal is subjected to a force of gradually increasing intensity, the atoms
within the individual grains are simply displaced from their initial positions in an elastic fashion;
if the force is reduced, the atoms will return to their initial positions.
When the force becomes sufficiently large, the atoms essentially shift from their elastically
distorted initial positions to new, identical positions within the lattice, resulting in permanent
plastic strain. Thus, in the ordinary tensile test of a metal, we may think of the elastic region as the
region where the atoms are slightly displaced from initial positions, whereas the plastic region,
planes of atoms are actually slipping over one another, resulting in permanent, plastic deformation.
10 - 7
The actual mechanism by which the atoms slip over one another is generally thought to involve
the idea of dislocations or imperfections in the otherwise perfect crystalline arrangement of the
atoms. These imperfections (Fig 6) make the slipping (Fig 7) relatively easy.
The two general types of locations are edge dislocations and screw dislocations (Fig. 8). Edge
dislocations exist when an extra plane of atoms exists in a portion of the lattice structure. Screw
dislocations result when a plane of atoms shifts its orientation from one plane to another.
Force analyses show that slipping of the plates can occur with much lower stress than would be
required if the arrangement of atoms was perfect. In fact, theoretical calculations accounting for
a given dislocation density are consistent with those found experimentally.
Fig. 6 Interstitial Atoms and Vacancies
Fig.7 Movement of Edge Dislocations
10 - 8
Fig. 8 Edge Dislocation (a) and Screw Dislocation (b)
10.4 True stress and strain measures.

Instead of using engineering stress and strain measures, we may use the so-called "true" stress, σ T,
and "true" strain, εT, measures. These are defined as
F
σT =
A
𝑑𝑙 𝑙
𝜀 = = ln ( )
𝑙 𝑙
where A denotes the deformed area of the specimen at load F and l denotes the deformed
length of the specimen associated with a small increase in extension dl.
In the elastic region, the difference between A and Ao and l and lo is very small so that the "true"
stress and strain values are essentially the same as the engineering stress and strain values. When
large plastic deformation takes place, the difference is no longer negligible and it is necessary to
specify which stress and strain measures are used.
For plastic deformations where the difference between A and l and Ao and lo are important, it has
been shown experimentally that the volume of the specimen remains essentially constant such that
Al = Aolo
We also have by the definition of engineering strain

l = lo(1 + ε)
Combining the above equations gives us "true" stress and strain expressible as
σT = σ(1 + ε), εT = ln(1 + ε)
10 - 9
These equations are useful for determining "true" stress-strain response. Notice, however, that
since the "true" stress and strain are expressible in terms of the engineering stress and strain, in the
real sense one is no more true than the other. One set can be obtained from the other simply by
computation. Figure 9 shows a comparison between engineering and "true" stress-strain response.
Figure 9: True vs Engineering Stress
10.5 Yielding of a Ductile Metal under a General Stress State - Mises Yield Condition.
The work done per unit volume of material by the stress in a simple tension test is determined by
the area under the stress-strain curve up to the level of strain experienced at max stress. In the
elastic region (assuming linear behavior) this is given by
W = ½σε
All of this work is stored as "elastic energy" in the material, which is released at unloading. Once
the material is in the plastic region a part of this work is associated with changing the materials
volume while the rest is associated with changing its shape (distorting).
This same division is also true in a general stress loading and forms the basis for the Mises Yield
Condition expressing the condition for yielding, or elastic break-down, under general stress
loading. The Mises Condition can be interpreted as requiring that yielding occurs when the strain
energy of distortion in a general stress state equals the strain energy of distortion at yielding in a
simple tension test on the same material.
To develop the mathematical form of this expression, we consider a general cubic element of
material having principal stresses σ1, σ2, and σ3 acting on it and experiencing associated strain ε1,
10 - 10
ε2, and ε3. The work (or strain energy) per unit volume of material associated with these strains is
given, by analogy with eq. (17) as
w = ½σ1ε1 + ½σ2ε2 + ½σ3ε3
Now, from the elastic stress strain relations we have

σ1 v
ε1 = - (σ + σ3)
E E 2
σ2 v
ε2 = - (σ + σ3)
E E 1
σ3 v
ε3 = - (σ + σ2)
E E 1
where E is Young's modulus, v is Poisson's ratio, both constants for a given material. Substituting
we find
1 v
w = (σ12 + σ22 + σ32) - (σ1σ2 + σ2σ3 + σ1σ3)
2E E
Now consider the volume of the small element of material. If its sides were initially of length a,
its initial volume was a3 and its final volume is a3(1 + ε1)(1 + ε2)(1 + ε3). For small strains, we
thus have for the change in unit volume per unit initial area
Δ = (1 + ε1)(1 + ε2)(1 + ε3) - 1  ε1 + ε2 + ε3

We also see that
1-2v
Δ= (σ + σ2 + σ3)
E 1
Hence, the change in volume is related directly to the sum σ 1 + σ2 + σ3. We may then define the
mean stress σ as
1
σ = (σ1 + σ2 + σ3)
3
The element stresses can be thought of as being composed of this mean stress σ and an additional
stress dependent on the mean stress; i.e.
Figure 10
10 - 11
It is clear that the stress _ is associated with the total volume change of the material and the stresses
σ1 - σ , σ2 - σ , σ3 - σ are associated only with its distortion. Hence, the strain energy
associated with the volume change is determined by substituting σ 1 = σ , σ2 = σ , σ3 = σ
into the equation, we find
3
wv = (1 - 2v) σ 2
2E
Subtracting this from the total strain energy w we find (using the definition of σ ) the strain
energy of distortion given by
1+v 2 2 2
w - wv = (σ + σ + σ - σ1σ2 - σ2σ3 - σ1σ3)
3E 1 2 3
The value of this expression at yielding in a simple tension test (σ 1 = σY, σ2 = σ3 = 0) is

1+v 2
w - wv = σ
3E Y
By equating the right-hand side of the previous equation to this value, we have the Mises Yield
Condition in the form
2 2 2 2
σ + σ + σ - σ1 σ 2 - σ2 σ 3 - σ1 σ 3 = σ
1 2 3 Y
An alternate form of this equation is
2
(σ1 - σ2)2 + (σ2 - σ3)2 - ( σ3 - σ1)2 = 2σ
Y
In two dimensional plane stress

2 2 2
σ1 + σ2 - σ1σ2 = σY
If we use for the two-dimensional case
σx + σ y σ x - σy 2
σ1,2 = ± 2+τ
2 2 xy
we find the Mises Yield Condition expressible as
10 - 12
2 2 2 2
σ + σ - σxσy + 3τ = σ
x 2 xy Y
Example: A cylindrical pressure vessel of 15 inches in diameter has a wall thickness of 0.25 inches.
The yield stress of the material is 40,000 psi. Determine the internal pressure required for yielding
using the Mises Condition.
The stresses are

σθ = (pr)/t, σz = (pr)/2t
where r = radius, t = wall thickness, p = net internal pressure. (these formulae are valid
only for thin walled pressure vessels).
Solution: On substituting for r and t we have σθ = 30 p and σz = 15p. These are principal stresses.
Substituting into the Mises Condition, eq. (30), we have
900p2 + 225p2 - 450p2 = σY = 40000 psi

or p = 1540 psi
This is the net internal pressure, i.e. the difference between internal and external pressure.
Example. A circular steel shaft is subjected to a bending moment of 3000 in-lb and a twisting
moment of 8000 in-lb. Determine the required diameter of the shaft if the yield stress in a simple
tension test in 36,000 psi.
10-13
Solution: the maximum bending stress σx is
σx = Mbr/I
where Mb is the bending moment and I is the moment of inertia of the cross-sectional area with
respect to the neutral axis. Therefore
σx = (3000)(r)/(πr4/4) = 3820/r3
The maximum shear stress is τxy = (Mtr)/(J) where Mt is the twisting moment and J is the polar
moment of inertia. Therefore
τxy = (8000)(r)/(πr4/2) = 5093/r3
Substituting into the Mises relation

2 2 2 2
σx + σ2 _ σxσy + 3τxy = σY
with σy = 0, we have
3820 2 5093
( 3 ) + 3( 3 )2 = (36000)2
r r
Hence r6 = 0.0710 in6 => r = 0.644 in and the required diameter is 1.29 in.
10.6 Maximum shear stress condition.
A theory for the yielding of ductile metals which is sometimes used in place of the Mises Yield
Condition is the maximum shear stress condition. This theory assumes that elastic breakdown will
occur when the maximum shear stress in a general stress state reaches the value it has in a simple
tension test at yielding. Thus, if τmax denotes the maximum shear stress in a general stress loading,
we have
τmax = ½σY
where ½σY is the maximum stress in a simple tension loading at yield.
Example: Reconsider the pressure vessel from the previous example using the maximum shear
stress condition.
Solution: The principal stresses are σ1 = pr/t, σ2 = pr/2t, σ3 = 0. The maximum shear stress is thus
σ1 - σ3 pr
τmax =  =
 2  2t
pr/2t = ½σY or p = 1333 psi.
10-14
Note that this is about 13% less than that given by the Mises equation and thus provides a more
conservative estimate. Experimentally, the Mises Yield Condition has proven to be the most
accurate.
10.7 Creep
“Creep” is a progressive plastic deformation that increases with time, even when the stress is below
the yield stress of the material. The effects of creep increase as temperature increases, and
generally noticeable at above 35% of the melting temperature of the material. Figure 11, below
shows the effects of creep, in term of strain versus time.
 Initially, the rate of creep (slope) is high. In this initial stage, effects analogous to strain
hardening reduce the rate of creep with time.
 In the second stage, the creep achieves a constant rate due to a balance between work
hardening and annealing (thermal softening).
 In the third stage, necking of the material causes the strain to increase more rapidly,
leading to rupture.
Figure 11: Stages of Creep
A well-known example of material creep was the World Trade Center Collapse. The stress on the
materials increased because of the reduced structural capacity of the damaged building and added
weight due to planes. The burning jet fuel increased the temperature, causing thermal creep. This
lead to a reduced cross section of the material and eventual failure.
10-15
Figure 12: Creep and Failure mechanism in World Trade Center Collapse
(From TLC “Anatomy of Collapse”)
10-16
CHAPTER 11
FRACTURE OF MATERIALS
11.1 Brittle vs. Ductile Fracture

11.2 Temperature Effects and Fracture Analysis Diagram
11.3 Griffith Theory of Brittle Fracture.
11.4 Fracture mechanics of high-strength materials
11.6 Stress Corrosion Cracking (SCC)
11.7 Non-Destructive Testing
11.1 Brittle vs. Ductile Fracture
Fracture involves the forced separation of a material into two or more parts. Brittle Fracture
involves fracture without any appreciable plastic deformation (i.e. energy absorption). Ductile
Fracture in the converse and involves large plastic deformation before separation. The difference
between brittle and ductile fracture is illustrated in figures 1 and 2. Remembering that the area
under the σ - ε curve, Fig. 1, represents energy, we can see that much less energy is expended in
brittle fracture than in ductile fracture.
Figure 1: Brittle vs Ductile Fracture
11 - 1
Figure 2. Photos of Tensile Specimens (Left) Brittle, (right) Ductile
Toughness is a measure of a material’s ability to absorb energy before failure. Formally, the
toughness is the integral under the stress-strain curve up to the fracture point. The Charpy
Impact test is often used to determine the material toughness (or relative fracture performance) in
the presence of notches and temperature changes. This test is illustrated in figure 3, along with
typical result for steel and aluminum.
The figure also shows the temperature effect. Reducing the temperature below the “transition
temperature” causes the material to change from ductile to brittle. For example, during World
War II, Liberty Ships operating in the North Atlantic had failures due to brittle fracture because
the water temperature was below the transition temperature of the material (Fig. 4).
Figure 3: Charpy Impact Test
11 - 2
Figure 4. Fractured Hull of Liberty Ship
Brittle fracture is generally governed by the maximum tensile principal stress. A simple
illustration is a piece of chalk in torsion (Fig 5). Here the plane of the principal tensile stress is
at 45o to the axis of the chalk, and thus is the plane of the fracture.
Fig 5: Principal Tensile Stress for Chalk in Torsion
11.2 Temperature Effects and Fracture Analysis Diagram
We have noted earlier in connection with the Charpy impact test that the fracture of steel is very
sensitive to the testing temperature within certain temperature ranges. To illustrate this point
further, consider a material with a certain crack size present. Below a certain temperature, called
the Nil Ductility transition Temperature (NDT temperature), the fracture will be completely
brittle in nature. At temperature greater than the NDT temperature, some plastic deformation
will accompany the fracture and the fracture will be ductile to some degree. The stress required
to cause brittle fracture will be less than that to cause ductile fracture in the same material and
the fracture stress vs temperature behavior will be as indicated in figure 25 for a given crack size.
If we consider a number of crack sizes, we will generate a family of curves similar to that shown
in figure 6. This leads to the Fracture Analysis Diagram. Such a diagram is shown in Figure 7
for a low carbon construction steel. The vertical axis in the fracture analysis diagram represents
the operating stress and the horizontal axis represents the operating temperature, expressed as:
11 - 3
T = TNDT + ΔT therefore, T - TNDT = ΔT
with the origin denoting ΔT = 0.
Figure 6 Illustration of Typical Fracture Behavior for Steels
Figure 7: Fracture Analysis Diagram
Example. The walls of a submersible are to be made of a steel having a yield stress of 45 ksi.
For this material, under a nominal stress of 30 KSI and an operating temperature of 35 oF,
determine the NDT temperature where a "stress concentration factor" of 1.5 exists, if (a) 8 inch
cracks are to be arrested and (b) if 12 inch cracks are to be arrested.
11 - 4
Solution. Because of the stress concentrations the operating stress will be (1.5)(30) = 45 ksi,
which equals the yield stress. Reading horizontally across from the yield-stress value in Figure
28 to the intersection of the dotted curve for 8inch cracks we find the temperature NDT + 25.
This must equal the operating temperature of 35 oF so the required NDT temperature is 10oF.
Similarly for 12 inch cracks, NDT = 0oF.
Example. For a steel with an operating stress equal to 1/2 of the yield stress, determine the
smallest crack size which could cause brittle fracture at 40 oF if the steel has a NDT of 15oF.
Solution. Reading horizontally in Figure 28 at a stress level of 1/2 yield stress and vertically up
from a temperature of NDT + 25oF, we find the intersection at an interpolated fracture curve of
about 18 inches. Hence cracks of 18 inches or longer would result in fracture.
11.3 Griffith Theory of Brittle Fracture.
Consider a thin plate of length l having a thru-crack of length 2c, as shown in figure 8.
Fig. 8 Fracture with Thru Crack
For a non-extending crack of length 2c, the force-deflection curve will be as the upper curve
shown in Figure 8. For a non-extending crack of length 2(c + Δc), the curve will be as the lower
curve in Figure 8. The area between these curves thus represents the energy released as the crack
extends from 2c to 2(c + Δc).
11 - 5
Using elasticity theory Griffith showed that the energy released per unit thickness during a crack
growth of 2Δc is
2πσ2
Δwe = cΔc
E
The creation of additional crack surface requires surface energy per unit thickness given by
Δws = 2γs(2Δc) = 4γsΔc
where γs is the surface energy per unit area.
Now, if Δwe < Δws the crack will not grow since the released energy will be less than that
required to create a new surface.
11 - 6
If, on the other hand, Δwe  Δws, crack growth will occur since adequate energy is available for
creating the new surface. Hence for crack growth and subsequent failure we must have the
condition
2πσ2
cΔc  4γsΔc
E
The critical fracture stress σc is therefore

2Eγs
σc =
πc
Thus, the critical stress is inversely proportion to c ½. Hence, the smaller the flaw, the greater the
value of σc. (The entire concept of crack growth is based on the natural tendency for things
(midshipmen included) to seek the lowest energy state available. Thermodynamically, there is a
'hump' which must be overcome in order for this to occur. In the case of crack propagation the
energy supplied during the loading of the material must be sufficient enough to overcome the
'hump').
The Griffith theory is good for very brittle material, such as glass, in which no plastic
deformation accompanies the fracture. When there is some plastic deformation associated with
the crack extension, we must add the plastic work γp expended in making the surface to the
surface energy term γs to obtain
2E(γs + γp)
σc =
πc
This formula forms the starting point for modern fracture mechanic analysis as discussed below.
11 - 7
11.4 Fracture mechanics of high-strength materials
High strength steels (yields stresses of 100,000 psi or greater) have a low toughness, and designs
using theses materials must employ fracture mechanics methods of analysis to insure against
failure by brittle fracture. Using the Griffith formulation, we define a new property of the
material:
σc πc = kc
where Kc is known as the fracture toughness of the material and is equal to
kc = 2E(γs + γp)
and has the units of

lbf
inch2 inch
The quantity Kc can be measured in a simple test by simply measuring the size of an induced
crack and measuring the resulting fracture stress, as shown in Figure 9(a). Figure 9(b) shows the
effect of varying the thickness of the test plate. It will be seen that K c falls of as plate thickness
increases and reaches a constant value of K IC at thickness t*. This critical thickness, t*, varies
with the material under consideration but usually is 1/2 inch or less. In using equation for an
interior crack, it is always conservative to take K c = KIC. The variation of KIC with the yield
strength for steels is illustrated in figure 10.
The critical defect size is

