How the biodegradability of wheat gluten-based agromaterial can be modulated by adding nanoclays

Polymer Degradation and Stability 96 (2011) 2088e2097

Contents lists available at SciVerse ScienceDirect

Polymer Degradation and Stability journal homepage: www.elsevier.com/locate/polydegstab

How the biodegradability of wheat gluten-based agromaterial can be modulated by adding nanoclays Anne Chevillard a, *, Hélène Angellier-Coussy a, Bernard Cuq b, Valérie Guillard a, Guy César c, Nathalie Gontard a, Emmanuelle Gastaldi a a b c

UMR IATE, Université Montpellier II, CC023, pl. E Bataillon, 34095 Montpellier Cedex, France UMR IATE, Montpellier SupAgro, Bat 37, 2 place Viala, 34060 Montpellier, France SERPBIO, Laboratoire LIMATB-L2PIC, Université de Bretagne Sud, rue Saint Maudé, 56321 Lorient Cedex, France

a r t i c l e i n f o

a b s t r a c t

Article history: Received 9 August 2011 Received in revised form 23 September 2011 Accepted 28 September 2011 Available online 6 October 2011

The objective of this work was to investigate the influence of clay nanoparticles on the biodegradability of wheat gluten-based materials through a better understanding of multi-scale relationships between biodegradability, water transfer properties and structure of wheat gluten/clay materials. Wheat gluten/ clay (nano)composites materials were prepared via bi-vis extrusion by using an unmodified sodium montmorillonite (MMT) and an organically modified MMT. Respirometric experiments showed that the rate of biodegradation of wheat gluten-based materials could be slowed down by adding unmodified MMT (HPS) without affecting the final biodegradation level whereas the presence of an organically modified MMT (C30B) did not significantly influence the biodegradation pattern. Based on the evaluation of the water sensitivity and a multi-scale characterization of material structure, three hypotheses have been proposed to account for the underlying mechanisms. The molecular/macromolecular affinity between the clay layers and the wheat gluten matrix, i.e. the ability of both components to establish interactions appeared as the key parameter governing the nanostructure, the water sensitivity and, as a result, the overall biodegradation process. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Biodegradation Wheat gluten Montmorillonite Nanocomposite Water sensitivity

1. Introduction The huge development of conventional plastics made from petroleum-based synthetic polymers unable to degrade in landfill or compost-like environment had led to serious environmental issues. In response to this increasing awareness pushed by governments and societies, the use of polymers stemming from renewable and sustainable resources to develop bioplastics constitutes an innovative and promising alternative. Among biosourced polymers, proteins such as wheat gluten are natural heteropolymers constituted by different amino acids which offer a large spectrum of chemical functionalities and thus, various polymer network structures [1]. Wheat gluten is a by-product of the wheat starch industry available at a reasonable price (around

* Corresponding author. Tel.: þ33 467 144 235; fax: þ33 467 144 990. E-mail addresses: [email protected], [email protected] (A. Chevillard), [email protected] (H. Angellier-Coussy), cuq@ supagro.inra.fr (B. Cuq), [email protected] (V. Guillard), cesar.guy@ neuf.fr (G. César), [email protected] (N. Gontard), [email protected] (E. Gastaldi). 0141-3910/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymdegradstab.2011.09.024

1.3 V/kg) and displaying functional properties interesting for packaging or agricultural applications [2]. Wheat gluten is mainly constituted of two main storage proteins that are gliadins (monomeric proteins with molecular weight ranging from 15 to 85 kDa) and glutenins (macropolymer with molecular weight ranging from 150 to more than 103 kDa). Gluten proteins can undergo disulphide interchange upon heating, which leads to the formation of a threedimensional macromolecular network [3]. Owing to good thermoplastic properties, wheat gluten can be processed by extrusion at temperature as low as 60
C in the presence of hydrophilic plasticizers [4]. Domenek et al. [5] have demonstrated the high biodegradability and non ecotoxicity of wheat gluten-based materials. Even if covalent cross-linking induced by thermal treatments allowed to significantly improve water resistance and mechanical properties of wheat gluten-based materials [6,7], it was shown that the biodegradability was not affected when evaluated in a liquid medium (modified Sturm test) [5]. Nevertheless, under composting conditions, Zhang et al. [8] have recently reported that the biodegradability of wheat gluten-based materials can be affected by chemical modification.

A. Chevillard et al. / Polymer Degradation and Stability 96 (2011) 2088e2097

Besides the application of thermal and chemical treatments, the creation of a nanocomposite structure through the introduction of layered silicates constitutes another promising route to modulate properties of wheat gluten materials [7,9e12]. Wheat gluten-based nanocomposites are commonly prepared using either a solvent (casting) or a thermomolding process but not yet by extrusion. It has been shown that the introduction of unmodified montmorillonite led to a significant decrease in water sensitivity, water vapor permeability [7,9,10] and liquid water diffusivity [11], together with an increase in rigidity and resistance [7,9] of wheat gluten-based materials. Although many studies have focused on the effect of layered silicates on some of functional properties (especially mechanical, thermal and transfer properties) of wheat gluten-based materials, new insights are still required to understand how the biodegradability can be modulated through the introduction of nanoclays. Some literature is already available concerning the biodegradability of some bio-sourced polymers and their related nanocomposites (gelatin [13], casein [14], methyl cellulose [15], polylactic acid [16e18] or polyhydroxybutyrate valerate [19]). However, the effects resulting from nanoclays addition on the rate, the level and the underlying mechanisms of biodegradability are not yet clearly elucidated and often contradictory. This state of the art can be explained by (i) a wide range of raw materials (clays and polymer matrices) differing in chemical and physical features, (ii) a large number of structures achieved at the nanometric scale, (iii) a large variety of methods used to evaluate biodegradability, and (iv) the fact that biodegradation data are not always normalized by the total carbon content, making difficult the comparison between studies. Moreover, most of the assumptions proposed to account for the biodegradability behavior are often not sufficiently supported by other complementary experimental approaches. The objective of this work was to investigate the influence of nanoclays on the biodegradability of wheat gluten-based materials by focusing on a better understanding of multi-scale relationships between biodegradability, water transfer properties and structure of resulting materials. For this purpose, wheat gluten/clay (nano) composites were prepared via bi-vis extrusion, a process largely used at the industrial scale, by using two types of montmorillonites (MMT): an unmodified sodium MMT and organically modified MMT. Biodegradation patterns were evaluated through respirometric experiments and water sensitivity was assessed by soaking experiments. The multi-scale structural characterization involved complementary technical approaches enabling to evaluate the cross-linking degree of the matrix, the glass transition temperature of materials, the nano-scale structure i.e. the level of dispersion of nanoclays within the matrix, and to finish with, the macroscale structure.

