Printer Friendly

Sheet molding compound resins from soybean oil: thickening behavior and mechanical properties.

INTRODUCTION

Unsaturated polyesters (UPE) are formed by the reaction of petroleum-based acids, such as iso-phthalic, orthophthalic, and maleic acid with diols such as ethylene glycol. The C=C double bonds remaining after the ester condensation reaction are capable of free radical polymerization to form a highly crosslinked network. Such resins are commonly used in fiberglass composites, sheet molding (SMC), and bulk molding compounds (BMC), and have enjoyed a long term leadership in petroleum-based composites since 1941 [1]. Their versatility and low cost have made these resins very popular in construction, transportation, and electronic applications. However, the rising cost of petroleum has created significant market stress on these materials. SMC is used in the automobile industry because of its light weight, high strength, dimensional stability, and very good surface quality. Because of finite petroleum reserve lifetimes (~25 years) and the impact of fossil fuel based materials on global warming [2], much research has been conducted on the development of new green materials from renewable resources, such as plant oil [3-6].

In a previous study, we reported the synthesis and characterization of new soy-based resin for SMC applications [7, 8]. The main component of soybean oil is triglycerides, which are three fatty acids connected by a glycerol center through ester linkages, as shown in Fig. 1. The triglyceride has an average of 4.6 carbon-carbon double bonds that can be functionalized [9]. The introduction of free radical polymerizable vinyl groups and low acid functionality to the triglyceride molecule made it possible to be used for SMC applications. Figure 1 shows a typical scheme to prepare SMC resins from soybean oil via a maleated hydroxylated soybean oil (MHSO) and maleated acrylated epoxidized soybean oil (MAESO) [4]. These resins when combined with styrene (~30-50%) react by free radical polymerization to form rigid polymers with very high crosslink density, suited for high volume composite applications such as SMC and BMC [2-4].

It is common in SMC applications to thicken the resin for easy handling and good fiber-carrying capability. The high viscosity increases the stability of the sheet-like structure and reduces the segregation of reinforcing agents during molding and polymerization shrinkage [1]. The most common thickeners for UPE are alkaline earth metal oxides or hydroxides and diisocyanate compounds. Although diisocyanate compounds can produce a high stable viscosity, the formation of covalent urethane bonds that are stable at the molding temperatures can adversely impact the viscosity, mold flow, and sheet welding of both petroleum and bio-based resins [10]. For thickening with divalent cations, magnesium oxide (MgO) is a popular choice because of its low cost and high reactivity [1]. During SMC processing, the initial viscosity should be low enough to permit a good fiber wet-out; during thickening, the resin viscosity should then increase to form a composite sheet (the equivalent of a metal sheet), which should remain stable with a long shelf life. To make a composite part, the SMC sheet is placed in a heated mold under pressure, the viscosity decreases as the divalent cation-acid linkages break at ~70[degrees]C, and the SMC resin flows and fills the mold. A similar synthesis scheme is used to make BMC.

[FIGURE 1 OMITTED]

The thickening mechanism for UPE has been extensively investigated and is a subject of controversy [11-16]. A two-stage reaction theory [15-17] (Fig. 2, Scheme A) postulates that a basic salt is formed first by the reaction of the alkaline earth oxide and polyester acid group. The basic salt then forms a coordination complex by interacting with carbonyl groups of the ester linkages, which possibly results in a three dimensional network, and this reaction is responsible for the large increase in viscosity. In the second theory, Burns et al. [11, 12] have proposed a polymerization theory, whereby the formation of a very high molecular weight species by the reaction of the dicarboxylic acid groups on a polyester chain and magnesium oxide results in high viscosity (Fig. 2, Scheme B). In our study, the thickening behavior of functionalized triglycerides with MgO is examined along with the thermal and mechanical properties of the cured resins.

Epoxidized triglycerides have long been used as a toughening agent for rigid epoxy resins and polyvinyl chloride because of the flexibility and plasticizing effect of the long fatty acid chains [18, 19]. It has been shown that the fracture toughness for acrylated triglycerides when copolymerized with styrene increased with increasing crosslink density when the crosslink density is low [20]. The relationship between the fracture energy [G.sub.IC] and crosslink density v has been well studied [21-23], but remains ambiguous. It is generally agreed that the fracture stress [sigma] should increase with v, but the fracture energy [G.sub.IC] should decrease with increasing crosslink density. As the strength and rigidity of the network increases with v, the strength of the plastic zone increases via [sigma] ~ [v.sup.1/2], but its critical crack opening displacement [delta] decreases because of lower plasticity, approximately as [delta] ~ [v.sup.-1]. Since [G.sub.IC] behaves generally as [G.sub.IC] ~ [sigma][delta], we expect that [G.sub.IC] decreases as [G.sub.IC] ~ [v.sup.-1/2]. Other factors, such as the chemical composition of the network, intermolecular and intramolecular packing, and the perfection of the network can affect the fracture toughness [23]. In this work, the mechanical properties of these new polymers, such as strength, modulus, and fracture toughness are reported and their relationship with the molecular structure is examined.

