Content uploaded by Mohini Sain
Author content
All content in this area was uploaded by Mohini Sain on Jun 26, 2014
Content may be subject to copyright.
Structure of Glycerol and Cellulose Fiber
Modified Water-Blown Soy Polyol-Based
Polyurethane Foams
INDRANIL BANIK AND MOHINI M. SAIN*
Faculty of Forestry
Centre for Biocomposites and Biomaterials Processing
33 Willcocks Street, University of Toronto
Toronto, Ontario, Canada M5S 3B3
ABSTRACT: In this brief report, the effects of reinforcing agents, glycerol and refined cellulose
fiber, have been studied on the foaming reactions of polyurethane based on polyol derived from soy
bean oil. The presence of glycerol and cellulose fiber has been observed to have a considerable
influence on the density of the foams. Changes in density of different foams have been correlated
with the chemical structures generated in the presence of the components, and analyzed by fourier
transform infrared (FTIR) spectroscopic study and thermogravimetric analysis (TGA). Scanning
electron microscopic (SEM) observations are in line with the modifications induced by the
reinforcing components.
KEY WORDS: glycerol, cellulose fiber, foaming, density, polyurethane, spectroscopy, thermal
analysis.
INTRODUCTION
SOY POLYOL-BASED POLYURETHANE foams are gradually gaining industrial importance
owing to both economic and environmental reasons. Like the conventional
polyurethanes, soy oil-based ones are produced via the reaction of soy polyol with an
isocyanate. For water-blown polyurethanes, foaming basically involves two basic reactions
where the isocyanate reacts with (1) polyol to generate the urethane linkage leading to
curing and (2) with water to form urea and carbon dioxide, the latter expanding the matrix
polymer i.e.:
RNCO þR=OH !RNH COO Rð1Þ
2R NCO þH2O!RNH CO NH RþCO2:ð2Þ
In polyurethane foaming controlling, the above two reactions are critical to obtain the
target morphology and the final physical properties. It is well known that ingredients in
foam formulation, like the catalyst and blowing agent, significantly affect the reactions
*Author to whom correspondence should be addressed. Email: m.sain@utoronto.ca
Figure 2(a–c) appear in color online: http://jrp.sagepub.com
Journal of REINFORCED PLASTICS AND COMPOSITES,Vol. 27, No. 16–17/2008 1745
0731-6844/08/16–17 1745–14 $10.00/0 DOI: 10.1177/0731684407081442
ßSAGE Publications 2008
Los Angeles, London, New Delhi and Singapore
leading to curing and blowing, and affect the final properties. Materials like glycerol
have been reported to reinforce foam, improving the strength [1]. Its role on foam modi-
fication is understood to be as a crosslinker, enhancing the rigidity of the foam system [2].
Though its effect on foam properties is reported, little is known about the relative effect
glycerol exerts on the reactions leading to foaming and curing, particularly involving
soy polyols of different isocyanate (NCO) indices, where an index denotes ratio of
the equivalents of isocyanate to polyol. In this brief communication, an attempt has
been made to identify the effect of glycerol on the foaming behavior of polyurethane
foams based on soy polyol. In addition, some preliminary studies involving an interesting
effect of cellulose fiber, obtained through mechanical refining, on soy polyurethane
foaming have been reported. Chemical structures generated in various foams have
been analyzed by FTIR measurements and complemented by TGA and SEM
observations.
EXPERIMENTAL
Trifunctional soy polyol R3-170, has been obtained from Urethane Soy Systems, Volga,
SD. The hydroxyl number of the polyol is reported to be 170. The silicone surfactant,
DABCO-DC 5357, and tin and amine catalysts, DABCO T-12 and Dabco 33-LV
respectively, have been supplied by Air Products and Chemicals Inc., USA. The aromatic
diisocyanate, PAPI-27, has been procured from Woodbridge Foam Corporation, Ontario,
Canada. It has a functionality of 2.7 and an equivalent weight of 134 g/eqiv.