2089

in C30B) on nanoclays are given in Chevillard et al. [20]. Chemicals, unless specified separately, were purchased from Sigma Aldrich in per analytical quality. 2.2. Preparation of wheat gluten-based materials Extrusion was performed using a co-rotating twin screw extruder (Coperion, ZSK25, Stuttgart, Germany) connected to a computer interface and controller unit (Brabender, O.H.G, Duisburg, Germany). The barrel consisted of twelve zones, each zone being equipped with an independent temperature control and a die head constituted of two 5-mm diameter holes. The total length of the screw was 42D. The first and second heating zones were constantly set at 40
C and the other heating zones at 60
C. The screw speed was set at 150 rpm. Wheat gluten and nanoclays powders were fed using two distinct weight feeders (Brabender, Duisburg, Germany) leading to a cumulate powder feed rate of 4.5 kg h1. The ratio nanoclays/wheat gluten was adjusted to have a final inorganic filler content corresponding to 5 wt%. Water was fed with a weight pump (Movacolor, WL Sneek, Netherlands) at a flow rate of 2 kg h1. Immediately after extrusion, extrudates were cooled and air-dried in ambient conditions during approximately 30 min before being cut using a Scheer SGS 50E pelletizer (Scheer Reduction Engineering GmbH, Stuttgart, Germany). Granulates were allowed to dry in ambient room conditions until constant weight. Water content of final granulates was 9 wt% (measured after drying 24 h at 105
C). They were characterized by a height of 2.3  0.2 mm, a diameter of 5.1  0.4 mm, and a weight of 53  2 mg. Samples were packed in polyethylene hermetic bags and stored in dark room at 4
C until experiments. 2.3. Biodegradation tests Respirometric tests were conducted in aerobic conditions to evaluate the biodegradability of wheat gluten-based materials. Method was adapted from the US standard ASTM D5988-96, which is a Standard Test Method for Determining Aerobic Biodegradation in Soil of Plastic Materials. The released CO2 being proportional to the percentage of biodegraded substrate, CO2 evolution measures ultimate degradation (i.e. mineralization) in which a substance is broken down to its final products. Beforehand, materials were ground with a domestic blender to obtain particles around 1e2 mm and carbon contents were measured with an elementary analyser (ThermoQuest NA 2500). Biodegradation tests were carried out in cylindrical hermetic glass vessels (1000 mL capacity) containing three small open polypropylene flasks (60 mL capacity) (Fig. 1). The

2. Materials and methods 2.1. Materials Commercial vital wheat gluten was kindly supplied by Syral (Belgium) under the reference AMYGLUTEN 110. Its moisture and protein content was approximately 10% and 80%, respectively. Two types of nanoclays were used as received in this study: an unmodified sodium montmorillonite provided by Laviosa (Italy) under the reference HPS and an organically modified montmorillonite carrying a methyl, tallow, bis-2-hydroxyethyl quaternary alkylammonium salt, supplied by Southern Clay under the reference CloisiteÒ30B (C30B). CEC (cation exchange capacity) values are 129 meq 100 g1 for HPS and around 93 meq 100 g1 for C30B. Further information (interlayer distance, organic content, interlayer cation organic cation saturation 3D models organic cations present

Fig. 1. Schematic representation of the glass vessel used for respirometric test for biodegradation evaluation.

2090

A. Chevillard et al. / Polymer Degradation and Stability 96 (2011) 2088e2097

first flask contained 25 g of dry soil (previously passed through a 2 mm sieve) mixed with ground material samples whose weight corresponded to 50 mg of carbon. The water content of soil samples was adjusted to reach 80% of the soil water retention capacity. Soil characteristics were as follows: pH 6.8 (H2O), 2.3 wt% of organic matter, 16.85 wt% of clay, 26.85 wt% of lime and 56.3 wt% of sand. The second flask contained 30 mL NaOH solution (0.1 M) to trap the CO2 produced by microorganisms. The third flask contained distilled water, in order to maintain the relative humidity at 100% inside the vessel. The glass vessels were hermetically closed and incubated in the dark at 28  1
C. At selected time, glass vessels were open to ensure the back titration of the excess of NaOH which has not reacted with CO2. Before titrating the residual NaOH with HCl solution (0.1 M) in the presence of thymophthaleine 0.10% in ethanol 95
, 5 mL of barium chloride solution (20% in water) was added in each flask to precipitate carbonate ions. The flasks containing soil were weighted and appropriate amount of water was added (if necessary) in order to keep constant the water content (initially fixed at 80% of the soil water retention capacity). During this procedure, the glass vessels were left open during 2 min in order to air the vessel. At each time interval, a new polypropylene flask containing 30 mL NaOH 0.1 M was placed in each glass vessel before being closed and put in the dark at 28
C until the next measurement. Biodegradation experiments included a control and a blank. For the control, cellulose was chosen as reference because of its well-known degradation characteristics. For the blank, the test was conducted without addition of a carbon source to measure both the CO2 produced by the soil carbon substrate and the CO2 present in the air of the glass vessel. All biodegradation tests were measured in triplicate. Results were calculated by subtracting the CO2 production of the blank. The theoretical maximum CO2 potential (CO2 (max) (mg)) produced by total oxidation of the material is calculated from CO2 max ¼ C  44.01/12.01, where C is the amount of carbon of the sample introduced in the soil for the test (mg). The percentage of degradation Deg is calculated by using equation (1):

Deg ¼

CO2 material  CO2 blank CO2 max

(1)

2.4. Liquid water sensitivity 2.4.1. Maximum water uptake, swelling and dry matter losses at equilibrium Liquid water uptake was determined gravimetrically for all materials in triplicate at 20
C. Four days before experiment, materials were stored in a desiccator upon silicagel (relative humidity close to 0%). After drying, granulate samples (3 granulates per sample) were weighted using a four-digit balance (m0) and their dimensions (thickness and diameter) were measured with a digital caliper to calculate the volume of the samples (V0). Then, samples were immersed in 50 mL distilled water (stirred solution) and removed at specific time t in order to be weighted again (mt) after having carefully removed the excess of water using tissue paper. The water uptake at time t (WUt) was calculated using equation (2).

WUt ¼

mt  m0 m0

(2)

When the equilibrium was reached, i.e. when materials absorbed no more water, samples were weighed, granulate dimensions (thickness and diameter) were measured with a digital caliper to calculate the volume of the samples at equilibrium (Veq). Then they were stored again 4 days in a desiccator upon silicagel before being weighed (meq dried). The maximum water uptake (WUeq) was

calculated at this specific time using equation (2) while the swelling of the materials (SWeq) as well as the dry mass losses (DMLeq) were calculated using equations (3) and (4), respectively.

