Sei sulla pagina 1di 10

Role of Silicone Surfactant in Flexible Polyurethane Foam

X. D. Zhang,* C. W. Macosko,*
,1
H. T. Davis,* A. D. Nikolov, and D. T. Wasan
*Department of Chemical Engineering and Materials Science, University of Minnesota, 421 Washington Ave. SE, Minneapolis, Minnesota 55455;
and Department of Chemical Engineering, Illinois Institute of Technology, Chicago, Illinois 60616
Received August 11, 1998; accepted March 26, 1999
Grafted copolymers which consist of a polydimethylsiloxane
backbone and polyethylene oxide-co-propylene oxide pendant
groups are used as surfactants to stabilize the foam cells in the
exible polyurethane foaming process. The mechanical properties
of the cured polyurethane foam such as air permeability and foam
cell size are affected signicantly by the structure of the silicone
surfactant used in the formulation. It is shown that silicone sur-
factant has an important impact on both the bubble generation
and the cell window stabilization stage. A series of silicone surfac-
tants with different structures was tested. Surfactants with higher
silicone content will provide lower surface tension and thus help
increase the number of air bubbles introduced during mixing.
These air bubbles serve as the starting point for foam cell growth.
As a result, the cured polyurethane foam made with higher sili-
cone content surfactant has a smaller bubble size. It is also shown
that silicone surfactant can reduce the cell window drainage rate
due to the surface tension gradient along the cell window. The
Gibbs lm elasticity, the dynamic lm elasticity, and the lm
drainage rate were measured for the rst time versus surfactant
composition. Surfactants with longer siloxane backbones are
shown to give higher lm elasticity. Using the vertical lm drain-
age and foam column tests, it is shown that surfactants with higher
lm elasticity will yield slower drainage rate and better foam cell
stability. 1999 Academic Press
KeyWords: polyurethane foam; silicone surfactant; lm tension;
lm elasticity; nucleation; foam stability; lm drainage.
INTRODUCTION
The addition polymerization reaction of diisocyanates with
alcohols was discovered by Bayer and co-workers in 1937 (1).
The discovery provided the fundamental chemistry for the
polyurethane industry. Of all the polyurethane products, ex-
ible polyurethane foam has the largest production and is most
widely used. During exible polyurethane foam formation,
water reacts with the isocyanate group and produces carbon
dioxide gas. The carbon dioxide gas will provide volume for
bubble expansion and occupy over 95% of the nal volume of
the product. Figure 1 shows the four stages of the foaming
process: (1) bubble generation and growth, (2) packing of the
bubble network and cell window stabilization, (3) polymer
stiffening and cell opening, and (4) nal curing (2, 3).
Silicone surfactants, which consists of a polydimethylsilox-
ane (PDMS) backbone and polyethylene oxide-co-propylene
oxide (PEO-PPO) random copolymer grafts, are used as sur-
factants in exible polyurethane foaming systems. It was
shown that these surfactants do not alter the reaction kinetics in
the exible polyurethane foaming process (4). In the absence
of these surfactants, a foaming system will experience cata-
strophic coalescence and eventually cause foam collapse.
These surfactants are efciently adsorbed at the airpolyol
interface and thus may have a signicant effect on both the
bubble generation and the cell window stabilization steps in the
polyurethane foaming process. The silicone surfactant struc-
ture is shown to have a great impact on the foam cell size and
air permeability of the nal foam product which is directly
related to the open cell window percentage of a foam (4). Many
physical properties of exible polyurethane foam are highly
affected by its porosity and cell size (5). However, porosity as
well as foam cell size are not perfectly regulated in industry
because the effect of the surfactant is not well understood.
Traditionally, it is believed that the silicone surfactant can
lower surface tension, emulsify incompatible formulation in-
gredients, promote generation of bubbles during mixing, and
stabilize cell windows (3). Among these functions, cell win-
dow stabilization is the most important. During foaming, the
bubbles initially introduced by mechanical mixing will grow.
