Comparison of the Mechanical Properties of Cellulose and Starch

126

Biomacromolecules 2010, 11, 126–132

Comparison of the Mechanical Properties of Cellulose and Starch Films Shaomin Sun,*,†,‡ John R. Mitchell,‡ William MacNaughtan,‡ Timothy J. Foster,‡ Valeria Harabagiu,§ Yihu Song,† and Qiang Zheng† Department of Polymer Science and Engineering, Zhejiang University, Hangzhou 310027, China, Division of Food Sciences, School of Biosciences, University of Nottingham, Sutton Bonington Campus, Loughborough, LE12 5RD, United Kingdom, and Petru Poni Institute of Macromolecular Chemistry, Aleea Grigore Ghica Voda, nr. 41A 700487 Iasi, Romania Received August 27, 2009; Revised Manuscript Received October 23, 2009

Some effects of water at levels up to 25% (dry solids basis db) on regenerated cellulose films with a thickness of 100 µm were investigated by dynamic vapor sorption (DVS), X-ray diffraction, tensile testing, and protonnuclear magnetic resonance (NMR). The sorption isotherm fitted by the D’Arcy and Watt model and the increase in NMR T2 with water content suggest that a mobile water fraction appears at water contents above 10%db. Water absorption increased the crystallinity of cellulose films from 31% (dry) to 38% (25%db) and altered the dimensions of crystallites. Mechanical measurements on planar and notched samples at all the water contents used here showed ductile fracture behavior. Although the properties of water in these cellulose films are comparable to previously reported data on starch, cellulose films at low water content are much less brittle than starch. The reasons for this difference are explored.

Introduction As the most abundant natural biopolymer in the world, cellulose has attracted considerable attention. Cellulose and its derivatives have seen extensive industrial exploitation in the areas of paper and fabric, and derivatives such as nitrocellulose, cellulose acetate, methyl cellulose, and carboxymethyl cellulose have an important economic significance.1 Intermolecular hydrogen bonds are different in different polymorphs of cellulose. Cellulose I is a natural polymorph, which can be divided into IR and Iβ crystal structures. In cellulose I (which has parallel arrangement of the chains), the hydrogen bonds exist only between chains belonging to the same sheet.2,3 Cellulose II (which features antiparallel arrangement of chains) can be obtained by recrystallization of native cellulose. In this polymorph, the hydrogen bonds are found also between sheets, that is, they form a three-dimensional (3D) network.2,4 Through the conversion from cellulose I into II, IIII, and IVI, the integral crystallinity index and the crystallite size decreased, and the internal surface area increased. Concomitantly, the Young’s modulus of the fiber decreased and the strain at breaking point increased.5 Due to cellulose’s hydrophilic nature, water plays an important role, mainly through the dissociation and reformation of intra- and intermolecular hydrogen bonds, thus leading to alterations in cellulose structure.6-10 Amorphous cellulose can recrystallize into cellulose II at ambient temperature upon contact with water or water vapor at relative humidities (RHs) above 90%.2,11,12 Cellulose I can be transformed to the cellulose II polymorph by a ball-milling method with a specific amount of water.7 * To whom correspondence should be addressed. Tel.: +86 571 87953970. E-mail: [email protected]. † Zhejiang University. ‡ University of Nottingham (member of the European Polysaccharide Network of Excellence (EPNOE; www.epnoe.eu)). § Petru Poni Institute of Macromolecular Chemistry (member of the European Polysaccharide Network of Excellence (EPNOE; www.epnoe.eu)).

The amorphous phase is proposed to play an important role in determining cellulose properties. The internal strain of the amorphous region of cellulose I decreases and the molecular chain is transformed into a more regular arrangement when water molecules break hydrogen bonds. On the other hand, the amorphous structures present in cellulose II relax and adopt a more expanded arrangement in the presence of water molecules which break hydrogen bonds in the crystalline region of cellulose II.6 Cellulose, especially in the form of cellulose derivative films,13-19 has been widely studied due to the many areas of potential application, however, there has been little research on the mechanical or fracture properties of cellulose films, particularly with respect to the water content dependence of these properties,16,20-24 and no attempt to compare the results with starch films. In contrast, the water content dependence of the mechanical properties of starch has been extensively studied. However, although the water content at which a glass rubber transition occurs at ambient temperature for both polysaccharides is about 20%db,25 in the case of starch, a brittle to ductile transition occurs at about 10%db water. This results in extremely brittle behavior in dry starch films in the absence of a plasticiser such as glycerol26 and a loss in sensory crispness of starchbased food products when the water content exceeds this value.27-29 In this article, we investigate the effects of water on crystalline structure and mechanical properties of regenerated cellulose films with a thickness of about 100 µm and compare the results with previously published work on starch. Despite the chemical similarities between the two molecules there has been a divide between the cellulose and starch worlds. It is our view that a consideration of the similarities and differences between these two important materials may benefit the subject.