EXPERIMENTAL

Chemicals

Acrylated epoxidized soybean oil (AESO) has approximately 3.4 acrylates per triglyceride and an average molecular weight of 1200 g/mol. Soybean oil, maleic anhydride (MA), N,N-dimethylbenzylamine (DMBA), hydroquinone, and styrene were obtained from the Aldrich Chemical Co. and used as received. Hydrogen peroxide was obtained from Fisher Scientific in a 30% aqueous form. Formic acid for the epoxidation reaction was obtained from Fluka with 97% purity. Deuterated chloroform for [.sup.1.H] NMR analysis was obtained from Cambridge Isotopes Ltd. The initiator was butyl peroxy benzoate (Luperox(P) from Elf Atochem).

[FIGURE 2 OMITTED]

Synthesis of MAESO

Three different ratios of MA to AESO (Table 1) were examined to understand their effect on the polymer properties. To synthesize the different MA-modified AESO monomers, 50 g AESO and 0.01 wt% hydroquinone were heated to 70[degrees]C at a rate of 1-2[degrees]C/min, while being stirred. The necessary amount of MA (Table 1) was ground up finely and added to the reaction at 70[degrees]C. The reactants were then heated to 80-85[degrees]C and MA was dissolved to form a homogeneous solution. The DMBA catalyst was then added in the amount of 1 g and the reaction was maintained at 80[degrees]C for 2 h under stirring to produce the MAESO.

Synthesis of MHSO

Thousand milliliters of soybean oil was mixed with 1000 ml of 97% formic acid and 550 ml 30% [H.sub.2][O.sub.2]. Ice-water was used externally to keep the temperature below 45[degrees]C. The reaction was vigorously stirred overnight. The resulting emulsion was poured into a separatory funnel and extracted with ether (Fisher Scientific, ACS certified). The water layer was discarded and the ether layer was washed with water, diluted sodium bicarbonate solution (Fisher Scientific, ACS certified), and saturated sodium chloride solution (Fisher Scientific, ACS certified). The resulting ether solution was dried over anhydrous sodium sulfate (Fisher Scientific, ACS certified) and the ether was removed by a rotary evaporator. The resulting product was hydroxylated soybean oil (HSO) with an average 6.8 hydroxyl groups per triglyceride, as determined from a [.sup.1.H] NMR analysis.

HSO was maleinized as follows: 400 g of HSO and 0.2 g hydroquinone were added to a 1-liter round-bottom flask. MA (144.4 g) was added as the mixture was warmed to 60[degrees]C with stirring. The temperature was then raised to 80[degrees]C and 15 ml of DMBA was added in two portions. The reaction was maintained at 80[degrees]C for 5 h under stirring to obtain the maleated product with a 4:1 molar ratio of MA to hydroxylated soybean oil (MHSO).

Acid Number

The acid number of functionalized triglycerides was determined following ASTM D974-95. Approximately 0.2 g of the sample was weighed to the nearest milligram into a 250-ml Erlenmeyer flask. Then 100 ml of titration solvent consisting of toluene, anhydrous isopropyl alcohol and water, and 0.5 ml p-naphtholbenzein indicator were added. After the sample was dissolved, the mixture was titrated with 0.1 N potassium hydroxide solution to the end point (the orange color changes to a green or green brown color). The acid number was calculated as follows:

Acid number, mg of KOH/g = [(A - B) x M x 56.1]/W, (1)

where A is the KOH solution concentration required for titration of the sample, B is the KOH solution concentration required for titration of the blank, M is the molarity of the KOH solution, and W is the sample weight. The free MA was subtracted based on the [.sup.1.H] NMR analysis.

Preparation of Thickened Resins and Viscosity Measurement

The mixture of triglyceride-based monomers and styrene were thickened with magnesium oxide paste (PLASTIGEL[R] liquid thickener PG-9033, Plasticolors, Inc.), which is 38 wt% magnesium oxide in an unsaturated, nonmonomer containing polyester vehicle. In a 600 ml beaker, 100 g triglyceride based monomer was mixed well with 50 g styrene and the desired amount of MgO paste. The mixture was well sealed and kept at room temperature, and to inhibit polymerization 1 wt% of hydroquinone was also added. The viscosity change during the maturation and heating process was monitored using a Brookfield DV-I+ viscometer. When the viscosity of the resin was low, a HB1 spindle was used, and the T-bar spindles (TA91-96) were used for the high viscosity. All measurements were done at room temperature. For the heating process, the beaker with the thickened resin was placed in a 150[degrees]C silicon oil bath and the viscosity changes were recorded.

Curing Study by Differential Scanning Calorimeter

The curing behavior of thickened resins was measured by a DuPont differential scanning calorimeter (DSC) with a DSC cell at atmospheric pressure. All the reactions were conducted in hermetically sealed aluminum sample pans with sample weight ranging from 6 to 10 mg. Twenty grams of MHSO was well mixed with 10 g styrene, 0.6 g t-butyl peroxy benzoate, and 1.18 g MgO paste and the mixture was placed in 10 DSC sample pans. After sealing the DSC sample pans, they were placed into several small vials. The vials were placed in a thermostatically controlled water bath, which was controlled at 25[degrees]C. At different degrees of thickening, the sample pan was taken out and placed into the DSC sample cell. Temperature scanning was done from room temperature to 200[degrees]C at a heating rate of 5[degrees]C/min.

