Soybean-Oil-Based Waterborne Polyurethane Dispersions: Effects of

Oct 20, 2008 - Soybean-Oil-Based Waterborne Polyurethane Dispersions: Effects of Polyol Functionality and Hard Segment Content on Properties. Yongshan...
1 downloads 11 Views 11MB Size
3332

Biomacromolecules 2008, 9, 3332–3340

Soybean-Oil-Based Waterborne Polyurethane Dispersions: Effects of Polyol Functionality and Hard Segment Content on Properties Yongshang Lu and Richard C. Larock* Department of Chemistry, Iowa State University, Ames, Iowa 50011 Received September 14, 2008

The environmentally friendly vegetable-oil-based waterborne polyurethane dispersions with very promising properties have been successfully synthesized without difficulty from a series of methoxylated soybean oil polyols (MSOLs) with different hydroxyl functionalities ranging from 2.4 to as high as 4.0. The resulting soybean-oilbased waterborne polyurethane (SPU) dispersions exhibit a uniform particle size, which increases from about 12 to 130 nm diameter with an increase in the OH functionality of the MSOL from 2.4 to 4.0 and decreases with increasing content of the hard segments. The structure and thermophysical and mechanical properties of the resulting SPU films, which contain 50-60 wt % MSOL as renewable resources, have been studied by Fourier transform infrared spectroscopy, differential scanning calorimetry, dynamic mechanical analysis, thermogravimetric analysis, transmission electron microscopy, and mechanical testing. The experimental results reveal that the functionality of the MSOLs and the hard segment content play a key role in controlling the structure and the thermophysical and mechanical properties of the SPU films. These novel films exhibit tensile stress-strain behavior ranging from elastomeric polymers to rigid plastics and possess Young’s moduli ranging from 8 to 720 MPa, ultimate tensile strengths ranging from 4.2 to 21.5 MPa, and percent elongation at break values ranging from 16 to 280%. This work has addressed concerns regarding gelation and higher cross-linking caused by the high functionality of vegetable-oil-based polyols. This article reports novel environmentally friendly biobased SPU materials with promising applications as decorative and protective coatings.

Introduction Polyurethanes (PUs) are one of the most versatile polymeric materials with regard to both processing methods and mechanical properties.1 PUs that range from high-performance elastomers to tough rigid plastics can be easily synthesized by the proper selection of reactants. This wide range of achievable properties makes PUs an indispensable component in coatings, binders, adhesives, sealants, fibers, and foams.2,3 However, the conventional PU products usually contain a significant amount of organic solvents and sometimes even free isocyanate monomers.4 The increasing need to reduce volatile organic compounds (VOCs) and hazardous air pollutants (HAPs) has led to increased efforts to formulate waterborne PUs for use as coatings, adhesives, and related end uses.5 Waterborne PUs present many advantages relative to conventional solventborne PUs, including low viscosity at high molecular weight and good applicability, and are now one of the most rapidly developing and active branches of PU chemistry and technology.6 Similar to polyethylene, polypropylene, polystyrene, poly(ethylene terephthalate), and poly(vinyl chloride), PUs are also generally based on fossil feedstocks, the reserves of which are predicted to last for only approximately 80 more years.7 In an age of increasing oil prices, global warming, and other environmental concerns (e.g., waste), a change from fossil feedstocks to renewable resources is necessary for sustainable development into the future.8 Therefore, academic and industrial researchers are increasingly devoting their attention and efforts to the possible use of renewable feedstocks as raw materials for the production of chemicals and polymeric materials.9 * To whom correspondence should be addressed. Tel: +1-5152944660. Fax: +1-5152940105. E-mail: [email protected].