SWeq ¼

Veq  V0 $100 V0

DMLeq ¼

(3)

m0  meq dried m0

(4)

2.4.2. Liquid water diffusivity Identification of liquid water diffusivity was done from water uptake kinetics data (equation (2)). The solution of Fick’s second law for diffusion from a finite cylinder of diameter 2r and height 2l immersed in a stirred solution of infinite volume was obtained by the superposition of the analytical solution for an infinite cylinder of diameter 2r (equation (5)) and that for an infinite slab of thickness 2l (equation (7)).

jr ¼ 1 

  4 2 a $exp  D $ $t app n r 2 $a2n n¼1 N X

(5)

where the an are the positive roots of:

J0 ðr an Þ ¼ 0

(6)

where J0 (x) is the Bessel function of the first kind of order zero. Roots of equation (6) are tabulated in tables of Bessel functions [21].

jz ¼ 1 

N 8 X

1

p2 n ¼ 0 ð2n þ 1Þ2

$exp 

ð2n þ 1Þ2 $p2 ð2lÞ2

! Dapp $t

(7)

In equations (5) and (7), jr and jz , are the quantities of water which have entered in the theoretical infinite cylinder and infinite plane sheet at time t to the corresponding quantity after infinite time and D is the effective liquid water diffusivity (supposed constant) in a stirred solution (considering no external resistance to transport). The hypothesis of a negligible external mass transfer coefficient, which is the case for a well-stirred solution or a Biot number greater than 100 [22] was first tentatively validated. To do this, the external mass transfer coefficient (k) was evaluated using the methodology extensively detailed in Mascheroni et al. [23]. and the Biot number was calculated. The external mass transfer coefficient, k, was found varying between 1.33  105 and 1.68  105 m s1 for all the liquid water sorption kinetics and the corresponding Biot numbers ranged from 513 to 536. It is obvious from these results that the external mass transfer resistance at the interface material/water could be neglected (Bi > 100) confirming the effectiveness of the stirring. In this approach, the swelling of the materials was neglected. Therefore, the diffusivities identified were considered as apparent diffusivity values called (Dapp). As demonstrated in Carslaw and Jaeger [24], solution of Fick’s second law for finite cylinder can be written down as products of the solutions obtained for simple geometries, i.e. infinite cylinder of 0 < r < a and infinite slab of l < x < l. This product solution is given by equation (8). This equation has been successfully used by different authors for modeling mass transfer in finite cylinder geometry [25,26]. The quantity of water entering in the finite cylinder at time t jt , is then calculated as follows:

jðtÞ ¼

WUt ¼ ðjr Þ$ðjz Þ WUN

(8)

2.4.3. Identification procedure Apparent diffusivity parameters were identified from the experimental curve by minimizing the root mean square deviations

A. Chevillard et al. / Polymer Degradation and Stability 96 (2011) 2088e2097

between simulated and experimental results using the Levqffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ^  yÞ2 =N  P enbergeMarquardt procedure [27] RMSE ¼ ðy ^ where y and y represent experimental and predicted values respectively, N is the number of experimental measurements and p is the number of estimated model parameters. If RMSE tends toward 0, that means that the calculated concentrations are very close to the experimental ones and the model is thus able to represent experimental data. All analytical solutions were programmed using MATLABÒ software (The Mathworks Inc., Natick, MA, USA) and the LevenbergeMarquardt procedure was used via a dedicated routine “lsqnonlin” developed in MATLABÒ. 2.5. Gluten network cross-linking degree The degree of network cross-linking of wheat gluten-based materials was evaluated through the determination of the sodium dodecyl sulfate (SDS)-insoluble protein fraction (Fi) according to Domenek et al. [28]. Ground samples (160 mg) were stirred for 80 min at 60
C into 20 mL 0.1 M sodium phosphate buffer (pH ¼ 6.9) containing 1% SDS and subsequently centrifuged at 18,000 rpm for 30 min. The supernatant contained the SDS-soluble protein fraction (Fs). The SDS-insoluble protein fraction (Fi) was extracted with 5 mL of SDS-phosphate buffer containing 20 mmol L1 dithioerythriol (DTE) under stirring for 60 min at 60
C, then tip sonicated for 3 min and finally centrifugated 30 min at 18,000 rpm. 500 mL of the resulting supernatant was mixed with 500 mL of SDS-phosphate buffer containing 40 mmol L1 iodoacetamide (IAM), (a sulfhydryl-reactive alkylating agent) used to block reduced cysteine residues. Both Fs and Fi extracts were submitted to size-exclusion HPLC [28]. Fi was expressed in percent as the ratio of the SDSinsoluble protein fraction on the total protein fraction (Fi þ Fs). 2.6. Differential scanning calorimetry (DSC) Differential scanning calorimetry was used to measure the glass transition temperature (Tg) of wheat gluten-based materials. Ground samples (around 12 mg) were placed in open aluminum pans (Tzero pan, TA Instruments New Castle, USA) and stored at aw ¼ 0.753 over a saturated salt solution of NaCl. After equilibration, pans were immediately and hermetically sealed. Measurements were done with a thermo-modulated calorimeter (Q200 modulated DSC, TA Instruments, New Castle, USA). Each sample was heated from 40
C to 130
C at a heating rate of 3
C min1. The period and the amplitude of modulation were respectively 100 s and 0.796
C. The glasserubber transition was characterized by three different temperatures in the DSC traces (heat flow curves), i.e. Tg onset, corresponding to the onset of the specific heat increment ascribed to glass transition, in other words, the temperature at which some polymer chains start to undergo the transition; Tgi the temperature at the inflexion point, corresponding to the temperature at which the differential heat flow is maximum; and Tg offset corresponding to the offset of the glasserubber transition. Tg values were measured in triplicate. 2.7. Wide angle X-ray scattering analysis Ground material samples and pristine nanoclays were characterized by wide angle X-ray scattering (WAXS) at room temperature and relative humidity. Experiments were performed using a PHILIPS X’Pert MPD diffractometer (diffractometer qe2q) with a X’celerator detector and a nickel filter operating at 40 kV and 20 mA with a Cu-Ka radiation (l ¼ 1.5418 Å). The spectra were recorded between 2
and 25
by using a scan speed of 0.035368
s1 and a step size of 0.0334226
.