Polymer Synthesis and Mechanical Tests

Resins were prepared by blending triglyceride-based monomers with styrene and a free radical initiator. The MAESO or MHSO was mixed with styrene in the ratio of 100 g monomer to 50 g styrene (33 wt% styrene). The free radical initiator, t-butyl peroxy benzoate was then added in the amount of 1.5 wt% of the total resin weight. Polymer samples were prepared by casting the resin into a vertical gasket mold, curing at 110[degrees]C for 2 h, and post-curing at 150[degrees]C for 2 h. To prevent oxygen free radical inhibition, the resin was purged with nitrogen gas prior to curing.

For the flexural tests, samples were polished and cut into dimensions of approximately 63.5 x 12.7 x 3.2 [mm.sup.3] and measured using ASTM method D790-95a with a span of 50 mm. For the tensile test, the resins were cured in the dog-bone mold at 90[degrees]C for 2 h, 100[degrees]C for 4 h, and post-cured at 150[degrees]C for 1.5 h. The slower cure cycle for the tensile samples was necessary to produce small samples without cracking. Samples were machined down to a thickness of ~3.2 mm. The samples were then conditioned at 50[degrees]C for 24 h to remove any residual water. Tensile properties were measured using ASTM method D638. For flexural modulus and tensile modulus measurements, strain gauges CEA-06-125UN-350 and CEA-06-125UT-350 (Measurements Group) were used to measure the imposed strain, respectively. A labview (National Instruments Corporation) data analysis program recorded the measurements from the strain gauges. The Poisson ratio was calculated from the initial slope of the longitudinal strain versus transverse strain. Flexural and tensile strength measurements were taken without strain gauges. Both tensile and flexural tests were conducted at a constant machine cross-head speed of 1.27 mm/min on an Instron 4502. At least five samples were tested in each case.

For the fracture toughness tests, the molded polymer samples were cured in a silicon rubber mold at 100[degrees]C for 3 h and postcured at 150[degrees]C for 2 h. The samples were polished with final dimensions of 55.9 x 12.7 x 6.4 [mm.sup.3]. Fracture mechanics samples were first prenotched with a diamond cutter and then notched by sliding a sharp razor blade into this prenotch. The crack length, a, was in the range of 5.7-7.0 mm (0.45 < a/W < 0.55, where W is the width of the specimen). The fracture test was performed in single edge notched bend geometry according to ASTM D 5045-99. The testing speed was 1.27 mm/min. The fracture surfaces of the samples were coated with a thin layer of gold and observed by a JEOL JSM-7400F Field Emission Scanning Electron Microscope, operating at 1 kV.

RESULTS AND DISCUSSION

Thickening Behavior

The thickening process is an essential step in the SMC application. The carboxylic acid groups in the functionalized triglycerides are able to react with magnesium ions, which cause at least a 1000-fold increase in viscosity in 2-3 days. Figure 3 shows the viscosity changes of MAESO2 and MHSO system when thickened with 1.5 wt% MgO paste. The starting viscosity for both resins is ~1200 cP. It requires less than 40 h for the viscosity of both resins to reach more than [10.sup.6] cP, which is in a typical viscosity range for the SMC process. After that, the viscosity fluctuates a little, which may result from the humidity change in the environment as the water content can affect the thickening behavior [24].

Compared with the thickening behavior of the commercial UPE, the triglyceride-based resins need a smaller amount of thickener and less time to achieve the same saturated viscosity. This is expected because of the crosslinking architecture of the triglycerides compared with linear UPE during complexation with MgO. The acid numbers of the triglyceride resins are in the range of 40-100 mg of KOH/g, as shown in Table 1, which is higher than that of typical UPE (normally in the range of 30-50). In addition, because of the distribution of carboxylic acid groups on the fatty acid backbones, triglyceride molecules may have dicarboxylic functionalities, or even tricarboxylic acid, which results in the formation of a more crosslinked network, as shown in Fig. 4. Another possible reason is that the triglyceride monomers that have unreacted hydroxyl groups (~1.2 hydroxyl groups/triglyceride), based on the two-stage reaction theory, can coordinate with the magnesium basic salt to form a 3-D network, as shown in Fig. 2, Scheme A. The linkages between the magnesium ion and carbonyl oxygen are weak and they break up at higher temperatures (~70[degrees]C).

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

To understand the viscosity reduction changes during heating, the thickened sheet was placed in a 150[degrees]C silicon oil bath and the viscosity of the sheet was followed by the Brookfield viscometer. Figure 5 shows the viscosity and temperature changes versus time during heating. With the temperature of the resins increasing, the viscosity decreases dramatically from [10.sup.7] cP to 380 cP or even lower, which means that all the thickening bonds are broken, and this whole process takes about 15 min. In the presence of an initiator, the viscosity may not drop that much as the exothermic curing reaction starts.