KIC2
Cc =
πσ2
where σ is the applied stress and 2Cc is the flaw or crack size needed for brittle fracture.
For surface and interior cracks that do not extend through the plate thickness, we have the
geometry shown in figure 11.
11 - 8
(a) (b)
Figure. 9
Figure 10: Fracture Toughness vs Yield Strength for Steel
Fig. 11 Flaw Geometry
11 - 9
Approximate formulas for the critical flaw size for cracks which do not initial extend thru the
plate thickness are:
for a surface crack:
KIC2 φ2 - 0.2(σ/σy)2
ac = [ ]
πσ2 1.2
for an interior crack:
KIC2
ac = [φ2 - 0.2(σ/σy)2]
πσ 2
where φ depends on the ratio a/c as given below.
a/c φ
0 1.0
0.2 1.05
0.4 1.15
0.6 1.28
0.8 1.42
1.0 1.57
Example. For a "thumb-nail" surface flaw, a/c = 0.4. Hence φ = 1.15. Suppose the operating
stress to be equal to ½σy, we thus have the critical flaw size for fracture given as
2
K
IC (1.15)2 - (0.2)(0.5)2 KIC2
ac = [ ] = 1.06
πσ2 1.2 πσ2
Example. A pressure vessel is to be made using a steel with K IC = 41 ksi - (in)½ and σy = 220 ksi.
Assuming surface cracks of 0.05 inches or less and σ = σy, determine whether the chosen
material is adequate from fracture considerations.
Assume first that a/c = 0.4 (thumb nail flaw). then φ = 1.15 and σ = σ y, we have
1.2acπσy2
KIC2
= = 8130
φ2 - 0.2
Hence KIC = 90 ksi-in½. But for the material, KIC is only equal to 41 ksi-in½. Thus fractures will
occur for thumb-nail cracks.
11 - 10
Next assume a/c = 1.0; we then find that K IC = 66 ksi-in½. This is still higher than KIC for the
material. Hence to ensure against fracture, we must select a material with a greater fracture
toughness or reduce the operating stress.
Example. A pressure vessel has material properties σy = 65 ksi and KIC = 35 ksi -in½. If the
diameter is 20 in and the wall thickness is 1 inch, determine the maximum pressure the tube can
withstand before failure if thumbnail cracks of depth 0.1 inches exist.
Solution:
Pa Pa
σθ = = 10p, σz = = 5p
t 2t
First check the maximum pressure required for yielding. The Mises condition gives
σθ2 + σz2 - σθσz = σy2
Hence
75p2 = σy2
and
p = 7.50 ksi (for yielding).
Next check the maximum pressure for fracture. Here only σθ enters since fracture is determined
by the greatest tensile principal stress. We have
11 - 11
Therefore, fracture will occur when σθ = 0.939 σy or when p = 0.1(0.939)σY and p = 6.10 ksi
which is before the yielding.
Example. A thin-walled cylindrical pressure vessel is to have a wall thickness of 0.5 inches and a
radius of 10 inches. The material used has K IC vs σy fracture characteristics as shown. Surface
cracks in the material may be assumed no deeper than 0,025 inches. Select the appropriate σ y
value which will allow the greatest safe pressure loading. What is the pressure value?
Solution:
Fracture Characteristics for Example
We assume thumbnail flaws with a/c = 0.4 and φ = 1.15. From the surface flaw equation we
have
1.2aπσθ2
2
Kic =
σθ
φ2 - 0.2( )2
σy
From the Mises yield condition we also have

σθ2 + σz2 - σθσz = σz2
Therefore for yielding we have (with σθ = 20p, σz = 10p )

p = 0.0577σy
Constructing the following table:

σy P σθ Kic Remarks
200 11.55 231 68.8 too low
240 13.85 277 82.4 too high
220 12.69 254 75.6 ok
Thus at σy = 220 ksi the allowable pressure is 12.69 ksi.
11 - 12
The above results imply, in terms of modern fracture mechanics, that the "fracture toughness" (or
critical stress intensity factor) varies with the relative temperature T - T NDT. Such variation has
been given in the ASME (American Society of Mechanical Engineers) Boiler and Pressure
Vessel Code used in the design of such vessels.
This variation is shown in Figure 12 for steels used in pressure vessels such as containment
vessels for nuclear reactors. ( Yield stresses  50 - 70 ksi at room temperature). The values
plotted are the minimum values from individual tests showing scatter.
Figure 12. Variation in Fracture Toughness with Temperature
Using the above fracture toughness data (K IC = KIR), curves such as shown earlier on the Failure
Analysis Diagram can be constructed for this steel using the thru-crack equation. (Fig 13)
KIC
σ=
π C
11 - 13
Figure 13: Fracture Toughness with Temperature
Example. A confining vessel for a nuclear reactor is designed to withstand a net pressure of
2000 psi. The outside radius in 80 inches and the wall thickness is 9 inches. The material is
steel with a yield stress of 40 ksi at an operating temperature of 500 oF. A hypothetical surface
flaw is postulated (as part of the safety code requirements) with a depth of 1/9 the wall thickness
and a length 2/3 times the thickness.
(a) What is KIC?

(a) If the operating temperature is 500oF, what is the required NDT?
(b) What about at 300oF?
Solution. We have the equation

KIC φ2 - 0.2(σ - σy)2
σ= [ ]12
πa 1.2
The values of the crack dimensions are a = (1/9)(9) = 1 in and 2c = (2/3)(9) = 6. Therefore
a/c = 0.33 and φ = 1.1. The stress is
Pr (2000)(80)
σ= = = 17,800(p.s.i)
t 9
or 17.8 ksi. Solving for KIC, we find KIC = 47.9 ksi in½. From the curve of fracture toughness
(figure 13) and the intersection of a horizontal line from σ = 17.8 KSI and the 6 inch curve, we
then have T - TNDT  55
(a) for T = 500oF we require TNDT= 445oF (or lower).
11 - 14
(b) for T = 300oF we require TNDT= 2450oF (or lower).
Initially TNDT may be low ( say -50oF), but with radiation embrittlement it can increase to
250 - 350oF over a 20 year period. If it has increased to 280 oF and a sudden cooling of the
reactor water to 300oF occurred, brittle fracture could result. This problem is very obviously
one of great concern. This phenomenon is known as "pressurized thermal shock".
11.6 Stress Corrosion Cracking (SCC)
Stress corrosion cracking refers to the fracture of metals and alloys under tensile stress in a
corrosive environment when the applied stress is below the yield stress for the material. The
fracture does not occur immediately, but instead occurs after a finite amount of time. During this
time localized corrosion, caused by a specific combination of electrolyte and metal, results in the
formation of cracks. Continued corrosion and solid formation within the crack combine to, in
effect, increase the applied tensile load. As shown in figure 14, the allowable stress decreases
the longer the particular environment exists. The phenomenon can be expressed in terms of
modern fracture mechanics in terms as follows: A "stress intensity" factor is defined for a thru
crack as
Kq = σπ(c)½
where σ is the applied stress and c is one-half the width of the crack. We have already seen that
fracture will occur when K = Kc.
Figure 14: Stress Corrosion Cracking
11 - 15
Experiments have shown that with certain high-strength steels, as well as stainless steels and
aluminum alloys, crack growth can occur when the material is under stress and in a corrosive
environment. In this case, the stress intensity, which is a function of flaw geometry, can increase
with time until it reaches the critical Kc value and failure occurs. If the stress intensity K is
initially less than a critical value Kscc for the material, no crack growth will occur. This is
illustrated in figure 15.
Figure 15: Stress Corrosion Cracking
11.7 Non-Destructive Testing
Nondestructive testing (NDT) methods are inspections for material defects, such as the cracks
discussed in this chapter.
External (Surface) Inspection Techniques
The three most commonly used external (surface) inspection techniques currently in use are the
Visual Test, Dye Penetrant Test, and Magnetic Particle Test.
 Visual Testing (VT) should be done during all phases of maintenance. It can usually be
performed quickly and easily and at virtually no cost. Sometimes photographs are made
as a permanent record. Visual inspections only allow the inspector to examine the surface
of a material.
 Dye Penetrant Testing (PT) uses dyes in order to make surface flaws visible to the
naked eye. It can be used as a field inspection for glass, metal, castings, forgings, and
welds. The technique is simple and inexpensive and is shown schematically at Figure
5.10. Only surface defects may be detected, and great care must be taken to ensure
cleanliness.
11 - 16
Fig 16: Dye Penetrant Testing
 Magnetic Particle Testing (MT) is only used on ferromagnetic materials. This method
involves covering the test area with iron filings and using magnetic fields to align the
filings with defects. Figure 17 shows the deformation of a magnetic field by the presence
of a defect. Magnetic particle tests may detect surface and shallow subsurface flaws, and
weld defects. It is simple and inexpensive to perform, however a power source is required
to apply the magnetic field.
Figure 17 Magnetic Particle Testing
11 - 17
Internal (Sub-surface) Inspection Techniques
The three most common internal (subsurface) techniques are the Ultrasonic Testing,
Radiographic Testing, and Eddy Current Testing.
 Radiographic Testing (RT) is accomplished by exposing photographic film to gamma

or x-ray sources. This type of testing detects a wide variety if internal flaws of thin or thick
sections and provides a permanent record. These methods of testing require trained
technicians and present radiation hazards during testing.
 Ultrasonic Testing (UT) utilizes a transducer to send sound waves through a material. It
may be used on all metals and nonmetallic materials. UT is an excellent technique for
detecting deep flaws in tubing, rods, brazed and adhesive-joined joints. The equipment is
portable, sensitive and accurate. Interpretation of the results requires a trained technician.
Figure 18 shows an ultrasonic transducer configuration.
Figure 18 Ultrasonic Testing
 Eddy Current Testing involves the creation of a magnetic field in a specimen and reading
field variations on an oscilloscope. It is used for the measurement of wall thicknesses and
the detection of longitudinal seams and cracks in tubing. Test results may be affected by a
wide variety of external factors. This method can only be used on very conductive
materials, and is only good for a limited penetration depth. Once very common, it is being
replaced by the increasing usage of ultrasonic testing.
Figure 19 summarizes these non-destructive testing techniques.
11 - 18
Fig. 19 Comparison of Non-Destructive Testing Techniques
11 - 19
CHAPTER 12
FATIGUE
12.1 Fatigue Failure of Materials

12.2 Fatigue Testing Procedures
12.3 S-N Fatigue Curves
12.4 Effect of Mean Stress using Goodman-Soderberg Relation
12.5 Stress concentrations
12.6 Miner's Rule for Fatigue Under Varying Stress Levels.
12.7 Fatigue Under Combined Stress
12.8 Specialized Formulae for Two Dimensions
12.9 Equivalent Stresses in Terms of Maximum Shear
12.10 Connection Between Fatigue and Failure.
12.1 Fatigue Failure of Materials

Fatigue failure of materials refers to their failure under the action of cyclic elastic stress.
Fatigue generally involves the formation and gradual growth of cracks and ultimately to fracture
as a result of reduced load carrying capacity.
Definitions. We consider uniaxial cyclic stress loading as shown below in figure 1.
Figure 1: Definitions associated with cyclic stress
The mean stress σm and stress amplitude σa are defined as
σm = ½(σmax + σmin)
σa = ½(σmax - σmin)
When the mean stress σm equals zero we have completely reversed stress. The stress variation
Δσa is referred to as the range of stress. For sinusoidal loading, the stress-time relation is given
simply as
12 - 1
σ = σm + σasin(2π/T)t
where T denotes the period of the stress cycle. The stress-time variation need not necessarily be
sinusoidal for fatigue failure, though in many practical cases it is.
12.2 Fatigue Testing procedures

The are two basic testing procedures used to study fatigue: the Rotatory-Bending Test and the
Deflection-Bending Test. These are illustrated in figure 2.
Figure 2: Fatigue Testing Procedures
In the rotatory-bending test, the internal moment M in the specimen between the slip-rings is
constant and equal to pa/2. Therefore the maximum stress is
Where I denotes the moment of inertial of the cross-section about the neutral axis and y denotes
the vertical distance from the section centroid. for a rotating beam with non-rotating forces, as in
this case, y is related to the radius r of a ring of material by
y = r sinθ = r sin (2π/T)t.
Figure 3: Coordinates for Rotary Bend Test
12 - 2
Taking r = R = outside radius, we have the stress in the rotating beam experiment given by
MR 2π
σ= sin( )t
I T
Thus, the mean stress σm is zero and the amplitude is equal to MR/I.
In the case of the Deflection - Bending test, the tip of the cantilever beam is subjected to
a sinusoidal force variation and we have (at the clamped end)
PolC 2π
σ= sin( )t
I T
Where po is the amplitude of the applied force, l is the length of the beam, and C the distance
from neutral axis to outer surface. Again the mean stress σ m is zero and the amplitude σa is given
as PolC/I.
12.3 S-N Fatigue Curves.

Using this test, the stress amplitudes associated with fatigue failure after N cycles can be
determined and plotted as shown for the case of steel in Figure 4.
Figure 4: Example S-N Curve for Steel
The curve of figure 4 is sometimes referred as the S-N curve. We denote the value of σ a
for failure after N cycles as σN. Note that for steels, there is a stress level below which no failure
is observed. This is known as the endurance limit. It generally does not exist for other materials
such as aluminum, etc. Figure 5 shows a comparison of two S-N curves for steel, with an
endurance limit, and aluminum, without an endurance limit.
12 - 3
Figure 5: Comparison of Aluminum and Steel SN curves
It is to be emphasized that fatigue failure can occur even though the stress never exceeds
the elastic limit of the material. In the elastic range, the fatigue failure usually occurs after 10 5
cycles of stress (high cycle fatigue). For large stresses causing plastic deformation, much fewer
cycles of stress are needed for failure (low cycle fatigue). In low-cycle fatigue, the strain vs.
number of cycles is plotted, since the calculation of stress by elastic formulas is no longer valid.
12.4 Effect of mean stress using Goodman-Soderberg Relation
The effect of mean stress on modifying the stress amplitude σ a needed for failure after N cycles
is estimated using the Goodman-Soderberg relation. This assumes that when the mean stress
equals the failure stress σu of the material, the additional stress amplitude σ a for failure after N
cycles is zero. It also uses the fact that the stress amplitude for failure at N cycles will be that
from the usual S-N curve (σN) when the mean stress is zero and it assumes a linear relation
between these two extremes as shown in figure 6.
σN
σa = σN - |σm|
σu
12 - 4
Figure 6: Mean stress effect on fatigue strength. The solid line denotes predictions from the
Goodman-Soderberg relation
Example. A rod of 0.5 in2 is subjected to static mean tensile load of 10 kips. What fatigue stress
amplitude σa will produce failure after 106 cycles? Assume σn = 32 ksi, σu = 60 ksi.
Solution. We have σm = 10/.5 = 20 ksi. The Goodman-Soderberg relation then give

σn 32
σa = σn - |σm| = 32 - 20 = 21.3 ksi thus FA = 21.3 ksi
σu 60
12.5 Stress Concentrations.

In the case of a hole, fillet (a welding term), or other geometric differences, the applied
stresses tend to concentrate and we must account for this. Letting k s be the static stress
concentration factor, we have:
σa = ksσφ
where σa = the actual tensile stress and σφ = the nominal or applied stress.
In the case of fatigue, we have a fatigue strength reduction factor k f such that
σNφ = σN/kf
where σNφ = the maximum allowable stress at N cycles and σ N = the strength at N cycle
regardless of the fatigue strength reduction factor (from the S-N curve).
As a result the Goodman-Soderberg relation can now be written as
σN
σN kf σNφ
σa = - |σm| = σNφ- |σ |
kf σu σu m
ks ks
12 - 5
Example. A threaded steel rod of 2 in root diameter is subjected to a mean stress of 20 ksi and a
variable stress σa. Determine the allowable σa if σN = 300 ksi, σu = 60 ksi, ks = 2.0, and kf = 1.5.
Solution. Using the Goodman-Soderberg relation,
30
30 1.5
σa = - 20 = 6.7 ksi
1.5 60
2.0
12.6 Miner's Rule for Fatigue Under Varying Stress Levels.