Experimental Section Sample Preparation. Cellulose films were provided by the Innovia Film Company Ltd. Viscose was prepared in the standard manner and contained 9% by weight of cellulose and 5.8% by weight of sodium

10.1021/bm900981t CCC: $40.75  2010 American Chemical Society Published on Web 11/12/2009

Mechanical Properties of Cellulose and Starch Films

Biomacromolecules, Vol. 11, No. 1, 2010 M)

Figure 1. Sorption isotherms for cellulose film and fits of GAB and D’Arcy and Watt models. Also shown are water contents in the monolayers and multilayers according to the D’Arcy and Watt model at 25 °C.

hydroxide. It was degassed and ripened to give a Hottenroth number in the range of 7.8-8.2. The viscose was cast on to a glass plate and immersed in an aqueous solution containing 20% by weight of sodium sulfate at 40-50 °C until partial coagulation of the coating was achieved. The resulting film was stripped from the plate and immersed in an aqueous solution containing by weight 12% of sulphuric acid and 20% of sodium sulfate at ambient temperature until coagulation and regeneration was complete. The resulting regenerated film was washed with cold running water until a uniform opaque sheet was observed. The film was dried under isotropic tension at 75 °C. Cellulose films were milled into small pieces using liquid nitrogen for DVS, proton-NMR, and X-ray diffraction tests. The samples were hydrated for at least one week in RH (relative humidity) chambers controlled by saturated salt solutions at 25 °C. A total of 11 saturated salt solutions, LiCl (∼14% RH), MgCl2 (∼34.9% RH), K2CO3 (∼43.2% RH), Mg(NO3)2 (∼52.1% RH), NaBr (∼57.3% RH), KI (∼67.8% RH), NaCl (∼74.6% RH), (NH4)2SO4 (∼79.7% RH), KCl (∼83.3% RH), KNO3 (∼94.6% RH), K2SO4 (∼97.2% RH), and P2O5 powder, were used to maintain specified relative humidities in the closed chambers. Dynamic Vapor Sorption. The water sorption isotherm of cellulose film was measured using a Dynamic Vapor Sorption Analyzer DVS-1 (Surface Measurement Systems Ltd., London, U.K.) equipped with a Cahn D200 microbalance. The experiments were carried out at 25 °C and RH values from 0 to 90%. The sample was predried in the DVS-1 by exposure to dry air (RH ) 0%) until the dry weight of the sample showed no further change. The dry sample was subsequently hydrated stepwise in 10% RH steps by equilibrating the sample weight at each step. The sample was considered equilibrated when the slope of ∆m/ ∆t was no higher than 0.0005 mg/min or, alternatively, if the equilibration time exceeded 700 min. The Guggenheim-Anderson-DeBoer (GAB) and D’Arcy and Watt models were chosen to describe the vapor sorption process of cellulose film and the fitted curves are shown in Figure 1. The GAB model30,31 is expressed as follows:

M0CKaw M) (1 - Kaw)(1 - Kaw + CKaw)

(1)

where M is the equilibrium moisture content (g water/100 g dry solids), M0 is the water content in the monolayer (g water/100 g dry solids), aw is the water activity, and C is a constant related to the multilayer heat of sorption. The constants C and K are temperature-dependent. The modified D’Arcy and Watt model31,32 is expressed as