[FIGURE 5 OMITTED]

The effect of molecular structure and the amount of MgO on the thickening behavior was also examined. Figure 6 shows that MAESO3 initially had a faster viscosity increase than MAESO2, but they finally reach a similar viscosity. Figure 6 also shows that with increasing thickener, the viscosity rises faster with a higher final viscosity. Note that 1.0 wt% of MgO is not enough to reach a moldable viscosity, but more than 2 wt% MgO may be excessive for the thickening process. The compound viscosity design can thus be achieved by combining the resins and different amount of thickeners.

Curing Behavior

The effect of thickening on the curing behavior of triglyceride-based resins was investigated by DSC using the MHSO system. Figure 7 shows the DSC traces for MHSO system, with and without 1.5 wt% MgO. The curing reaction occurs by free radical crosslinking of styrene with the C=C functional groups on the triglycerides, which is reflected in a sharp exothermic peak (Fig. 7). For nonthickened MHSO resins, the initial, peak, and end temperature for curing occur at 117, 127, and 148[degrees]C, respectively (Table 2). With the addition of MgO, the exothermic peak gets broader; the initial curing temperature reduces to 111-113[degrees]C and the end temperature increases up to 171[degrees]C, whereas the peak temperature of curing remains about the same. The thickened samples also exhibit a more prominent second exothermic peak in the range 150-171[degrees]C, as shown in Fig. 7.

[FIGURE 6 OMITTED]

The total heat of reaction, [H.sub.rxn], was determined by (computer) integration of the exothermic peak using a linear baseline. A value of 322 J/g was obtained for the MHSO resin. The thickened samples show reduced heat of reaction when compared with the unthickened samples, as shown in Table 2. This is similar to the curing behavior observed for UPE [25-27]. Thickening MHSO resins with MgO may lead to the formation of microdomains, which results from the aggregation of fatty acid chains with MgO. In these microdomains, the molar ratios of styrene to MHSO would be lower than in the original resin, which could cause a reduction in reactivity of styrene and functional groups on the triglycerides. Lower final conversions of styrene and functional vinyl groups may cause the decrease of the total heat of curing [26]. The effect of thickening on the mechanical properties of these resins was also examined and it was found that there is a slight increase in modulus but no significant effect on strength [28].

[FIGURE 7 OMITTED]

Mechanical Properties and Their Relationship to Molecular Structure

Strength and Modulus. Figure 8 shows the stress-strain behavior of the triglyceride-based polymers from flexural tests. Basically they all show the typical deformation behavior of brittle plastics with an initial high modulus and yielding followed by some plastic deformation. Beyond the yield point, the deformation of MHSO and MAESO1 polymers shows plastic strains up to about 10%, whereas the failure of MAESO2 and MAESO3 occurs at lower strains (~5-6%) but at higher stresses. Table 3 shows that these polymers have a flexural strength in the range of 60-90 MPa and modulus in a range of 1.6-2.4 GPa. The tensile stress-strain behavior of these new polymers also shows a typical deformation of brittle plastics with tensile strength in a range of 27-44 MPa, and tensile modulus in a range of 1.6-2.5 GPa. The Poisson ratio is ~0.4, which is in the range of typical thermosets (Table 3). The Poisson ratio slightly decreases with increasing MA modification. Table 3 also compares the mechanical properties of triglyceride-based polymers with those of typical commercial resins, such as ortho-UPE and iso-UPE, which are used in the SMC and BMC industry. The mechanical properties of these new polymers are comparable with the commercial resins. The primary advantage gained here is in the amount of renewable material in the resins, which contains up to 50 wt% soybean oil.

[FIGURE 8 OMITTED]

It is obvious that with more MA modification, both flexural and tensile properties increase with increasing the molar ratio of MA to AESO. The modification with MA increases the functional groups on triglyceride molecules, resulting in an increase in crosslink density, as shown in Table 1, which increases the mechanical properties. The crosslink density v, was calculated from the rubbery modulus at temperatures well above [T.sub.g] as shown in Table 1 [8]. Figure 9 shows the effect of the crosslink density on the Young's modulus E of triglyceride-based polymers. Initially, E increases rapidly with increasing v, but a further increase in crosslink density has very little effect on the Young's modulus. It has been suggested that the high crosslink density can result in the development of submicroscopic cracks from the internal stress as the mobility of the molecular segments decreases [29].

A percolation model has been used to predict the properties of these new polymers in terms of their structure [30, 31], and is based on rigidity percolation theory developed by Feng and Sen [32], Kantor and Webman [33], and Wool [34]. The critical stress, [[sigma].sub.c], required to break a network is given by Wool as [30]:

[[sigma].sub.c] = {Ev[D.sub.o][p - [p.sub.c]]}[.sup.1/2], (2)

where E is the Young's modulus, [D.sub.o] is the C-C bond rupture energy, v is the crosslink density, p is the perfection of the network, and [p.sub.c] is the percolation threshold. This relation is derived from the concept that the stored strain energy density U = [[sigma].sup.2]/2E becomes critical when it releases to break a percolation fraction [p - [p.sub.c]] of the network bonds, with energy [U.sub.f] = v[D.sub.o][p - [p.sub.c]]. When [p - [p.sub.c]] is essentially constant for perfect nets at about 0.5, then we have the simple relation for the fracture stress as [30]:

[FIGURE 9 OMITTED]

[sigma] ~ [Ev][.sup.1/2]. (3)