Suppose a material is subjected to n1 cycles of stress at stress amplitude σ1 and n2 cycles at
amplitude σ2, etc. Miner's Rules states that
where ni = number of cycles at a particular stress and Ni = the fatigue life at that stress. figure 7
shows the relationship between allowable stress σ N and N cycles.
Figure 7: Example S-N curve
12 - 6
Example. A beam is subjected to 5 x 102 reversed cycles of stress at σ1 = 30 ksi. How many
additional cycles can the beam withstand at stress σ2 = 20 ksi? (N30ksi = 104, N20ksi= 105).
Solution:
5 x 102 n2
+ =1 n2 = 5 x x104
104 105
Example. An off-shore structure is subjected to reversed stress loading by surface waves as

shown in the table. Determine the total number of waves for failure.
σ (ksi) 30 20 10
N 104 105 106
% of waves 5% 15% 80%
From Miner's Rule
0.05N 0.15N 0.80N

+ + =1 N = 1.37 x 105Waves
104 105 106
Example. Suppose that in the previous example, the periods of the waves associated with σ 1, σ2,
and σ3 are 5, 3, and 1 seconds respectively. Determine the life of the structure.
Solution. If T equals the life of the structure in seconds, we then have
T = (0.05N)(5) + (0.15N)(3) + (0.80N)(1)

T = 2.055 x 105 sec = 57.08 hours
12.7 Fatigue Under Combined Stress.

The above considerations have been limited to fatigue under the action of a single stress
component. To handle fatigue under combined stress, we use a Mises-type condition as in the
case of yielding. For the case where the mean stress of each principal stress component is zero
and the amplitudes of the principal stresses are σ1a, σ2a, and σ3a we have
(σ1a - σ2a)2 + (σ2a - σ3a)2 + (σ3a - σ1a)2 = 2σN2
(This formula is merely resolving the magnitude of the combined Von Mises stress)
12 - 7
Example: If σN = 30 ksi, for failure after 106 cycles, determine the amplitude of a completely
reversed shear stress for failure after 106 cycles.
Solution: The principal stresses are: σ1a = τa, σ2a = -τa, σ3a = 0. Substituting into the above
relation we have
1
τa = σN = 17.3 ksi
3
Next consider the more complex case of the non-zero mean stresses σ 1m, σ2m, σ3m together with
amplitudes σ1a, σ2a, σ3a. From the Mises-type relation we then define
where σ*m is the resolved mean stress and σ*a is the resolved stress amplitude. Substituting into
the Goodman-Soderberg relation,
*
σN *
σa = σ N - σ
σu M
We thus obtain an equation relating the "equivalent" stress amplitude σ a* to the "equivalent mean
stress σM*.
12 - 8
Example. A rod of circler cross section is subjected to tension and torsion such that the tension
stress is constant and equal to 20 ksi and a shear stress which cyclically varies from 0 to S. If σ N
= 40 ksi for failure after 106 cycles, determine the allowable value of S for failure after this
number of cycles. Assume σu = 80 ksi.
Solution.
The mean and amplitude stresses are σm = 20 ksi, σa = 0, τm= S/2, and τa = S/2. We also have
the principal stress formula
σ σ
σ1, 2 = ± ( )2 + τ2
2 2
Substituting for the "equivalent" mean stress σm* and the stress amplitude σa* we find
S S
σ1m = 10 + 100 + ( )2 σ2m = 10 - 100 + ( )2
2 2
S S
σ1a = σ2a = -
2 2
Using the Goodman-Soderberg relation we find
* 1 3 1 * 1 3 1 3
σ = (800 + S2)2 σ = ( S2)2 = S
m 2 2 a 22 4
3 σN 1 3 21
S = σ - (800 + S)
4 N σ
u 2 2 2
With the above values of σN and σu, this reduces to
S2 - 123.2S + 2669 = 0
12 - 9
The solutions to which are

S = 28 ksi, S = 95.2 ksi
to decide which root is appropriate, again consider the Goodman-Soderberg relation
*
σN *
σa = σN - σ
σu m
Taking σa = (3/4)½S and solving for σm with S = 28 ksi, we find that σm = 31.5 ksi. For S = 95.2
ksi we find that σm = - 84. 9 ksi. Since σm must be positive in the Goodman-Soderberg relation,
the only acceptable solution is S = 28 ksi.
Example. A proposed scuba air bottle may be idealized as a closed cylinder of 8 in. diameter
with a wall thickness of 0.10 inches. Determine the maximum safe internal pressure loading for
104 cycles of loading if σN = 30 ksi and σu = 60 ksi.
Solution. We assume the pressure varies from 0 to P during a cycle of loading. We have
Pa Pa
σθ = = 40P = σ1 σz= = 20P = σ2
t 2t
Hence
σ1m = 20P, σ2m = 10P

σ1a = 20P, σ2a = 10P
The equivalent σm* and σa* stresses are
σm* = σa* = 17.32P
Substituting into the Goodman Soderberg relation
17.32P = 30 - (30/60)(17.32P) or P = 1.15 ksi
12.8 Specialized Formulas for Two-Dimensions

The equivalent stress formulas are in terms of principal stresses. For two-dimensional
stress problems, it is convenient to express the formulas in terms of non-principal stress
components. Those relations may be written as:
Example Use the above relations to solve the problem of the earlier example where a rod is
12 - 10
subjected to a constant tensile stress of 20 ksi and a cyclic shear stress between 0 and S, with the
value of S required for failure after 10^6 cycles.
Solution: With 𝜎 = 20 𝑘𝑠𝑖 𝜎 = 0 𝜏 = 𝑆/2 and 𝜏 = 𝑆/2, we find directly
And the Goodman-Soderberg relation then provided the relation
The solution for S is 28 ksi, as before.
12.9 Equivalent Stresses in Terms of Maximum Shear
12 - 11
12.10 Connection Between Fatigue and Failure.

We have previously noted that fatigue failure ultimately results from the growth of small cracks
and flows with cyclic loading until such time that the cracks are of such size that fracture occurs.
It is of interest to establish a quantitative relationship to describe the crack growth. Experimental
measurements indicate a relation of the form
da n
dN = C(Δk)
where a denotes crack size after N cycles of stress, C and n denote constants and Δk, for a
thumb-nail surface flaw is given approximately by
Δk = Δσ πa
da n
dN = c(Δσ πa)
Example. A structural member contains surface flaws of initial depth 0.01 inches. Determine
the life of the member in cycles if it is repeated loaded from 0 to 70 ksi. Assume K 1C = 90 ksi -
(in)½, n = 2, and C= 4 x 10-10 (ksi2-cycle)-1.
Solution.
da
= C(Δσ)2πa
dN
12 - 12
ac
ln( ) = c(Δσ)2πN
ao
Since Δσ = 70 ksi, ao = the initial crack depth, and ac denotes the crack depth at fracture and is
given by
2
K1C 902
ac = = = 0.526inches
πσ
2 π702
max
using this value we thus find that with ac/au = 52.6, N = 6.4 x 105 cycles.
Table of Value
(units of kips, inches)
Material n C
mild steel 3 3.6 x 10-10
High strength steel 2.25 6.6 x 10-9
12 - 13
CHAPTER 13
WOOD
13.1 Directionality (isotropy)

13.2 Mechanical Properties
13.3 Typical Uses in the Marine Environment
13.4 Structural Analysis Equations
13.5 Fastening
13.6 Control of Degradation
13.7 Plywood
Wood has been used as a shipbuilding and offshore structure material since the earliest
days of each industry. The USS Constitution and the Standard Oil Company’s California
Rig #1 are both examples of wood’s long-term presence in the industry. On the
shipbuilding side, over 40,000 wooden ships and small craft were built for the US Navy
in WWII. The last Navy wooden-hull purchases were the YP’s in the ‘80’s and the
Avenger-class minesweepers (Figure 1) in the ‘90’s. At 224 feet, the Avengers are the
longest wooden hulls built for the Navy.
Figure 1: USS Chief, Avenger-class wooden minesweeper

Commissioned in 1994
Although it is unlikely that the Navy will ever again purchase wooden hulled vessels
(mostly because of high initial costs compared to composites), wood’s properties make it
a viable material for structural use in the Navy for many years to come. In EN380 we will
briefly explore these properties.
First is a description of the microscopic properties. Wood consists mainly of hollow fiber
cells (sort of like short soda straws) made of lignin and cellulose (Figure 2). The lignin
acts as an adhesive holding the cells together. The cellulose is a fibrous material that is
hygroscopic, or attracts water like a sponge. (And just like a sponge, wood will swell
when moist.) The lignin is very strong, so much so that wood typically fails by crushing
of the cellulose structure rather than failure of the lignin.
13 - 1
Figure 2: Wood microstructure – tt = tg = transverse, rr = radial, wr = wood ray
Wood is typically grouped into two classes: hardwoods and softwoods, although
unfortunately there is no correlation between these terms and the wood’s surface
hardness. Hardwoods have broad leaves and small fiber (cell) sizes (about 0.01” diameter
and 0.05” length). Softwoods are conifers (have cones) and large cell sizes (0.05” x
0.13”). Balsa and oak are hardwoods, Douglas fir, pine and cedar are softwoods.
13.1 Directionality (isotropy)
Continuing the soda straw analogy will explain wood’s directional properties. If a soda
straw is held in the hand and squeezed between two fingers, the straw’s wall crushes.
This is the radial direction of the wood and like the straw it is not very strong. If the
straw is held between the hands in the long direction and pulled or pushed, it is strong as
long as it does not buckle. This is the longitudinal (axial) or with-the-grain direction of
the wood and is its strongest direction. The last property direction is the tangential or
hoop direction. This direction can be visualized by the pressure built up in a straw if one
end is capped and the other is blown in to. If the straw splits (or “checks”), it is a
tangential failure, and this is typically the weakest of wood’s three directions. The radial
direction is slightly stronger than the tangential direction due to the presence of wood
“rays” (wood fibers). Few species have rays in the tangential direction, but those that do
are highly preferred in marine construction due to the increased splintering resistance.
The most notable are Sitka Spruce, Live Oak, and Black Locust. The last two are fairly
rare. On an historical footnote, the 1797 frigates were specified for Live Oak frames and
planking. Due to a shortage, only two were built using Live Oak rather than the inferior
White Oak. One of those was the Constitution.
In material science, a material which has the same properties in all directions is isotropic
(metals are usually considered isotropic). One which has different properties in every
direction is anisotropic. If it has the different properties in three perpendicular planes it is
orthotropic (composites are orthotropic). If one of the planes has the same material
13 - 2
properties in all directions it is transversely isotropic (an example is a single fiber in a

resin). These terms are important when using Hooke’s Law. Wood is technically
anisotropic but is usually modeled as transversely isotropic. Composites are technically
orthotropic but are also modeled as transversely isotropic. Analytically, wood and
composites share many characteristics.
The three directional properties of wood vary widely, but in general, if a wood species
has a longitudinal strength of 100, its radial strength will be about 5-10 and its tangential
strength about 2-10. Given the large difference in properties it is clear that making sure
the load is arranged with the grain is critical to a structure’s success! This is often not
easy as most trees do not grow perfectly straight and lumber mills do not usually cut
lumber for maximum strength.
Figure 3 shows a typical log undergoing the milling process. Depending on what part of
the log the plank comes from it will have dramatically different properties when exposed
to moisture. The “A” plank is quartersawn (or vertical grain – VG). This cut goes through
the tree’s center and gives planks that are warp stable. (See Figure 4.) Another advantage
to this cut is that it minimizes the weakest hoop direction fibers. The major disadvantage
is that if small trees are used, planks of a useful width for marine construction will often
include sapwood, the outer rings that are rot prone. Plainsawed planks of low density
woods like pine, fir and cedar will often warp as much as a plank thickness. The bottom
line is that for any structural application exposed to moisture, quartersawn heartwood is
specified.
Figure 3: Quartersawed (A) and plainsawed (B) boards cut from a log. (From Ref. 2)
13 - 3
Figure 4: Distortion due to sawing location of planks from the log. (From Ref. 2)
13.2 Mechanical Properties
Material properties of wood are generally given for the longitudinal direction and are
presented for a standard moisture content of 12%. Radial and tangential properties are
occasionally given but are usually estimated from the previously mentioned ratio’s. If
given, then usually the tangential, being lower, is given. “Green wood”, where the
moisture content is above 18%, generally has strength properties about ½ of the dry wood
properties. Wood can be “green” either because it has not dried sufficiently after
harvesting, or because of exposure to water after construction. As the structures designed
by naval architects and ocean engineers are often in a wet environment, it is also
important that they understand what will be the equilibrium moisture content of the wood
in their designs! The equation for moisture content is:
wet weight  oven dry weight

MC 
oven dry weight
Table 1 shows the material properties for common woods used in ships and offshore
structures. Tensile properties parallel to the grain are usually taken as equal to the
compressive strength, although they are typically higher.
13 - 4
Species green dry flexural tensile compressive modulus of

specific specific strength strength* strength elasticity
gravity gravity psi psi psi msi
douglas fir 0.45 0.48 12,200 340 7,430 1.95

white oak 0.6 0.68 15,200 800 7,440 1.78
western red cedar 0.31 0.32 7,500 220 4,560 1.11
sitka spruce 0.37 0.4 10,200 370 5,610 1.57
mahogany 0.45 0.45 11,500 - 6,780 1.5
teak 0.55 0.55 14,600 - 8,410 1.55
lignumvitae 1.05 1.06 - - 11,400 -
*tangential
all properties are for oven dry samples
Table 1: Mechanical Properties of Common Woods used
in the Marine Environment (from Refs. 1 & 2)
13.3 Typical Uses in the Marine Environment
Apart from recreational craft, wood is still commonly used in fishing vessels and other
small commercial vessels worldwide. Its natural buoyancy, abundancy, and ability to be
worked with simple tools lend itself to marine construction in developing countries. It has
significant disadvantages for large commercial and naval applications however,
including:
 Low modulus of elasticity
 Low fire resistance
 Susceptibility to biodegradation
 Splintering
In the US Navy, wood is still specified for well-decks of amphibious ships and as a
damage control material. Figure 5 shows “4x4’s” (four inch square lumber) stored on
DDG-53. Typical species are White oak, Eastern pine and Douglas fir.
Figure 5: Damage Control Material on USS Ramage (DDG-51)
13 - 5
13.4 Structural Analysis Equations
Because wood is essentially transversely isotropic, equations governing stress and strain
are different from isotropic materials. The most accurate way to analyze a wood structure
would be to use the same methods as for composites. In the case of plywood this is
especially true as it is a wood laminate. The mechanics approach for plywood and
composites uses the plane stress assumption which assumes that stresses through-the-
thickness are zero.
In general practice though, as lumber is typically used in plank form, one axis dominates
the properties and the stresses in the other two directions are assumed zero. Little
additional error is created in using the resulting isotropic beam formulas.
For example, in axial loading, the deflection is given by:
PL
 
AE
where P is the axial force, L is the length, A is the cross-sectional area and E is the
Young’s modulus of the axial direction.
In bending the deflection of a straight beam of constant cross-section is:
k bWL3 k sWL
  
EI GA
where the bending and shear constants (kb and ks) are based on the loading and boundary
conditions (see Table 2), W is the total beam load acting perpendicular to the beam
neutral axis, L is the span, I is the moment of inertia, G is the shear modulus and A’ is a
modified shear area.
bh 3 d 4
I for rectangular beams, I  for circular beams
12 64
A  5 6 bh for rectangular beams, A  9 40 d 2 for circular beams
Table 2: kb and ks for beam deflection
13 - 6
Again, due to the beam assumption, stress formulas for wood are the same as those of
isotropic materials:
P My V
  for axial loads, and   for bending, and  max  k for shear,
A I A
where k is 3/2 for rectangular cross-sections and 4/3 for circular cross-sections
Typical lumber dimensions lead to the possibility of buckling under compressive loading.
The critical buckling stress is:
 2 EL b
 cr  2
where r is the radius of gyration, r  for a rectangular section with b as
 L 12
 
r
d
the smaller dimension and, r  for a circular section.
4
As wood will creep under long-term load it is customary to design to half the allowable
deflection or stress.
13.5 Fastening
Wood is relatively easy to join compared to other materials. And as with other materials,
joining can be broadly classified into two methods: bonded joints and mechanically-
fastened joints. The trade-off between the two are that bonded joints are typically 1 ½ - 3
times stronger, but take longer to assemble and are regarded as permanent.
Due to fastener degradation, adhesive bonding is required for long-term, efficient joints
in the marine environment. Adhesives transfer load from one member (called an
adherend) to another, through the adhesive. In most cases the load transfer is designed as
a shear load, as adhesives are typically much stronger in shear than in tension. Factors
involved in adhesively bonded joints include:
 Type of wood
 Surface quality
 Adhesive
 Bonding process
 Joint geometry
 Service environment
All of the woods noted above can be glued, although teak, oak and lignumvitae all
require specialized methods for satisfactory results. Figure 6 shows typical edge and axial
joints and Figure 7 shows corner joints.
13 - 7
Figure 6: Edge (A-plain, B-tongue and groove) and