As the volume fraction of the gas bubbles exceeds 74%, the
spherical bubbles will distort into multisided polyhedrals and
cell windows with Gibbs Plateau Borders are formed. Due to
capillary pressure, pressure inside the plateau borders is lower
than that in the cell windows. This pressure difference will
cause liquid in the cell window to drain into the struts. Without
adding silicone surfactant, this drainage rate will be very fast
so that lm rupture and bubble coalescence occur rapidly. It is
shown in this work that due to the surface tension gradient
generated by the silicone surfactant the cell window drainage
rate is slower. Different surfactant structures will have differ-
ent effects on cell window drainage, resulting in a distribution
of different cell window thicknesses at the time of cell opening
(4). Thus the porosity or the air permeability of the solid
1
To whom correspondence should be addressed.
Journal of Colloid and Interface Science 215, 270279 (1999)
Article ID jcis.1999.6233, available online at http://www.idealibrary.com on
270
0021-9797/99 $30.00
Copyright 1999 by Academic Press
All rights of reproduction in any form reserved.
polyurethane foam is signicantly affected by the structure of
the silicone surfactant used in the formulation.
Surface elasticity is known to be caused by a surface
tension gradient along the cell window upon expansion and
will retard the cell window drainage rate. This effect is
called the GibbsMarangoni effect and was reviewed by
Scriven and Sternling (6). Owen and co-workers studied the
dynamic surface tension of a series of surfactant solutions
and stressed the importance of surface elasticity (7). Similar
work was performed by Zhang and co-workers (8). In this
work, the drop volume method has been used to evaluate the
effect of surfactant composition on dynamic surface tension
in the process of bubble growing. However, surfactant be-
havior in thin lms may be signicantly different from the
behavior of a single interface (9). No lm elasticity or lm
drainage measurements on model polyurethane foam sys-
tems have been reported in the literature.
In this work, a series of silicone surfactants was used by
varying surfactant structure, namely, silicone backbone length
and polyether percentage. Different foam cell sizes and nal
foam porosity have been observed for foams made with these
surfactants. A surfactant solution in polyether polyol is used as
a model nonreactive system to evaluate surfactant perfor-
mance. The effects of surfactant structure on surface tension,
surface elasticity, and vertical lm drainage rate were mea-
sured to characterize the surfactant properties. Effects of sur-
factant chemical structure on nal polyurethane foam proper-
ties were discussed as well.
MATERIALS
All surfactants used in this study were synthesized by the
OSi Group of Witco Corporation. Structures of the surfac-
tants are shown in Fig. 2 and Table 1. Typical nomenclature
of silicone surfactant is MDxDyM, where M is trimethyl-
siloxane ((CH
3
)
3
SiO) or ((CH
3
)
3
Si), D is dimethylsiloxane
((CH
3
)
2
SiO), D is methylsiloxane with polyoxyalkylene
substitute (CH
3
RSiO) where R is composed of polyethylene
oxide and polypropylene oxide, and x and y are the number
of units of D and D, respectively. Siloxane backbone
length ( x y 2) and silicone/polyether ratio ((M
Wsurfactant

y*M
W(polyether branch)
)/y*M
W(polyether branch)
) are the two variables that
are studied in this work. The solvent used in this work is
Voranol 3137 supplied by Dow Chemical. Voranol 3137 is a
polyether consisting of 87% ethylene oxide and 13% propylene
oxide with a molecular weight of 3100.
EXPERIMENTAL
Effect of Surfactant on Bubble Formation
Polyurethane foam formulation and preparation. Foam
formulation used in this work is listed in Table 2. Procedures
FIG. 1. Macroscopic view of different stages during foaming. The four
regions of the process are: (1) Bubble generation and growth, (2) packed
bubble network (foam expansion), (3) urea microphase separation (polymer
stiffening) and cell opening, and (4) nal curing. [Adapted from Ref. (2)]
FIG. 2. The structure of a typical silicone surfactant used in polyurethane
foam systems.