Vegetable oils and animal fats play an important part in renewable resources because of their ready availability and their many versatile applications.10 There have been many studies on the synthesis and characterization of a wide variety of polymers based on vegetable oils, such as alkyd resins,11 epoxy resins,12 polyamides,13 alkenyl succinic anhydrides,14 vinyl polymers,15-18 and PUs.19-25 With the exception of castor and lesquerella oils, hydroxyl groups must be introduced at the unsaturated sites to synthesize PU thermosets based on vegetable oils. This has been accomplished by hydroformylation, followed by hydrogenation;26 epoxidation, followed by oxirane opening;20 ozonolysis, followed by hydrogenation;19,25 and microbial conversion.27 Petrovic´ et al. have extensively studied the use of various soybean-oil-based polyols to synthesize PU thermosets whose structures depend on the type of triglyceride used,21 the nature of the isocyanate group,28 and the degree of cross-linking.29 Recently, Lligadas et al.22-24 have described the synthesis of PU thermosets, which behave like hard rubbers or rigid plastics, from polyether polyols (through a combination of cationic polymerization of epoxidized methyl oleate and reduction of carboxylate groups to hydroxyl moieties) and triols (prepared by the cyclotrimerization of methyl 10-undecynoate and methyl 9-octadecynoate). Because of the hydrophobic nature of triglycerides, the PUs derived from vegetable oils exhibit excellent chemical and physical properties, including enhanced hydrolytic and thermal stability.23-26 However, limited attention has been paid to the synthesis and characterization of environmentally friendly waterborne PUs using vegetable oils as a renewable resource. One of us recently reported the synthesis of waterborne PU dispersions from castor oil and a chlorinated rapeseed-oil-based

10.1021/bm801030g CCC: $40.75  2008 American Chemical Society Published on Web 10/21/2008

Soybean-Oil-Based Waterborne PU Dispersions

polyol.30,31 These novel vegetable-oil-based waterborne PU dispersions have been used to modify starch for the preparation of biodegradable plastics. In our previous work,32,33 a soybeanoil-based waterborne PU dispersion has also been successfully prepared from toluene-2,4-diisocyanate, DMPA, and a chlorinated soybean-oil-based polyol. By the use of this dispersion, novel urethane-acrylic hybrid latexes were synthesized by emulsion polymerization using different amounts of the acrylate monomers butyl acrylate and methyl methacrylate. As such, grafting copolymerization of the acrylate monomers onto the soybean-oil-based PU network occurs, leading to a significant increase in the thermal and mechanical properties of the resulting hybrid latexes. Inexpensive, readily available vegetable-oil-based polyols are good candidates for the synthesis of environmentally friendly waterborne PU dispersions from renewable raw materials, but they also face an important challenge when polyols with high hydroxyl functionality are employed. The high functionality of some of these polyols may lead to gelation and higher crosslinking and therefore present potential difficulties in dispersing the resulting highly cross-linked PU prepolymers into water. Therefore, only soybean-oil-based polyol with a relatively low average hydroxyl functionality of about 2.3 has previously been successfullyusedforthesynthesisofwaterbornePUdispersions.32,33 To enlarge the potential applications of vegetable oils as raw materials for the development of waterborne PUs with high performance significantly, the successful utilization of highly functionalized vegetable-oil-based polyols and a systematic investigation into the effects of the polyol functionality on the synthesis and properties of the resulting vegetable-oil-based waterborne PU dispersions are essential. However, to the best of our knowledge, this has not been reported in the literature thus far. In this work, a series of soybean-oil-based polyols with average hydroxyl functionalities ranging from 2.4 to as high as 4.0 have been prepared by the ring-opening of ESOs with methanol and have been successfully used to synthesize environmentally friendly, soybean-oil-based waterborne PUs with high performance. The effects of the polyol functionality, as well as the hard segment content, on the synthesis, structure, and properties of the resulting soybean-oil-based waterborne polyurethane (SPU) dispersions and cast films have been extensively investigated, providing results that are very promising for the development of environmentally friendly waterborne PU dispersions from renewable raw materials.