Axial (A-plain, B-scarf, D-finger) Bonded Wood Joints
Figure 7: Corner Bonded Wood Joint
Mechanically-fastened joints in wood use a staggering variety of fasteners. (Just walk

down an aisle at Home Depot.) Marine joints however, rely primarily on two types: wood
screws and bolts. In general, screws are used for lightly loaded, watertight connections.
Bolts are used for high-load conditions, but are significantly more difficult to make
watertight. Thin planking on small craft is occasionally fastened with copper rivets.
An important corrosion issue is the reaction of wood fasteners. Traditionally the way to
avoid fastener corrosion was to use wood trunnels (dowels). In the US the preferred
trunnel material was Black Locust. These are relatively low-strength however and silicon
bronze and monel are preferred. If cost is an issue stainless or galvanized steel is used,
although service life is reduced. If cost is no object then titanium can be used.
13.6 Control of Degradation
Under ideal conditions wood structures can survive for centuries. The primary reasons
wood structures fail include fungi, insects, bacteria and marine borers. In some cases
large structures have been known to fail within a month of infestation.
Fungi (rot) requires moisture, oxygen and mild temperatures and is considered the largest
source of marine decay. To get rot the moisture content must be above the fiber
saturation point (30%), and the temperature must be between 50 and 90 oF. The best
prevention methods include ventilation, reduction of leaks, and preservatives. Note that
most preservatives are toxic and make painting difficult.
Insect attack includes termites, beetles and carpenter ants. Rarely are marine structures
infected by insects. Some vessels stored out of the water, or along infested docks, have
become infected, and alongshore structures are susceptible.
13 - 8
Damage by marine borers is worldwide, and can occur in seawater, brackish and even
fresh water. Attack can be rapid, with pilings destroyed in less than six months. The
primary borers are the toredo (shipworm) which can grow to lengths of four feet in ideal
conditions, and various species of bankia, pholads, and limnoria.
In all cases preservation techniques attempt to either impregnate the wood with toxic
chemicals or provide an impregnable barrier. Preservatives include copper compounds
and creosote tars. The allowable (by EPA or OSHA) versions change frequently. The US
Navy currently favors a combination of epoxy/glass coatings and copper antifouling paint
for vessels and concrete coverings for stationary structures.
13.7 Plywood
Plywood is a glued-up lamination of many layers of wood. Figure 8 shows typical 3, 4,

and 5-ply construction. Compared to lumber, plywood is more stable and uniform. It’s
axial and transverse properties are similar and it is available in wider widths. Its primary
disadvantage is that it is not easily formable. Compound curves (bending about two axes)
is not possible. If formed in-place into a curved shape (such as a boat hull) it is
commonly called “cold-molded” construction.
Figure 8: Plywood construction.
The individual plies are held together with various glues. Exterior, marine, and aircraft
grade plywood are required to use waterproof grade. The difference between the three
grades are the number of flaws in each ply, and the variation is quite wide. In general, the
price is reflective of the ply quality and wood species. Many marine applications can use
exterior-grade. Aircraft grade is rarely justified except in weight critical structures, and
even then may not be desirable as many woods used in aircraft ply are not rot resistant.
Recommended References:
1. Wood: A Manual for it Use as a Shipbuilding Material, 4 Volumes, Bureau of Ships,
Department of the Navy, 1957-1962
2. The Encyclopedia of Wood, Revised Edition, Sterling Publishing, New York, 1989
(Also called, Wood Handbook: Wood as an Engineering Material by the US Government
Printing Office)
13 - 9
CHAPTER 14
CONCRETE
14.1 What is Concrete?

14.2 Variations of Concrete
14.3 Mechanical Properties of Concrete:
14.4 Design Options
Concrete has a wide range of uses in the marine environment, including applications for
decks, buildings, seawalls, piles, pipelines, ships/barges.
Concrete Applications
14.1 What is Concrete?
A typical concrete mixture has 5 main components:

Portland Cement 7-15%
Water 14-21%
Air 0.5-8%
Fine Aggregate 24-30%
Coarse Aggregate 31-50%
14 - 1
Advantages Disadvantages
Low Cost Low tensile strength
High Durability Heavy
High Fatigue Resistance Develops strength slowly
Easy to form/Shape Dimensions can change with time
Concrete ≠ Cement!
1. Concrete is a composite material composed of coarse granular material
(Aggregate) within a matrix of cement paste (Portland Cement and water)
2. Cement is a mixture of lime, silica, alumina and iron oxide that forms the
binder for concrete.
3. Portland cement is the most common type of cement, and has five types:
a. Type I: Common/General Purpose - used for general construction,
when not significant contact with soils or groundwater is expected.
Examples:
b. Type II: Moderate Sulfate Resistance - for concrete in contact with soil
or groundwater and for large structures that may be subject to high
temperatures.
c. Type III: High Early Strength - develops significant compressive

strength in ~7 days (same as 28 day strength for Types I and II). Often
used when laying concrete in cold weather.
14 - 2
d. Type IV: Low Heat of Hydration - low heat of hydration as concrete

strengthens. Often used for dams and massive structures.
e. Type V: Sulfate Resistant - prevents sulfate attack in highly alkaline

soil or seawater.
14.2 Variations of Concrete
Reinforced Concrete - since the tensile strength of plain concrete is 10-15 times lower
than its compressive strength, reinforcement can be added to carry the tensile load.
1. Examples of reinforcement: Rebar (steel rods), wires, mesh, glass fibers,
carbon fiber, etc.
Rebar
Glass Fiber Reinforcement
14 - 3
Mesh Reinforcement
Pre-stressed Concrete: place concrete under compression before supporting any loads
(besides its own dead weight).
1. The process involves tensioning steel “tendons” that are located within or
adjacent to the concrete form.
2. This improves structural capacity and serviceability.
3. Applications: long-span bridges, high rise buildings, names, foundation
systems
Prestressed Concrete
14 - 4
Prestressed concrete
Post-tensioned concrete: tendons are tensioned after the concrete has been cast.
Post Tensioning
Post Tensioning
14 - 5
Shotcrete: Concrete is reinforced with 1-3 inch long glass, fibers, etcetera, and is sprayed
through a nozzle/hose.
1. Wet and dry mix variations
2. Gunite is another variation that is a dry mix of sand and cement
3. Many applications in the mining industry
Shotcrete
Grout: a very fluid form of concrete

1. Used to fill gaps
2. The mixture includes water, cement, and fine aggregate (sand) - no coarse
aggregate.
3. Applications: tiling
Grout
14 - 6
Tremie concrete: for underwater concrete placement.

1. Process involves placing concrete below the water level using a pipe. The
lower end of the pipe is submerged in fresh concrete, and rising concrete
from the bottom will displace water without washing out the cement content.
Tremie Concrete
14.3 Mechanical Properties of Concrete

Strength increases with time from 1-6 months after placement. It is often tested at 3, 7,
and 28 days. The compressive strength varies from ≈ 3-14 ksi The tensile strength is
about 10% of compressive strength.
The following empirical Equation can be used to estimate elastic modulus based on
compressive strength:
𝐸 = 𝛾 . 33 𝜎
Where,
𝐸 = elastic modulus (psi)
𝜎 = compressive strength (psi)
𝛾 = unit weight of concrete (54-150 lb/cu ft)
14 - 7
Concrete Compression Test
14.4 Design Options
The tensile strength of concrete is a significant drawback. How can we increase the
tensile strength of a structure?
Add reinforcement
Pre/Post Stress
High-Strength Mix
To increase the durability of a concrete structure, the following may be considered:

Minimum 2” cover for rebar to prevent corrosion of steel
Low water/cement ratio (ph increases, permeability decreases)
Coatings for rebar or the exterior of the concrete
Admixtures that promote air entrainment – pores for capillary water in
freeze-thaw cycles.
14 - 8
CHAPTER 15
COMPOSITES
15.1 Fiber Reinforced Polymers

15.3 FRP Fiber Orientation
15.4 FRP Fabrication
15.5 Vacuum Bagging
15.6 Mechanical Properties of FRP
A composite is simply a thing made up of different elements. Reinforced concrete, for

instance, is a composite of steel rebar and concrete. They used to build “composite”
wooden boats, with iron frames. Typically when the word “composite” is used today, it
relates to some type of fiber reinforced polymer.
15.1 Fiber Reinforced Polymers
Fiber Reinforced Polymers are a composite material made of a polymer matrix that is
reinforced with fibers.
The matrix is a resin that holds the fibers together. Three commonly used resins are:
 Epoxy – Strong and flexible to resist cracking. This resin is good for FRP-wood
construction, in which the swelling and shrinking of the wood can cause delamination
 Polyester – This is very low cost, but somewhat more brittle than epoxy.
 Vinylester – This is in between epoxy and polyester for cost and performance.
The reinforcing fibers also vary, including

 Glass – inexpensive “fiberglass” Available in different types, like E-glass (standard) and S-
glass (with higher strength)
 Carbon Fiber – Very high tensile strength, and high modulus of elasticity (stiffness)
 Kevlar – Similar tensile strength to carbon fiber, but more deflection
The matrix, or resin that holds the fibers together is made of polymers. Polymers are
“repeating chains of organic material.” A mer is the basic unit and there are many of
them in a repeating structure. Natural polymers include wood, rubber, wool, leather and
methane. Man-made polymers are made from small organic molecules and include
plastic, rubber and fibers. Figure 1 represents a simple polymer. Because of the
connection details, the polymer can bend in 3 dimensions.
15 - 1
Figure 1: Notional Hydrocarbon
Polymers form very large molecules because of the repeating structure. The carbon atom
is the backbone of the polymer. The average molecular weight of a polymer changes the
melting temperature. For instance:
g
M  100 Liguid/Gas
mole
g
M  1000 Waxy solids
mole
g
M  10 4 to10 6 Solids (high polymers)
mole
Figure 2 shows a typical polymer chain. The intertwining gives it elastic properties.
Figure 2: Typical Polymer Chain
15 - 2
Unlike metals, polymers do not create completely crystalline structures. There may be
regions of crystallinity, along with amorphous regions. These variations affect how hard
the structure is. Figure 3 shows the structure of a polymer solid.
Figure 3: Crystalline and Amorphous structures of Polymers
15.3 FRP Fiber Orientation
can be varied, meaning composites are not isotropic. The strength can be increased in the
direction needed by the design. Fiber can come in the following orientations:
 Chopper Gun – These are short fibers that are sprayed onto the inside of the mold along
with the resin. This is the lowest strength-to-weight ratio, but the cheapest (many
production boats are made this way)
 Chopped Strand Mat – This is like a roll of cloth, but with randomly oriented strands.
 Bi-axial – comes in a roll with two directions. Can be laid up “on the bias” at 45 degrees
as well.
 Uni-axial – Strength in one-direction
15 - 3
Figure 4: Laying up a fiberglass boat with a chopper gun (Fibers and resin are sprayed
on)
Figure 5: Chopped Strand Mat
Figure 6: Woven Roving – Bi-Axial Figure 7:Uniaxial Fibers
15.4 FRP Fabrication
For production boats, pools, hot tubs, etc, there is generally a female mold with a smooth
interior surface.
 They first spray it with a release agent to keep the hull from sticking to the mold.
15 - 4
 Then they spray on “gel coat” which is a hard shiny surface that you see on the outside
of the boat or the inside of the hot tub.
 If the boat is cheap (like most) then they get the chopper gun and spray. The inside will
then be very rough.
 More expensive hand layup requires a layer of chopped strand mat first, so that the bi-
axial fibers don’t “print through.” Then generally they alternate woven roving and
chopped strand mat for each subsequent layer (this keeps the woven parts from
bunching up and leaving voids).
 In specialized custom design boats, like racing yachts, they will use uniaxial fibers to
increase the strength in particular directions. This requires careful design and
understanding of the loads.
 Hulls sometimes include a core material so that there is strong fiberglass on the outside
and the inside, with a core to space them apart. This increases the moment of inertia
and hence the stiffness and strength.
Figure 8: Boat Molds in Shop
15.5 Vacuum Bagging
Once there is enough resin to bond the fibers together, more resin doesn’t add strength. It
costs money to buy and adds weight. As a result, efforts are sometimes taken to reduce
the amount of resin in a particular layup. Vacuum bagging is used to put a suction over
the layup to compress it and reduce the amount if resin. This works well, but increases
the difficulty of the job. For very thin layups, it can sometimes result in pinhole leaks.
15 - 5
Figure 9: Vacuum Bagging (From Jamestown Distributers)
Figure 10: Small Vacuum Bagging Figure 11:Vacuum Bagging a Wind Turbine Blade
15 - 6
15.6 Mechanical Properties of FRP
The matrix and the fibers have different strengths and elasticity. When the forces are in
the same direction as the fibers, assume isostrain – the fibers and the matrix elongate the
same amount.
Figure 11: Elongation of Uni-axial Fiber
Using isostrain and the volume fraction of the fibers and matrix, we can use the Rule of
Mixtures
𝐸 =𝐸 𝑉 +𝐸 𝑉
Where, 𝐸 , 𝐸 , 𝐸 is the modulus of elasticity of the FRP, the matrix and the fibers.
𝑉 and 𝑉 are the volume fractions of the matrix and the fibers.
𝑉 +𝑉 =1
The ratio of the force carried by the fibers and the force carried by the matrix
𝐹 𝐸 𝑉
=
𝐹 𝐸 𝑉
Where, 𝐹 and 𝐹 = the force carried by the fibers and the matrix, in N or lb. The total
load is then,
𝐹 =𝐹 +𝐹
15 - 7
What if the loads are in the transverse direction? The fibers do not have significant
strength in this direction, so it is not isostrain. Instead, assume isostress - -the same force
per unit area.
𝐸 𝐸
𝐸 =
𝐸 𝑉 +𝐸 𝑉
Example 1a: A continuous uniaxial glass FRP has 40% by volume fibers (𝐸 = 10 × 10
psi) in a polyester resin (𝐸 = 0.5 × 10 psi). Compute the longitudinal elastic modulus
of the composite.
Answer
𝐸 =𝐸 𝑉 +𝐸 𝑉
𝑉 = 0.4 so 𝑉 = 1 − 𝑉 = 0.6
𝐸 = 0.5 × 10 𝑝𝑠𝑖 × 0.6 + 10 × 10 𝑝𝑠𝑖 × 0.4 = 4.3 × 10 𝑝𝑠𝑖
Example 1b: If the cross-section area of the composite in the previous example of 0.4
square inches and a longitudinal stress of 7250 psi is applied, find the load carried by the
fibers and by the matrix.
Answer:
15 - 8
Example 1c: For the previous example, what is the strain sustained by each of the fibers
and the matrix?
Answer:
Example 1d: What if we load in the transverse direction?
Solution:
15 - 9
Example 2: For a uniaxial FRP that has a 50-50 ratio by volume of fibers (𝐸 =
10 × 10 psi) in a polyester resin (𝐸 = 0.5 × 10 psi).
a. Find the strain felt by the composite fibers and matrix if the applied force is 4500 lb
and the total cross-section area is 0.6 sq inches.
b. Find the load carried by the fibers in the matrix.
c. The longitudinal elastic modulus of the composite
d. The transverse elastic modulus of the composite.
15 - 10
APPENDIX
CATHODIC PROTECTION DESIGN

from Swain Classnotes (1996)
STRUCTURE
Metal, Design Life, Dimensions, Coatings, Other
LOCATION
Environmental Conditions, Other Structures
C.P. CRITERIA
Potential for Cathodic Protection
C.P. CURRENT DEMAND

Initial, Mean, Final
C.P. TYPE
Impressed Current, Sacrificial Anode
ANODE AND HARDWARE SELECTION