KKaw baw + caw + 1 + K′aw 1 - baw

127 (2)

where K and K′ are the affinity and number of strong binding sites, respectively, c is a function of the number and strength of weak binding sites and b is a function of the number and strength of multimolecular sorption sites. The D’Arcy and Watt model is composed of three terms, each representing a different type of water binding in the material. The first term denotes monolayer-region 1 describing binding sites where individual water molecules bind strongly (Langmuir sorption). The second term denotes monolayer-region 2 describing sites where clusters of water bind weakly (Henry sorption). The last term denotes multilayerregion 3 describing sites where water is arranged as a collection of molecules in the bulk phase.31,32 The three types of absorption are also displayed in Figure 1. Water Content Determination. Water contents of cellulose film equilibrated at different RH values were determined by drying the samples in an oven at 105 °C. The water content of the sample was calculated on a dry basis. Low Resolution Proton NMR. Low resolution proton NMR was carried out on a Resonance Instruments (RI) Maran benchtop spectrometer operating at 23 MHz. The 90° pulse lengths were approximately 3 µs with repeat times of 4 s. A total of 4K of data were normally collected with the dwell times for simple free induction decay measurements being 1 µs. The transverse relaxation time (T2) values were determined using a Carr Purcell Gill Meiboom pulse sequence with a single Tau value of 32 µs and typically 2048 echoes being recorded. Data was fitted to single exponential decay curves using RI software. The longitudinal relaxation time T1 measurements were made using an inversion recovery pulse sequence with appropriately chosen delay times. Wide Angle X-ray Diffraction. X-ray diffractograms were recorded on milled samples for 2θ values between 4 and 38° at 0.1° intervals with a scanning time of 6 s per interval, using a Bruker D5005 (Bruker AXS, U.K.) diffractometer at 20 °C. The diffractometer was equipped with a copper tube operating at 40 kV and 40 mA producing Cu KR radiation of 1.5418 Å wavelength. Crystallinity was calculated according to literature methods.33 Crystallite sizes corresponding to each sharp diffraction peak and at right angles to the diffracting planes were calculated using the Debye-Scherrer equation:

crystallite size ) [0.9 × X-ray wavelength]/[corrected peak width × cos(θ)] where θ is the angle of diffraction. There are several contributions to the peak width, including instrumental broadening (Binst), lattice distortions (Bdist), and the finite size of the crystals (Bsize). As in Ibbett et al.,34 a correction was only made for the instrumental broadening using a value of 0.15 for the integral breadth of the 28.4° peak of the silicon powder standard.

(Bobserved)2 ) (Binst)2 + (Bdist)2 + (Bsize)2

(3)

Tensile Tests. The tensile tests were performed on a Texture Analyzer (TA.XT plus, Stable Micro Systems, U.K.) using the tensile grips with a 30 kg Load Cell. The tests were conducted with an extension rate of 10 mm min-1 at ambient temperature (about 25 °C), with a 50 mm gauge length (specimen size: thickness ∼ 100 µm, width ) 10 mm, and length ∼ 100 mm). At least five specimens for each sample were used. Single edge notched tensile (SENT) specimens with a notch length of 3 mm were also used. The notches were inserted using sharp razor blades. At least three SENT specimens at the same RH were tested.

128

Biomacromolecules, Vol. 11, No. 1, 2010

Sun et al.

Table 1. GAB Model and D’Arcy and Watt Model Parameters for the Cellulose Film at 25 °C GAB model Mo (g/100 g) 6.31

c

K

D’Arcy and Watt model R2

K

K′

c

b

R2

4.39 0.81 0.9987 2.13 2.05 12.14 1.02 0.9994

Results and Discussion Water Sorption. The sorption isotherm of cellulose film determined by Dynamic Vapor Sorption (DVS) is shown in Figure 1. The water content of the cellulose sample measured using the DVS at all RHs is higher than that of native cellulose or microcrystalline cellulose particularly in the high RH region.2,35 A highly crystallized sample exhibits little moisture sorption because of the highly ordered and compact hydrogen bonded structure. The unit cell of both cellulose I and cellulose II is anhydrous. Therefore, there must be amorphous regions in the cellulose film, which can absorb water. Two classical models, namely, the GAB and D’Arcy and Watt models, were used to fit the experimental data. Both gave good fits to the sorption isotherm of the cellulose film. The fitting parameters for both models are listed in Table 1. According to the D’Arcy and Watt model, multilayer sorption begins around RH ) 60%. The curve for monolayer region 1 is almost level, close to 0% water content, suggesting that there is very little strongly bound or disrupted water. The appearance of multilayer sorption at a water content of about 10%db is consistent with the NMR results shown in Figure 2, where the spin-spin relaxation time T2 starts to show an increase at this water level. Hills et al.36 proposed that the water activity above the RH associated with the appearance of bulk water is the result of a rapid exchange between a component of water associated with the biopolymer and a component, which is effectively bulk water. The increase in both T1 and T2 in the region of 10%db indicates the onset of exchange between protons in a bulk phase of water that is only present above 10%db and the exchangeable protons on the cellulose. In addition, there is the exchange of water molecules between the surface and the bulk environments. The resultant T2 is then effectively an average of these three values weighted by the respective content of each fraction. Consequently, the T2 value increases as the amount of bulk water increases. Three site models have been proposed recently for cellulose with varying morphologies and hydroxyl functional polymer gels.37,38 A similar argument is put forward here for the T1 increase on the wet side of the minimum value. In addition