Figure 10 shows a comparison of the percolation theory with the experimental tensile strength data. We can see that the percolation theory fits the experimental data very well. From these results, the modulus and strength of triglyceride-based polymers can be increased by simply increasing the chemical functionalities on the triglycerides. This can be achieved by functionalizing highly unsaturated oils such as high linolenic fatty acid, or by the addition of multiple functionality to a more saturated oil, as illustrated in Fig. 1. For example, the SMC resins derived from linseed oil, which contains more than 56 wt% linolenic acid, exhibit much better mechanical properties than corn, soy, hemp, or olive oil when functionalized with just acrylic acid [35]. However, the other oils can be made to give a comparable performance to linseed oil if the equivalent functionality is added through maleinization. The economics of the oil can suggest the optimal chemical pathway. For example, linseed oil is about twice the price of soybean oil, but the cheaper soyoil can be additionally functionalized to give comparable properties to the linseed oil at lower cost, at an approximate price of about $1.00/lb. Thus, it is possible to develop a new low cost bio-based SMC and BMC resins from plant oils with excellent properties by optimizing the fatty acid composition of the starting oil.

[FIGURE 10 OMITTED]

[FIGURE 11 OMITTED]

Fracture Toughness. Fracture toughness, which can be expressed by the critical stress intensity factor [K.sub.IC] and critical strain energy release rate [G.sub.IC], quantify the resistance of a polymer to initiate and propagate cracks. In the Linear Elastic Fracture Mechanics (LEFM) approximation, [K.sub.IC] and [G.sub.IC] are related via the modulus E by,

[K.sub.IC] = [[G.sub.IC]E][.sup.1/2]. (4)

For these triglyceride-based polymers, with an increase in the crosslink density v, a transition from ductility to brittleness happens because of the restriction of the molecular mobility. Figure 11 shows the microstructure of the fracture surfaces near the notch for AESO and MAESO2 based polymers. The fracture surface of the AESO polymer is very rough whereas the MAESO2 polymer has a smooth mirror-like surface, which indicates a brittle structure. Table 4 lists the [G.sub.IC] and [K.sub.IC] values and qualitative surface roughness. [K.sub.IC] for these polymers varies from 0.5 to 1.5 MPa [m.sup.1/2] and [G.sub.IC] is in the range of 110-1371 J/[m.sup.2]. Thus, for a typical SMC resin from soybean oil (MAESO2), values are 0.60 MPa [m.sup.1/2] and 167.9 J/[m.sup.2] for [K.sub.IC] and [G.sub.IC], respectively, which are slightly lower than those of commercial resins, such as vinyl ester. The relationship between fracture toughness and crosslink density is shown in Figs. 12 and 13. Surprisingly, both [K.sub.IC] and [G.sub.IC] do not follow the trend predicted by other studies. LeMay et al. [36] claimed a proportionality of [G.sub.IC] ~ [v.sup.-1/2] based on the fracture behavior with molecular weight between crosslink points, [G.sub.IC] ~ [M.sub.c.sup.1/2]. Levita et al. [37] and Pearson and Yee [38] found a linear relationships between the fracture toughness and [M.sub.c], suggesting [G.sub.IC] ~ [v.sup.-1]. For these triglyceride-based polymers, the fracture toughness decreases sharply (from AESO to MAESO1) with increasing v, and then slows down (from MAESO1-MAESO3) in accord approximately with [G.sub.IC] ~ [v.sup.-2]. The higher crosslink density restricts the mobility of the polymer chains in the network. In addition, the chemical nature of maleate half-esters imparted rigidity to the network, which may reduce the toughness of the network. In Fig. 9, we see that the modulus increases almost linearly with v, as E ~ v, in the region of interest, 3000-6000 mol/[m.sup.3]. The new Twinkling Fractal Theory (TFT) [39] of [T.sub.g] and yield stress predicts that the glass transition temperature increases as [T.sub.g] ~ v [2, 30]. The yield stress for a glass composed of molecules interacting as anharmonic oscillators is predicted to be [39]:

[FIGURE 12 OMITTED]

[FIGURE 13 OMITTED]

[[sigma].sub.y] = [0.16E[D.sub.o](p - [p.sub.c])/[V.sub.m]][.sup.1/2], (5)

where [D.sub.o] is the intermolecular energy (~2-5 kcal/mol) of the anharmonic potential and [V.sub.m] is the molar volume of the oscillators. Essentially, the term [D.sub.o]/[V.sub.m] is equivalent to the cohesive energy density of the polymer in the glassy state. The fraction of oscillators in the glassy state is given by the rigidity percolation parameter p. Basically, when a glass is deformed mechanically, the stored strain energy releases to cause flow by providing the energy to overcome the anharmonic intermolecular potentials and effectively raises the temperature to the glass transition. At [T.sub.g], the number of molecules in the rigid state is in dynamic equilibrium with those in the liquid state and there exists a density of states [phi]([omega]) ~ [[omega].sup.[d.sub.f]-1] of phonon frequency [omega] ([d.sub.f] is the fracton dimension), which describes the transitions of the solid-to-liquid frequencies of the percolating rigidity cluster at [T.sub.g]. These are the Twinkling Frequencies from which the theory derives its name and these frequencies control important properties such as the rate dependence of yield and structural relaxation in the glass. For glassy polymers such as thermosets, whose fracture behavior is determined by yield rather than the breakdown of a craze network at a crack tip, the TFT equation gives the correct magnitude of the yield stress and the proper scaling with respect to crosslink density via [[sigma].sub.y] ~ [v.sup.1/2].