Current Output, Design Life, Placement
COST AND IMPLEMENTATION
Appendix: Cathodic Protection Design - 1

Table of Contents
1.0 Introduction .................................................................................................................................. 3
2.0 Structure ...................................................................................................................................... 3
3.0 Location ....................................................................................................................................... 3
4.0 CP Criteria ................................................................................................................................... 4
4.1 Potential Values....................................................................................................................... 4
4.2 300 mV Shift ............................................................................................................................ 5
4.3 100 mV Shift ............................................................................................................................ 5
4.4 E-log-I Curve ........................................................................................................................... 6
4.5 Anodic Current Discharge Points ............................................................................................ 6
5.0 Cathodic Protection Current Demand .......................................................................................... 6
5.1 Recommended Practice RP B401, Det Norske Veritas .......................................................... 6
5.1.1 Uncoated Steel ............................................................................................................... 7
5.1.2 Coated Steel ................................................................................................................... 7
5.1.3 Pipeline Coatings ............................................................................................................ 9
5.1.4 Concrete ......................................................................................................................... 9
5.2 Current Requirements for Pipelines in Soils of Different Types ............................................ 10
5.3 Current Requirements for Ship Protection ............................................................................ 11
6.0 Cp Type ..................................................................................................................................... 12
7.0 Anode Selection ......................................................................................................................... 13
7.1 Anode Resistance to Ground ................................................................................................ 16
7.2 Anode Ground Beds .............................................................................................................. 18
7.3 Anode Current Output ........................................................................................................... 18
7.4 Anode Size, Weight, Number, Distribution and Design Life .................................................. 18
8.0 Cost and Implementation ........................................................................................................... 21
8.1 Oil Platform Example ............................................................................................................. 21
8.1.1 Structural Details .......................................................................................................... 21
8.1.2. Current Demand ........................................................................................................... 21
8.1.3. Sacrificial Anode Design for Uncoated Structure ......................................................... 22
8.1.4. Sacrificial Anode Design Coated Structure .................................................................. 26
8.1.5 Impressed Current Anode Design ................................................................................ 27
8.2 Ship Hull Protection ............................................................................................................... 29
2
1.0 INTRODUCTION
Metallic structures in contact with water, soil, concrete, and moist air are subject to
corrosion. Cathodic protection (CP) is one of the few methods that successfully
mitigates corrosion. It can be applied in any situation where the environment
surrounding the metal acts as a conductor for electric current. It has been successfully
applied to offshore structures, ships, boats, propellers, moorings, pipelines, storage
tanks, piers, jetties, bridges, aquaria, instrumentation etc.
This handout is designed as an introduction to CP design. As such, it does not cover all
aspects of the subject. Therefore, the student should realize the limitations of his/her
knowledge and consult other literature or experts in the field when necessary.
2.0 STRUCTURE
CP design begins with a thorough understanding of the structure to be protected. This
includes the following information:
 Metal type(s)
 Operating conditions
 Dimensions and surface area
 Coatings
 Data from previous structures and CP systems
 Design life
3.0 LOCATION
The environmental conditions are determined by the location of the proposed
installation. Factors such as climate, electrolyte conductivity and chemistry, physical
loading, and biological activity, all impact CP requirements. These factors are generally
allowed for in the CP current demand and polarization potential criteria.
3
4.0 CP CRITERIA
Potential measurements are the most commonly used criteria to ascertain the level of
CP afforded to metals and alloys. CP potential values vary according to the metal and
the environment. Corrosion is likely to occur at potentials which are more positive than
the protected value. Damage may also occur if the metal is overprotected (i.e. the
potential too negative). The most common error associated with potential
measurements is a result of IR drop. This is the displacement of measured metal
potential due to current flow through the electrolyte. High electrolyte resistivity and high
current densities can cause significant differences between the measured and actual
metal potential.
4.1 Potential Values

The measurement of potential with respect to a standard reference electrode is
probably the most common method of evaluating the degree of cathodic protection
afforded to a structure. Typical cathodic protection potentials for commonly used metals
ref. Ag/AgCl reference electrode (seawater) are provided in Table 4.1. A more detailed
summary of protection potentials for steel in seawater is provided in Table 4.2.
Table 4.1 Approximate freely corroding and protected potentials of metals in

seawater (may vary according to velocity and conditions).
Metal or Alloy Freely Corroding Protected Potential
Potential (V) ref. Ag/AgCl
(V) ref. Ag/AgCl
316,304 Stainless (passive) -0.10 -0.75
Copper Alloys -0.35 -0.70
316,304 Stainless (active) -0.50 -0.75
Steel -0.60 -0.80
Aluminum Alloys -0.75 -1.00
Zinc/Aluminum Anodes -1.05
Magnesium Anodes -1.50
4
Table 4.2 Potential values for corrosion and protection of steel in seawater
V ref. Ag/AgCl Condition V ref. Zn
Heavy Corrosion
-0.60 Freely Corroding Steel +0.50
-0.70 Some Protection +0.40
-0.80 Cathodic Protection +0.30
-0.90 Some Over +0.20
-1.00 Protection +0.10
-1.10 0.00
-1.20 Over Protection -0.10
-1.30 May Cause -0.20
-1.40 Paint Blistering and Flaking -0.30
-1.50 -0.40
4.2 300 mV Shift

The NACE Standard, RP-02-85 states that a minimum negative (cathodic) voltage shift
of 300mV, produced by the application of protective current should provide CP to iron
and steels. The voltage shift is measured between structure surface and a stable
reference electrode contacting the electrolyte. This criteria does not apply to structures
in contact with dissimilar metals.
4.3 100 mV Shift

The NACE Standard, RP-02-85 states that a minimum negative (cathodic) voltage shift
of 100mV measured between the structure surface and a stable reference electrode
contacting the electrolyte should provide CP to iron and steel. This polarization voltage
shift is determined by interrupting the protective current and measuring the instant off
and polarization decay. The instant off value is obtained immediately following the
interruption of the CP current. The voltage shift is equivalent to the IR drop created by
the CP current and electrolyte resistance. The polarization decay is measured as the
change in potential over a period of time from the instant off value.
5
4.4 E-log-I Curve
The NACE Standard, RP-02-85 states that a structure-to-electrolyte voltage at least as
negative (cathodic) as that originally established at the beginning of the Tafel segment
of the E-log-I-curve should provide CP to iron and steel. This structure-to-electrolyte
voltage shall be measured between the structure surface and a stable reference
electrode contacting the electrolyte at the same location where voltage measurements
were taken to obtain the E-log-I curve.
4.5 Anodic Current Discharge Points

The NACE Standard, RP-02-85 states that a net protective current from the electrolyte
into the structure surface as measured by an earth current technique applied to
predetermined current discharge (anodic) points of the structure should provide CP to
iron and steel.
5.0 CATHODIC PROTECTION CURRENT DEMAND

The cathodic protection current demand is the amount of electricity required to polarize
the structure to a level that meets the criteria described in Section 4. This may be
obtained from a trial polarization of the structure at the installation site, a trial
polarization of a metal test coupon at the installation site, or from conservative
estimates obtained from historical information obtained from previous structures
operating under the prescribed conditions.
For planning and design purposes, it is often possible to rely on conservative estimates
provided by recommended practice. There are several sources for this information.
The most current one is Recommended Practice RP B401, Cathodic Protection Design,
Det Norske Veritas Industri Norge AS, 1993.
5.1 Recommended Practice RP B401, Det Norske Veritas

The CP current densities are calculated for different environmental conditions and
conditions of the steel (i.e. uncoated, coated, concrete reinforcing steel, pipeline).
6
5.1.1 Uncoated Steel
Three design current densities are given: initial, final, and average.
Initial This is the current density required to effect polarization of the initially
exposed bare steel surface. It assumes some atmospheric rusting and/or
millscale. The initial current density is higher because of lack of
calcareous scales (cathodic chalks). A proper initial current density
enables rapid formation of protective calcareous scales.
Final This is the current density required to protect the metal surface with
established marine growth and calcareous layers. It takes into account
the current density required to repolarize the structure in the event of
removal of these layers by storms, cleaning operations etc.
Average This is the anticipated current density required once the cathodic
protection system has reached its steady state. The average or
maintenance current density is used to calculate the minimum mass of
anode material required to protect the structure throughout the design life.
Table 5.1 Initial, final, and average current densities for various climatic
conditions and depths (climatic conditions are based on yearly range
of average surface water temperatures).
Design Current Densities (A/m2)
Tropical >20oC Sub-Tropical 12o-20oC Temperate 7o-12oC Arctic <7oC
Depth Initial Final Average Initial Final Average Initial Final Average Initial Final Average
(m)
0 - 30 0.150 0.090 0.070 0.170 0.110 0.080 0.200 0.130 0.100 0.250 0.170 0.120
>30 0.130 0.080 0.060 0.150 0.090 0.070 0.180 0.110 0.080 0.220 0.130 0.100
5.1.2 Coated Steel

The use of coatings on steel dramatically reduces the current demand on the cathodic
protection system. This can save on the cost and structural weight associated with
sacrificial anode systems. The CP current demand of a coated offshore jacket may be
estimated by multiplying the bare steel current demand by a coating breakdown factor
(fc). The coating breakdown factor does not allow for mechanical damage to paint
7
coatings. These areas are treated as bare metal surface. For CP design purposes the
average and final coating breakdown factors for a design life of t r years are as follows:
tr
f c (average)  k1  k 2
2
f c ( final )  k1  k 2 t r
When the design life of the CP system exceeds that of the coating system then fc
(average) is calculated as follows:
f (average)  1 
1  k  1
2
c
2k 2 t r
If the calculated value exceeds 1, then fc = 1 shall be applied to the design.
Table 5.2 Constants (k1 and k2) for calculation of paint coating breakdown
factors.
Category Description k1 k2 k2
0-30m >30m
I One layer of primer coat, about 50 m nominal 0.10 0.10 0.05

DFT.
II One layer of primer coat, plus minimum one layer 0.05 0.03 0.02
of intermediate top coat, 150 - 250 m nominal
DFT.
III One layer of primer coat, plus minimum two 0.02 0.015 0.012
layers of intermediate/top coats, 300 m nominal
DFT.
IV One layer of primer coat, plus minimum three 0.02 0.012 0.012
layers of intermediate/top coats, 450 m nominal
DFT.
8
5.1.3 Pipeline Coatings
The coating breakdown factors as shown in table 5.2 apply equally to both buried and
non-buried pipelines. It is assumed that coatings and field joint systems have been
chosen to be compatible with the maximum design temperature of the pipeline.
For pipelines with the following coating systems, another coating breakdown factor is
calculated.
 asphalt + concrete weight coating
 fusion bonded epoxy + adhesive + polyethylene or polypropylene
 polychloroprene rubber
 equivalent coating systems based on an inner layer dedicated to corrosion
protection and one or more outer layers for mechanical protection.
This is as follows:
f c (average)  0.05  0.002(t r  30)
f c ( final )  0.07  0.004(t r  20)
5.1.4 Concrete
It is now recognized that cathodic protection of concrete reinforcing steel is necessary to
ensure the long term integrity of the structure. Also, any CP system designed to protect
metallic appendages and components must be designed to allow for current drain from
CP to the reinforcement. The cathodic current density is determined by transport of
oxygen to the steel by capillary action of pore water driven by evaporation in the
atmospheric zone and internal dry compartments. The current densities are, therefore,
dependent on depth and climatic conditions.
9
Table 5.4 Design current densities for concrete reinforcing steel.
(NOTE: design currents refer to the area of the reinforcing steel)
Design Current Densities (A/m2)

Depth (m) Tropical >20oC Sub-Tropical Temperate 7o- Arctic <7oC
12o-20oC 12oC
5 to -10 0.0030 0.0025 0.0015 0.0010
<-10 0.0020 0.0015 0.0010 0.0008
5.2 Current Requirements for Pipelines in Soils of Different Types

The current demands for steel pipelines are determined by the soil type (conductivity,
pH, moisture, temperature) and the condition of the steel (coating type). An example of
typical CP current demand for a pipeline with different coating conditions is presented
as follows.
Table 5.3 Range of current required to protect 10 miles of 36" diameter pipe in
soil with average resistivity of 1000 ohm-centimeters. Current
required is that needed to cause a 0.3 Volt drop across the effective
resistance between pipeline and remote earth. [from A.W.Peabody,
Control of Pipeline Corrosion,NACE, 1967]
Effective Coating Resistance in Ohms Current Required, Amps
for One Average Square Foot
Bare Pipe (minimum 1 mA/ft2) 500
10,000 14.91
25,000 5.964
50,000 2.982
100,000 1.491
500,000 0.2982
1,000,000 0.1491
5,000,000 0.0298
Perfect Coating 0.000058
10
5.3 Current Requirements for Ship Protection
Information with regard to current density requirements for ship hull protection is limited.
One source of information is the Technical and Research Report R-21, Fundamentals
of Cathodic Protection for Marine Service, The Society of Naval Architects and Marine
Engineers, January 1976. It must be remembered that this was compiled before the
development of modern day bottom coatings. It may, therefore, be better to use the
DNV practice for coated steel and to include an allowance for damaged surfaces.
Table 5.5 Protective current densities for ships. [from Technical and Research
Report R-21, Fundamentals of Cathodic Protection for Marine
Service, The Society of Naval Architects and Marine Engineers,
January 1976]
Specific Area Current Density, mA/m 2
External Hull 22-54
Rudders (Coated and for velocities not exceeding 5 knots. 490
Current demand maybe 3 or more times greater underway)
Propellers (For velocities not exceeding 5 knots. Current 150 -170
demand maybe 3 or more times greater underway)
Coated Tanks 11
Segregated Ballast 150
Washed Cargo / Clean Ballast 130
Dirty Ballast Tanks 86
11
6.0 CP TYPE
The CP type determines how the cathodic current is supplied to the structure. CP can
be applied by either an impressed current system or by a sacrificial anode system.
Impressed current CP systems use an external DC current source and a variety of
anode materials to supply the cathodic current. Sacrificial anode CP systems generate
the cathodic current from the corrosion of metals less noble than the metal to be
protected.
The choice between impressed and sacrificial cathodic protection depends many factors
and may be just personal preference. There are, however, situations where one or the
other provides the correct choice. The advantages and disadvantages of each type of
CP system are described in Table 6.1.
Table 6.1 Advantages and disadvantages of impressed current and sacrificial

anode CP systems.
Impressed Current Sacrificial Anodes
Advantages
Variable control of current and potential Self contained
Can be automated Can be self adjusting
Light weight and fewer anodes Polarity of connections always correct
Varied anode geometry Needs no supervision
Long life with inert anodes Simple to install
Disadvantages
Complex installation and maintenance Expensive method of generating electricity
Requires external power source No variable control
Anodes require dielectric shields Anodes add weight
Anodes may be damaged Anodes have finite life
Probability of stray current corrosion Small lead resistance reduces current
12
7.0 ANODE SELECTION
Anodes, for both impressed current and sacrificial anodes, are selected according to
their size and chemical composition. This determines the current output and design life.
Specifications for impressed current anodes are provided in Table 7.1 and for sacrificial
anodes in Tables 7.2 and 7.3.
Table 7.1 Impressed current anodes.

Anode Material Recommended Maximum Consumption Comments
Current density Voltage, V Rate, g/A-yr
2
A/m
Scrap Steel Varies - 200 - 9,000 Difficult life
prediction
Graphite 10 - 30 - 450 Very brittle
Silicon-Chromium- 10 - 100 - 90 - 250 Very brittle
Cast Iron
Lead-Silver 250 - 500 - 30 - 90 Heavy, Poor
mechanical
properties
Lead-Platinum 100 - 2 - 60
Magnetite 10 - 500 - 40 Very Brittle
Platinized 250 - 700 9 0.01 5 m thick Pt film
Titanium provides 10 year
life
Tantalum provides 10 year
life
Columbium provides 10 year
life
Lithium-Ferrite 15 - 2000 9.7 1-2 Lightweight and
Ceramic tough
13
Table 7.2 Sacrificial anode types and use.
Anode Preferred Use Approx. Potential
Volts ref. Ag/AgCl
Magnesium, High Soils with resistance > 2000 -cm -1.75
Potential
Magnesium, Standard Soils with resistance < 2000 -cm, and in aqueous -1.50
environments with controllers if necessary
Zinc, Hi-Amp Seawater, brackish water, saline mud. Temps < -1.05
60oC
Zinc, Hi-Purity Underground, fresh water, and saline environments -1.05
> 60oC
Galvalum I Submerged seawater, max. temp 25 oC -1.05
Galvalum II Saline mud -1.04
Galvalum III Seawater, brackish water, saline mud -1.10
Reynode -1.05
Al-Sn-In Alloy -1.05
Table 7.3 Sacrificial anode properties.

