568
Biomacromolecules 2008, 9, 568–573
Imaging and Thermal Studies of Wheat Gluten/Poly(vinyl alcohol) and Wheat Gluten/Thiolated Poly(vinyl alcohol) Blends Jing Dong,† Rebecca Dicharry,† Eleanor Waxman,‡ Richard S. Parnas,*,†,‡ and Alexandru D. Asandei‡,§ University of Connecticut, Department of Chemical, Materials and Biomolecular Engineering, Storrs, Connecticut, University of Connecticut, Institute of Materials Science, Storrs, Connecticut, University of Connecticut, Department of Chemistry, Storrs, Connecticut Received October 5, 2007; Revised Manuscript Received November 30, 2007
The morphology of wheat protein (WG) blends with polyvinyl alcohol (PVA) and respectively with thiolated polyvinyl alcohol (TPVA) was investigated by atomic force (AFM) and transmission electron microscopy (TEM) as well as by modulated dynamic scanning calorimetry (MDSC). Thiolated additives based on PVA and other substrates were previously presented as effective means of improving the strength and toughness of compression molded native WG bars via disulfide-sulfhydryl exchange reactions. Consistent with our earlier results, AFM and TEM imaging clearly indicate that the addition of just a few mole percent of thiol to PVA was sufficient to dramatically change its compatibility with wheat protein. Thus, TPVA is much more compatible with WG and phase separates into much smaller domains than in the case of PVA, although there are still two phases in the blend: one WG-rich phase and another TPVA-rich phase. The WG/TPVA blend has phase domains ranging in size from 0.01 to 0.1 µm, which are roughly 10 times smaller than those of the WG/PVA blend. MDSC further illustrates the compatibilization of the protein with TPVA via the dependence of the transition temperatures on composition.
Introduction Plastics made from abundant and inexpensive natural proteins, such as wheat gluten,1–5 soy protein,6–8 pea,9 corn zein,10 and so on, are potential substitutes for petroleum-based plastics because they are nontoxic, biodegradable, and environmentally friendly. However, many protein-based plastics have important mechanical property limitations (e.g., brittleness) and readily absorb water after being processed. Two approaches are widely used to improve the mechanical performance and water resistance of protein-based plastics. One approach is based on chemical and radiation treatments. For example, acetylation and esterification11 can modify soy protein side chains, and denaturation can alter the configuration of soy protein.12 Another approach is to blend other materials with the protein, such as plasticizer,13–15 filler,16 or other polymers,7,8,17 to improve properties. Among all the cereal and other plant proteins, wheat gluten (WG) is unique in its ability to form a cohesive blend with viscoelastic properties once plasticized.18 WG consists of a lowmolecular-weight gliadin fraction and a high-molecular-weight glutenin fraction. The glutenin fraction is further made up of high-molecular-weight subunits (HMW-GS) and low-molecularweight subunits (LMW-GS). The x-type HMW-GS has Mw between 80 000 and 90 000, while the y-type HMW-GS has Mw between 65 000 and 75 000. Glutenin also contains negligible levels of free sulfhydryl group.19,20 Thus, all or nearly all sulfhydryl groups are involved in disulfide bonds. The disulfide bonds play an important role in stabilizing the glutenin structure. * Corresponding Author. E-mail:
[email protected]. † University of Connecticut, Department of Chemical, Materials, and Biomolecular Engineering. ‡ University of Connecticut, Institute of Materials Science. § University of Connecticut, Department of Chemistry.
Disulfide mapping of glutenin has provided the direct proof of interchain disulfide bonds between glutenin subunits.21–24 Generally, WG-based plastics require plasticizing agents because of their poor toughness and water resistance. The intermolecular interactions and phase structures of plasticized wheat protein materials have been proposed recently. The hydrogen bonding interactions between plasticizer and protein and the soluble protein components are important factors in determining molecular motion in wheat protein materials. Both factors play key roles in the formation of a cohesive and continuous matrix in the materials to provide excellent mechanical properties.25 Well-known plasticizers of WG include water,26–31 glycerol,32,33 and sorbitol.34 Saturated fatty acids,18 diethanolamine, and triethanolamine35 were also reported to plasticize WG. The use of a plasticizer normally enhances the elongation of WG but significantly reduces the strength. Blending with polymers (e.g., polycaprolactone (PCL),17 poly(hydroxyesterether),36 maleic anhydride-modified polycaprolactone,37 poly(ethylene-covinyl acetate)/poly(vinyl chloride),38 cassava starch,39 poly(butylene succinate) and poly(lactic acid)40) is an alternative approach to improve the properties of WG; however, some of these polymers are too expensive, while others improve the mechanical strength at the expense of the loss of polymer elongation. To address these problems, we have synthesized,41,42 using the esterification of 3-mercaptopropionic acid with poly (vinyl alcohol) (PVA), a multifunctional macromolecular thiol (TPVA), as a WG reactive modifier. This approach appears effective for improving the properties of molded WG plastics through the interaction of the additive with the disulfide bonds between the wheat protein subunits. Earlier, a low-molecular-weight (Mw ) 1247), three-arm, thiol-terminated poly(ethylene oxide) was synthesized to modify WG, but it is too expensive to use in commodity plastics.1
10.1021/bm7011136 CCC: $40.75 2008 American Chemical Society Published on Web 01/16/2008
Wheat Gluten/PVA and Wheat Gluten/TPVA Blends
This paper is intended as a complement to our previous report,41 which illustrated the TPVA/WG interactions with rheology, size exclusion high-performance liquid chromatography (SE-HPLC), three-point bending tests, and water absorption measurements. The present objective is to support the existence of chemical interactions between TPVA and WG by examining the blend morphology. Here, TPVA is shown to interact strongly with the WG through an examination of its microscopic properties using AFM and TEM. Modulated differential scanning calorimetry (MDSC) was also used to determine the strength of interactions between polymer-rich and WG-rich phases.