Since in the vicinity of [T.sub.g], at temperature T < [T.sub.g], the modulus E behaves as E(T) ~ ([T.sub.g] - T), we expect that E ~ v. Thus, the molecular mobility is affected by both the shift in [T.sub.g], as shown in Table 1, with respect to the test temperature and the changing crosslink density, as predicted by the TFT. To examine polarity effects induced by additional acid group in the MAESO polymers, a highly acrylated triglyceride (AELO) was synthesized from linseed oil with ~5.7-5.8 acrylates per triglyceride [35]. Compared with the maleated triglyceride-based polymers with the same crosslink density, acrylated polymers show very similar fracture toughness (Figs. 12 and 13), although slightly higher. Thus, the decrease of the fracture toughness mainly results from the increase in the crosslink density. Since [G.sub.IC] [approximately equal to] [sigma][delta], where [delta] is the critical crack opening displacement, with [sigma] ~ [Ev][.sup.1/2] ~ v and with [G.sub.IC] ~ [v.sup.-2], it follows that [delta] ~ [v.sup.-3], in which the exponent of -3 manifests the rapidly decreasing ductility of the network with increasing crosslink density.

There is an apparent threshold for the LEFM fracture properties in the crosslink density range of 3000-3500 mol/[m.sup.3] (Figs. 12 and 13). With molecular weight [M.sub.c] in the range of 310-367 g/mol, the triglyceride molecules go through a transition from highly flexible to nonflexible chains, after which, the fracture toughness is not very sensitive to the crosslink density. The size of the plastic zone, [r.sub.p] can be calculated from [40]

[r.sub.p] = [1/2[pi]] ([K.sub.IC]/[[sigma].sub.y])[.sup.2], (6)

[FIGURE 14 OMITTED]

where [[sigma].sub.y] is the yield or critical fracture stress. Both Fischer [41] and Bos and Nusselder [42] have shown the size of plastic zone is directly proportional to the crosslink density of epoxy resins. Figure 14 shows a plot of the plastic zone size versus crosslink density. The size of plastic zone for AESO polymer is 772 [micro]m, and it significantly decreases to 110-20 [micro]m with increasing v for the MAESO polymers. We can expect the plastic zone size variation with v to behave as follows; [r.sub.p] ~ [[K.sub.IC]/[sigma]][.sup.2] ~ [G.sub.IC]E/Ev ~ [G.sub.IC]/v, and since [G.sub.IC] ~ [v.sup.-2], then [r.sub.p] ~ [v.sup.-3], which is similar behavior as expected for [delta] ~ [v.sup.-3]. The data in Fig. 14 support the result that [r.sub.p] ~ [v.sup.-3] in the crosslink density range of 3000-6000 mol/[m.sup.3], where the experimental slope for the MAESO polymers is -3.08. Additional work is being done with these bio-based resins to improve the toughness using liquid rubber, nanoclays [43], and lignin [44].

CONCLUSIONS

In conclusion, the new thermosetting resins were successfully synthesized from soybean oil with polymerizable functional groups and sufficient acid functionality for SMC applications. This work provided experimental evidence that these resins can be used for SMC and BMC applications. When thickened with MgO paste, these resins exhibited a substantial rise in viscosity at room temperature, which is because of complexation of MgO with the acid groups on the triglyceride molecules. The viscosity of these resins also remained stable during room temperature storage. The thickening behavior is affected by the monomer structure and the amount of thickener. The degree of thickening slightly affected the curing behavior of the resin system but had no effect on the mechanical properties of the resulting cured materials. The flexural strength and moduli of these polymers varied from 61 to 87 MPa and 1.6 to 2.4 GPa, respectively. The tensile strength and moduli varied from 27 to 44 MPa and 1.6 to 2.5 GPa, respectively. The elongation at break was approximately 5.0%. The properties of these polymers can be predicted using the percolation theory, [sigma] ~ [Ev][.sup.1/2] and the Twinkling Fractal theory. The fracture toughness of these new polymers dramatically decreased, which results from the increase in the crosslink density, chemical composition, and intermolecular packing. These new polymers possessed mechanical properties comparable with those of commercially available UPE, which are commonly used in SMC applications.

ACKNOWLEDGMENTS

The authors would like to thank the Center for Composite Materials at the University of Delaware for use of their mechanical testing facility. We also appreciate the thickening agent donation from Plasticolors, Inc.