Property Anode Material Type
Magnesium Zinc Galvalum 1 Galvalum II Galvalum
III
Density, kg/m3 1940 7130 2700 2700 2700
Electrochem Equiv, g/coulomb 0.126E-3 0.339E-3 0.093E-3 0.093E-3 0.093E-3
Theoretical Ah/Kg 2,205 819 2,987 2,987 2,987
Current Efficiency % 0.55 0.95 0.95 0.57 0.85
Actual Ah/Kg 1,212 780 2,830 1,698 2,535
Actual Kg / Amp / Year 7.95 11.25 3.10 5.16 3.46
Potential V, ref. Ag/AgCl -1.75 -1.05 -1.05 -1.04 -1.10
14
Table 7.4 Sacrificial Anode Composition.
Percent of Total Weight
Mg Zn Al Cd Cu Fe Hg In Mn Ni Pb Si
Magnesium, rem 0.01 0.02 0.03 0.50 - 0.001
High Potential 1.30
Magnesium, rem 2.5 - 5.3 - 0.05 0.003 0.15 0.003 0.30
Standard 3.5 6.7
Zinc, Hi-Amp rem 0.1 - 0.025 - 0.005 0.006
0.4 0.060
Zinc, Hi-Purity rem 0.003 0.0014 0.003
Galvalum I 0.35 - rem 0.035 - 0.14 -

0.48 0.048 0.21
Galvalum II 3.5 - rem 0.035 -
5.0 0.048
Galvalum III 2.8 - rem 0.01 - 0.08 -
3.5 0.02 0.12

7.1 Anode Resistance to Ground
The current output from an anode is determined by its shape, electrolyte resistance, and
driving potential. The shape and electrolyte resistance determine the anode resistance
to ground which is calculated from standard anode resistance formulae. The most
commonly used formulae are presented in Table 7.4 and seawater conductance values
in Table 7.5.
Table 7.4 Anode resistance to ground formulae.

Anode Type Resistance Formula
Long Slender stand-off L  4r    4L  
R  ln   1 (Modified Dwight)
2L   r  
Long Slender stand-off L  4r   2
   2 L   r 
2
 r  r  
R ln  1 1      1  
2L   r   2L   2 L  2L  
    
Long flush mounted 
R (Lloyds)
L  4 x width and thickness 2S
Short flush-mounted, bracelet 0.315 
R (McCoy)
and other flush mounted shapes A
where: R is anode resistance, ohms

 is electrolyte resistivity, ohm-cm
L is anode length, cm
ab
S is the mean of the anode sides 
2
anode cross - sectional area
r is equivalent radius, cm, 

2
A is the exposed surface area of anode, cm

Table 7.5 Specific Conductance of Seawater
Note: Resistivity, , is the reciprocal of conductance. Tabled values are expressed in in
(-1-cm-1).
Chlorinity, ppt Temperature, oC

0 5 10 15 20 30
1 0.001839 0.002134 0.002439 0.002763 0.003091 0.003431
2 0.003556 0.004125 0.004714 0.005338 0.005971 0.006628
3 0.005187 0.006016 0.006872 0.007778 0.008702 0.009658
4 0.006758 0.007845 0.008958 0.010133 0.011337 0.012583
5 0.008327 0.009653 0.011019 0.012459 0.013939 0.015471
6 0.009878 0.011444 0.013063 0.014758 0.016512 0.018324
7 0.011404 0.013203 0.015069 0.017015 0.019035 0.021121
8 0.012905 0.014934 0.017042 0.019235 0.021514 0.023868
9 0.014388 0.016641 0.018986 0.021423 0.023957 0.026573
10 0.015852 0.018329 0.020906 0.023584 0.026367 0.029242
11 0.017304 0.020000 0.022804 0.025722 0.028749 0.031879
12 0.018741 0.021655 0.024684 0.027841 0.031109 0.034489
13 0.020167 0.023297 0.026548 0.029940 0.033447 0.037075
14 0.021585 0.024929 0.028397 0.032024 0.035765 0.039638
15 0.022993 0.026548 0.030231 0.034090 0.038065 0.042180
16 0.024393 0.028156 0.032050 0.036138 0.040345 0.044701
17 0.025783 0.029753 0.033855 0.038168 0.042606 0.047201
18 0.027162 0.031336 0.035644 0.040176 0.044844 0.049677
19 0.028530 0.032903 0.037415 0.042158 0.047058 0.052127
20 0.029885 0.034454 0.039167 0.044114 0.049248 0.054551
21 0.031227 0.035989 0.040900 0.046044 0.051414 0.056949
22 0.032556 0.037508 0.042614 0.047948 0.053556 0.059321
17
7.2 Anode Ground Beds
Anode ground beds are used to increase the anode current output in soils. They
typically comprise an excavation which is filled with low conductance carbonaceous
material into which the anode(s) are placed. The total resistance of the system then
becomes the resistance of the anode to the carbonaceous backfill plus the resistance to
earth of the backfill itself. The anode resistance is reduced by the low resistance of the
backfill (typically 50 ohm-cm for coke breeze), and the resistance of the backfill to earth
is reduced by the large surface area of the backfill in contact with the soil. Standard
anode to ground resistance formulae are used to obtain the resistance values.
Because of the variables involved in ground bed sites, experience is invaluable in

attaining competence in their design. They are designed with regard to the current
demand of the structure to be protected, to the soil resistance, and to other structures
and stray current effects.
7.3 Anode Current Output

Anode current output is calculated using Ohm’s Law:
V
I
R
This is where V is the driving potential, and R is the anode resistance. The driving
potential is determined by the anode type. The driving potential for sacrificial CP
systems is determined by the environment (Table 7.2), but for impressed current
systems it is determined by the rectifier and controller voltage output (Table 7.1). The
anode resistance to ground is found from the anode resistance formulas (Table 7.4).
7.4 Anode Size, Weight, Number, Distribution and Design Life

The CP system must be designed to provide the required current to every part of the
structure for the required design life. This requires determining anode size, weight,
number, and distribution.
18
The calculations for impressed current CP systems are relatively simple. In this case, it
is only necessary to match the number of anodes of known current output to the total
current demand of the structure, and to be sure that the anode distribution insures an
even and well balanced current distribution.
The calculations for sacrificial CP systems are a little more complex. Not only must the
number of anodes satisfy the current demand of the structure, but they must also have
sufficient mass to provide electricity for the design life of the structure.
Anode size and shape are determined by the following factors:

 Requirements for minimum and maximum current output
 Requirement for mounting and attachment
 Requirement for streamlining
 Requirement for weight of anode material (sacrificial)
 Commercial availability
The minimum and maximum current outputs are calculated as described in sections 7.1
and 7.2. The types of mounting methods range from welding steel cores, cast into the
anode materials, directly to the structure, to mounting complex dielectric shields with
screw in platinized titanium impressed current anodes. Where streamlining is required,
recesses may be built into the structure to house both impressed and sacrificial anode
types.
Sacrificial CP design requires that the weight of anode material is sufficient to supply
current for the design life of the structure. This is calculated by the following formula:
19
8760 YC
h
yr
W 
ZU
where :
W  weight of anode material
Y  design life (yrs)
C  current demand (Amps)
Z  anode capacity
U  utilization factor (0.9 for aluminum and zinc)
Finally, due to practical considerations, anode selection may ultimately be determined

by commercial availability. It is often too expensive to customize anode size and
geometry for one job. Therefore, except for large and specialized requirements, CP
design centers around standard, commercially available anode types.
20
8.0 COST AND IMPLEMENTATION
There are many permutations possible in CP design, however to be successful it must

satisfy economic constraints and be easy to install and operate. Examples of CP
designs are presented for an oil platform and ship.
8.1 Oil Platform Example

8.1.1 Structural Details
Water Depth 110m
No. of Legs 4
No. Horizontal Frames 5
No. of Nodes Below Surface 75
Total Submerged Surface Area 63,000 m2
Total Pile Surface Area in Mud 6,000 m2
Allowance for Risers, Conductors, Wells 220 Amps
Design Life 35 years
8.1.2. Current Demand

Total  (submerged S. A. )(CP current density) + (pile S. A. )(CP current density) + allowance
Submerged Steel (A) Piles (A) Allowance Total (A)

(A)
North Sea
Initial 11,340 150 220 11,710
Mean 5,670 120 220 6,010
Final 7,560 90 220 7,870
Gulf of Mexico
Initial 6,930 150 220 7,300
Mean 3,780 120 220 4,120
Final 5,040 90 220 5,350
21
The following design current densities were used for calculating the current demand
shown in the previous table:
Submerged Steel Piles

(mA/m2) (mA/m2)
North Sea
Initial 180 25
Mean 90 20
Final 120 15
Gulf of Mexico
Initial 110 25
Mean 60 20
Final 80 15
Example calculation (for North Sea, Initial)

CP current for submerged steel  (submerged S. A. )(CP current density) = (63,000m 2 )(0.180 mA2 )
CP current for submerged steel  11340
, A
CP current for piles  (pile S. A. )(CP current density) = (63,000m 2 )(0.025 mA2 )
CP current for piles  150A
Total CP current  11,340A  150A  220 A

Total CP current  11710
, A
8.1.3. Sacrificial Anode Design for Uncoated Structure

The following table is the weight required for 35 year CP design life (using mean
current density)
Location Zinc, kg Aluminum, kg
NORTH SEA 2,624,880 723,462
GULF OF MEXICO 1,799,419 495,953
22
Example calculation for the weight of zinc required, North Sea:
W
8760 YC
h
yr
ZU
W
8760 35 yr 6,010 A
h
yr
780 0.9
Ah
kg
W  2,624,880kg
Number of Anodes Required to Provide Initial Current Demand to Polarize the

Structure
Assume the following anode dimensions: 2,500 mm long

250 mm width
207 mm thick
NORTH SEA GULF OF MEXICO

Seawater Resistivity, ohm-cm 30 20
Anode Resistance, Ohms 0.0641 0.0427
Anode Current Output, Amps 3.900 5.855
Number of Anodes for Initial Current 3003 1247
Demand
Mass of Zinc Anodes, kg 2,768,766 1,149,734
Mass of Aluminum Anodes, kg 1,048,047 435,203
23
The previous table was created using the following methodology:
Find Mass of One Anode, Zinc

m   zinc 
m  (7,130 mkg3 )(2.5m)(0.25m)(0.207 m)
m  922kg
Find Mass of One Anode, Aluminum
m   Al 
m  (2,700 mkg3 )(2.5m)(0.25m)(0.207 m)
m  349kg
Find the Resistance of the Anode Using Table 7.4 (example for North Sea)
   4L  
R  ln   1
2L   r  
   
   
30cm   4(250cm)   1
R  ln
2 (250cm)  (25cm)(20.7cm)  
   
    
R  0.0641
Calculate single anode current output (example for North Sea):

V
I
R
(0.800V  (1.05V ))
I
0.0641
I  3.90 A
Find the # of Anodes Necessary to Produce the Initial Current (example for North Sea):
Initial current demand
# of anodes 
current output per anode
11,710A
# of anodes   3002.5 anodes = 3003 anodes
3.90 anode
A
24
Find the Mass of the Anodes (example for zinc anodes, North Sea):
Total Mass of Zinc Anodes = (# of anodes)(mass per anode)

kg
Total Mass of Zinc Anodes = (3003 anodes)(922 anode ) = 2,768,766 kg
Let us then compare the mass of anodes required for the design life and the mass
required for the initial polarization of the structure:

Mass of Zinc to protect the 2,624,880 1,799,419
structure for 35 years, kg
Mass of Zinc required to 2,768,766 1,149,734
provide the initial current, kg
Mass of Aluminum to protect 723,462 495,953
the structure for 35 years, kg
Mass of Aluminum required to 1,048,047 435,203
provide the initial current, kg
# of Zinc Anodes Needed 3003 1952
# of Aluminum Anodes 3003 1421
Needed
Anode Distribution and Spacing

For sacrificial anodes this maybe based on the current demand of the structure and the
maximum current output of the anode.
NORTH SEA
Maximum Current Output 3.900 Amps
Maximum Current Demand 180 mA/m2
Maximum Area Protected 3.900 A/(0.180A/m 2) = 21.7m2
25
GULF OF MEXICO
Maximum Current Output 5.855 Amps
Maximum Current Demand 110 mA/m2
Maximum Area Protected 5.855 A/(0.110 A/m 2) = 53.2m2
If the structural member is relatively large, say 3 m diameter, then a single anode
placed in the center of a 22 or 53 m2 area will not be too far from the extremities of the
cathode it is protecting. For smaller members, allowances have to be made for
attenuation, and anode sizes must be selected to ensure that the anode protects half
way to the next anode.
8.1.4 Sacrificial Anode Design Coated Structure

The effect of coating a structure on CP design can be seen by applying the DNV criteria
to Example 1. It can be seen that significant savings in anode material can be achieved
if a Category 2, 3 or 4 coating is used. The category 1 coatings are only helpful for
short design life. This interpretation of the interaction between coatings and CP is still
open to debate with many experts in the Oil Industry questioning the low performance
criteria assigned to Category I coatings.
26
Structure Area, m2: 63,000
Allowance, Amps 220
Piles, Amps: 120
Zinc Anodes, Ah/kg 780 Cost, $/tonne, Feb 1995 1,025
Aluminum Anodes, Ah/kg 2,830 Cost, $/tonne, Feb 1995 2,000
CD mA/m2 tr, years Category k1 k2 f (ave) f (final) CP (ave) mA/m 2 CP (final) mA/m2
North Sea
90 35 1 0.1000 0.1000 1.8500 3.6000 90 90
90 35 1 0.1000 0.0500 0.9750 1.8500 88 90
90 35 2 0.0500 0.0300 0.5750 1.1000 52 90
90 35 2 0.0500 0.0200 0.4000 0.7500 36 68
90 35 3 0.0200 0.0150 0.2825 0.5450 25 49
90 35 3 0.0200 0.0120 0.2300 0.4400 21 40
90 35 4 0.0120 0.0120 0.2220 0.4320 20 39
90 35 4 0.0120 0.0120 0.2220 0.4320 20 39
Category I II III IV
Total Current Demand, Amps 6,010 3,600 1,942 1,599
Wt Aluminum, kg 723,465 433,387 233,745 192,451
Cost Aluminum 1,446,931 866,774 467,490 384,903
Number of Anodes 3,003 3,000 2,500 2,500
Weight/Anode, kg 482 289 187 154
Cost, $ / Anode 1,156 693 449 370
Cost Installation $ / Anode 450 400 400 400
Total Cost CP, $ 4,823,983 3,280,258 2,121,975 1,923,767
Gulf of Mexico
60 35 1 0.1000 0.1000 1.8500 3.6000 60 60
60 35 1 0.1000 0.0500 0.9750 1.8500 59 60
60 35 2 0.0500 0.0300 0.5750 1.1000 35 60
60 35 2 0.0500 0.0200 0.4000 0.7500 24 45
60 35 3 0.0200 0.0150 0.2825 0.5450 17 33
60 35 3 0.0200 0.0120 0.2300 0.4400 14 26
60 35 4 0.0120 0.0120 0.2220 0.4320 13 26
60 35 4 0.0120 0.0120 0.2220 0.4320 13 26
Category I II III IV
Total Current Demand 4,120 2,514 1,408 1,179
Wt Aluminum 233,535 142,473 79,802 66,839
8.1.5 Impressed Current Anode Design

Assuming the use of platinized titanium anodes with the following dimensions.
length = 1000 mm
diameter = 25 mm
Maximum driving potential = 9V
Then the surface area of the anode is:
27
dL = 0.078m2

Anode Resistance, Ohms 0.2277 0.1518
Maximum Anode Current Output, 39.5 59.3
Amps
Maximum current density on anode 506 760
surface, A/m2
Number of anodes required 296 123
The distribution of the anodes is critical.
28
8.2 Ship Hull Protection
Ship Hulls require protection from both corrosion and the development of biofouling
accumulations. The former is achieved by both coating and cathodic protection
systems and the latter by antifouling coatings. In addition to the basic costs of the
coating and cathodic protection systems, allowances must be made for dry dock costs,
loss of revenue, and increased fuel consumption and lost performance due to the
increase in skin friction drag caused by poor hull maintenance and biofouling.
SHIP HULL COATING
GRIT BLAST
NACE #1, with 1 - 2 mil anchor profile
BARRIER COAT
Hard Boiled Mastic, two component epoxy amine
Two coats, 3 mils D.F.T.
Coverage per one mil dry - 564 sq.ft./gal.
Cost, $23.00/gal.
Cost for 6 mils D.F.T. $0.24/sq.ft.
CATHODIC PROTECTION DESIGN