A typical formulation, as shown in Table 1, has been used to make a brief comparative
study of the effect of glycerol and cellulose fiber on foaming reactions. The concentration
of ingredients is expressed in parts per hundred parts polyol, php. As known, the silicone
surfactant aids in the dispersion of hydrophilic blowing agent, water, into the hydrophobic
soy polyol system. Tin catalyst mainly accelerates the curing reaction leading to urethane
formation, whereas amine functionality catalyzes both the gelling and the blowing
reaction. To justify the effect of cellulose fiber, it has been added as an aqueous suspension
to the soy polyol system. In order to understand the effect of glycerol on foaming, it has
been incorporated at levels of 0, 10 and 25 php, as distinct changes in foaming have been
noted at these levels.
To prepare foams, polyol and all other ingredients have been mixed in the required
proportions, under ambient conditions, in a formable mold for 5 min, and then the
isocyanate has been added and mixed for an additional minute. To see the effect of
cellulosic fiber on foaming, the blowing agent water was substituted by a dilute aqueous
Table 1. Formulation used in the foaming
of polyurethane.
Ingredients Concentration, php
Soy polyol 100.0
Water 2.0
Dabco-DC 5357 1.0
Dabco-T 12 0.4
Dabco-33 LV 0.4
Glycerol 0, 10, 25
NCO index vary
1746 I. BANIK AND M.M. SAIN
suspension of the mechanically disintegrated fiber by the same amount (2 php).
The concentration of the refined fiber used has been extremely low (0.1%), and the
fibers have been obtained through a combination of high shear and impact [2]. The reason
for using such a low concentration of fiber has been to attempt to reduce the possible
agglomeration of the same at higher concentrations [3]. The dimensions of the fiber were
measured and reported to be minute (few microns). The time of stirring has been observed
to be very important in affecting the changes in foam properties, and varied from 5 to
20 min. Sample designation of the fiber-filled foam samples are listed in Table 2. Identical
preparation conditions have been followed for each system and at least four foams were
replicated for each experiment. Foams have been characterized by density, FTIR and SEM
measurements. In some cases, rise height and fiber time has been used to monitor the
foaming reactions. Rise height indicates the ease of foaming, whereas fiber time denotes
the time to gelling and is counted from the moment when isocyanate is added to the
mixture of polyol with other ingredients, and allowed to stir. In polyurethane foaming,
both urethane and urea reactions, as stated above, contribute to blowing.
Apparent densities of all the foam samples have been measured according to the ASTM
D 1621-94 method. At least three samples have been considered for density measurements.
The error in the measurement has been 5%. FTIR spectra of the foam samples have been
recorded on KBR pellets using a Brucker infrared spectrometer. The resolution of the
spectra is 4 cm
1
and 64 scans have been recorded for signal averaging. Comparison of the
spectral absorbances have been obtained by constructing a base line and normalizing the
absorbance of interest by dividing each absorbance by the phenyl absorbtion at 1602 cm
1
[4]. The concentration of the functional groups was expressed as the abosorbance ratio
(AR). Morphology of the samples have been recorded with a Hitachi S-2500 scanning
electron microscope. The samples have been fractured in liquid nitrogen and then coated
with an alloy of gold and platinum using a sputter coater. TGA analysis has been carried
out using TGA analyzer Q 500, TA Instruments, at a heating rate of 108C/min in a
nitrogen atmosphere. Decomposition temperature of the foam samples at a certain weight
loss has been reported to correlate with the structures generated in the foaming process [4].
RESULTS AND DISCUSSION
Figure 1(a,b) shows the variation of the rise height and density of the foam systems,
based on trifunctional polyol with the level of glycerol at NCO indices of 150 and 200.
It is observed that for each foam system there is a decrease in rise height, with
a concomitant increase in the density of the foam indicating a reduction in the ease of
foaming with increase in level of glycerol. However, as the NCO index is raised, there
is an increase in rise height, accompanied by a decrease in the density of the foam at any
level of glycerol. This indicates enhanced foaming at higher NCO index.
Table 2. Designation of different cellulose fiber-filled foam samples.
Blowing agent System
(2 php of water þ5 min stirring) S1
(2 php of aqueous suspension of microfiber þ5 min stirring) S2
(2 php of water þ20 min stirring) S3
(2 php of aqueous suspension of microfiber þ20 min stirring) S4
Structure of Glycerol and Cellulose Fiber 1747
To verify the observed facts, FTIR observations have been made on the above foam
systems both in presence and absence of glycerol. Figure 1(c) shows the IR spectra of a
typical polyurethane foam based on trifunctional soy polyol. The peak assignments are
reported elsewhere [4–6]. Foams show the presence of N-H bending absorption at
1542 cm
1
[6] and urethane and urea carbonyl absorptions at 1743 [5] and 1664 cm
1
[4],
respectively. For simplicity, curing tendency has mainly been monitored by changes in IR
peak related to the urethane carbonyl functionality at 1743 cm
1
, while that corresponding
to blowing by the urea carbonyl group at 1664
1
. Figure 1(d) depicts the variation of
peaks corresponding to urethane and urea functions for glycerol modified PU foam at an
isocyanate (NCO) index of 150. It is observed that the concentration of the urethane
functionalities increases, while that due to urea group is decreased with rise in level of
glycerol. This supports enhanced networking reactions leading to urethane formation at
increasing levels of glycerol. Presence of glycerol increases the hydroxyl functionality of a
polyurethane system [2]. As the glycerol level is increased, an enhanced isocyanate-glycerol
reaction leads to urethane formation, which results in reduced foaming. To check how the
foaming behavior varies with NCO index, IR features have been compared at two NCO
indices. Figure 1(e–f) represents the comparative variation of the urethane and urea
functionalities at isocyanate indices of 150 and 200. It is observed that with increase in the
Gl
y
cerol
(p
h
p)
0 5 10 15 20 25 30
Rise height (cm)
7
8
9
10
11
12
13
NCO index: 150
NCO index: 200
(a)
Figure 1. (a) Rise height of glycerol modified soy polyurethane foams at different isocyanate (NCO) indices.
(b) Density of various soy polyurethane foams at different levels of glycerol and NCO indices. (c) Typical IR
spectra of a polyurethane foam based on soybean oil. (d) Variation of absorbance ratio due to urethane and
urea functional groups of various soy polyurethane foams modified with different levels of glycerol at a fixed
NCO index of 150. (e) Variation of AR due to urethane carbonyl group of various soy polyurethane foams with
levels of glycerol and at different NCO indices. (f) Plot depicting change in AR due to urea carbonyl group of
soy polyurethane foams at different levels of glycerol and NCO indices. (g) Fiber time of different glycerol
modified soy polyurethane foams at various isocyanate indices. (h) SEM photograph of a soy polyurethane
foam mixed without glycerol at an NCO index of 150. (i) SEM micrograph of soy polyurethane foam modified
with glycerol at a level of 10 php and NCO index of 150. (j) SEM photograph of soy polyurethane foam mixed
with glycerol at a dosage of 25 php and an NCO index of 150. (k) SEM photograph of soy polyurethane foam
modified with glycerol at level of 25 php and an NCO index of 200. (l) TGA thermograms of glycerol modified
soy polyurethane foams.
1748 I. BANIK AND M.M. SAIN
isocyanate index there is a decrease in AR of the urethane group, with a concomitant
increase in the concentration of the urea group. This indicates that blowing reactions
involving the urea group predominate over urethane formation with an increased NCO
index. The reason is not clearly understood at this point. As urea reactions are enhanced,
1743
1664
1602
1542
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
(c)
Absorbance units
1800 1750 1700 1650 1600 1550 1500 1450 1400
Wavenumber (cm−1)
Figure 1. Continued.
Glycerol (php)
0 5 10 15 20 25 30
Density (kg/m3)
100
150
200
250
300
350
400(b)
NCO index: 150
NCO index: 200
Structure of Glycerol and Cellulose Fiber 1749
increased matrix expansion favors reduction in density of the foam at a higher NCO index.