Experimental Section Materials. Wesson soybean oil was purchased at the local supermarket and used directly without further purification. Isophorone diisocyanate (IPDI) and dimethylol propionic acid (DMPA) were purchased from Aldrich Chemical. Hydrogen peroxide (30%), formic acid (88%), triethylamine (TEA), magnesium sulfate, methyl ethyl ketone (MEK), and ethyl acetate were purchased from Fisher Scientific. All materials were used as received without further purification. Synthesis of the Epoxidized Soybean Oils and Polyols. Epoxidized soybean oils (ESOs) with differing numbers of epoxide groups have been prepared by reaction of the unsaturation sites of the soybean oil with a mixture of formic acid and hydrogen peroxide according to a literature procedure.18 In brief, the soybean oil (100 g) was added to a 500 mL flask; then, certain amounts of 30% hydrogen peroxide were added, followed by the addition of formic acid under vigorous stirring. The weight ratio between the hydrogen peroxide and the formic acid was held at 1:1.1. The reaction was carried out at room temperature for 24 h. Then, 150 mL of ethyl acetate and 100 mL of distilled water were added, resulting in two layers. The organic layer was washed

Biomacromolecules, Vol. 9, No. 11, 2008

3333

Table 1. General Properties of Methoxylated Soybean Oil Polyols (MSOLs)

polyol MSOL-135 MSOL-149 MSOL-176 MSOL-190 MSOL-200

physical equiv state OH number weight hydroxyl at room (mg KOH/g) (g/equiv) functionality Mw temperature 135 149 176 190 200

417 375 318 295 282

2.4 2.8 3.3 3.7 4.0

1030 1045 1050 1091 1126

liquid liquid liquid liquid liquid

with aqueous sodium bicarbonate solution until a slightly alkaline pH was obtained, and the organic layer was then dried over MgSO4 and filtered. Finally, the clear viscous ESOs were obtained after removal of the organic solvent under vacuum. By an adjustment of the molar ratio of the hydrogen peroxide and the carbon-carbon double bonds in the triglyceride from 2.5 to 3.0, 3.4, 4.1, and 5.0, ESOs averaging 2.0 to 2.3, 2.7, 3.1, and 3.7 epoxide groups per triglyceride (as determined by 1H NMR spectroscopy,18 Varian Associates, Palo Alto, CA) have been successfully obtained. 1H NMR (CDCl3, δ): 0.8-1.1 (CH3 of the fatty acids), 1.2-1.8 (CH2 of the fatty acids), 1.9-2.4 (-CH2CdO-), 2.7 (-CdC-CH2-CdC-), 2.8-3.2 (-CH of the oxirane rings), 4.1-4.3 (-CH2-O-CdO), 5.2-5.6 (-CHdCH-). The methoxylated soybean oil polyols (MSOLs) have been prepared by the ring opening of ESO with methanol.20 Briefly, methanol (100 g), water (10 g), isopropanol (100 g), and fluoroboric acid (48% in water, 4.0 g) were mixed in a flask equipped with a magnetic stirrer and a dropping funnel. The resulting mixture was maintained at 40 °C and stirred vigorously while the ESO (100 g) was added dropwise. The reaction mixture was stirred for an additional 2 h at 50 °C, at which time ammonia (30% in water, 6 mL) was added to quench the reaction. After purification using the same methods that were used for the ESO mentioned above, the clear and viscous polyols with different hydroxyl numbers were obtained. The OH number of the MSOL was determined according to the Unilever method,23 and the results are collected in Table 1. Synthesis of the SPU Dispersions. Scheme 1 depicts the approach used to prepare the SPU dispersions. The MSOL (15 g), IPDI, and DMPA were added to a four-necked flask equipped with a mechanical stirrer, nitrogen inlet, condenser, and thermometer. The molar ratio between the NCO groups of the IPDI, the OH groups of the MSOL, and the OH groups of the DMPA is summarized in Table 2. The reaction was carried out at 78 °C for 1 h under a dry nitrogen atmosphere, and 30 g of MEK was added to reduce the viscosity of the system. After an additional 2 h of reaction, the reactants were cooled to about 40 °C and then neutralized by the addition of TEA (1.2 equiv per DMPA), followed by dispersion at high speed with distilled water to produce the SPU dispersions with a solid content of ∼20 wt % after removal of the MEK under vacuum. Two groups of SPU dispersions have been synthesized in this work. In one, we have maintained a constant ratio between the diisocyanate, the polyol, and the DMPA (entries 1, 2, 5, 6, and 7 in Table 2), which leads to SPUs with an increased polyol functionality. In the other, we have varied the molar ratio of the three components (entries 2, 3, and 4 in Table 2), which affords SPUs with the same polyol functionality but different hard segment contents. The corresponding SPU films were obtained by drying the SPU dispersions at room temperature in a glass mold. Characterization. The FT-IR spectra of the SPU films were recorded on a Nicolet 460 FT-IR spectrometer (Madison, WI). The morphology of the SPU dispersions was observed on a transmission electron microscope (JEOL 1200EX). The dispersions prepared were diluted with distilled water to about 0.1 wt %. One drop of the diluted dispersion was placed on the coated side of a 200-mesh nickel grid in a Petri dish and was characterized after drying. The soluble fraction (SF) of the SPU films was determined as follows. Samples of ∼1 g were cut from the SPU films used in this study, weighed, and then immersed in an excess of N,N-dimethylfor-