2091

2.8. Microscopic observations 2.8.1. Scanning electron microscopy (SEM) The morphology of the wheat gluten-based extrudate sections was observed after platinum coating (5 nm thickness) using a field emission scanning electron microscope (FESEM S-4500, Hitachi, Japan) at magnification 30 and an acceleration voltage of 2000V. SEM observations were coupled with image analysis (Image J software) to measure the diameter of the holes observed on extrudate sections. For each formulation, three samples were analyzed, corresponding to a total number of holes ranging between 260 and 700 depending on the material. The average number of holes (nb holes), the median diameter (d50) and the cumulated hole area (Sholes) were calculated for each granulate. The cumulated hole area was expressed as a percentage of the total observed area. 2.8.2. Transmission electron microscopy (TEM) Materials (about 1 mm3) were initially fixed in glutaraldehyde 2.5% (v/v), dehydrated in an ethanol gradient, then impregnated in propylene oxide and finally embedded in epoxy resin epon-812 substitute, (Electron Microscopy Science, England). After 3 days of incubation at 60
C, ultra-thin sections of 70 nm were cut with an ultramicrotome diamond and mounted on 100 mesh grids covered by a colodion film. Samples were examined with a Jeol JEM-1200EX II TEM (Jeol ltd., Tokyo, Japan) using magnifications from 10 to 100 K. 3. Results and discussion Layered silicates (montmorillonites) were introduced within the wheat gluten matrix with the aim to modulate the biodegradability of materials. The approach consisted in selecting two different montmorillonites: (i) a hydrophilic MMT (HPS) and (ii) an organically modified MMT containing a quaternary alkylammonium with two hydroxy-ethanol groups as interlayer cation (C30B). Indeed, since unmodified MMT is hydrophilic and negatively charged, it was supposed to be naturally compatible with the hydrophilic and positively charged wheat gluten matrix and thus, appeared suitable for nanocomposite preparation. However, the very low charged groups frequency of gluten proteins combined with a rather high frequency in non-polar side chain [29] dropped hints that the compatibility between protein and clay might be improved by using an organically modified MMT (OMMT). The interest of using an organically modified MMT such as C30B is the presence on its surfactant of apolar (tallow chain and methyl group) and polar (hydroxyl groups) groups, which confer both hydrophobicity and hydrophilicity. Moreover, since the degree of penetration of the polymer into the clay gallery was expected to be enhanced by a greater interlayer distance, C30B (which displays a d001 of 18.3 Å instead of 12.7 Å for HPS) could favor the formation of an intercalated/exfoliated structure. 3.1. Biodegradation pattern of wheat gluten-based materials under indoor soil conditions Respirometric tests were used to evaluate the biodegradability of the wheat gluten-based materials. CO2 evolution provides an indicator of the ultimate biodegradability ascribed to mineralization of the test samples. The experimental degradation data were modeled with the Hill equation [5,30]:

tn Deg ¼ Degmax $ n ðk þ t n Þ

(9)

where Deg [%] is the percentage of degradation at time t [days], Degmax [%] the percentage of degradation at infinite time, k [days]

2092

A. Chevillard et al. / Polymer Degradation and Stability 96 (2011) 2088e2097

into the matrix and enzymes to cleave the protein chains [32]. Once the size of polymer fragments is small enough, they are transported into the cell where they are ultimately mineralized. The products of the mineralization process are gasses (CO2, CH4, N2, H2), water, minerals, and biomass [31]. As compared to pure cellulose (i.e. the control), wheat glutenbased materials biodegraded very rapidly since the maximum degradation rate was observed after only 4 days (Fig. 2b) as indicated by Timerate max values, and it took less than 6 days to reach 50% of Dmax as indicated by k values (Table 1). The high biodegradability of proteins, and especially wheat gluten, has already been reported [5,30]. It worth noting that experiments have been conducted using ground materials (1e2 mm) in accordance with normative tests. Indeed, even if our objective was only to investigate the influence of the material formulation, we could remind that sample size and shape, and notably its specific surface area, are very important factors in determining biodegradation pattern. The biodegradability of the materials was not significantly influenced by the presence of organically modified MMT (C30B) (Fig. 2, Table 1). As already reported in literature, the effect of organically modified montmorillonites on biodegradation is often attributed to a catalytic effect involving the hydroxyl groups located at the edges of the clay layers [18,33e36]. It is worth noting that investigated polymers are generally water resistant polymers notably poly(lactic acid) (PLA) [16e18,33e35,37,38] making difficult the comparison with protein based materials. In the case of these hydrophobic polymers, the use of OMMT usually favors the layers dispersion/exfoliation within the matrix and results in a higher exchange surface for catalytic effect. It has been notably reported that C30B enabled enhancement of PLA biodegradability under compost conditions [16,37]. In the present study, the lack of evidence of such a potential catalytic effect could be due to a low level of dispersion/exfoliation of C30B within the wheat gluten matrix due to a probable low compatibility between these two components. On the contrary, the unmodified MMT (HPS) significantly decreased the biodegradation kinetic of wheat gluten-based materials without changing the final biodegradation (Fig. 2, Table 1). Both the time required to reach the maximum degradation rate (Timerate max) as well as the time to attain 50% of Dmax (k) were doubled. In addition, the maximum degradation rate was twice lower (4.5% instead of 9.4% degradation day1). These results highlighted that the biodegradation phenomenon of wheat glutenbased materials was clearly slowed down in the presence of unmodified MMT, as already observed for methyl cellulose films [15] or gelatin materials [13]. Three hypotheses can be proposed to explain the decrease of biodegradability in the presence of MMT:

Fig. 2. Kinetic of biodegradation (a) and biodegradation rate (b) of neat wheat gluten-based materials (WG B), and wheat gluten-based materials filled with HPS (WG-HPS 6) and C30B (WG-C30B >) in the respirometric test. Symbols are experimental data points. Solid and dot lines correspond to the degradation curves calculated with the Hill equation of wheat gluten-based materials and cellulose, respectively. Error bars represent standard deviation.

the time for which Deg ¼ ½Degmax and n the curve radius of the sigmoid function. The degradation kinetic and the biodegradation rate of wheat gluten-based materials are presented in Fig. 2 and Hill parameters are reported in Table 1. The biodegradation curves of all wheat gluten-based materials displayed a sigmoidal shape, which is characteristic of biodegradation measurements (Fig. 2). Indeed, two key steps occur in the microbial polymer degradation process: first, depolymerization or chain cleavage, and second, mineralization. The first step normally occurs outside the organism thanks to extracellular enzymes [31]. As a result, a lag phase can sometimes be observed on sigmoid curves, corresponding to the time needed for water to penetrate

(i) a reduced water adsorption capacity of the materials in the presence of such fillers [13,15],

Table 1 Hill parameters (Degmax, k, n) and related biodegradation indicators (Timerate max, Degrate max) of wheat gluten-based materials (WG), and wheat gluten-based materials filled with HPS (WG-HPS) and C30B (WG-C30B) in respirometric test. Sample

Control (cellulose) WG WG-HPS WG-C30B

Hill parametersa 2

Timerate [days]

max

b

Degrate maxc [% day1]

Degmax [%]

k [days]

n

R

100

20.6 (0.6)

1.5 (0.1)

0.97

8

3.0

87 (1.6) 87 (1.9) 92 (3.0)

5.8 (0.2) 12.6 (0.5) 6.3 (0.4)

1.9 (0.2) 2.1 (0.1) 1.7 (0.2)

0.98 0.99 0.95

4 8 3

9.4 4.5 8.9

Values in parentheses represent the confident intervals. a Degmax: percentage of degradation at infinite time; k: constant of the Hill equation representing the time for which Deg ¼ ½ Degmax; n: constant of the Hill equation representing the curve radius of the sigmoid function. b Timerate max: time to reach the maximum biodegradation rate. c Degrate max: maximum degradation rate.