TABLE 1
Experimental Silicone Surfactants
Surfactant
structure x y
Silicone/polyether
ratio
Siloxane
backbone length
MD
x
D
y
M
a
119 21 0.324 142
MD
x
D
y
M 85 15 0.326 102
MD
x
D
y
M 51 9 0.331 62
MD
x
D
y
M 54 6 0.501 62
MD
x
D
y
M 57 3 1.013 62
MD
x
D
y
M
b
51 9 0.124 62
a
Branched copolymer of polydimethylsiloxane and polyether where M
[(CH
3
)
3
SiO] or [(CH
3
)
3
Si], D (CH
3
)
2
Si-O, D (CH
3
) RSi-O, R 60%
propylene oxide and 40% ethylene oxide polyether, 1500 mW, acetoxy capped.
b
D (CH
3
) RSi-O, R 60% propylene oxide and 40% ethylene oxide
polyether, 4000 Mw, acetoxy capped.
271 SILICONE SURFACTANT IN FLEXIBLE POLYURETHANE FOAM
to make polyurethane foam are provided in our previous paper
(4). The polyurethane foams were allowed to cure at room
temperature for at least 48 h before foam specimens were cut
out from the foam bun for further testing.
Surface tension measurements. Surface tension measure-
ments were made by the Wilhelmy method using digital ten-
siometer K10ST. Surfactant solutions in polyether polyol
Voranol 3137 were measured. Prior to each run the surface
tension of pure water was measured. Samples were prepared at
least 48 h before the measurements.
Bubble formation tests. In the polyurethane foam system,
air bubbles are introduced in the mixing step using a high speed
mixer. These air bubbles will serve as starting points for the
foam growth. This is the bubble dispersion stage. No more new
bubbles form during the bubble expansion (10). The energy
required for formation and growing of a bubble is related to the
surface tension of the surfactant solution. Silicone surfactant
can reduce surface tension and thus reduce the energy needed
to form a bubble. As a result, the number of bubbles introduced
into the system as well as the nal foam cell size will be
determined by the surface tension lowering ability of the added
surfactant.
Since the actual polyurethane foam system is reactive and it
is hard to perform initial bubble count, model systems of 1
wt% surfactant solutions in polyol Voranol 3137 were used for
the bubble generation test. This surfactant solution was agi-
tated using a high speed mixer. A small sample of the agitated
mixture was transferred onto a glass slide with a dened gap of
0.3 mm between the cover glass and the glass slide. Optical
microscope images were then obtained and analyzed to give
the number of bubbles per unit volume of solution. A sample
image is shown in Fig. 3.
Scanning electron microscopy. Scanning electron micros-
copy was used to take images of cured solid polyurethane
foams. The size of the foam cell can be evaluated from these
images. All images were taken at an accelerating voltage of 15
kV. All samples were coated with 5 nm of Au/Pt. A sample
image is shown in Fig. 4.
Effect of Surfactant on Foam Stabilization
Air ow. Air permeability of the cured polyurethane foams
was measured according to American Society for Testing and
TABLE 2
Polyurethane Foam Formulation
Chemicals Parts (pphp
a
)
Voranol
TM
3137
b
100
Water 4.0
Dabco
TM
33LV
c
0.30
Dabco
TM
T-9
d
0.20
Silicone surfactant
e
1.0
Voranate
TM
T-80
f
index 110
a
Parts per hundred parts polyol Voranol
TM
3137 (by weight).
b
A 1000 equivalent weight triol containing 13% ethylene oxide, 87%
propylene oxide and all secondary hydroxyl groups (Dow Chemical).
c
33% triethylenediamine in dipropylene glycol (Air Products and Chemi-
cals).
d
Stabilized stannous octoate (Air Products and Chemicals).
e
Polysiloxanepolyoxyalkylene block copolymer (Witco Corp.).
f
An 80:20 mixture of 2,4-, and 2,6-toluene diisocyanate (Dow Chemical). FIG. 3. Images of air bubbles in 1 wt% MD
51
D
9
M solution in Voranol
3137 after high speed mixing.