Figure 2. Effect of water content on the relaxation times T1 and T2 of cellulose films.

Figure 3. X-ray patterns of cellulose films at ambient temperature with different water contents (dry basis).

to exchange, there are through-space averaging possibilities for T1 processes. Hence, the entire broad minimum region may represent the contribution of several of these processes. The initial decrease in T1 represents the gradual increase in mobility for all processes as the minimum is approached. The subsequent increase in T1 probably represents the dominant contribution of bulk water in accordance with well established theories of T1 and T2 relaxation in NMR. From a combination of the NMR results and the D’Arcy and Watt region 3 multilayer plot, we can conclude that in the region of 10%db or so (RH approx 0.6-0.7) bulk water exists in the sample. Effect of Water Content on Crystallinity. In addition to water mobility, an important factor that would be expected to influence mechanical properties is the degree of crystallinity. The X-ray diffractograms of cellulose films with different water contents at ambient temperature are illustrated in Figure 3. The peaks at approximately 2θ ) 12, 20, and 22°, corresponding to planes in the sample with Miller indices of 11j0, 110, and 020 (direction c of the unit cell is along the chain axis of the polymer), show that the crystalline component is cellulose II.5,39 For the cellulose II crystalline structure, models have been proposed where the crystals consist of two antiparallel and crystallographically independent chains.40 An increase in water content does not affect the positions of the three peaks. The diffractograms in Figure 3 have been fitted by a series of Gaussian curves, and the crystallinity index, while still plagued by the usual problems of obtaining an absolute value for the index, nevertheless returns convincing evidence for a gradual increase in crystallinity with water content (Figure 4). This is almost certainly a consequence of the increased mobility allowing disordered chains to slot into the correct crystal orientation. The crystalline phase is anhydrous; therefore, the water present concentrates in the amorphous phase and raises the water activity. The water then moves through the atmosphere into the salt solution as the system tends toward equilibrium and the amount of water in the total solid material becomes less. As the water content is expressed relative to the amorphous + crystalline material, it changes; therefore, the sorption isotherm becomes dependent on the degree of crystallinity. Perhaps of more interest and relevance to the mechanical properties considered here are the measures of crystallite dimension. Figure 5 shows the surprising result that there is an



129

Figure 4. Effect of water content on the crystallinity of cellulose films. Figure 6. Stress-strain curves of cellulose films with different water contents at ambient temperature.

Figure 5. Effect of water content on the crystalline sizes of cellulose films.

asymmetric change in crystal dimension. On consideration of the aforementioned Miller indices, we can allocate a dimension perpendicular to the Miller planes for each peak and so give an indication of the change in shape of the crystallites due to differential changes in the growth of the crystal face. Of particular interest is that the dimension perpendicular to the 11j0 plane appears to decrease while that perpendicular to the 110 plane appears to increase. This implies a change in shape and an increase of surface area of the crystallites. Mechanical Properties. Figure 6 shows the stress-strain (σ-) curves for cellulose films with different water contents at ambient temperature. All the samples showed thermoplasticlike behavior with stress increasing rapidly at small strains and more slowly after a yield point. Cellulose films show this ductile behavior even at the relatively low hydration level of 5%db. This has been attributed in previous work to the amorphous region. Okajima and Kai41 found that the Bast fiber Ramie could be divided into amorphous, intermediate, and crystalline cellulose. The intermediate region has an important role in determining the effect of moisture on the structures of cellulose fibers. Water molecules can enter the intermediate regions causing swelling of the cellulose molecules and converting the badly ordered crystalline regions into amorphous structures.42 The amorphous chains of cellulose II adopts a less dense conformation in the presence of water which breaks hydrogen bonds in the crystalline region,6 thus increasing the ductility of cellulose films.43 Although the water effect on the strain/

Figure 7. Effect of water content on the elongation/strain at break for cellulose films at ambient temperature.