SHIP ORIANA
TYPE PASSENGER
LENGTH, m 245
BREADTH, m 30
DRAFT, m 9.75
BLOCK COEFFICIENT, Cb 0.6
WETTED SURFACE AREA (1.7xLxD)+(CbxLxB) m2 8,471
CURRENT DEMAND FOR HULL, (@22mA/m 2) Amps 186
CURRENT DEMAND FOR RUDDERS, Amps 15
CURRENT DEMAND FOR PROPELLERS, Amps 35
TOTAL CURRENT DEMAND, AMPS 236
29
Sacrificial Anode Calculations
Current Output per Anode
Assume anode dimensions: 500mm x 115mm x 65mm
V
The current output, I
R
R= the anode resistance calculated using Lloyds' Formula for plate anodes
Lloyds' Formula

R
2S
20  cm
R
2(30.75cm)
R  0.325
Current output per anode,

V
I
R
0.25V
I  0.769 A
0.325
Number of anodes required to supply current,

Total current required
# of anodes needed 
Current output per anode
236 A
# of anodes needed   307 anodes
0.384 anode
A
The total current capacity for one anode is:

Volume of Anodes * Density * Capacity of Galvalum 1
= 3.738x10-3 m3 * 2,700 kg/m3 * 2,830 Ah/kg = 28,558 Ah
The design life of the system will be

= 28,558 A hrs * 307 anodes * (0.9) = 33,435 hrs or 3.8 years
236 Amps
30
APPENDIX
Some Commonly Used Marine Materials

from Dexter, S.C. (1985), Handbook of Oceanographic Materials, Krieger Publishing, Malabar, FL.
Aluminum Alloys
Material Composition Density,  (lb/in3)
Aluminum alloy 5052 97.25% Al, 2.5% Mg, 0.25% 0.097
Cr
Elastic Modulus, E (psi) Yield Strength, y (ksi) Tensile Strength, u (ksi)
10 x 106 31 (H34) 38 (H34)
Potential in Seawater, ref. Corrosion Types Suffered Uses
Ag-AgCl (V)
-0.92 to -1.1 Crevice and sometimes pitting Applications demanding good
attack. corrosion resistance and fatigue
strength. These include fuel
lines, tanks, and sheets.
Special Notes: Severe galvanic attack can occur when placed in contact with steel, stainless
steel, copper alloys, nickel alloys, and titanium. Corrosion protection is desired for submerged
applications.

Aluminum alloy 6061 97.95% Al, 1.0% Mg, 0.6% Si, 0.098
0.25% Cu, 0.20% Cr
10 x 106 40 (T6) 45 (T6)
Ag-AgCl (V)
-0.72 to -1.07 Crevice and pitting attack. Applications demanding
May also undergo adequate corrosion resistance
intergranular attack and SCC. and good mechanical properties.
The most versatile aluminum
alloy for marine use.
steel, copper alloys, nickel alloys, and titanium. Cathodic protection and coating is desired for
submerged applications.

Aluminum alloy 7075 90% Al, 5.6% Zn, 2.5% Mg, 0.101
1.6% Cu, 0.3% Cr
10.4 x 106 73 (T6) 83 (T6)
Ag-AgCl (V)
-0.72 to -0.83 Severe crevice and pitting Applications requiring high
attack. Also susceptible to strength and low weight.
SCC and exfoliation.
steel, copper alloys, nickel alloys, and titanium. Must be anodized, coated, and cathodically
protected for submerged applications.
Appendix: Material Properties - 1

Copper and Copper alloys

Copper 99.9% Cu 0.322
17 x 106 50 (Fully hardened) 55 (Fully hardened)
Ag-AgCl (V)
-0.12 to -0.30 Uniform attack and sometimes Electrical and architectural
localized attack as a result of applications. Resists biofouling
metal ion concentration cells. at corrosion rates > 1 mpy.
Velocity effects can also be
quite marked.
Special Notes: None

Beryllium-Copper, CDA 172 97.9% Cu, 1.9% Be, 0.2% Co 0.298
18 x 106 150 - 190 195 - 205
(HT and 75% Cold Worked) (HT and 75% Cold Worked)
Ag-AgCl (V)
-0.10 to -0.25 Uniform corrosion and slight Applications requiring good
crevice attack. corrosion resistance and high
strength. These include springs,
bearings, and bushings.
Special Notes: As in most copper alloys, metal ion concentration cells may form.

Red Brass, CDA 230 85% Cu, 15% Zn 0.316
17 x 106 49 57 - 72
(50% Work Hardened) (50% Work Hardened)
Ag-AgCl (V)
-0.20 to -0.40 Uniform corrosion and slight Applications requiring good
dezincification. corrosion resistance.

Inhibited Admiralty Brass 71% Cu, 28% Zn, 1% Sn, 0.308
0.6% As, Sb, or Pb
16 x 106 72 88 - 97
(Fully Hardened) (Fully Hardened)
Ag-AgCl (V)
-0.16 to -0.25 Uniform corrosion and crevice Heat exchanger and condenser
attack. Erosion corrosion at tubes and plates.
velocities > 6 fps.

Naval Brass 60% Cu, 39% Zn, 1% Sn 0.304
15 x 106 66 88
(Hard-Drawn) (Hard-Drawn)
Ag-AgCl (V)
-0.20 to -0.27 Uniform corrosion and Condenser plates, prop shafts,
dezincification (not completely fasteners.
eliminated with CP).
Special Notes: Dezincification of this alloy may be severe, and, therefore, should be used with
caution in submerged applications.

Aluminum Bronze D, CDA 91% Cu, 7% Al, 2% Fe 0.281
614
18 x 106 40 - 55 75 - 85
(Hardened) (Hardened)
Ag-AgCl (V)
-0.09 to -0.26 Uniform corrosion and some Corrosion resistant tubing, tanks,
dezincification and crevice fasteners, and sheathing.
corrosion

High Silicon Bronze A, CDA 94.8% Cu, 3.3% Si, 1.5% Mn, 0.308
655 < 1.5% Fe and Zn
15 x 106 45 - 57 78 - 98
(Half-Hard) (Half-Hard)
Ag-AgCl (V)
-0.17 to -0.23 Uniform corrosion and crevice Marine hardware, fasteners,
corrosion. shafting, and heat exchanger
tubing.

90-10 Copper-Nickel 88.7% Cu, 10% Ni, 1.3% Fe 0.323
18 x 106 57 60
(Light Drawn) (Light Drawn)
Ag-AgCl (V)
-0.15 to -0.30 Uniform corrosion and some Excellent resistance to marine
surface attack. fouling if allowed to freely
corrode. Used for seawater
tubing and boat hulls.
Special Notes: Susceptible to sulfide attack. Velocity effects at velocities > 10 fps.

70-30 Copper-Nickel 68.9% Cu, 30% Ni, 0.5% Fe, 0.323
0.6% Mn
22 x 106 79 85
(Cold Drawn) (Cold Drawn)
Ag-AgCl (V)
-0.17 to -0.23 Uniform corrosion and some Good strength. Used in heat
surface attack. exchangers with high water
velocities.
Special Notes: Susceptible to sulfide attack. Velocity effects at velocities > 15 fps.

Cast Silicon Brass and 82 - 91% Cu, 5 - 14% Zn, 4% Si 0.302
Bronze
15 x 106 - 18 x 106 22 - 35 55 - 70
(As Cast) (As Cast)
S.C.E. (V)
~ -0.27 Uniform corrosion. Bearings, impellers, gears,
props, pumps, fittings.
Special Notes: None

Cast Aluminum Bronze 81 - 88% Cu, 9 - 13% Al, 0.272 - 0.281
1 - 5% Fe, others
14 x 106 - 20 x 106 40 - 80 80 - 124
(Heat Treated) (Heat Treated)
S.C.E. (V)
-0.3 to -0.4 Uniform corrosion and Pump housings, bearings,
dealloying. impellers, gears, props, fittings.
Special Notes: None

Nickel Alloys

Monel 400 66.25% Ni, 31.5% Cu, 1.35% 0.319
Fe, 0.9% Mn
26 x 106 90 - 130 100 - 140
(Fully Hardened) (Fully Hardened)
Potential in Seawater, Corrosion Types Suffered Uses
ref. S.C.E. (V)
-0.04 to -0.14 Uniform corrosion, pitting, and Valves, pumps, prop shafts,
crevice attack. fixtures, fasteners.
Special Notes: Resists erosion corrosion to high velocities. May cause severe galvanic attack of
less noble metals when coupled.

Inconel 625 65.3% Ni, 18.6% Cr, 9% Mo, 4% 0.305
Cb, 3% Fe, 0.05% C
29.8 x 106 201 219
(70% Cold Worked) (70% Cold Worked)
ref. S.C.E. (V)
-0.04 to +0.10 Highly resistant to most forms of Wire rope, propeller blades,
attack. fittings, springs, fasteners.
Parts where little to no corrosion
can be accepted.
Special Notes: Resists erosion corrosion to high velocities.

Incoloy 825 41.8% Ni, 21.5% Cr, 30% Fe, 3% 0.294
Mo, 1.8% Cu, 1% Ti, 0.03% C
28 x 106 35 - 45 85 - 101
(Cold Drawn) (Cold Drawn)
ref. S.C.E. (V)
-0.03 to +0.05 Crevice corrosion and pitting. Components in desalination
plants and heat exchangers.
Special Notes: Resistant to chloride SCC.

Iron and Steels

Ductile Cast Iron 3.3 - 4.0% C, 2 - 3% Si, 0.2 - 0.257
0.6% Mn, < 2.5% Ni, > 0.15%
total P and Mg, remainder Fe
23 x 106 - 25 x 106 40 - 150 60 - 175
ref. S.C.E. (V)
-0.60 to -0.72 Mostly uniform with some shallow General machinery parts, props,
pitting. piping.
Special Notes: Cathodic protection is needed for longer term submerged exposure.

AISI 1040 Steel 99.6% Fe, 0.4% C 0.283
30 x 106 86 (Heat Treated) 113 (Heat Treated)
ref. S.C.E. (V)
-0.60 to -0.70 Mostly uniform with slight crevice Multiple structural and
corrosion. mechanical uses.
Special Notes: Coatings and/or cathodic protection is needed for marine use. Galvanic attack
may occur if this alloy is immersed in seawater and in contact with copper alloys, nickel alloys,
stainless steels, or titanium alloys.

AISI 1080 Steel 99.2% Fe, 0.8% C 0.283
30 x 106 142 (Heat Treated) 190 (Heat Treated)
ref. S.C.E. (V)
-0.60 to -0.70 Mostly uniform with slight crevice Multiple structural and
corrosion. Also susceptible to mechanical uses.
SCC and hydrogen
embrittlement.

HY-80 Steel 2 - 3.25% Ni, 1 - 1.8% Cr, 0.2% - 0.6% 0.284
Mo, 0.15 - 0.35% Si, <0.25% P and S,
0.1 - 0.4% Mn, 0.18% C, remainder Fe
30 x 106 80 - 100 103
(Quenched and Tempered) (Quenched and
Tempered)
ref. S.C.E. (V)
-0.63 Mostly uniform. Hull plating, offshore
platforms, tanks pressure
vessels, cranes, booms

HY-100 Steel 2.25 - 3.5% Ni, 1 - 1.8% Cr, 0.284
0.2% - 0.6% Mo, 0.15 - 0.35%
Si, <0.25% P and S, 0.1 - 0.4%
Mn, 0.2% C, remainder Fe
30 x 106 100 - 105 110 - 118
(Quenched and Tempered) (Quenched and Tempered)
ref. S.C.E. (V)
-0.63 Mostly uniform. Some Hull plating, offshore platforms,
tendency towards SCC and tanks pressure vessels, cranes,
hydrogen embrittlement. booms

Low Alloy-High Strength 0.18 - 0.22% C, 0.5 - 1.5% (each) Mn, 0.283
Steels (ASTM A-242 and Ni, Cr, ~0.25% (each) P, Si, S, Cu,
A-441) remainder Fe
30 x 106 40 - 60 (Annealed) 60 - 80 (Annealed)
ref. S.C.E. (V)
-0.57 to -0.63 Mostly uniform with some crevice Structural sections and
corrosion and pitting. members.

Maraging 300 Steel 18 - 19 % Ni, 8.5 - 9.5% Co, 4.7 - 5.2% 0.290
Mo, 0.5 - 0.7% Ti, 0.05 - 0.15% Al, <
0.03% C, remainder Fe
29 x 106 295 - 303 (Heat Treated) 297 - 306 (Heat Treated)
ref. S.C.E. (V)
-0.57 to -0.58 Uniform, SCC, and hydrogen High strength weldable
embrittlement. SCC and embrittlement structural pieces.
can be controlled with cathodic
protection.
Titanium and Titanium Alloys

Unalloyed Titanium 98.9 - 99.5% Ti 0.163
15 x 106 up to 90 up to 100
(Cold Worked) (Cold Worked)
ref. S.C.E. (V)
-0.05 to +0.06 None. Structural members, marine
parts requiring immunity,
impressed current anodes.
Special Notes: SCC is possible if titanium contains higher levels of oxygen.

Titanium 6Al-4V 5.5 - 6.5% Al, 3.5 - 4.5% V, 0.160
<0.25% Fe, remainder Ti
16.5 x 106 155 165 - 170
(Age Hardened) (Age Hardened)
ref. S.C.E. (V)
-0.05 to +0.06 None, except some tendency for Pumps, impellers, structural
SCC. members, marine hardware
and parts requiring immunity.
Special Notes: None

Stainless Steels

302 Stainless Steel 70.85 - 74.85% Fe, 17 - 19% Cr, 0.290
8 - 10% Ni, 0.15% C
28 x 106 75 125
(Quarter Hard) (Quarter Hard)
ref. S.C.E. (V)
-0.05 to -0.10 (passive) Crevice and pitting corrosion. General purpose in non-
-0.45 to -0.57 (active) Susceptible to local attack in submerged applications.
areas covered by fouling
Special Notes: Local attack minimized in water velocities > 5 fps, but not generally
recommended for submerged use due to pitting.

303 Stainless Steel 17 - 19% Cr, 8 - 10% Ni, > 0.15% 0.290
S or Se, 0.15% C, remainder Fe
28 x 106 75 110
ref. S.C.E. (V)
-0.05 to -0.10 (passive) Severe crevice and pitting Generally not recommended for
-0.45 to -0.57 (active) corrosion. Susceptible to local seawater application.
attack in areas covered by fouling
Special Notes: This grade may be substituted for others by suppliers with devastating
consequences. Non-magnetic, austenitic alloy.

9 - 12% Ni, 0.08% C
28 x 106 75 110
ref. S.C.E. (V)
-0.09 to -0.15 (passive) Crevice and pitting corrosion. Topside wire rope and general
-0.20 to -0.57 (active) Susceptible to local attack in purpose submerged use where
areas covered by fouling. Heat velocities are > 5 fps.
effected zones may be
sensitized.
Special Notes: Cathodic protection is necessary for submerged structural applications when
exposure is greater than two months. Non-magnetic, austenitic alloy.

10 - 14% Ni, 2 - 3% Mo, 0.08% C
28 x 106 30 - 42 80 - 90
(Annealed) (Annealed)
ref. S.C.E. and Ag-AgCl
(V)
-0.00 to -0.15 (passive) Crevice and pitting corrosion. Topside wire rope and general
-0.35 to -0.60 (active) Susceptible to local attack purpose where velocities are > 5
especially in areas covered by fps.
fouling. Heat effected zones may
be sensitized.
Special Notes: Cathodic protection is necessary for submerged structural applications when
exposure is greater than six months. This is the most corrosion resistant 300 series stainless.
Non-magnetic, austenitic alloy.

17-4 PH Stainless Steel 16.5% Cr, 4% Ni, 4% Cu, 0.3% Nb 0.280 - 0.282
and Ta, 0.07% C, remainder Fe
28.5 x 106 178 - 185 200
(Hardened) (Hardened)
ref. Ag-AgCl (V)
-0.10 to -0.20 (passive) Crevice and pitting corrosion. Parts with moderate corrosion
-0.20 to -0.40 (active) Weld bead attack. resistance and high strength to
weight ratio.
Special Notes: Cathodic protection should be from impressed current or mild steel sacrificial
anodes. Problems arise with aluminum, zinc, and magnesium anodes. Cathodic protection may
lead to hydrogen embrittlement and cracking.