REFERENCES

1. H.G. Kia, Ed., Sheet Molding Compounds Science and Technology, Hanser/Gardner Publications, Cincinnati (1993).

2. R.P. Wool and X.S. Sun, Bio-Based Polymers and Composites, Elsevier, Burlington, MA (2005).

3. R.P. Wool, S.H. Kusefoglu, G.R. Palmese, R. Zhao, and S.N. Khot, U.S. Patent 6,121,398 (2000).

4. R.P. Wool, J. Lu, and S.N. Khot, U.S. Patent 6,900,261 (2005).

5. A. Guo, D. Demydov, W. Zhang, and Z.S. Petrovic, J. Polym. Environ., 10, 49 (2002).

6. F.K. Li and R.C. Larock J. Appl. Polym. Sci., 80, 658 (2001).

7. J. Lu and R.P. Wool, Abstr. Pap. Am. Chem. Soc., 221, U575 (2001).

8. J. Lu, S.N. Khot, and R.P. Wool, Polymer, 46, 71 (2005).

9. E.H. Pryde, Ed., Fatty Acids, The American Oil Chemists' Society. Champaign, IL (1979).

10. E.G. Melby and J.M. Castro, "Low Temperature Cure of Unsaturated Polyester Resins," in Comprehensive Polymer Science, G. Allen and J.C. Bevington, Eds., Pergamon Press, Oxford, 51 (1989).

11. R. Burns, K.S. Gandhi, A.G. Hankin, and B.M. Lynskey, Plast. Polym., 43, 228 (1975).

12. K.S. Gandhi and R. Burns, J. Polym. Sci. Part A: Polym. Chem., 14, 793 (1976).

13. K.B. Rao and K.S. Gandhi, J. Polym. Sci. Part A: Polym. Chem., 23, 2133 (1985).

14. K.B. Rao and K.S. Gandhi, J. Polym. Sci. Part A: Polym. Chem., 23, 2305 (1985).

15. I. Vancso-Szmercsanyi and A. Szilagyi, J. Polym. Sci. Part A: Polym. Chem., 12, 2155 (1974).

16. I. Vancso-Szmercsanyi and A. Kallo, J. Polym. Sci. Part A: Polym. Chem., 20, 639 (1982).

17. I. Vanscosz, Kautschuk Gummi Kunststoffe, 24, 129 (1971).

18. I. Frischinger and S. Dirlikov, Eds., Toughening of Epoxy Resins by Epoxidized Soybean Oil, American Chemical Society, Washington, DC (1993).

19. D. Ratna, Polym. Int., 50, 179 (2001).

20. J.J. Lascala, The Effects of Triglyceride Structure on the Properties of Plant Oil-Based Resins, PhD Dissertation, Chemical Engineering, University of Delaware (2002).

21. L. Shan, C.G. Robertson, K.N.E. Verghese, E. Burts, J.S. Riffle, T.C. Ward, and K.L. Reifsnider, J. Appl. Polym. Sci., 80, 917 (2001).

22. L. Plangsangmas, J.J. Mecholsky, and A.B. Brennan, J. Appl. Polym. Sci., 72, 257 (1999).

23. V.B. Gupta, L.T. Drzal, C.Y.C. Lee, and M.J. Rich, Polym. Eng. Sci., 25, 812 (1985).

24. F. Alve, J. Polym. Sci. Part A-1: Polym. Chem, 9, 2233 (1971).

25. H. Kubota, J. Appl. Polym. Sci., 19, 2279 (1975).

26. Y.J. Huang and Y.S. Wen, Polymer, 35, 5259 (1994).

27. T.L. Yu and S.C. Ma, J. Macromol. Sci. Pure Appl. Chem., A30, 293 (1993).

28. J. Lu, Synthesis and Characterization of SMC Resins and Nanocomposites from Plant oils, PhD Dissertation, Chemical Engineering, University of Delaware (2004).

29. L.E. Nielsen, J. Macromol. Sci. Rev. Macromol. Chem., C3, 69, (1969).

30. R.P. Wool, J. Polym. Sci. Part B: Polym. Phys., 43, 168 (2005).

31. J.J. La Scala and R.P. Wool, Polymer, 46, 61 (2005).

32. S. Feng and P.N. Sen, Phys. Rev. Lett., 52, 216 (1984).

33. Y. Kantor and I. Webman, Phys. Rev. Lett., 52, 1891 (1984).

34. R.P. Wool, Polymer Interfaces: Structure and Strength, Hanser, New York (1995).

35. J. Lu and R.P. Wool, J. Appl. Polym. Sci., 99, 2481 (2006).

36. J.D. Lemay, B.J. Swetlin, and F.N. Kelley, ACS Symp. Ser., 243, 165 (1984).

37. G. Levita, S. Depetris, A. Marchetti, and A. Lazzeri, J. Mater. Sci., 26, 2348 (1991).

38. R.A. Pearson and A.F. Yee, J. Mater. Sci., 24, 2571 (1989).

39. R.P. Wool, "The Twinkling Fractal Theory of [T.sub.g] and Yield", in the 13th International Conference on Deformation, Yield and Fracture of Polymers, Kerkrade, NL, April 10-13 (2006).