Figure 8. Effect of water content on the tensile strength and Young’s modulus for cellulose films at ambient temperature.

elongation at break is weak (Figure 7), a small increase could still be observed, which agrees with previous work.43 The strain at break levels off when water content reaches about 10%db. The increase in strain at break is probably caused by the breaking of hydrogen bonds allowing more extension. Figure 8 shows the effect of water content on the tensile strength and Young’s modulus of cellulose films. The tensile parameters are of the same order as those reported for uniaxially

130


Sun et al.

Figure 9. Hydrogen bond density of cellulose as a function of water content.

drawn cellulose films21 and cotton and wood based cellulose fibers.44 An increase of water content has been reported to decrease tensile strength and Young’s modulus due to water plasticization, and the values reported here are consistent with those reported by Hatakeyama et al.6 The decrease of tensile strength and Young’s modulus begins again at a water content of approximately 10%db. We propose that these changes are probably due to a general loosening effect of water on a composite like structure of crystallites embedded in an amorphous phase. The water will be in the form of linear arrays which can weaken the average hydrogen bonding in the whole cellulose structure by a process of exchange from cellulose/cellulose to cellulose/water hydrogen bonds. This water could be in the amorphous phase or concentrated in the interfacial region around the surface or in clefts of cellulose crystallites bearing in mind that the unit cell contains no water. A sufficiently long array of water molecules, which is necessary for this exchange process, may only occur above a certain water content ∼ 10%db (Figure 12). The dependence of mechanical properties, such as the Young’s modulus, on the level of hydrogen bonding in hydrogen bond dominated solids has been investigated previously.45 The equation

E ) kN1/3

(4)

where E is Young’s modulus measured in dyn/cm2, N is the number of effective hydrogen bonds per cm3, and k for cellulose is approximately 8 × 103, has been proposed to describe typical hydrogen bond densities in solids such as cellulose. A plot of the hydrogen bond density as a function of water is shown in Figure 9. This shows a transition once again around the 10%db water content and a second transition at about 20%db possibly associated with the glass transition. Figure 10 gives the stress-strain (σ-) curves for SENT specimens of cellulose films with different water contents at room temperature. The total specific energy to fracture, WT, represents the energy required to deform and break the material at a given notch length and can be calculated as the area under the σ- curves

WT )

ε)ε σ · dε ∫ε)0 b

(5)

Figure 10. Stress-strain curves for SENT specimens of cellulose films with different water contents.

Figure 11. Strain at maximum stress for SENT specimens of cellulose films with different water contents.

where σ is the stress, ε is the strain during stretching, and εb is the strain at break. As the water content is increased from 5.31%db to 8.31%db, the energy for fracture of the regenerated cellulose films also increases (data not shown). This is a weak effect with much scatter on the data. Above approximately 10%db a limiting value for WT is reached. For ductile fracture, the total energy includes the energy required for stable crack propagation and the formation of a plastic zone.46,47 According to Broberg, the fracture of a material can be divided into three distinct phases: loading without crack growth, stable crack growth, and unstable crack growth.48 The strain at maximum stress approximates the transition between the stable crack growth and unstable crack growth. There is an increase in the strain at maximum stress as the water content increases, suggesting that an increase of water content pushes the occurrence of unstable crack growth to higher strains (Figure 11). This effect is much stronger than that in Figure 7. Comparison of the Properties of Starch and Cellulose. An important difference between starch and cellulose is that the crystalline unit cell of the former contains water, the amount depending on the whether there is an A or B type organization,49 whereas the cellulose unit cell is anhydrous. A consequence of this is that the crystallinity of starch is strongly water content dependent and completely dry starch has a crystallinity of around 10%.50 Presumably, the water has a structural role in starch crystals. For the semicrystalline cellulose films used here, there is a water content dependence, as shown in Figure 4, but this is weaker than observed for starch and, as discussed previously, may be the result of further crystallization as a result of