3334

Biomacromolecules, Vol. 9, No. 11, 2008

Lu and Larock

Scheme 1. Synthesis of Soybean-Oil-Based Waterborne Polyurethane Dispersions

Table 2. Chemical Composition, Soluble Fraction (SF), and Cross-Link Density νe of the SPU Films molar ratio sample

NCO (IPDI)

OHa

OHb

HS (wt %)c

DMPA (wt %)

SF (%)

νe (mol/m3)

SPU-135 SPU-149 SPU-149I SPU-149II SPU-176 SPU-190 SPU-200

1.7 1.7 1.85 2.0 1.7 1.7 1.7

1.0 (135) 1.0 (149) 1.0 (149) 1.0 (149) 1.0 (176) 1.0 (190) 1.0 (200)

0.69 0.69 0.84 0.99 0.69 0.69 0.69

39.3 41.9 44.8 47.4 45.8 47.8 49.0

5.0 5.4 6.4 6.9 5.9 6.1 6.3

30.4 23.3 19.4 17.9 22.4 21.6 18.9

67 69 73 70 85 116 127

a Hydroxyl molar ratio of the MSOL (the number in parentheses denotes the OH number of the MSOL). segment content ) [mass (IPDI + DMPA + TEA)]/[mass (MSOL + IPDI + DMPA +TEA)].

mamide (DMF) for 48 h at room temperature under stirring. Then, the samples were removed from the solvent and dried under a vacuum at 80 °C for at least 24 h. The dried samples were then weighed, and the SF was calculated by

SF (%) )

Wi - We × 100% Wi

where Wi and We are the weight of the initial and the extracted SPU films, respectively. The dynamic mechanical behavior of the specimens was determined using a dynamic mechanical analyzer (TA instrument DMA Q800) with tensile mode at 1 Hz and a heating rate of 5 °C/min in the temperature range of -80 to 150 °C. The glass-transition temperature (Tg) of the samples was obtained from the peaks of the tan δ curves. Differential scanning calorimetry (DSC) was carried out on a thermal analyzer (TA instrument DSC Q20). The samples were heated at a rate of 10 °C/min from 25 to 100 °C to erase thermal history, cooled to -70 °C at a cooling rate of 10 °C/min, and then heated again to 150 °C at a heating rate of 5 °C/min. The glass-transition temperature (Tg) of the samples was determined from the midpoint temperature in heat capacity change of the second DSC scan. Samples of ∼10-15 mg were cut from the films and used. A Q50 thermogravimeter (TA instrument TGA Q50) was used to measure the weight loss of the SPU films under an air atmosphere.

b

Hydroxyl molar ratio of the DMPA. c Hard

The samples were heated from 100 to 650 °C at a heating rate of 20 °C/min. Generally, 10-15 mg samples were used for the thermogravimetric analysis. The mechanical properties of the SPU films were determined using an Instron universal testing machine (model 4502) with a crosshead speed of 50 mm/min. Rectangular specimens of 80 × 10 mm2 (length × width) were used. An average value of at least five replicates of each sample was taken. The toughness of the polymer, which is the fracture energy per unit volume of the sample, was obtained from the area under the corresponding tensile stress-strain curve.