A. Chevillard et al. / Polymer Degradation and Stability 96 (2011) 2088e2097

(ii) the presence of a tortuous path induced by the nanodispersion of layered silicates resulting in a slower diffusion of penetrants (water, enzymes, microorganisms) [15,19,39,40], (iii) the establishment of specific interactions between layered silicates and the matrix leading to a lower availability for the matrix to be biodegraded [15,19,39]. 3.2. Water sensitivity of wheat gluten-based materials Because changes in polymer biodegradability are often explained by modification of water sensitivity, soaking experiments have been carried out to verify the first assumption. This approach was conducted to allow evaluating both equilibrium parameters (notably water adsorption capacity) and kinetic parameters (diffusivity) with the objective to evaluate the contribution of each phenomenon on the overall biodegradation pattern. Fig. 3 gives an example of liquid water sorption kinetic in wheat gluten material at 20
C. The mathematical model fit very well the data from 0 to 0.5 days then slightly overestimated the water uptake between 0.5 and 1.5 days. This discrepancy could be related to the swelling of the material (not taken into account in the modeling) that modifies the apparent water diffusion rate. Nevertheless, in spite of its simplifications, the mathematical model used here is satisfying to predict the water uptake at equilibrium. Equilibrium parameters and apparent diffusivity values (Table 2) of wheat gluten-based materials were in agreement with values previously reported in literature [6] which reflected a high sensitivity to liquid water. This has been already highlighted in previous works [6,7,9,41] and is generally explained by the hydrophilic nature of the wheat gluten proteins (high content of polar amino acid) [29]. The presence of organically modified MMT (C30B) did not impact the water sensitivity, since equilibrium parameters were not significantly different from values obtained for the neat matrix (Table 2). On the contrary, the presence of unmodified MMT (HPS) significantly increased water resistance leading to a reduction of the liquid water uptake of 25% while concomitantly, the dry mass losses and swelling values were reduced of respectively 28% and 53% (Table 2). This improvement occurred in spite of the high water retention capacity of unmodified MMT, known to absorb water up to 30 times their weight [42]. These results were consistent with Tunc et al. [9] who also reported a decrease in water uptake of wheat gluten nanocomposite films while increasing the filler content. Such a reduction in water sensitivity in presence of unmodified MMT could be explained either by the establishment of

Fig. 3. Kinetic of water uptake of wheat gluten-based materials. Symbols are experimental data points. Solid lines correspond to the model.

2093

Table 2 Liquid water sensitivity of wheat gluten-based materials: equilibrium parameters (water uptake (WUeq), dry matter losses (DMLeq), swelling (SWeq) and kinetic parameter (water diffusivity (D)). Sample

WUeq [g g1]

DMLeq [g g1]

SWeq [%]

D [1010 m2 s1]

RMSE [g g1]

WG WG-HPS WG-C30B

1.36 (0.01) 1.03 (0.01) 1.34 (0.02)

0.21 (0.02) 0.15 (0.03) 0.25 (0.06)

142 (15) 67 (8) 134 (21)

0.61 (0.09) 0.76 (0.09) 0.61 (0.09)

0.07 0.07 0.07

RMSE: roots mean square error. Values in parentheses represent the confident intervals.

hydrogen bonds between HPS and protein hydrophilic sites that would become less available for interactions with water molecules, or by a potential exfoliated structure giving rise to a tortuous pathway able to restrict water penetration. Nevertheless, diffusivity values remained unchanged whatever the presence and the type of nanoclays (HPS or C30B). In spite of a better water resistance in the presence of HPS, the layered silicates did not slow down the kinetic of water penetration through the material. Thus, the second hypothesis mentioned above should be invalidated: even if a nanocomposite structure leading to the creation of a tortuous pathway was supposed to be obtained in the presence of HPS, it would not be effective enough to decrease the rate of water penetration. These results evidenced a direct relationship between biodegradability and water sensitivity parameters at equilibrium, but not with kinetic parameters. Indeed, it was demonstrated that the introduction of nanoclays able to reduce material water adsorption were also assumed to decrease biodegradation rate. Indeed, given that water is required for enzymatic hydrolysis reaction occurring in the first stage of the biodegradation process and also for the transport of solutes and microorganisms, a significant decrease in water content would be expected to slow down the biodegradation process. At this stage, further investigation was required to deepen our understanding of multi-scale relationships between biodegradability, water transfer properties and structure of resulting materials. 3.3. Multi-scale structure characterization of wheat gluten-based materials: from the macromolecular to the macroscopic level This multi-scale approach aimed at exploring the structure of the materials from the macromolecular to the macroscopic level through the characterization of the network structure, the potential establishment of specific interactions between nanoclays and wheat gluten, the level of dispersion of nanoclays within the matrix and to finish, the presence of macroscopic defects in the materials. Such an approach would enable to settle on the three hypotheses proposed to explain changes in biodegradation pattern induced by the addition of nanoclays. The macromolecular structure of wheat gluten-based materials has been investigated through the evaluation of the degree of covalent cross-linking between protein chains, as revealed by the fraction of SDS-insoluble proteins (Fi) [28,43]. As shown in Fig. 4, the extrusion process led to an increased Fi value of the wheat gluten-based material as compared to the raw powder (from 9% to 29%). During extrusion, a considerable amount of thermomechanical energy is conferred to the material, resulting in a reversible change in protein conformation followed by aggregation reactions involving disulphide interchanges and leading to new intermolecular disulphide cross-links [28]. Among the two main protein fractions constituting wheat gluten, glutenins start to cross-link above 60e70
C, whereas for gliadins, the reactive zone

2094

A. Chevillard et al. / Polymer Degradation and Stability 96 (2011) 2088e2097

Fig. 4. Fi values (%) of wheat gluten raw powder, wheat gluten-based materials (WG) and wheat gluten-based materials filled with HPS (WG-HPS) and C30B (WG-C30B).

is clearly evidenced around 90
C. In the present study, the processing conditions (combining heating and shearing treatments) would be sufficient to reduce the activation energy of gluten crosslinking and enable new intermolecular bonds [44]. Indeed, even if the barrel temperature was set at 60
C, the temperature at the core of the product was supposed to be shifted toward higher temperatures, as already observed upon wheat gluten extrusion [4]. In close conditions of setting temperature and screw speed (60
C and 100 rpm), these authors reported a huge increase of the temperature (reaching 101
C) in the converging section of the die. It could be noted that the Fi increase (from 9 to 29%) appeared consistent with the high reactivity of gluten proteins upon heating and shearing [45,46] even if it was quite moderate as compared to the Fi values that might be reachable when applying a more drastic heating treatment. For example, Domenek et al. [5] obtained Fi values up to 80% for wheat gluten materials thermo-pressed during 35 min at 120
C and 150 bar. The introduction of unmodified MMT (HPS) led to a slightly higher Fi value (33%) as compared to the neat wheat gluten matrix (29%), whereas the introduction of organically modified MMT (C30B) did not influence this parameter (Fig. 4). This implied that the presence of unmodified MMT might favor protein cross-linking via disulphide bonds, contrary to the use of organically modified MMT. An increased value of Fi is known to reflect biochemical changes induced by temperature suggesting the occurrence of heat-activated reactions during the extrusion process. In the presence of unmodified MMT (HPS), it could be supposed that an increase in thermo-mechanical energy (due to an increased viscosity) might occur at the core of the product in the extruder converging section even if the barrel temperature was maintained at 60
C. An increase in Fi value was also reported by Angellier et al. [7] when applying a thermo-mechanical process to a wheat gluten matrix in the presence of unmodified MMT. This was ascribed to a concomitant temperature increase of the material during the extrusion process due to intense shearing effects. The increase in Fi value could also result from an anti-plasticizing effect of unmodified MMT (HPS). The hydrophilic character of HPS might favor the establishment of hydrogen bonds with the plasticizer (water), this latter becoming less available for the plasticization of the wheat gluten matrix. A decrease in plasticizer content has already been found to increase Fi values [7]. We could thus conclude that the macromolecular structure of wheat gluten-based materials, which is directly related to the rheology of the melt upon extrusion process, was affected by the nature of the reinforcing filler, and more precisely, by the interactions that might be established between the different constituents.

DSC measurements have been carried out to show how the presence of nanoclays can affect the glass transition temperature of wheat gluten-based materials. This method would enable to evidence the potential establishment of interactions between nanoclays and proteins which were expected to affect the polymer segmental motion at the interface. In the case of wheat glutenbased materials reinforced with unmodified MMT (HPS), a significant shift of Tg values toward higher temperatures (around 4.5
C higher) was noted, indicating that the presence of HPS strongly restricted the protein chain mobility (Fig. 5). This result was consistent with the increase in Fi values and can also be ascribed to the establishment of strong interactions between unmodified MMT and the wheat gluten matrix. The fact that the amplitude of the glasserubber transition was not significantly affected by the presence of HPS suggested that only one type of interaction would be involved. No change in Tg values was observed for materials filled with organically modified MMT (C30B), indicating that the mobility of protein chains was not influenced by the introduction of such MMT. It can thus be deduced that no specific interaction was formed between the two components in that case. Since Fi values and DSC measurements highlighted that the affinity between nanoclays and the gluten matrix strongly depended on the nature of the clay, it could be assumed that different levels of clay dispersion would be achieved according to the type of clay used. Based on results presented above, a low dispersion level was expected in the presence of organically modified MMT (C30B) due to its low affinity for the wheat gluten matrix. On the opposite, a good dispersion/exfoliation of clays was supposed to be obtained in the case of unmodified MMT (HPS) owing to its good affinity for wheat gluten. To go further in this investigation, characterization of the nanostructure by TEM and WAXS appeared essential to conclude about all the hypotheses proposed above. The nano-scale structure of wheat gluten-based materials was evaluated using WAXS analysis (Fig. 6) combined with TEM observations (Fig. 7). The neat wheat gluten-based matrix displayed a typical amorphous structure characterized by two very broad peaks centered around 2q ¼ 8
and 2q ¼ 20
on WAXS patterns (Fig. 6). In the presence of organically modified MMT (C30B), a microcomposite structure was achieved, as demonstrated by both WAXS (Fig. 6) and TEM analyses (Fig. 7). TEM pictures were characterized by the presence of huge agglomerates of clays (micrometer sized) and very little dispersed particles. Nevertheless, the peak around 2q ¼ 4.8
characteristic of the pristine C30B (d001 ¼ 18.3 Å) was slightly shifted toward lower angles (2q ¼ 4.60


Fig. 5. Glass transition parameters (Tg onset, Tgi and Tg offset) of wheat gluten-based materials (WG) and wheat gluten-based materials filled with HPS (WG-HPS) and C30B (WG-C30B).

A. Chevillard et al. / Polymer Degradation and Stability 96 (2011) 2088e2097

Fig. 6. Wide angle X-ray diffractograms of pristine unmodified MMT (HPS) and organically modified MMT (C30B), wheat gluten-based materials (WG) and wheat gluten-based materials filled with HPS (WG-HPS) and C30B (WG-C30B).

corresponding to d001 ¼ 19.2 Å), confirming that organically modified MMT were not unable to well disperse within the wheat gluten matrix until exfoliation, even if C30B nanoclays were slightly intercalated by partial penetration of the protein chain in the

Fig. 7. TEM pictures of wheat gluten-based materials filled with HPS (a) and C30B (b).

2095

interlayer (Fig. 6). Although C30B displayed a higher interlayer distance and a certain hydrophilicity brought by hydroxyl groups, it seemed that its hydrophobic character was preponderant, leading to a poor compatibility with the wheat gluten matrix. It should be noted that our findings were different from those of Zhang et al. [10] who concluded that a nanocomposite structure was achieved for wheat gluten-C30B material prepared by casting. Nevertheless, the presence of the d001 peak of the C30B on their XRD pattern of the corresponding nanocomposite suggested that the nanoclays would be not really exfoliated. Based on WAXS diffractograms (Fig. 6), the introduction of unmodified MMT (HPS) resulted in the disappearance of the diffraction peak around 2q ¼ 7
, corresponding to the basal interlayer spacing value of HPS (d001 ¼ 12.7 Å). The presence of the group of peaks (at around 20
) ascribed to the crystallographic planes of the MMT on the diffraction pattern of wheat gluten material containing HPS, demonstrated that the WAXS analysis was sufficiently sensitive to detect the presence of MMT (5 wt%) in the nanocomposite. These results were supported by TEM observations (Fig. 7) since almost all nanoclays appeared well dispersed. This demonstrated that such materials displayed a well intercalatedexfoliated nanocomposite structure in favor of the creation of a tortuous pathway for penetrants. However, even if it has been shown that the water adsorption capacity was reduced in the presence of HPS, the water diffusivity was not affected in spite of a well-exfoliated nanocomposite structure achieved in the presence of HPS. Thus, the tortuosity effect, which is often mentioned in literature but rarely supported by structural characterization combined with water diffusivity measurements, seemed ineffective to reduce water diffusion in the present study. In addition, it is worth noting that the characteristic times for water sorption and biodegradation kinetics were not in the same order of magnitude in our experimental conditions. The time required to reach 50% of maximum water uptake (WUeq) was around 3 h, whereas the time to reach 50% of maximum degradation (Degmax) was up to 5 days. Nevertheless, in soil conditions, the temperature is usually lower (<10
C) and the water is less available; thus the water uptake kinetic is expected to be slower. Moreover, once the sine qua non water activity condition would be reached for microbial growth, it would still require at least 24e48 h of incubation to reach the growing stage of soil microflora. Thus, in soil conditions, the kinetic of water uptake could be considered as a limiting factor and consequently, the tortuosity effect might be a key parameter for limiting biodegradation. Finally, the analysis of the micro/macroscopic structure of the wheat gluten materials revealed that granulates were porous as reflected by the presence of holes on SEM pictures (Fig. 8). The porosity was evaluated by image analysis of SEM scans (Fig. 8). For unfilled wheat gluten-based materials, around 90 holes per granulate were counted, with a median hole diameter around 36 mm. As shown in Fig. 8, the granulate porosity was affected by the addition of nanoclays. The presence of organically modified MMT (C30B) led to a two-fold increase of the number of holes and a decrease of their size, also by a factor of 2. In the case of unmodified MMT (HPS), a more contrasted macrostructure was observed: holes were 2.7 fold more numerous and 1.5 fold bigger with a larger diameter distribution. As a result, these latter materials displayed a relatively important total hole area (4.6% of the total area as compared to 1.1% for the neat matrix). Whatever the formulation, these holes could be ascribed to combined phenomena including air incorporation during the extrusion process and/or a rapid water evaporation in the die, resulting in material expansion. In the presence of hydrophilic MMT (HPS), this latter phenomenon was probably

2096

A. Chevillard et al. / Polymer Degradation and Stability 96 (2011) 2088e2097

Fig. 8. SEM pictures of wheat gluten-based materials (WG) and wheat gluten-based materials filled with HPS (WG-HPS) and C30B (WG-C30B) and corresponding hole area distribution.

emphasized by the supposed temperature increase at the core of the product. 4. Conclusion The rate of biodegradation of wheat gluten-based materials has been reduced by adding unmodified MMT (HPS) without affecting the final biodegradation level whereas the presence of an organically modified MMT (C30B) did not significantly influence the biodegradation pattern. Three hypotheses have been proposed to explain how the presence of MMT could slow down biodegradation patterns of wheat gluten-based materials: (i) a reduced water adsorption capacity of the materials in the presence of such fillers, (ii) the establishment of interactions between MMT and the matrix, resulting in a lower availability for the matrix to be biodegraded, and/or (iii) the presence of a tortuous path induced by the nanodispersion of layered silicates leading to a slower diffusion of penetrants. In the case of organically modified MMT (C30B), no change in biodegradation pattern was observed based on the three

hypotheses proposed above (no change in water sensitivity, poor compatibility between C30B and wheat gluten and no tortuous pathway due to a bad dispersion of nanoclays). On the contrary, the presence on unmodified MMT (HPS) led to a significant reduction in biodegradation rate which is fully consistent with the two first hypotheses (decrease of liquid water adsorption and good affinity between HPS and wheat gluten). Concerning the third hypothesis, a good dispersion/exfoliation of nanoclays was achieved, but the resulting tortuosity effect appeared not effective to significantly reduce the liquid water diffusion even if it might be sufficient to limit diffusion of other penetrants like enzymes and microorganisms. To conclude, biodegradation pattern of wheat gluten-based materials can be modulated by incorporating nanoclays at filler content as low as 5 wt%. Among the three hypotheses proposed to explain the underlying mechanisms, all of them were validated by the results obtained. The molecular/ macromolecular compatibility between the clay layers and the wheat gluten matrix, i.e. the ability of both components to establish interactions, appeared as the key parameter governing

A. Chevillard et al. / Polymer Degradation and Stability 96 (2011) 2088e2097

the nanostructure, the liquid water sensitivity and, as a result, the biodegradation process. References [1] Cuq B, Gontard N, Guilbert S. Proteins as agricultural polymers for packaging production. Cereal Chemistry 1998;75:1e9. [2] Angellier-Coussy H, Guillard V, Guillaume C, Gontard N. Wheat gluten-based materials for food packaging. In: Lagaron DJM, editor. Multifunctional and nanoreinforced polymers for food packaging. Woodhead; 2011. p. 750. [3] Schofield JD, Bottomley RC, Timms MF, Booth MR. The effect of heat on wheat gluten and the involvement of sulfhydryl-disulfide interchange reactions. Journal of Cereal Science 1983;1:241e53. [4] Redl A, Morel MH, Bonicel J, Vergnes B, Guilbert S. Extrusion of wheat gluten plasticized with glycerol: influence of process conditions on flow behavior, rheological properties, and molecular size distribution. Cereal Chemistry 1999;76:361e70. [5] Domenek S, Feuilloley P, Gratraud J, Morel M-H, Guilbert S. Biodegradability of wheat gluten based bioplastics. Chemosphere 2004;54:551e9. [6] Angellier-Coussy H, Gastaldi E, Gontard N, Guillard V. Influence of processing temperature on the water vapour transport properties of wheat gluten based agromaterials. Industrial Crops and Products 2011;33:457e61. [7] Angellier-Coussy H, Torres-Giner S, Morel MH, Gontard N, Gastaldi E. Functional properties of thermoformed wheat gluten/montmorillonite materials with respect to formulation and processing conditions. Journal of Applied Polymer Science 2008;107:487e96. [8] Zhang X, Gozukara Y, Sangwan P, Gao D, Bateman S. Biodegradation of chemically modified wheat gluten-based natural polymer materials. Polymer Degradation and Stability 2010;95:2309e17. [9] Tunc S, Angellier H, Cahyana Y, Chalier P, Gontard N, Gastaldi E. Functional properties of wheat gluten/montmorillonite nanocomposite films processed by casting. Journal of Membrane Science 2007;289:159e68. [10] Zhang XQ, Do MD, Dean K, Hoobin P, Burgar IM. Wheat-gluten-based natural polymer nanoparticle composites. Biomacromolecules 2007;8:345e53. [11] Angellier-Coussy H, Gastaldi E, Correa Da Silva F, Gontard N, Guillard V. Nanoparticle size and water diffusivity in nanocomposite agro-polymer based films. Journal of Membrane Science, Unpublished results. [12] Olabarrieta I, Gallstedt M, Ispizua I, Sarasua JR, Hedenqvist MS. Properties of aged montmorillionite-wheat gluten composite films. Journal of Agricultural and Food Chemistry 2006;54:1283e8. [13] Martucci JF, Ruseckaite RA. Biodegradation of three-layer laminate films based on gelatin under indoor soil conditions. Polymer Degradation and Stability 2009;94:1307e13. [14] Pojanavaraphan T, Magaraphan R, Chiou BS, Schiraldi DA. Development of biodegradable foamlike materials based on casein and sodium montmorillonite clay. Biomacromolecules 2010;11:2640e6. [15] Rimdusit S, Jingjid S, Damrongsakkul S, Tiptipakorn S, Takeichi T. Biodegradability and property characterizations of methyl cellulose: effect of nanocompositing and chemical crosslinking. Carbohydrate Polymers 2008;72: 444e55. [16] Fukushima K, Abbate C, Tabuani D, Gennari M, Camino G. Biodegradation of poly(lactic acid) and its nanocomposites. Polymer Degradation and Stability 2009;94:1646e55. [17] Fukushima K, Tabuani D, Abbate C, Arena M, Ferreri L. Effect of sepiolite on the biodegradation of poly(lactic acid) and polycaprolactone. Polymer Degradation and Stability 2010;95:2049e56. [18] Sabet SS, Katbab AA. Interfacially compatibilized poly(lactic acid) and poly(lactic acid)/polycaprolactone/organoclay nanocomposites with improved biodegradability and barrier properties: effects of the compatibilizer structural parameters and feeding route. Journal of Applied Polymer Science 2009;111:1954e63. [19] Wang SF, Song CJ, Chen GX, Guo TY, Liu J, Zhang BH, et al. Characteristics and biodegradation properties of poly(3-hydroxybutyrate-co-3-hydroxyvalerate)/ organophilic montmorillonite (PHBV/OMMT) nanocomposite. Polymer Degradation and Stability 2005;87:69e76. [20] Chevillard A, Angellier-Coussy H, Peyron S, Gontard N, Gastaldi E. Investigating ethofumesate e clay interactions for pesticide controlled release. Soil Science Society of America Journal, in press. [21] Crank J. The mathematics of diffusion. Oxford University Press; 1976.

2097

[22] Vergnaud JM, Rosca ID. Assessing food safety of polymer packaging. Rapra Technology; 2006. [23] Mascheroni E, Guillard V, Nalin F, Mora L, Piergiovanni L. Diffusivity of propolis compounds in polylactic acid polymer for the development of antimicrobial packaging films. Journal of Food Engineering 2010;98:294e301. [24] Carslaw HS, Jaeger JC. Conduction of heat in solids. Oxford: Clarendon Press; 1997. [25] Rossello C, Simal S, SanJuan N, Mulet A. Nonisotropic mass transfer model for green bean drying. Journal of Agricultural and Food Chemistry 1997;45: 337e42. [26] Senadeera W, Bhandari BR, Young G, Wijesinghe B. Influence of shapes of selected vegetable materials on drying kinetics during fluidized bed drying. Journal of Food Engineering 2003;58:277e83. [27] Gill PE, Murray W, Wright MH. Practical optimization. London: Academic Press; 1981. [28] Domenek S, Morel M-H, Bonicel J, Guilbert S. Polymerization kinetics of wheat gluten upon thermosetting. A mechanistic model. Journal of Agricultural and Food Chemistry 2002;50:5947e54. [29] Singh H, MacRitchie F. Application of polymer science to properties of gluten. Journal of Cereal Science 2001;33:231e43. [30] Calmon A, Silvestre F, Bellon-Maurel V, Roger JM, Feuilloley P. Modelling easily biodegradability of materials in liquid medium-relationship between structure and biodegradability. Journal of Environmental Polymer Degradation 1999;7:135e44. [31] Bastioli C. Handbook of biodegradable polymers. Smithers Rapra Technology; 2005. [32] Park SK, Hettiarachchy NS, Were L. Degradation behavior of soy proteinwheat gluten films in simulated soil conditions. Journal of Agricultural and Food Chemistry 2000;48:3027e31. [33] Ray SS, Okamoto M. Biodegradable polylactide and its nanocomposites: opening a new dimension for plastics and composites. Macromolecular Rapid Communications 2003;24:815e40. [34] Ray SS, Yamada K, Okamoto M, Ueda K. Control of biodegradability of polylactide via nanocomposite technology. Macromolecular Materials and Engineering 2003;288:203e8. [35] Sinha Ray S, Yamada K, Okamoto M, Ueda K. New polylactide-layered silicate nanocomposites. 2. Concurrent improvements of material properties, biodegradability and melt rheology. Polymer 2003;44:857e66. [36] Dutta S, Karak N, Saikia JP, Konwar BK. Biocompatible epoxy modified biobased polyurethane nanocomposites: mechanical property, cytotoxicity and biodegradation. Bioresource Technology 2009;100:6391e7. [37] Paul MA, Delcourt C, Alexandre M, Degée P, Monteverde F, Dubois P. Polylactide/montmorillonite nanocomposites: study of the hydrolytic degradation. Polymer Degradation and Stability 2005;87:535e42. [38] Nieddu E, Mazzucco L, Gentile P, Benko T, Balbo V, Mandrile R, et al. Preparation and biodegradation of clay composites of PLA. Reactive and Functional Polymers 2009;69:371e9. [39] Zhuang H, Zheng JP, Gao H, Yao KD. In vitro biodegradation and biocompatibility of gelatin/montmorillonite-chitosan intercalated nanocomposite. Journal of Materials Science-Materials in Medicine 2007;18:951e7. [40] Someya Y, Kondo N, Shibata M. Biodegradation of poly(butylene adipate-cobutyleneterephthalate)/layered-silicate nanocomposites. Journal of Applied Polymer Science 2007;106:730e6. [41] Domenek S, Brendel L, Morel MH, Guilbert S. Swelling behavior and structural characteristics of wheat gluten polypeptide films. Biomacromolecules 2004;5: 1002e8. [42] Grimshaw RW, Searle AB. The chemistry and physics of clays and allied ceramic materials. Wiley-Interscience; 1971. [43] Morel MH, Redl A, Guilbert S. Mechanism of heat and shear mediated aggregation of wheat gluten protein upon mixing. Biomacromolecules 2002; 3:488e97. [44] Redl A, Guilbert S, Morel MH. Heat and shear mediated polymerisation of plasticized wheat gluten protein upon mixing. Journal of Cereal Science 2003; 38:105e14. [45] Redl A, Morel MH, Bonicel J, Guilbert S, Vergnes B. Rheological properties of gluten plasticized with glycerol: dependence on temperature, glycerol content and mixing conditions. Rheologica Acta 1999;38:311e20. [46] Kokini JL, Cocero AM, Madeka H, Degraaf E. The development of state diagrams for cereal proteins. Trends in Food Science & Technology 1994;5: 281e8.