40. J.G. Williams, Fracture Mechanics of Polymers, Halsted Press, New York (1984).

41. M. Fischer, Adv. Polym. Sci., 100, 313 (1992).

42. H.L. Bos and J.J.H. Nusselder, Polymer 35, 2793 (1994).

43. J. Lu and R.P. Wool, Compos. Sci. Technol., in press.

44. W. Thielemans and R.P. Wool, Biomacromolecules, 6, 1895 (2005).

Jue Lu, Richard P. Wool

Department of Chemical Engineering, University of Delaware, Newark, Delaware 19716

Correspondence to: Richard P. Wool; e-mail: wool@udel.edu

Contract grant sponsor: National Research Initiative of the USDA Cooperative State Research, Education and Extension Service; contract grant number: 2005-35504-16137; contract grant sponsor: the EPA-STAR Program.
TABLE 1. The components of maleinization reaction and characteristics of
the resulting monomers and polymers.

 Molar ratio Weight ratio Acid C=C functional
 (AESO:MA) (AESO:MA) number groups added

MAESO1 1:1 100:8.17 43.15 4.4
MAESO2 1:2 100:16.34 65.24 5.2
MAESO3 1:3 100:24.52 106.89 5.9

 Glass transition Crosslink density
 temperature [T.sub.g] ([degrees]C) v (mol/[m.sup.3])[.sup.8]

MAESO1 96 3645
MAESO2 115 4657
MAESO3 133 6322

TABLE 2. Effects of thickening on initial, peak, and final temperature
([T.sub.onset], [T.sub.peak], and [T.sub.end]) of curing and the amount
of exothermic heat of curing.

Thickening time [T.sub.onset] ([degrees]C) [T.sub.peak] ([degrees]C)

No thickener 117.608 127.676
1 h 112.504 122.050
2 days 111.632 128.349
5 days 112.058 127.840
7 days 113.127 127.340

Thickening time [T.sub.end] ([degrees]C) [DELTA][H.sub.rxn] (J/g)

No thickener 147.525 -322.2
1 h 152.588 -231.5
2 days 149.663 -296.1
5 days 171.962 -304.9
7 days 171.962 -297.9

TABLE 3. Mechanical properties of triglyceride-based polymers and their
comparison with the isophthalic unsaturated polyester (Iso-UPE) and
orthophthalic unsaturated polyester (Ortho-UPE).

Polymer sample AESO MAESO1

Flexural strength (MPa) 54.84 [+ or -] 1.53 69.55 [+ or -] 0.72
Flexural modulus (GPa) 1.53 [+ or -] 0.05 1.79 [+ or -] 0.03
Tensile strength (MPa) 21 29.52 [+ or -] 2.66
Tensile modulus (GPa) 1.63 1.81 [+ or -] 0.08
Elongation 0.040
Poisson ratio 0.43 0.412

Polymer sample MAESO2 MAESO3

Flexural strength (MPa) 77.06 [+ or -] 1.50 87.24 [+ or -] 5.24
Flexural modulus (GPa) 2.14 [+ or -] 0.04 2.44 [+ or -] 0.05
Tensile strength (MPa) 39.70 [+ or -] 3.27 44.08 [+ or -] 3.03
Tensile modulus (GPa) 2.18 [+ or -] 0.13 2.47 [+ or -] 0.0.26
Elongation 0.052 0.0439
Poisson ratio 0.394 0.392

Polymer sample MHSO4 Iso-UPE Ortho-UPE

Flexural strength (MPa) 61.42 [+ or -] 0.75 80 130
Flexural modulus (GPa) 1.57 [+ or -] 0.04 3.45 3.59
Tensile strength (MPa) 27.04 [+ or -] 5.71 55 75
Tensile modulus (GPa) 1.61 [+ or -] 0.05 3.45 3.38
Elongation 0.038
Poisson ratio 0.394

TABLE 4. Fracture properties of triglyceride-based polymers.

Polymer [K.sub.IC] [G.sub.IC]
sample Failure type (MPa [m.sup.1/2]) (J/[m.sup.2])

AESO Rough 1.462 [+ or -] 0.068 1371.74 [+ or -] 206.94
MAESO1 Smooth 0.804 [+ or -] 0.053 339.73 [+ or -] 38.74
MAESO2 Smooth 0.597 [+ or -] 0.018 167.90 [+ or -] 13.32
MAESO3 Smooth 0.499 [+ or -] 0.088 110.41 [+ or -] 33.55
COPYRIGHT 2007 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lu, Jue; Wool, Richard P.
Publication:Polymer Engineering and Science
Geographic Code:1USA
Date:Sep 1, 2007
Words:6695
Previous Article:The effect of density profile on the flexural properties of structural foams.
Next Article:Influence of UV absorber on photodegradation processes of poly(vinyl chloride) with different average degrees of polymerization.
Topics:


Related Articles
SMC goes for lighter weight, easier flow.
CPVC.
STARCH BASED POLYMERS.
Photocurable polyanhydrides engineered for orthopedic applications.
Starch Based Polymers. (Buyers' Guide to Thermoplastics).
Properties of rapid prototype injection mold tooling materials.
Self-emulsifiable soybean oil phosphate ester polyols for low-VOC corrosion resistant coatings.
The effect of epoxy-polyester hybrid resin on mechanical properties, rheological behavior, and water absorption of polypropylene wood flour...
A novel biobased resin-epoxidized soybean oil modified cyanate ester.
Effect of resin compositions on microwave processing and thermophysical properties of benzoxazine-epoxy-phenolic ternary systems filled with silicon...

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters