- Email: [email protected]
Preparation and characterization of cross-linked canola protein isolate films
European Polymer Journal 89 (2017) 419–430
Contents lists available at ScienceDirect
European Polymer Journal journal homepage: www.elsevier.com/locate/europolj
Preparation and characterization of cross-linked canola protein isolate films
MARK
Shuzhao Lia, Elizabeth Donnera, Michael Thompsonb, Yachuan Zhangc, ⁎ Curtis Rempeld, Qiang Liua, a
Guelph Research and Development Centre, Agriculture and Agri-Food Canada, 93 Stone Road West, Guelph, ON N1G 5C9, Canada Department of Chemical Engineering, McMaster University, Hamilton, ON L8S 4L8, Canada c Lethbridge Research and Development Centre, Agriculture and Agri-Food Canada, Lethbridge, AB T1J 4B1, Canada d Canola Council of Canada, Winnipeg, MB R3B 0T6, Canada b
AR TI CLE I NF O
AB S T R A CT
Keywords: Canola protein Chemical modification Cross-linking reaction Mechanical property Curing effect
Cross-linked canola protein isolate (CPI) films were prepared using a bisepoxide as a chain extender by wet cast followed by heat compression. By DSC measurements, it was verified that the cross-linking reaction took place in the CPI matrix and was completed in 10 min at 90 °C. FTIR analysis confirmed that the chain extenders were covalently bonded with CPI chains. The reaction led to the formation of a cross-linked architecture which was revealed by an increase in molecular weight of the modified CPI films. The newly formed chain architecture increased the thermostability of CPI as evaluated by TGA; furthermore, the thermostability increased with the increase in cross-linking degree of the modified CPI films. DMA results of the unmodified CPI film showed that microcosmic phase separation occurred between two major proteins, cruciferin and napin in CPI; however, the extent of phase separation decreased with the increase of chain extenders and finally disappeared because of the increased compatibility between cruciferin and napin. Elastomeric mechanical behavior was observed with the introduction of cross-linked architecture in the modified CPI films. At low humidity of 55%, the cross-linked CPI films showed an increase in tensile strength and a decrease in elongation at break. At high humidity of 98%, the increase in hydrophobicity of cross-linked CPI matrix resisted the decreasing tensile strength seen in unmodified films.
1. Introduction With a growing interest in green products, the demand for protein-based industrial products is increasing worldwide in the fields of food, gels, adhesives and packaging [1–4]. Canola protein is obtained from the second largest oilseed crop after soy [5,6]; however, the main source of canola protein, canola meal, is not used in human food applications because of the presence of some antinutritional compounds, such as glucosinolates, erucic acid, phytates, and phenolics [7,8], making it an ideal candidate for industrial purposes lest its use be relegated to low value animal feed [9]. However, the poor processability, low mechanical properties, and high moisture sensitivity of protein-based biopolymers have limited their applications [10–12]. Accordingly, to increase their mechanical properties and processability, various efforts have been made, such as plasticizing with glycerol [13], propanediol [14–17], polyethylene glycol, or sorbitol [18]; modification of the protein with chemicals [19–21]; or mixing with hydrophobic materials [22–
⁎
Corresponding author. E-mail addresses: [email protected] (S. Li), [email protected] (E. Donner), [email protected] (M. Thompson), [email protected] (Y. Zhang), [email protected] (C. Rempel), [email protected] (Q. Liu). http://dx.doi.org/10.1016/j.eurpolymj.2017.03.001 Received 17 November 2016; Received in revised form 16 January 2017; Accepted 1 March 2017 Available online 02 March 2017 0014-3057/ Crown Copyright © 2017 Published by Elsevier Ltd. All rights reserved.
European Polymer Journal 89 (2017) 419–430
S. Li et al.
25]. The introduction of polar additives can increase mechanical properties of protein-based biopolymer to an extent [26]; on the other hand, this method can also lead to an increase in moisture sensitivity. Blending with hydrophobic materials usually generates immiscible/incompatible phase separation, which deteriorates the mechanical properties of the protein matrix. Chemical crosslinking modification can enhance the formation of covalent bonds in the protein matrix; thus, mechanical properties and barrier properties of the protein-based biopolymer can be further improved. Many cross-linkers, such as thiol [27], polyurethane and polyisocyanate [28–31], carbodiimide [32], genipin [33], or enzymes [21] can be used to modify the protein matrix. Of these chemicals, aldehydes including dialdehydes and monoaldehydes have been extensively used for cross-linking protein macromolecules, usually by the compression molding method [20,34–39]. Although the use of an aldehyde generally improved the mechanical properties and water-resistance properties of protein-based biopolymers, the optimal aldehyde type and cross-linking density should be experimentally determined. Zárate-Ramírez et al. [40] evaluated the mechanical properties of wheat gluten biopolymers with and without treatment using formaldehyde, glutaraldehyde, and glyoxal by compression molding. It was found that tensile strength of the aldehyde-modified samples was not improved compared to that of aldehyde-free samples. Vaz et al. [41] found that both the tensile modulus and strength of extruded soy protein isolate biopolymer decreased with the addition of glyoxal. They believed that the probability of glyoxal cross-linking via two amine groups of soy located on two adjacent chains was smaller than cross-linking along the same chain during extrusion. Accordingly, the covalent bonds were predominantly introduced within the polypeptide chains instead of between them. In fact, the compression molding method is usually conducted under high pressure and temperature in order to melt and compact the protein matrix [26], which is not easy to apply in industry. In canola, cruciferin and napin are the two major families of storage proteins, which constitute 60% and 20% of the total protein content in mature seeds, respectively [42]. Cruciferin is a 12S globulin with