To justify the increased matrix expansion at a higher NCO index, fiber times have been
noted. Figure 1(k) depicts the fiber time of glycerol modified PU foams at two different
NCO indices. It is observed that the fiber time of the glycerol modified foams at a higher
NCO index is lower than the corresponding foam samples at a lower NCO index. This
indirectly supports an enhanced blowing tendency at a higher NCO index. In this case,
gelling is understood to be promoted through urea reaction. Urea formation also
contributes to the gelling process in polyurethane foaming [2, 6].
To justify the foaming behavior of various glycerol modified foam samples, SEM
observations have been carried out on different foam systems. Figure 1(h–j) represents
SEM pictures at different levels of glycerol at a fixed NCO index. It is observed that for the
system without glycerol, void structures are generated in random; but as glycerol content is
progressively increased, formation of cellular structures are much restricted. This is
consistent with the IR observation that in presence of glycerol, urethane structures are
generated in amounts higher than the urea groups and a restricted blowing causes a
hindrance to matrix expansion. To compare the microstructure of the foam at a higher
NCO index, microscopic observation is made at a particular level of glycerol. Figure 1(k)
shows the SEM picture of a foam modified with glycerol at a level of 25 php. Comparison
with a similar foam system at a lower index, as shown in Figure 1(j), shows that cellular
structure in the former are more open, indicating an increased tendency towards expansion
of the foam matrix.
In order test the various structures formed, TGA measurements have been conducted.
Figure 1(l) shows the representative thermograms of foam samples, in this case glycerol
modified polyurethane foams. Decomposition characteristics of the foams are listed in
Table 3. The polyurethane foam degradation is reported to be a multi-step process, with
the first stage representing the decomposition of urethane at about 2608C [1]. Based on
Glycerol (php)
0 5 10 15 20 25 3
0
AR
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
2.2(d)
AR-1743
AR-1542
AR-1664
Figure 1. Continued.
1750 I. BANIK AND M.M. SAIN
TGA results, it is observed that the decomposition temperature at 5% weight loss,
T
5
appears to agree more closely with the thermal degradation of polyurethane. From the
table, it is observed that with increase in glycerol, at a certain isocyanate index, temp-
eratures at both 5 and 75% weight loss, T
5
and T
75
respectively, decrease with increase in
glycerol level. However, with an increased NCO index, the former is enhanced though the
later remains somewhat unaffected. Also, from the result, it appears that the influence of
NCO index and glycerol is larger on T
5
than the higher decomposition temperature T
75
.
As indicated by IR measurements, at a certain NCO index, glycerol addition promotes the
Glycerol (php)
0 5 10 15 20 25 30
AR
1.0
1.2
1.4
1.6
1.8
2.0(e)
AR-1743, NCO index=150
AR-1743, NCO Index=200
Glycerol (php)
0 5 10 15 20 25 30
AR
0.6
0.7
0.8
0.9
1.0
1.1(f)
AR-1664, NCO Index=150
AR-1664, NCO Index=200
Figure 1. Continued.
Structure of Glycerol and Cellulose Fiber 1751
urethane structures, and urethanes have low thermal stability [7]. Hence, decrease of T
5
and T
75
at higher glycerol levels may be attributed to the increased formation of unstable
urethane functionalities in the foam matrix. However, as the NCO index is enhanced, T
5
is
raised. This is most probably associated with the formation of thermally stable urea
structures in the foam at a higher NCO index. Urea formation has been reported to
increase the thermal stability of the polymer [8].
The effect of refined fibers has also been studied on soy polyurethane foaming.
Figure 2(a) depicts the density of the fiber modified soy polyurethane foams at an NCO
index of 120. It is observed that there is a large increase in density of the system, S4, made
with higher time of stirring (20 min). At intermediate time of mixing, such a change in
foam density has not been noted. To demonstrate and verify, if the observed changes are
due to effects of an altered mixing time, a system (S3) has been made without
incorporating suspension of the refined fiber. It is noted that there is an effect of mixing
time on the density of the resultant foam, but the observed values are much lower that the
foam incorporated with fiber. This indicates that there is an influence of mixing time on
the dispersion of the fiber. At low and intermediate time of mixing, there seems to be lack
of proper dispersion of the hydrophilic fiber suspension in the hydrophobic soy polyol
system.