Results and Discussion Structure and Morphology. The soybean oil employed in this work has 4.5 carbon-carbon double bonds per triglyceride. After epoxidization, some of the double bonds are converted into epoxy groups, which subsequently results in MSOLs with OH functionalities ranging from 2.4 to 4.0 after ring opening with methanol, as shown in Table 1. All MSOLs obtained are clear, slightly yellow liquids. It is important that the distribution of OH groups within the MSOL molecules varies from molecule to molecule. In fact, the OH numbers determined for the MSOLs are the average distributions of OH groups present in the triglycerides, which are responsible for the different properties

Soybean-Oil-Based Waterborne PU Dispersions

Figure 1. FT-IR spectra of (a) SPU-135, (b) SPU-176, (c) SPU-149I, and (d) SPU-149II.

and characteristics of these materials when compared with petrochemical-based polyols. This will play a key role in controlling the structure, morphology, and thermophysical and mechanical properties of the resulting SPU films, as discussed later. Hydrogen bonding is a very important feature in PUs and has a significant effect on the material properties. The hydrogen bonds involve the N-H bonds of the amide group as the donor and the urethane carbonyl, the ether oxygen, or the carbonyl group in the MSOL as the acceptor.34 Thus, FT-IR spectroscopy has been used to investigate the hydrogen bonding in the SPU films to understand the phase structure better. The corresponding FT-IR spectra of the SPU films are shown in Figure 1. A single stretching band is observed at ∼3336 cm-1, which corresponds to the hydrogen-bonded N-H stretching vibration.35 A small shoulder peak at 3400-3500 cm-1, which is attributed to nonhydrogen-bonded N-H stretching, is relatively weak, indicating that most of the amide groups in the SPU films are involved in hydrogen bonding. The IR spectra of the CdO stretching region appears to be composed of three bands at around 1740, 1723, and 1710 cm-1. The peak at 1740 cm-1 is assigned to free CdO stretching, whereas the peaks at 1723 and 1710 cm-1 are due to hydrogen-bonded CdO stretching.36 Hsu et al.37 and Sung et al.38 have attributed a urethane carbonyl stretch at roughly 1710 cm-1 to hydrogen bonding in disordered regions. For the stronger hydrogen bonds in ordered or crystalline regions, stretching occurs at an even lower frequency ranging from 1684 to 1702 cm-1. The ordered hydrogen-bonded carbonyl band is not observed in this work, which indicates the amorphous nature of these SPUs.39 The intensity ratio of the peaks corresponding to the hydrogen-bonded and the non-hydrogen-bonded carbonyl groups is found to increase for SPU-176 (Figure 1a) when compared with that of SPU-135 (Figure 1b), which suggests that more hydrogen bonds may have been formed for SPU-176 than for SPU-135; consequently, there is an increased intermolecular interaction of the hard segments with the soft segments. This is because increasing the OH functionality of the soft segments leads to higher cross-linking and an increased urethane content in the resulting materials.23,24 For the SPU-149 series, the intensity of the band attributed to the hydrogen-bonded carbonyl groups relative to the band attributed to the free