mobilization of the rigid amorphous regions on the crystallite surface. The change in crystallite shape would be consistent with different accessibilities to the crystallite faces. It is possible to argue that the lower GAB monolayer value reported above for cellulose compared with starch reflects the fact that for cellulose all the water will be distributed in the amorphous and thus there will be less polymer chain surface available for the nonmobile water than will be the case for starch where there is water in the crystalline phase. This argument should be treated with caution partly because of the uncertainties in parameters obtained from a three component fit to the isotherm data. In addition to the water dependence of crystallinity, the major difference between cellulose and starch is in the large deformation mechanical properties revealed by both of the test geometries used. Some reported values for the strain/elongation at break for un-notched starch films prepared in the absence of plasticisers are shown in the table in the Supporting Information. As can be seen, these are much lower than for cellulose (see also Figure 7). A consequence of this brittle behavior of starch films is that an additional plasticizer, generally glycerol, is included when starch is processed to produce films or other thermoplastic structures.26,51,52 The reasons for the difference in brittleness between starch and cellulose are therefore of considerable technological significance. The large change in facture properties of starch at 10% water has been discussed in terms of the onset of motion of portions of the molecule (secondary relaxation)27,53 but another interpretation would be the presence of the mobile water fraction as discussed above. The movement of this water would provide an energy dissipation mechanism which would allow ductile mechanical behavior. There is also the possibility of higher local concentrations of water in cellulose at the crystallite interface due to the expulsion of water from the anhydrous crystals to the amorphous region. However, cellulose shows ductile behavior at water contents where there is no mobile water. Chen et al.43 modeled the amorphous state of dry cellulose and predicted that yield should occur at approximately 7-8% strain, quite close to the extrapolated yield strain here for dry films. They proposed that hydrogen bond breakage followed by chain slippage would still be the mechanism for yield at low water content as a result of hydrogen bond exchange. Superficially it might be expected that starch, being a hydrogen bonded solid, would behave in a similar fashion, therefore, comparable modeling studies on dry amylose and amylopectin would have to be carried out to ascertain whether the predicted behavior was the same. Cellulose also has a ribbonlike and hence more linear three-dimensional structure than starch and offers the possibility of more extensive and indeed cooperative behavior of the hydrogen bonds. Therefore, although slippage may still occur, there may always be hydrogen bonds available somewhere along the chain to prevent rupture. The predominantly intramolecularly bonded helical structures in starch, with water mainly located inside the helix, may be more likely to rupture than the linear cellulose. Although this is admittedly a speculative hypothesis, it could lead the way to further understanding of the structural factors determining the mechanical properties of polysaccharide films, an understanding that in our view is badly needed.

Conclusions In an attempt to bring together all the data so far presented, it is proposed that at around a water content of 10%db a bulk phase of water appears. This is supported by the NMR and the


131

Figure 12. Demonstration on how the presence of a certain number of water molecules can weaken the mechanical properties of a hydrogen bonded polymer such as cellulose by breaking and replacing hydrogen bonds.

sorption isotherm measurements. This amount of water is thought to be necessary for a proton exchange mechanism to function (Figure 12). This exchange mechanism is proposed to shift protons from the cellulose to the water by the mutual flipping of proton bonds. This behavior is not too different from results reported for starch. For example, van Nieuwenhuizen et al.54 have shown a comparable increase in the T2 relaxation time at about 10% water for bread where the starch component is about 70% of the total material, and Yakimets et al.31 have reported values for the GAB parameters for cassava starch of 9.0 g/100 g (Mo), 4.1 (c), and 0.69 (K) compared with our reported values for cellulose of 6.3 g/100 g, 4.4, and 0.81, respectively. The presence of this mobile water has the effect of reducing the density of the hydrogen bonded network, particularly in the interfacial region between the crystallites and the amorphous matrix, resulting in reductions in the elastic moduli but more significantly in differences in the large deformation properties due to slippage of the polymer chains. In contrast to starch, cellulose shows a weaker dependence of crystallinity on water content. Whereas the water content dependence of the elastic modulus is similar for starch and cellulose, there is a very large difference in fracture behavior, with cellulose showing much larger elongations at break and a lower water content dependence. It is speculated that this could be due to differences in hydrogen bonding patterns allowing exchange and slippage for cellulose, even at low water content, perhaps due to the extensive network of hydrogen bonds along the linear structure. This does not occur for starch where ductile behavior requires the presence of mobile water. Understanding the fundamental mechanical properties of cellulose and starch is of significance for the development of biodegradable plastics and packaging to replace synthetic polymers. Acknowledgment. We thank Stephen Moore from the Innovia Film Company Ltd. for providing the cellulose films. Supporting Information Available. Some reported values for elongation at break for un-notched starch films and values recorded for cellulose in this work. This material is available free of charge via the Internet at http://pubs.acs.org.