To understand the effect of refined fiber on foaming, FTIR measurements have been
carried out. Figure 2(b,c) depicts the variation of peaks at 1743 and 1664 cm
1
.Itis
observed that, for the system S4, there is an increase in the absorbance ratio of
urethane carbonyl followed by a decrease in the urea carbonyl peak compared to the
systems, S1 and S3, without any fiber. This indicates that curing reactions leading to
urethane formation predominates over blowing with the addition of fiber, especially at
higher mixing times. The results are surprising, and the exact reason is not clearly
understood at this point and needs more detailed investigation. But it appears that
fibers act to supply the hydroxyl functionality necessary to promote urethane formation
Gl
y
cerol
(p
h
p)
0 5 10 15 20 25 30
Fiber time (s)
320
340
360
380
400
420
440
460(g)
NCO Index: 150
NCO Index: 200
Figure 1. Continued.
1752 I. BANIK AND M.M. SAIN
with the isocyanate [1]. Cellulosic materials have been reported to act as a source of
hydroxyl functionality, modifying the properties of the foam and stiffening the
polyurethane foam [9]. The refined fibers are extremely small in dimension and are
characterized by high surface area [2,10] having large number of hydroxyl functions.
Hence, a reduced fiber size is expected to cause lower chain entanglement and increase
the mobility of the fiber during processing, thus enhancing the possibility of reaction of
(h) (i)
(j) (
k
)
Figure 1. Continued.
Structure of Glycerol and Cellulose Fiber 1753
more hydroxyl groups on the fiber surface. Refined fibers at a higher concentration
have been observed to loose activity through hydroxyl functionality owing to
agglomeration [3]. More detailed studies would be carried out later to ascertain this
observed fact, using fibers disintegrated for different periods of time. In order to justify
enhanced urethane formation in presence of disintegrated fiber, thermal degradation
studies have carried out to complement the IR observations. Table 4 lists the thermal
decomposition feature of two foam samples in presence and absence of fiber. It is
observed that, in comparison to the sample S1, T
5
is reduced for the fiber-filled foam,
S4. This seems to justify the formation of relatively more, thermally unstable, urethane
groups in the latter.
120(l)
100
80
60
40
20
0
Weight (%)
0 100 200 300 400 500 600 70
0
Temperature (°C)
NCO index150-Glycerol-0 php
NCO index150-Glycerol-25 php
NCO index200-Glycerol-25 php
Figure 1. Continued.
Table 3. Weight loss temperatures of various polyurethane samples.
Foam system
Temperature at 5%
weight loss (T
5
), 8C
Temperature at 75%
weight loss (T
75
), 8C
NCO index:150/Glycerol:0 php 267 474
NCO index: 150/Glycerol: 10 php 253 469
NCO index: 150/Glycerol: 25 php 167 466
NCO index: 200/Glycerol: 0 php 284 474
NCO index: 200/Glycerol: 10 php 267 469
NC index: 200/Glycerol: 25 php 216 469
1754 I. BANIK AND M.M. SAIN
To justify the structure of the micro-fiber modified foams, SEM observations have been
made on two selected systems, S1 and S4, as distinct changes have been noted on these
systems. Figure 2(d,e) denotes the micrographs of two foams corresponding to S1 and S4,
respectively. It is observed that the foam mixed without refined fiber is characterized by
voids much larger in size than one incorporated with micro-fiber. Also, the generation of
void structures are relatively more hindered in presence of micro-fiber. This is due to the
expected preponderance of curing reaction involving urethane formation in presence of
fiber where the matrix is expanded less.