Experimental Section Materials. American vital WG is from Arrowhead Mills, Hereford, TX. Poly(vinyl alcohol) (PVA) (Mw ) 50k) and 3-mercaptopropionic acid are from Sigma-Aldrich. All were used as received. Techniques. Synthesis of TPVA. TPVA was synthesized by esterification of PVA with 3-mercaptopropionic acid in the presence of hydrochloric acid. PVA (5 g) was dissolved in 20 mL of water at 80 °C and 6 g of 3-mercaptopropionic acid and 1 mL of 7 N hydrochloric acid were added. The reaction was carried out for 8 h at 80 °C, after which the mixture was precipitated in methanol, washed repeatedly, and dried under vacuum at room temperature. Elemental analysis and 1 H NMR determined a thiolation level of 5.2–6 mol % replacement of hydroxyl with the 3-mercaptopropionate. Analytical details are provided in the earlier report.41 Modification of WG. TPVA or PVA was dissolved in 100 mL of 0.05 M acetic acid. TPVA (2 g) was added at an amount to give an equivalent number of thiols to those found in the cysteine residues of the 10 g of WG. Alternatively, for control samples, 2 g PVA were added. WG (10 g) was then dispersed into the mixture and the mixture was stirred overnight. The mixture was then freeze-dried and ground into a powder. The powders were compression molded at 150 °C and 20 000 lbs for 10 min in a 10-cavity mold to form bars of 4 cm × 0.5 cm × 0.2 cm. The mold was coated with mold release agent and cured before molding the WG blends to reduce adhesion between the blend and mold surfaces. Atomic Force Microscopy. Atomic force microscopy (AFM) was performed on compression-molded samples using an Asylum Research MFP 3D. An AC160 silicon tip was used. The tip was mounted on 160 µm long, single-beam cantilever with resonant frequency in the range of 250–350 kHz and corresponding spring constant of 20–60 N/m. Specimens of molded WG/PVA, WG/TPVA, and WG bars were ultramicrotomed by a diamond knife. Transmission Electron Microscopy. Transmission electron microscopy (TEM) was carried out with a Philips 300 electron microscope using an operating voltage of 80 kV. Sections microtomed from the center of compression molded WG, PVA, WG/PVA, and WG/TPVA specimens at room temperature are roughly 200–300 nm thick. They were cut on a dry glass knife and transferred to Formvar-coated copper grids with a Dalmatian-dog-hair probe. No external staining was used. MDSC Study of WG/TPVA and WG/PVA Blend Material. Samples were analyzed using TA Instruments modulated DSCQ100 from TA Instruments. Modulation amplitude was (0.5 °C every 60 s. The samples (10 mg) were loaded and sealed in aluminum pans. Samples were run at 5 °C/min ramp speed from 20 to 200 °C, held at 200 °C for 5 min, cooled at 5 °C/min back to 20 °C, held at 20 °C for 5 min, and then heated a second time to 200 at 5 °C/min. The second heating is reported here. All samples were powder without compression molding, and three replicates were performed for each sample. The Tg temperature and the onset temperature were obtained by analyzing the data using the Universal Analysis software from TA Instruments. TGA Study of WG. TGA study was conducted on a TGA 2950, equilibrated at 20 °C, followed by a ramp to 800 at 5 °C /min.
Biomacromolecules, Vol. 9, No. 2, 2008 569
Results and Discussion AFM Study. AFM has been previously used to study foodrelated systems, ranging from relatively large structures such as starch granules to the organization of secondary structures in proteins and protein interactions.43,44 AFM images of wheat proteins in both dry and hydrated states as well as of gliadins and glutenins were used to study the aggregate behavior in gluten and dough systems.45 In the current study, specimens of molded WG/PVA, WG/ TPVA, and WG bars were ultramicrotomed by a diamond knife and examined using AFM over areas ranging from 1 µm × 1 µm to 5 µm × 5 µm. Samples were kept in desiccators before scanning. Spatial variations of mechanical properties across the surface are shown as image contrast as the AFM tip scans across the sample surface (Figure 1). In tapping mode, the tip was oscillated in and out of contact with the surface while keeping the maximum force constant and small in magnitude to produce height, amplitude, and phase images. The phase contrast image is very sensitive to material surface properties such as stiffness, viscoelasticity, and chemical composition and can provide a qualitative mapping of mechanical properties.46,47 The tapping mode applies almost no lateral force and allows for imaging with very small normal force, thus creating little or no sample damage. Thus, the height image obtained using tapping mode is a direct measurement of the true surface topography even for soft samples. Figure 1 presents the height and, respectively, phase images of WG (a,d), 40% wt WG/PVA (b,e), and 40 wt % WG/TPVA (c,f), which are representative images out of more than 50 images collected for each sample. The concave-shaped features in the WG height image (Figure 1a) likely correspond to holes in the matrix, while the phase image (Figure 1d) provides little contrast. The height and phase images of WG/PVA (Figure 1b,e) also present little contrast because both WG and PVA have similar moduli (∼3.5 GPa for WG and ∼4.0 GPa for PVA), and thus the tapping tip cannot differentiate them well. Because a possible cause of the low contrast in Figure 1e is a phase domain size significantly larger than the 5 µm × 5 µm scan size, line scans of over 1 mm were also conducted but no contrast improvement was obtained. However, in accordance to the larger difference in modulus between TPVA (