FIG. 4. A typical cured polyurethane foam image by scanning electron
microscopy.
272 ZHANG ET AL.
Materials (ASTM) standard D 3574. Foam specimens of 5
5 2.5 cm in size were cut out from the top part of the foam
buns in the buckets. The 5 5 cm surface was always
horizontal, i.e., perpendicular to the blowing direction. Air
ow rate was measured by a rotameter with a pressure drop of
1.27 cm H
2
O across the specimen thickness (2.5 cm). It was
shown that the air ow of a foam increases linearly as the open
cell window percentage increases (4). Thus the open cell
window percentage of a foam can be estimated from its air
permeability.
Film elasticity. Details of this instrument are described
elsewhere (9). A diagram is shown in Fig. 5. A glass capillary
provides good wettability, and a surfactant solution in Voranol
3137 can form a stable bubble on the tip of the glass capillary.
For lm tension and dynamic lm elasticity measurements,
initial bubble radius was set to 0.39 mm which is slightly above
hemisphere. The capillary used has a diameter of 0.6 mm. The
pressure inside the bubble can be measured by a highly sensi-
tive pressure transducer. The bubble diameter can be obtained
by direct imaging with an optical microscope. Film tension and
surface tension can be obtained from the YoungLaplace equa-
tion. At equilibrium, the lm tension is twice the surface
tension.
f 2
pR
2
, [1]
where R is the lm radius, p is the measured capillary pressure,
is the surface tension of the lm, and f is the lm tension
(dyne/cm). The bubble can be expanded using a syringe pump
with a constant rate while the pressure inside the bubble is
recorded versus time.
Silicone surfactants with different structures were tested. All
the tests were performed with 0.2 wt% surfactant solution in
Voranol 3137. The following tests were performed.
(1) Equilibrium lm tensions for different surfactant solu-
tions were obtained and compared with results from the Wil-
helmy plate method.
(2) Bubbles were expanded at four different rates ranging
from 0.02 to 0.16 mm
2
/s. Dynamic lm tension was recorded.
Film elasticity can be calculated. An expansion rate of 0.08
mm
2
/s was used to obtain dynamic lm elasticity for different
surfactant solutions. The bubble expansion rates are of the
same magnitude as those in the actual polyurethane foam
process.
(3) Gibbs elasticity was obtained by sudden expansion of
the initially formed bubble to a different area expansion ratio
( A/A
0
). Immediately after expansion, lm tension was re-
corded. Gibbs elasticity was calculated using lm tension
FIG. 6. A typical vertical lm drainage curve obtained by light interfer-
ence microscopy.
FIG. 5. Schematics of the lm elasticity measurement.
273 SILICONE SURFACTANT IN FLEXIBLE POLYURETHANE FOAM
values obtained with four different area expansion ratios as
well as the initial tension without expansion.
Vertical lm drainage tests. There are many vertical lms
in the foam system. One way to evaluate is the vertical lm
drainage method. A 1 wt% surfactant solution was placed in a
vial with a Teon frame of size 1 2 cm. A lm was formed
in the Teon frame by pulling the frame out of the solution.
White light was reected off the lm with a 45 angle and a
color video camera was used to capture the interference pat-
terns. The lm thickness at 1 cm from the top of the frame was
obtained from the interference pattern (11). The lm thickness
change against time can be recorded for solutions made with
different surfactants. The lm thinning rate against lm thick-
ness can be evaluated by taking the slope of the curve. A
sample result is shown in Fig. 6.
Foam column test. A static foam column test was per-
formed in a glass column with a 2.5 cm I.D. glass frit (maxi-
mum pore size 1015 m) to evaluate foam stability (Fig. 7).
Forty milliliters of a 1 wt% surfactant solution were placed in
the column. Compressed air was then injected at rate of
22.525 ml/min until the foam height reached 14 cm, at which
time inlet air was stopped. As the foam collapsed, foam height
was recorded until it reached 7 cm. The time to reach 7 cm is
T
1/ 2
, the half time of the foam. A typical foam column result is
shown in Fig. 8.
Light interferometry. Cell window thickness was mea-
sured using light interferometry. Interference patterns from
white light were used to determine the thickness of a cell
window and the thin-point thickness of each window. Similar
results were obtained using the differential interference tech-
nique (12). The average values of thin-point thicknesses were
obtained for both the high air ow formulation and the low air
ow formulation.
RESULTS AND DISCUSSION
It is known that silicone surfactant can reduce the surface
tension of the foaming system (3). This reduction of surface
tension may have a signicant impact on the bubble dispersion
step. Thus, surface tensions of the surfactant with different
compositions were measured. These results are shown in Figs.
9 and 10. As the polyether percentage of the surfactant in-
creases or the silicone content decreases, the surface tension
increases signicantly. As the silicone/polyether ratio of the
surfactant increases, the surfactant becomes more surface ac-
tive and provides a lower surface tension. On the other hand, if
the silicone/polyether ratio is kept constant, the surface tension
of the surfactant solution does not seem to be affected by the
size of the surfactant or the silicone backbone length of the
surfactant.
A surfactant solution in Voranol 3137 was used as a model
system for the dispersion test. The result is plotted in Fig. 11.
It can be seen that as the surface tension decreases, more air
FIG. 7. Schematics of the foam column test.
FIG. 8. A typical static foam stability curve obtained by foam column test.
FIG. 9. Effect of silicone/polyether ratio of the silicone surfactants on
surface tension of 1 wt% surfactant solution in Voranol 3137. Silicone back-
bone length is kept constant at 62 (SiO) units.
274 ZHANG ET AL.
bubbles per unit volume of solution were generated in the
system. The energy required to generate the bubble surface
with radius r is 4r
2
, where is the surface tension of the
solution. Surfactants with higher silicone/polyether ratios will
give a lower surface tension value and thus reduce bubble
generation energy. As a result, a higher bubble count was
obtained after mixing for systems with a high silicone/poly-
ether ratio surfactant. On the other hand, if the silicone/poly-
ether ratio is kept constant, surfactants with different silicone
backbone lengths give similar surface tension values (24.4
24.6 dyne/cm). Therefore, it is not surprising to see little or no
effect of the surfactant backbone length on the model system
nucleation tests. The cell size for the cured solid polyurethane
foams made with different surfactants is shown in Fig. 12.
Surfactant performance in actual polyurethane foam conrms
the effect of silicone surfactant composition (silicone/polyether
ratio) on bubble dispersion observed in the model system. That
is, for a surfactant with a lower polyether percentage, more
bubbles can be introduced into the system and the resulting
polyurethane foam will have a ner cell size. Surfactants with
different backbone lengths do not affect the bubble count after
mixing, nor do they affect the nal cell size of the solid
polyurethane foam. These results suggest that to get smaller
size cells, surfactants with high silicone content should be
used. However, the surfactant with a very high silicone/poly-
ether ratio yields collapsed foams. This leads to the issue of the
effect of the surfactant composition on the foam porosity.
Another effect of the surfactant is on the rate of cell
window drainage. Polyurethane foam develops from a liquid
foam to a solid foam in a short period of time. It is very
important for the surfactant to stabilize the cell window
before the solidication of the foaming material. This is
accomplished by establishing a surface tension gradient
along the foam lamella and reducing the rate of thinning of
the cell window. A schematic diagram is shown in Fig. 13.
Due to the capillary suction in the plateau border, the liquid
inside the lamella will drain into the plateau border. Without
addition of the surfactant, the gasliquid interface cannot
support any stress and this interface will have a slip bound-
ary condition. The thin liquid lm will rapidly thin and
result in cell window rupture and foam collapse. When a
silicone surfactant is added, a surface tension gradient will
develop along the gasliquid interface and this interface can
have a nite surface stress. As the bubble size increases, the
cell window is formed and expanding. During the cell
window expansion, a surface tension gradient is generated
on the gasliquid interface. The surface tension gradient
FIG. 10. Effect of silicone backbone length of the surfactants on surface
tension of 1 wt% surfactant solution in Voranol 3137. Silicone/polyether ratio
is kept between 0.320.33.
FIG. 11. Effect of surface tension of 1 wt% surfactant solution on number
of air bubbles entrained after high speed mixing.
FIG. 12. Effect of surface tension of 1 wt% surfactant solution on cured
solid polyurethane foam cell size.
275 SILICONE SURFACTANT IN FLEXIBLE POLYURETHANE FOAM
will establish a surface stress and will reduce the drainage
ow rate of the thinning lm. This surface tension gradient
is characterized by lm elasticity. The effect of surface
elasticity on thin liquid lm drainage rate was investigated
by Malhotra and Wasan (13) using numerical simulation. A
signicant effect of the surface elasticity on lm drainage
rate was observed. Larger surface elasticity will cause
slower lm drainage rate. Measuring the lm tension of the
surfactant solution versus time at different expansion rates
can provide information on the lm elasticity and the sur-
face tension gradient.
The lm rheometer is used to measure the lm tension and
lm elasticity of different surfactants. First, the static surface
tensions of various surfactant solutions were measured and
compared with values obtained by the Wilhelmy plate method.
Good agreement was observed. The surface tension values
obtained by the two methods are typically within 1%.
Gibbs elasticity was obtained by recording the surface ten-
sion after the sudden expansion of a bubble. When the bubble
is expanded suddenly, the surface concentration of the surfac-
tant will decrease due to the increase in the surface area of the
bubble. Since the expansion is a sudden event, surfactant
diffusion is ignored. By expanding the bubble at a very rapid
rate, different surface tensions right after expansion will be
obtained. The Gibbs elasticity is dened as
E
f

df
d ln A/A
0

, [2]
where A and A
0
are the expanded and the initial lm areas,
respectively, and f is the lm tension. A typical result is shown
in Fig. 14. As we can see, good reproducibility is obtained. The
Gibbs elasticities for different surfactant solutions are listed in
Table 3. The Gibbs elasticity of the silicone surfactant solution
does not seem to be affected by the composition of the surfac-
tants.
In a polyurethane foam system, bubbles grow at a nite rate
of about 0.02 mm
3
/s rather than sudden expansion. Assuming
the average bubble diameter to be 1 mm, the area expansion
rate will be 0.12 mm
2
/s. The nite expansion rate allows some
surfactant diffusion and adsorption to reduce surface tension as
the bubble expands. As a result, the dynamic lm elasticity
obtained under the nite expansion rate is always lower than
the Gibbs elasticity which is obtained by sudden expansion.
Figure 15 shows that as the expansion rate increases, the
FIG. 13. Schematics of cell window drainage and surface tension gradient formation.
276 ZHANG ET AL.
dynamic lm elasticity will increase because of the decreased
time for surfactant diffusion and adsorption.
From above, it is shown that the lm rheometer can give
accurate, reproducible results. Next, the dynamic lm elastic-
ities of different surfactants were evaluated with an expansion
rate of 0.084 mm
2
/s, which is close to the actual polyurethane
foam system. The higher the dynamic lm elasticity, the higher
the surface tension gradient that can be obtained in the expand-
ing cell window.
As shown in Figs. 16 and 17, as the silicone backbone length
of the surfactant increases, the surface elasticity increases
linearly versus the backbone length of the surfactant. As a
result, a higher lm elasticity and lower cell window drainage
rate can be expected for foams made with these larger silicone
surfactants. This will lead to a smaller open cell window
percentage in the cured polyurethane foam. On the other hand
(Fig. 17), when the silicone/polyether ratio is varied, a less
clear trend is observed. The surfactant with very low lm
elasticity results in a collapsed foam due to its low drainage
retarding ability. The surfactant with elasticity higher than 4.6
dyne/cm will result in a stable foam. When the silicone/poly-
ether ratio of the surfactant is too low (0.124), the lm elas-
ticity test cannot be performed since the bubble on the tip of the
capillary is highly unstable. Due to the low stability of the
lamella, one can expect such a surfactant to result in an
unstable polyurethane foam. In experimental studies, polyure-
thane foams made with this surfactant experience foam col-
lapse.