References and Notes (1) Perez, S.; Mazeau, K. Conformations, structures, and morphologies of cellulose. In Polysaccharides: Structural diVersity and functional Versatility; Dumitriu, S., Ed.; Marcel Dekker: New York, 1998; pp 41-65. (2) Kocherbitov, V.; Ulvenlund, S.; Kober, M.; Jarring, K.; Arnebrant, T. J. Phys. Chem. B 2008, 112, 3728–3734. (3) Nishiyama, Y.; Sugiyama, J.; Chanzy, H.; Langan, P. J. Am. Chem. Soc. 2003, 125, 14300–14306. (4) Langan, P.; Nishiyama, Y.; Chanzy, H. J. Am. Chem. Soc. 1999, 121, 9940–9946. (5) Ishikawa, A.; Okano, T.; Sugiyama, J. Polymer 1997, 38, 463–468.

132


(6) Hatakeyama, T.; Ikeda, Y.; Hatakeyama, H. Macromol. Chem. Phys. 1987, 188, 1875–1884. (7) Ago, M.; Endo, T.; Hirotsu, T. Cellulose 2004, 11, 163–167. (8) Batzer, H.; Kreibich, U. T. Polym. Bull. 1981, 5, 585–590. (9) Zhou, S.; Tashiro, K.; Hongo, T.; Shirataki, H.; Yamane, C.; Ii, T. Macromolecules 2001, 34, 1274–1280. (10) Hatakeyama, H.; Hatakeyama, T. Macromol. Chem. Phys. 1981, 182, 1655–1668. (11) Hermans, P. H.; Weidinger, A. J. Am. Chem. Soc. 1946, 68, 1138– 1138. (12) Kimura, M.; Hatakeyama, T.; Nakano, J. J. Appl. Polym. Sci. 1974, 18, 3069–3076. (13) Tokonami, S.; Iimoto, T.; Ichiji, T.; Tanaka, N.; Kurosawa, R. Radiat. Meas. 1996, 26, 689–699. (14) Suto, S.; Suzuki, K. Polymer 1997, 38, 391–396. (15) Garnier, G.; Wright, J.; Godbout, L.; Yu, L. Colloids Surf., A 1998, 145, 153–165. (16) Hilyard, N. C. J. Appl. Polym. Sci. 1967, 11, 1315–1323. (17) Takahara, H.; Nomura, S.; Kawai, H.; Yamaguch, Y.; Okazaki, K.; Fukushim1, A. J. Polym. Sci., Part B: Polym. Phys. 1968, 6, 197– 221. (18) Borch, J.; Marchessault, R. H. J. Polym. Sci., Polym. Symp. 1969, 28, 153–167. (19) Holmberg, M.; Berg, J.; Stemme, S.; Berg, L.; Rasmusson, J.; Claesson, P. J. Colloid Interface Sci. 1997, 186, 369–381. (20) Hilyard, N. C.; Crum, J. L. J. Appl. Polym. Sci. 1970, 14, 1373–1375. (21) Gindl, W.; Martinschitz, K. J.; Boesecke, P.; Keckes, J. Biomacromolecules 2006, 7, 3146–3150. (22) Gindl, W.; Keckes, J. Polymer 2005, 46, 10221–10225. (23) Hermans, P. H. Physics and chemistry of cellulose fibres (with particular reference to rayon); Elsevier Publishing Co.: New York, 1949. (24) Hinterstoisser, B.; Akerholm, M.; Salmen, L. Biomacromolecules 2003, 4, 1232–1237. (25) Kalichevsky, M. T.; Jaroszkiewicz, E. M.; Ablett, S.; Blanshard, J. M. V.; Lillford, P. J. Carbohydr. Polym. 1992, 18, 77–88. (26) Rindlav-Westling, A.; Stading, M.; Hermansson, A. M.; Gatenholm, P. Carbohydr. Polym. 1998, 36, 217–224. (27) Roudaut, G.; Poirier, F.; Simatos, D.; Le Meste, M. Rheol. Acta 2004, 44, 104–111. (28) Pamies, B. V.; Roudaut, G.; Dacremont, C.; Le Meste, M.; Mitchell, J. R. J. Sci. Food Agric. 2000, 80, 1679–1685. (29) Roudaut, G.; Dacremont, C.; Le Meste, M. J. Texture Stud. 1998, 29, 199–213. (30) Van de Berg, C.; Bruin, S., Water actiVity: influence on food quality; Academic Press: New York, 1981.