Biomacromolecules, Vol. 9, No. 11, 2008

3335

carbonyl groups increases with an increase in the hard segment content from 44 to 47 wt %, which indicates that the carbonyl groups in SPU-149II (Figure 1c) are hydrogen bonded to a greater degree than those in SPU-149 (Figure 1d). In addition to this observation, it is clearly seen that the bonded N-H stretching shifts slightly to a lower wavenumber for SPU-149II when compared with that for SPU-149I, indicating an increase in the degree of association in the SPU films with higher hard segment content.23 The dependence of the SF of the SPU films on the OH number of the MSOL is summarized in Table 2. A trend toward a decreasing SF of the SPU films with an increasing OH number in the polyol is observed. As we would expect, when the OH number of the MSOL is increased, the resulting polyol with higher functionality has a greater chance of being incorporated into the network because it has more reactive sites than a lower functionality polyol.40 This results in SPU films with relatively high cross-link densities, which will be confirmed later. The SPU-135 from the lowest functionality polyol, MSOL-135, contains ∼30 wt % SF, whereas SPU-200 from the highest functionality polyol, MSOL-200, exhibits a SF of ∼19 wt %. For the SPU films made from MSOL-149 with different hard segment contents (Table 2, entries 2-4), the SF is found to decrease from 23 wt % for SPU-149 with a 40 wt % hard segment content to 18 wt % for SPU-149II with a 47 wt % hard segment content. As the hard segment content increases, the molar ratio of the diisocyanate to the hydroxyl of the polyols increases, which will ensure that more OH groups in the MSOL will react, leading to a relatively low value for the SF.41 Figure 2 shows the TEM image of the particle size for the SPU dispersions. Generally, several factors, such as the hydrophilicity, prepolymer viscosity, ionic group position, chain rigidity, and the chemical structure of the soft segment, play a key role in influencing the particle size of the PU dispersion.4,42 In general, the average particle size of the dispersions is not directly related to the physical properties of the resulting PU films. However, control of the average particle size is important with respect to the particular application of the PU dispersions. For example, dispersions of relatively larger particles are preferred in surface coatings for rapid drying, and smaller particle sizes are desirable when deep penetration of the dispersion into a substrate is essential.43 The SPU-135 dispersion is clear and slightly blue, and, as shown in Figure 2, it exhibits an average particle size of ∼12 nm diameter. The particle size of the SPU dispersion increases with an increase in the OH functionality of the MSOL, and the average particle sizes of approximately 60 and 130 nm diameter are observed for the SPU-149 and SPU-200 dispersions where the OH functionalities of the MSOLs are about 2.8 and 4.0, respectively. In this study, the particle size of the SPU dispersions from MSOLs with different hydroxyl functionalities can be controlled in two ways. First, higher cross-linking can be obtained for the dispersions by increasing the OH functionality of the polyols, which results in an increase in particle size. However, as shown in Table 2, when the OH number of the MSOL increases, the amount of diisocyanate and DMPA in the system also increases to maintain a constant molar ratio between the NCO and OH groups. The increased content of hydrophilic DMPA should have the opposite effect.43 Therefore, the increase in the particle size for the resulting SPU dispersions with different OH functionalities in this study is mainly due to the occurrence of higher crosslinking in the SPU dispersions. When compared with SPU-149, SPU-149II exhibits a smaller particle size of ∼30 nm diameter, which indicates that the DMPA content seems to be a major

3336

Biomacromolecules, Vol. 9, No. 11, 2008

Lu and Larock

Figure 2. TEM images of the dispersions of (a) SPU-135, (b) SPU-149, (c) SPU-200, and (d) SPU-149II.

governing factor in determining the particle size of the SPU dispersions from the same polyol.42,43 The relatively wide range of particle size from 12 nm diameter for SPU-135 to 130 nm diameter for SPU-200 suggests that the dispersions prepared by this technology should be quite promising for applications from adhesives to elastomeric or hard coatings. Thermal Properties. The DSC thermograms of the SPU films are shown in Figure 3. No melting or crystallization transition is observed in the DSC curves, indicating the amorphous nature of these SPUs, which is in good agreement with the IR results. All SPU samples show only one glasstransition temperature (Tg) from 8.9 to 33.5 °C. The Tg value increases with an increase in the OH number of the MSOL. This is mainly due to the higher cross-linking in the soft segment because of an increase in the MSOL OH functionality and the increased content of the hard segments that provide more physical cross-linking through hydrogen bonding.40 For the SPU films from MSOL-149 with different hard segment contents, as shown in Figure 3, the Tg value increases from 14.9 to 24.7 °C with an increase in hard segment content from 42 to 47 wt %. This increase can be attributed to the restricted mobility of the polymer chains as a result of the higher degree of hydrogen bonding between the hard segments and the soft segments.23 DMA has been used to investigate the dynamic mechanical behavior of the SPU films because it is more sensitive than DSC

Figure 3. DSC thermograms of the SPU films.

to the mobility of the soft segments through relaxation at the molecular level. Figure 4 shows the storage moduli and tan δ values of the SPU films from MSOLs with different OH numbers. At temperatures below 0 °C, the SPU films are in the glassy state, and their storage moduli (E′) decrease slightly with increasing temperature. Then, a rapid decrease in the E′ value of roughly 3 orders of magnitude is observed in the temperature range from 0 to 100 °C, corresponding to the primary relaxation

Soybean-Oil-Based Waterborne PU Dispersions

Figure 4. The storage modulus and loss factor (tan δ) as a function of the temperature for SPU films from MSOLs with different OH numbers.