As discussed above, a lm with higher elasticity will have a
higher surface tension gradient along the lm to retard drain-
age. As a result of the slower drainage rate of the foam cells,
thicker lm will be formed. The higher the lm elasticity, the
thicker the average cell window and the lower percentage of
rupture cell windows there will be. To verify this, the remain-
ing cell window thickness in the cured solid foam is measured
FIG. 15. Effect of bubble expansion rate on dynamic lm elasticity of 0.2
wt% MD
51
D
9
M solution in Voranol 3137.
FIG. 16. Effect of silicone backbone length of silicone surfactant on
dynamic lm elasticity of 0.2 wt% surfactant solution in Voranol 3137 as well
as air permeability of the cured solid polyurethane foam. Silicone/polyether
ratio is kept between 0.320.33.
FIG. 14. Reproducibility of Gibbs elasticity measurements of 0.2 wt%
MD
51
D
9
M solution in Voranol 3137.
TABLE 3
Gibbs Elasticity for 0.2 wt% Surfactant
Solution in Voranol 3137
MD
119
D
21
M MD
85
D
15
M MD
51
D
9
M MD
54
D
6
M
Gibbs elasticity
(dyne/cm) 22.6 23.1 23.1 26.0
277 SILICONE SURFACTANT IN FLEXIBLE POLYURETHANE FOAM
for foams made with different surfactants and different lm
elasticities. Surfactants with different backbone lengths
(MD
85
D
15
M and MD
51
D
9
M) are used since they result in
foams with the same cell size. Figure 18 shows a differential
interference pattern of a cell window; the thickness prole
shows a dimple formation, which is a strong indication of the
existence of a surface tension gradient. The thin part of the
window was measured. Fifty windows were tested for foams
made with these two surfactants. Figure 19 shows the thin-
point window thickness distribution of the remaining cell win-
dows. MD
85
D
15
M has a higher lm elasticity value (7.1 dyne/
cm) than that of MD
51
D
9
M (5.2 dyne/cm). A different
distribution is observed as well: foam with a higher elasticity
has higher lm density toward the thicker lms while foam
with a low elasticity has higher lm density toward the thinner
lms.
The short time foam stability for different surfactant solu-
tions can be evaluated by obtaining the actual thinning rate of
thin liquid lamella as well as the foam column test. Since there
are many vertical lamella in a foam, where gravity plays an
important role, a vertical lm drainage test is performed. A
typical graph is shown in Fig. 6. The lm thinning rate against
lm thickness is plotted as well. The driving force for vertical
lm thinning is the hydrodynamic pressure and the capillary
pressure in the lm. As the lm thickness decreases, the
drainage rate decreases signicantly.
The results of the vertical lm drainage test for different
FIG. 17. Effect of silicone/polyether ratio of silicone surfactant on surface
tension of 0.2 wt% surfactant solution in Voranol 3137 as well as air perme-
ability of the cured solid polyurethane foam. Silicone backbone length is kept
constant at 62 (SiO) units.
FIG. 18. A cell window thickness prole determined by differential light interference microscopy.
FIG. 19. A window thickness distribution for foams made with two
different surfactants. The less open foam is made with surfactant MD
85
D
15
M
while the more open foam is made with surfactant MD
51
D
9
M.
278 ZHANG ET AL.
surfactant solutions are shown in Fig. 20. All drainage rates
were taken when the lm thickness was 0.5 m. A similar
trend is observed for lm drainage rate at different lm thick-
ness. The higher the backbone length of the surfactant, the
slower the drainage rate. In vertical lms, the drainage ow
towards the plateau borders will create a stretching motion
along the surface. A surface tension gradient along the lm will
be created to counteract the stretching drainage ow. The
higher the surface tension gradient a surfactant can establish,
the slower the cell window drainage will be. It was shown that
with the silicone/polyether ratio constant, the surfactant with a
longer silicone backbone can provide a higher surface tension
gradient, and a slower cell window drainage rate will be
obtained. The lifetime of the foams in a static foam column
increase as the backbone length of the surfactant increases. It is
found that the surfactant composition is extremely important to
polyurethane foam quality. There is an optimal point of sili-
cone/polyether ratio for the best polyurethane foam quality
(Fig. 17).