410 Stainless Steel 85.35 - 87.35% Fe, 11.5 - 13.5% 0.280
Cr, 1% Mn, 0.15% C
29 x 106 140 - 145 180 - 190
(Heat Treated) (Heat Treated)
ref. S.C.E. (V)
-0.24 to -0.35 (passive) Severe crevice and pitting Applications where a high
-0.45 to -0.57 (active) corrosion. strength alloy is important.
Submerged applications
require cathodic protection to
prevent localized attack.
Special Notes: Magnetic, martensitic alloy.

Other Metals

Commercially Pure 99.98% Mg 0.063
Magnesium
5.7 x 106 - 6.5 x 106 3 (As Cast) 13 (As Cast)
ref. S.C.E. (V)
-1.60 to -1.63 Rapid uniform corrosion. Severe Sacrificial anodes and
galvanic effects when coupled with corrosive links.
all common marine metals.
Special Notes: Not suitable for structural applications in seawater.

Zinc 99.92% Zn, 0.08% Pb 0.258
---- ---- 19 - 23 (Hot Rolled)
ref. S.C.E. (V)
-0.98 to -1.03 Uniform corrosion with some Galvanizing, sacrificial anodes
pitting in anaerobic conditions. and corrosive links.
Special Notes: None.

Lead ---- 0.410
2.0 x 106 ---- 2.0 (As Cast)
ref. S.C.E. (V)
-0.19 to -0.25 Uniform corrosion. Galvanizing, sacrificial anodes
and corrosive links.

Gold 99.5 - 99.99% Au 0.698
10.8 x 106 - 11.6 x 106 30 (60% Cold Worked) 32 (60% Cold Worked)
ref. S.C.E. (V)
---- Completely resistant to marine Specialty applications and
corrosion. electrical contacts.

Platinum 99.85% Pt 0.775
21 x 106 - 25 x 106 27 (50% Cold Worked) 28 - 30 (50% Cold Worked)
ref. S.C.E. (V)
+0.20 to +0.35 Completely resistant to marine Impressed current anodes and
corrosion. electrical contacts.

Silver 99.9% Ag 0.379
10.3 x 106 - 11.3 x 106 44 (50% Cold Worked) up to 54 (50% Cold Worked)
ref. S.C.E. (V)
-0.09 to -0.14 Slight uniform corrosion. Electrical conductors and radar
Increased tarnish with sulfur applications. Solder when
compounds present. alloyed.

Polymers, Rubbers, and Elastomers
Material Description Specific Gravity

ABS, Medium Impact Thermoplastic (acrylonitrile butadiene 1.05 - 1.07
styrene)
Elastic Modulus, E (psi) Water Absorption Tensile Strength, u (ksi)
3.0 x 105 - 4.1 x 105 0.2 - 0.45% in 24 hrs. 5.9 - 8.0
Behavior in the Marine Environment Uses
Slight yellowing and embrittlement in direct sunlight. No chemical Multiple structural uses.
degradation in seawater. Not attacked by borers unless in contact Pipe, tubing, instrument
with wood. housings, bearings.

Acetal, Standard Thermoplastic (tradenames - Delrin, 1.43
Homopolymer Celcon)
5.2 x 105 0.25% in 24 hrs. 10.0
Slight chalking in sunlight. No chemical degradation in seawater. Gears, bushings, levers,
Usually not attacked by borers unless in contact with wood or tape. shafts, springs, hardware.
It is attacked by strong acids and bases. Excellent resistance to
many organic solvents.

Cast Acrylics Polymethyl methacrylate 1.17 - 1.28
(tradenames - Lucite, Perspex,
Plexiglas)
2.7 x 105 - 5.0 x 105 0.2 - 0.5% in 24 hrs. 5.5 - 8.0 (High Impact Sheet)
Usually not affected by sunlight. No chemical degradation in Lenses, windows, housings
seawater. Usually not attacked by borers unless in contact with and many general purpose
wood or tape. Water absorption can lead to a 10% reduction in applications.
hardness and up to a 30% reduction in tensile strength. It is
attacked by strong acids and bases. Soluble in ketones, esters,
and aromatic and chlorinated hydrocarbons. Excellent resistance
to many organic solvents.

Epoxy Diglycidal ether of bisphenol A 1.1 - 2.0
5 x 105 - 15 x 105 0.1 - 1.0% in 24 hrs. 5 - 15
Resistant to sunlight with UV inhibitors. Usually not attacked by Potting material for electical
borers unless in contact with wood or tape. May be attacked by components, castings,
sulfuric and acetic acid. Some attack by stong bases. Resistant marine coatings, adhesives,
to organic solvents and weak acids and bases. patching compounds.

Nylon, Type 6 Thermoplastic, polyamide 1.14
3.8 x 105 1.3 - 1.9% in 24 hrs. 9.5 - 12.5
Embrittled by prolonged exposure to sunlight. Usually not Bearings, gears, bushings,
attacked by borers unless in contact with wood. May be attacked housings, rods, ropes,
by strong acids, phenols, and formic acid. Resistant to bases, coatings.
weak acids, and most common solvents. Water absorption and
swelling may alter dimensions and mechanical properties

Polycarbonate (unfilled) Tradenames - Lexan, Merlon 1.19 - 1.25
3.0 x 105 - 4.5 x 105 0.12 - 0.19% in 24 hrs. 9.0 - 10.5
Sunlight may lead to a slight color change and some Lenses, windows, housings,
embrittlement. No chemical degradation in seawater. Usually not impellers and parts requiring
attacked by borers unless in contact with wood. Low water high impact resistance.
absorption leads to good dimensional stability and retention of
properties. It is attacked by strong acids and bases, organic
solvents, and fuels. Resistance to weak acids, oils, and greases.

Polyethylene, High Density Thermoplastic (Tradenames - 0.95 - 0.96
Marlex, Norchem, Rulan)
3.0 x 105 - 4.5 x 105 < 0.01% in 24 hrs. 4.4
Greatly degraded by sunlight if inhibitors are not used. Usually Wire and cable insulation,
not attacked by borers unless in contact with wood. Generally pipe, housings.
shows excellent resistance to the marine environment. Good
chemical resistance. Slowly attacked by strong acids.

Polypropylene General purpose and high impact 0.89 - 0.91
thermoplastic
1.6 x 105 - 2.2 x 105 < 0.01 - 0.03% in 24 hrs. 4.3 - 5.5
Greatly degraded by sunlight if inhibitors are not used. Usually Wire and cable coatings,
not attacked by borers unless in contact with wood. Generally film, packaging, hinges
shows excellent resistance to the marine environment. Good housings.
chemical resistance. Slowly attacked by strong acids.

Teflon, PTFE Polytetrafluoroethylene (tradenames 2.1 - 2.3
- Teflon, Fluon, Halon, Rulon)
0.38 x 105 - 0.65 x 105 0.01% in 24 hrs. 2.0 - 6.5
Not degraded by sunlight. Usually not attacked by borers unless Pipes, valves, bearings,
in contact with wood. Generally shows excellent resistance to the impellers, electrical
marine environment. Excellent chemical resistance. May be insulators, non-stick
attacked by the alkali metals. coatings.

Polyvinyl Chloride, PVC Polytetrafluoroethylene (tradenames 1.30 - 1.45
- Teflon, Fluon, Halon, Rulon)
3.5 x 105 - 6.0 x 105 0.03 - 0.04% in 24 hrs. 5.5 - 9.0
Sunlight produces minor effects. Generally shows good Pipes, tanks, molded and
resistance to the marine environment. Attacked by strong acids, extruded parts, housings.
ketones, esters, and aromatic hydrocarbons. Resistant to
alcohols, aliphatic hydrocarbons, oils, bases, and weak acids.

Butyl Rubber Isobutylene-isoprene 0.90
Tear Resistance Abrasion Resistance Tensile Strength, u (ksi)
Good Good to Excellent 2.5 - 3.0
Very good resistance to sunlight produces. Excellent resistance Flexible electrical insulation,
to swelling in water. Resistant to acids, oxidation and heat aging. hose, shock absorption,
Vulnerable to many solvents, oils, and fuels. diaphragms.

Natural and Synthetic Polyisoprene 0.93
Rubber
Excellent Excellent 2.5 - 4.5
Fair to poor resistance to sunlight. Marine exposure may cause Seals, gaskets, hose,
some swelling. Microorganisms and hydrogen sulfide may lead to chemical tank linings.
severe cracking. Resistant to oxidation and heat aging.
Vulnerable to many solvents, oils, and fuels. Shows fair to good
resistance to organic acids.

Neoprene Rubber Chloroprene 1.25
Fair to Good Good 3.0 - 4.0
Very good resistance to sunlight. Marine exposure may cause Seals, gaskets, chemical
some swelling. May be degraded by aromatic hydrocarbons. tank linings, wetsuits.
Good to excellent resistance to other organic solvents, oils, fuels,
acids, and heat.

Fluorocarbon Elastomers Trade names: Kel-F, Viton 1.40 - 1.95
Poor to Fair Good 1.5 - 3.0
Very good resistance to sunlight and swelling. May be degraded O-rings, seals, gaskets,
by alkalies, synthetic lubricants, hydraulic fluids containing hose, shaft seals.
phosphates. Excellent resistance to high temperature air and oils.

Urethane Elastomers Trade names: Adiprene, 1.07
Cyanaprene, Elastothane, Roylar
Excellent Superior 5.0 - 8.0
Very good resistance to sunlight and swelling (except at high Components requiring
temperatures). May be degraded by acids, alkalies, oxygenated superior abrasion
alcohols. Excellent resistance to hydrocarbon solvents, oils. resistance.

Concrete and Glass
Material Composition Density,  (lb/ft3)

Concrete Varied 140 - 150
Elastic Modulus, E (psi) Compressive Strength, (ksi) Flexural Strength, (ksi)
3 x 106 - 6 x 106 3.5 - 7.5 0.4 - 0.8
Porosity (% by volume) Marine Attack Suffered Uses
5 - 10 Water absorption up to 2.4% Large structural members (with
@ 550 ft head. Degraded by and without steel reinforcement),
high sulfate waters, causing ship hulls, moorings.
cracking and softening.
Special Notes: Deterioration of concrete can lead to lowering of pH at the rebar, causing
corrosion and spalling of the concrete. For longer term durability concrete, it is desirable to use
low permeability types with reduced alkalinity and low 3 CaO Al2O3 content.
Material Composition Specific Gravity

Glass (Borosilicate and Trade names: Kimax, Pyrex 2.13 to 2.55
Soda Lime)
Elastic Modulus, E (psi) Abrasion Resistance Tensile Strength, u (ksi)
7.4 x 106 - 10.0 x 106 Excellent 0.5 - 40.0
Refractive Index Behavior in Marine Environment Uses
1.468 - 1.525 Generally unaffected by Containers, plates, buoyancy
weathering or marine exposure. spheres, lenses, housings.
Special Notes: May be attacked by hydrofluoric acid and sodium hydroxide.

Wood
Material Moisture Content (%) Specific Gravity

Hardwood, Seasoned 12 0.63 to 0.68
Maple and Oak
Elastic Modulus in Modulus of Rupture (ksi) Behavior in Marine Environment
Bending (psi)
1.8 x 106 15.8 (maple) Left untreated these woods can be
14.3 - 15.2 (oak) severely damaged by marine borers in as
little as six months exposure when placed
within six feet of bottom sediments. The
borer attack is generally most rapid in
warm coastal and lower latitude waters.

Teak and Mahogany 52 (teak) 0.50 (mahogany)
Elastic Modulus in Modulus of Rupture (ksi) Behavior in Marine
Bending (psi) Environment
1.4 x 106 - 1.7 x 106 11.1 - 11.4 These woods are susceptible
to borer attack.

Softwood, Seasoned 12 0.46 to 0.47
Cedar and Cypress
Bending (psi)
0.9 x 106 - 1.4 x 106 8.8 (cedar) Left untreated these woods can be
10.6 (cypress) severely damaged by marine borers in as

Softwood, Seasoned 12 0.35 to 0.40
Pine and Spruce
Bending (psi)
1.2 x 106 - 2.0 x 106 8.6 - 10.1 (pine) Left untreated these woods can be
10.2 (spruce) severely damaged by marine borers in as

Fiber Reinforced Plastics (FRP)
from Agarwal, B.D and L.J. Broutman (1990), Analysis and Performance of Fiber Composites, Wiley Publishing, New York.

E-Glass Reinforced Epoxy 57% E-Glass, 43% Epoxy 1.97
Tensile Modulus, E (psi) Tensile Strength, u (ksi) Longitudinal Poisson’s Ratio
3.1 x 106 82.5 0.25
Uses:
Applications demanding excellent corrosion resistance and high strength to weight ratio. These
include structural members, boat hulls, tanks, specialty items.
Special Notes: Can be degraded by sunlight if inhibitors or barrier coat protection is not used.
May also be somewhat degraded by water absorption if not protected.

Kevlar 49 Reinforced Epoxy 60% Kevlar 49, 40% Epoxy 1.40
5.8 x 106 94.3 0.34
Uses:
Applications demanding excellent corrosion resistance and high strength to weight ratio. These
include structural members, boat hulls, tanks, specialty items.

Carbon Fiber Reinforced 58% Carbon Fiber, 42% Epoxy 1.54
Epoxy
12.0 x 106 55.1 0.38
Uses:
Applications demanding excellent corrosion resistance and high strength to weight ratio.
Excellent stiffness properties. Uses include structural members, boat hulls, tanks, specialty
items.

Naval Material Science and Engineering USNA PDF

Caricato da

Informazioni sul documento

Descrizione originale:

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Naval Material Science and Engineering USNA PDF

Caricato da

Copyright:

Formati disponibili

EN380

Naval Material Science and Engineering

With Contributions by:

Thomas Dawson, Victoria Johnson, Paul Miller, Michael Morabito, Michael

United States Naval Academy

Chapter 2: Corrosion and Electromotive Chapter 6: Corrosion Protection and

MATERIALS THE MARINE ENVIRONMENT

1.1 Characterization of Seawater

1.1 Characterization of Seawater

CHEMICAL COMPOSITION OF SEAWATER, 19ppt Chlorinity

Specific Conductance of Seawater

Table of the Specific Conductance of Seawater (-1cm-1)

Solubility of Oxygen in Seawater

An increase in temperature increases seawater conductivity and decreases oxygen concentration.

Typical Profiles of Oxygen, Temperature, and Salinity in the Ocean.

1.2 Organic Matter and Biology

Penetration of Water by Visible Light.

The Size and Shape of Typical Fouling Organisms

1.3 The Physical Environment

Beaufort Wind Scale and Sea State

Wind, Wave and Current Effects on Offshore Structures

1.4 Classification of Marine Environments

1.5 Corrosion factors for Carbon Steel in Marine Environments

Typical atmospheric corrosion rates for various locations

Splash Zone Corrosion

Condition Years Exposure Rate, mpy

1.6 Historical Perspective

Bronze and Copper

USS Virginia (SSN 774)

USS Independence (LCS 2)

Continued Use of Wood

U.S. Navy R2 Tower Offshore Oil Platform

U.S. Naval Base in Okinawa

2.1 Corrosion cell

Why do metals corrode?

Safety / Loss of Life

2.1 Corrosion cell

Schematic of the corrosion cell

Anode half-cell reaction

Cathode half-cell reaction

1. O2  2 H 2 O  4e   4OH  aerated neutral to alkaline water

2.2 Specific Examples of Electrochemical Reactions

Keeping in mind the definition of corrosion as a destructive electrochemical process, resulting in

Iron Bar in Hydrochloric Acid

Iron and Copper in Hydrochloric Acid

Iron and Copper in Hydrochloric Acid

Reduction occurs at the cathode according to the reaction:

2H+ + 2e-  H2(gas)

A compound also forms as described by:

Fe++ + 2Cl-  FeCl2

Iron and Copper in Water

The cathode reaction is, however, now mainly

O2 + 2H2O + 4e-  4(OH)-

Fe++ + 2(OH)-  Fe(OH)2

4Fe(OH)2 + O2 + 2H2O  4 Fe(OH)3

Local Anodes and Cathodes

At the cathode region

O2 + 2H2O + 4e-  4(OH)-

2.3 Measurement of Potential

Iron and Copper Bars with a Voltmeter Connection

ΔV = VFe - VCu = -0.789 volts

Standard Hydrogen Electrode