CONCLUSION
Effect of cellulose fiber and glycerol has been studied on the foaming reactions of
polyurethane based on polyol derived from soy bean oil. For refined cellulose fiber
modified samples, density of the foam is enhanced significantly, particularly at higher
mixing times. Infrared spectroscopic studies and thermal analysis indicate a possible
reduced foaming in presence of fiber. The effect appears to be related to enhanced
urethane formation in presence of refined cellulose fibers. In case of glycerol mixed foams,
0
50
100
150
200
250
300
350(a)
S1 S2 S3 S4
PU foams
Density (kg per cubic meter)
Figure 2. (a) Plot showing the density of different soy polyurethane foams in presence and absence of refined
cellulose fibers. (b) Variation of AR due to urethane functional group of various soy polyurethane foam
modified without and with refined fiber. (c) Plot depicting the variation of AR due to urea functional group of
different soy polyurethane foams mixed in presence and absence of refined cellulose fiber. (d) SEM
photograph of soy polyurethane foam mixed without refined fiber. (f) SEM micrograph of a soy polyurethane
foam modified with refined cellulose fiber.
Structure of Glycerol and Cellulose Fiber 1755
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6(b)
S1 S3 S4
PU foams
AR-1743
0.88
0.9
0.92
0.94
0.96
0.98
1
1.02
1.04
1.06(c)
S1 S2 S3
PU foams
AR-1664
Figure 2. Continued.
1756 I. BANIK AND M.M. SAIN
urethane formation has been observed to lead to an increase in the foam density with
increasing levels of glycerol at a particular isocyanate (NCO) index, and the density
decreased, due to formation of urea groups, as the index increased. IR and TGA
measurements justify the chemical groups formed in presence of glycerol. Structures of
different foams as verified using scanning electron microscopic (SEM) observations seem
to be consistent with the IR and TGA observations.
ACKNOWLEDGMENT
The authors acknowledge the financial support received from NSERC-CRD for
carrying out the work.
REFERENCES
1. Guo, A., Javni, I. and Petrovic, Z. (2000). Rigid Polyurethane Foams Based on Soybean Oil, J. Appl. Polym.
Sci., 77: 467–473.
2. Chakraborty, A., Sain, M. and Kortschot, M. (2005). Cellulose Microfibrils: A Novel method of Preparing
using High Shear Refining and Cryocrushing, Holzforschung,59: 102–107.
3. Chakroborty, A., Sain, M. and Kortschot, M. (2006). Reinforcing Potential of Wood-Pulp Derived
Microfibers in a PVA Matrix, Holzforschung,60: 53–58.
(
d
)(
e
)
Figure 2. Continued.
Table 4. Weight loss temperature of different fiber-filled samples.
Foam system
Temperature at 5%
weight loss (T
5
), 8C
Temperature at 75%
weight loss (T
75
), 8C
S1 276 464
S4 266 463
Structure of Glycerol and Cellulose Fiber 1757
4. Falke, P., Kudoke, C., Rotermund, I. and Zaschke, B. (1999). Some Aspects of Matrix Formation of
Flexible Polyurethane Foams, J. Cell. Plast., 35: 43–68.
5. Ning, L., Ning, W. D. and Kang, Y. S. (1997). Hydrogen Bonding Properties of Segmented Polyether
Poly(urethane-urea)Copolymer, Macromolecules,30: 4405–4409.
6. Modesti, L., Zanella, A., Lorenzetti, R. Bertani, R. and Gleria, M. (2005). Thermally Stable Hybrid Foams
Based on Cyclophosphazenes and Polyurethanes, Polym. Degrad. Stab., 87: 287–292.
7. Gaboriaud, F. and Vantelon, J. P. (1982). Mechanism of Thermal Degradation of Polyurethane Based on
MDI and Propoxylated Trimethylol Propan, J. Polym. Sci. Part A,20: 2063–2071.
8. Oprea, S. (2000). Effect of Structure on the Thermal Stability of Polyester Urethane Urea Acrylates, Polym.
Degrad. Stab., 75: 9–15.
9. Rivera Armanta, J. L., Heinze, T. and Martinez, A. M. M. (2004). New Polyurethane Foams Modified
with Cellulose Derivatives, Eur. Polym. J., 40: 2803–2812.
10. Bhatnagar, A. and Sain, M. (2005). Processing of Cellulose-Nanofiber Reinforced Composites, J. Reinf.
Plast. Comp., 24: 1259–1268.
1758 I. BANIK AND M.M. SAIN