SUMMARY
The effects of silicone surfactant composition (siloxane
backbone to polyether ratio) on the polyurethane foam quality
such as foam cell size and cell lamellae opening (stability)
were studied. When the siloxane backbone/polyether ratio was
varied, due to the effects of surface tension and lamella elas-
ticity, a strong impact of surfactant composition on foam
quality was observed. The surfactant siloxane backbone to
polyether ratio affects both the foam cell size and foam cell
lamella opening. Silicone surfactants with higher silicone con-
tent provide lower surface tension and this leads to a decrease
in bubble size and an increase in bubble generation rate.
However, due to low lamella elasticity these foams are unsta-
ble and contained cells with open lamella. The silicone surfac-
tants with a longer backbone improve the cell lamella elastic-
ity, but because of higher surface tension these foams have
cells with larger lamellae and as a consequence of larger
lamella size, these foams are unstable and contained open
lamellae. Due to the moderated balance of surface tension and
lamella elasticity, surfactants with siloxane backbone to poly-
ether ratio from 0.32 to 0.5 have a good performance on
polyurethane foam quality.
ACKNOWLEDGMENTS
The surfactants used in this study are synthesized by Witco Corp., Osi
Specialities Group. Support for this research was provided by the Center of
Interfacial Engineering, a NSF engineering research center at the University of
Minnesota. Part of the bubble formation and foam column tests were per-
formed by Ms. Lise Bertsch.
REFERENCES
1. Bayer, O., Mod. Plast. 24, 149 (1947).
2. Artavia, L. D., and Macosko, C. W., in Low Density Cellular Plastics
(N. C. Hilyard and A. Cunningham, Eds.), Chap. 2. Chapman & Hall,
London, 1994.
3. Herrington, R., and Hock, K., Flexible Polyurethane Foams. Dow
Chemical Co., Midland, MI, 1991.
4. Yasunaga, K., Neff, R. A., Zhang, X. D., and Macosko, C. W., J. Cell.
Plast. 32, 427 (1996).
5. Jones, R. E., and Fesman, G., J. Cell. Plast. 1(1), 200 (1965).
6. Scriven, L. E., and Sternling, C. V., Nature 187, 186 (1960).
7. Owen, M. J., Kendrick, T. C., Kingston, B. M., and Lloyd, N. C., J.
Colloid Interface Sci. 24, 141 (1967).
8. Zhang, X. D., Macosko, C. W., and Davis, H. T., in Polymeric Foams
(K. C. Khemani, Ed.), Chap. 9, ACS Symposium Series 669. American
Chemical Society, Washington, D.C., 1997.
9. Kim, Y. H., Koczo, K., and Wasan, D. T., J. Colloid Interface Sci. 187, 29
(1997).
10. Kanner, B., and Decker, T. G., J. Cell. Plast. 5(1), 48 (1969).
11. Mysels, K. J., Shinoda, K., and Frankel, S. P., Soap Films-Studies on
Their Thinning and a Bibliography. Pergamon Press, New York, 1959.
12. Nikolov, A. D., and Wasan, D. T., J. Colloid Interface Sci. 113, 1 (1989).
13. Malhotra, A. K., and Wasan, D. T., Chem. Eng. Commun. 55, 95 (1987).
FIG. 20. Effect of silicone backbone length of the surfactants on vertical
lm drainage rate as well as static foam stability of 1 wt% surfactant solution
in Voranol 3137. Silicone/polyether ratio is kept between 0.320.33.
279 SILICONE SURFACTANT IN FLEXIBLE POLYURETHANE FOAM