Biomacromolecules, Vol. 9, No. 11, 2008

3337

Figure 5. The storage modulus and loss factor (tan δ) as a function of the temperature for SPU films with different hard segment content.

process (R) of the resulting materials. This modulus decrease corresponds to an energy dissipation that is shown in the tan δ versus T curve, where a maximum is observed in the tan δ curve. At higher temperatures, the modulus reaches a rubbery plateau, which is assigned to the rubbery state. As the OH number of the MSOL increases, the storage modulus of the resulting SPU film increases over the entire temperature range. The rubbery plateau modulus can be explained by qualitative consideration of the cross-linking density (νe) of the thermosets according to the kinetic theory of rubber elasticity using the following equation16

E′ ) 3VeRT where Tg is the glass-transition temperature, E′ is the storage modulus at Tg + 40 °C, R is the gas constant, and T is the absolute temperature at Tg + 40 °C. The cross-link densities of the SPU films increase with increasing OH number of the MSOL and are typically in the range of 6.7 × 101 to 1.3 × 102 mol/ m3, as shown in Table 2. This indicates that the triglyceride arms are increasingly incorporated into the PU networks as the OH number of the MSOL increases, which results in a higher cross-linked SPU film with a resulting enhancement in the rubbery modulus. The tan δ versus T curves of the SPU films show only one relaxation process, which involves energy dissipation and cooperative chain motions. When the OH number of the MSOL is increased from 135 to 200 mg KOH/g, the Tg values of the resulting SPU films shift from 38 to 82 °C, mainly because of the restricted movement of soft segment chains caused by higher cross-linking, as mentioned above. Similar results have also been reported for the thermosets polymerized in bulk from soybeanoil-based polyols and MDI.29,40 In addition to higher crosslinking, there is an increase in the hard segment content because we have maintained a constant molar ratio between the NCO and OH groups. This is another reason for the increase in Tg of the resulting SPU films. Figure 5 shows the storage moduli and tan δ values for the SPU films with different hard segment contents from MSOL149. A shift of the Tg from 60 to 71 °C and an enhancement of the storage modulus are observed as the hard segment moieties increase from 42 to 47 wt %. This can be correlated to the increased number of urethane connections and the increase in intermolecular interactions caused by the hydrogen bonding, which indicates a role for hard segments as physical cross-links and fillers.23,44

Figure 6. The dependence of glass-transition temperature (Tg) of the SPU films on the OH number of the MSOL.

Figure 6 shows the dependence of the Tg of the SPU films on the OH number of the MSOL. According to the Fox-Losheak equation that relates the cross-linking density and Tg45

Tg ) Tg∝ +

K ) Tg∝ + kV Mc

where Tg∝ is the glass-transition temperature of the linear polymer of the same structure, V is the number of cross-links per unit of volume (density/Mc), and K and k are constants for a given system, V should be directly proportional to the OH functionality provided that the conversion is complete in the system. As expected, a linear relationship between the OH number of the MSOL and the Tg of the resulting SPU film is observed over the range of OH numbers from 135 to 200 mg KOH/g, offering respectable r values of 0.998 and 0.980 for the data from the DMA and DSC analyses, respectively. This linear behavior between the Tg value and the OH number of the MSOL for the SPU films is similar to what was reported earlier by Petrovic´ et al.29 and Wilkes et al.40 in their work on PU thermosets produced directly from soybean-oil-based polyols and MDI. The Tg values obtained from DMA analysis are relatively high when compared with those from DSC. Such a difference is very common because of the different nature between two methods. DSC measures the change in heat capacity from frozen to unfrozen chains, whereas DMA measures the change in the mechanical response of these chains.32

3338

Biomacromolecules, Vol. 9, No. 11, 2008

Figure 7. TGA curves for the SPU films from MSOLs with different OH numbers.

Typical TGA curves for the SPU films from MSOLs with different OH numbers are shown in Figure 7, and the TGA data are summarized in Table 3. Generally, PUs exhibit relatively low thermal stability because of the presence of labile urethane groups.26,29 The onset of urethane bond dissociation is somewhere