Sun et al. (31) Yakimets, I.; Paes, S. S.; Wellner, N.; Smith, A. C.; Wilson, R. H.; Mitchell, J. R. Biomacromolecules 2007, 8, 1710–1722. (32) D’Arcy, R. L.; Watt, I. C. T. Faraday Soc. 1970, 66, 1236–1245. (33) Segal, L.; Creely, J. J.; Martin, A. E.; Conrad, C. M. Text. Res. J. 1959, 29, 786–794. (34) Ibbett, R. N.; Domvoglou, D.; Phillips, D. A. S. Cellulose 2008, 15, 241–254. (35) Agrawal, A. M.; Manek, R. V.; Kolling, W. M.; Neau, S. H. J. Pharm. Sci. 2004, 93, 1766–1779. (36) Hills, B. P.; Manning, C. E.; Ridge, Y. J. Chem. Soc., Faraday Trans. 1996, 92, 979–983. (37) Ibbetta, R. N.; Schuster, K. C.; Fasching, M. Polymer 2008, 49, 5013– 5022. (38) McConville, P.; Pope, J. M. Polymer 2001, 42, 3559–3568. (39) Moharram, M. A.; Mahmoud, O. M. J. Appl. Polym. Sci. 2007, 105, 2978–2983. (40) Kondo, T. Hydrogen bonds in cellulose and cellulose derivatives. In Polysaccharides. Strutural diVersity and functional Versatility; Dumitriu, S., Ed.; Marcel Dekker: New York, 1998; pp 131-172. (41) Okajima, S.; Kai, A. Sen-I Gakkaishi 1972, 28, 387–391. (42) Ray, P. K.; Bandyopadhyay, S. B. J. Appl. Polym. Sci. 1975, 19, 729– 733. (43) Chen, W.; Lickfield, G. C.; Yang, C. Q. Polymer 2004, 45, 1063– 1071. (44) Kompella, M. K.; Lambros, J. Polym. Test. 2002, 21, 523–530. (45) Nissan, A. H. Macromolecules 1976, 9, 840–850. (46) Ni, B. Y. H.; Zhang, T. Y.; Li, J. C. M. J. Mater. Res. 1991, 6, 1369– 1373. (47) Wu, J. S.; Mai, Y. W.; Cotterell, B. J. Mater. Sci. 1993, 28, 3373– 3384. (48) Broberg, K. B. Cracks and Fracture; Academic Press: San Diego, 1999. (49) Buleon, A.; Colonna, P.; Planchot, V.; Ball, S. Int. J. Biol. Macromol. 1998, 23, 85–112. (50) Hartley, L.; Chevance, F.; Hill, S. E.; Mitchell, J. R.; Blanshard, J. M. V. Carbohydr. Polym. 1995, 28, 83–89. (51) Follain, N.; Joly, C.; Dole, P.; Bliard, C. J. Appl. Polym. Sci. 2005, 97, 1783–1794. (52) vanSoest, J. J. G.; Benes, K.; deWit, D.; Vliegenthart, J. F. G. Polymer 1996, 37, 3543–3552. (53) Roudaut, G.; Dacremont, C.; Pamies, B. V.; Colas, B.; Le Meste, M. Trends Food Sci. Technol. 2002, 13, 217–227. (54) Van Nieuwenhuijzen, N. H.; Primo-Martin, C.; Meinders, M. B. J.; Tromp, R. H.; Hamer, R. J.; van Vliet, T. J. Agric. Food Chem. 2008, 56, 6432–6438.

BM900981T

Comparison of the Mechanical Properties of Cellulose and Starch

Recommend Documents