a high molecular weight (300–310 kDa) and several subunits. Napin is a 2S albumin with a low molecular weight (12.5–14.5 kDa) [43]. Therefore, canola protein is very suitable for the preparation of protein-based films by the casting method because of its water solubility and the well balanced constitution of components with different molecular weight, in which cruciferin can increase mechanical properties of the matrix and napin can act as a plasticizer to increase processability. However, few studies have been conducted on the chemical modification of CPI until now. The objectives of this research are to use canola protein as an ingredient to develop a novel protein-based bioplastic film and to understand the factors that influence the properties of this newly developed product. In this research, 1,4-butanediol diglycidyl ether (BDDE) was used as a chain extender to prepare cross-linked CPI film for the first time. Dermal sheep collagen cross-linked with BDDE was prepared and evaluated after subcutaneous implantation in rats [44]. The results showed cross-linking of dermal sheep collagen with BDDE resulted in biocompatible materials in terms of non-cytotoxicity and non-antigenicity. Additionally, clinical and biocompatibility data spanning more than 15 years support the favorable clinical safety profile of BDDE-crosslinked hyaluronic acid (HA) and its degradation products [45]. Therefore, the usage of BDDE as a chain extender would be anticipated to offer novel properties to CPI in food and industrial applications. Furthermore, the casting method was employed to denature canola protein and unfold the protein chains to an extent in solvent by heating to form a cross-linked protein chain network. The preparation method and the investigation of properties help to increase the knowledge of protein chemical modification and broaden the application fields of canola protein. 2. Materials and methods 2.1. Materials and extraction of CPI from canola meal Glycerol and 1,4-butanediol diglycidyl ether were purchased from Sigma Aldrich and used as received. The protein extraction procedure was based on the work reported by Manamperi et al. [6] with some minor modifications. Conventional dark-seeded (Brassica napus) defatted canola meal was provided by Bunge Canada (Altona, Manitoba). The canola meal was tested for composition which showed 36.5% protein, 11.7% crude fibre, 4.1% crude oil (d.b), 9.8% moisture, and 7.3% ash, in the labs of the Department of Food Science at the University of Manitoba, Canada. The meal was ground into powder using a mill (A10 analytical mill, Ika Works, Wilmington, NC) to pass through a 250 μm sieve and then suspended in distilled water at a ratio of 1/10 (g/ml). The pH of the suspension was adjusted to 12 with NaOH (4 N) with vigorous stirring at 55 °C for 3 h. The first centrifugation took place at 10,000 rpm for 20 min in a Sorvall LYNX 4000 centrifuge (Thermo Scientific, Germany) and the supernatant was collected. The pH of the obtained solution was adjusted to 5 by drop-wise addition of HCl solution (2 N) to allow the canola protein to precipitate. The second centrifugation was conducted and the sediment (protein) was collected and then washed three times with distilled water. The extracted proteins were then dialyzed for 12 h with distilled water with frequent distilled water changes. The protein was recovered after another centrifugation, freeze dried and stored at 5 °C until further use. 2.2. Preparation of cross-linked CPI samples Cross-linked CPI samples were prepared as follows, using the solvent casting method followed by heat compression. A protein suspension was prepared by mixing 32 g of protein powder with 1200 mL of distilled water. After adding 13.6 g of glycerol, the protein suspension was homogenized via a magnetic stirrer for 30 min. The suspension was heated at 65 °C for 30 min and then cooled down to ambient temperature to obtain a protein solution. The solution was then separated equally into four parts, to three of which 16 mg, 32 mg, and 80 mg of BDDE were added to prepare three cross-linked CPI samples. The solution without added BDDE was used as the control. All of the solutions were stirred for 1 h, and then degassed in a vacuum oven at ambient temperature for 420
European Polymer Journal 89 (2017) 419–430
S. Li et al.
45 min, cast onto a Teflon-coated plate with a diameter of 20 cm, and finally dried at ambient temperature for 72 h in a fume hood. The films peeled off of the plates with 0, 16, 32 and 80 mg BDDE were named CP+BDDE-0, CP+BDDE-1, CP+BDDE-2, and CP +BDDE-3, respectively. All of the dried CPI samples were conditioned at ambient conditions (23.8 °C, 50% RH) for 48 h and then heat compression cured with a Carver hydraulic hot press (model 3925) at 90 °C for 10 min under pressure of 1.5 MPa. The thickness of the prepared protein film varied between 0.25 mm and 0.30 mm, and the prepared films containing 0, 16, 32 and 80 mg BDDE after heat compression were denoted as unmodified CPI, CP-BDDE-1, CP-BDDE-2, and CP-BDDE-3. 2.3. Differential scanning calorimetry (DSC) DSC analysis was performed using a differential scanning calorimeter (Q20 DSC, TA Instruments, New Castle, DE, USA) equipped with a refrigerated cooling system. A high-volume pan was loaded with a film sample and then hermetically sealed. The temperature scan range was from 20 to 180 °C at a heating rate of 10 °C/min with a nitrogen purge. The instrument was calibrated with indium and an empty pan was used as a reference. 2.4. Fourier transform infrared spectroscopy (FTIR) Infrared spectra of unmodified and cross-linked CPI films were recorded with a Digilab FTS 7000 spectrometer (Digilab USA, Randolph, MA) equipped with a thermoelectrically cooled deuterated tri-glycine sulfate (DTGS) detector using an attenuated total reflectance accessory at a resolution of 4 cm−1 by 64 scans. A half-band width of 15 cm−1 and a resolution enhancement factor of 1.5 with Bessel apodization were employed. Spectra were subjected to ATR correction (Algorithm 1, correction 1.000), deconvolution (k factor 1.9, half width 24, Bessel apodization), and a multipoint linear baseline correction using Win-IR Pro software. The measured unmodified and cross-linked CPI films were dissolved in distilled water to prepare solutions with a concentration of 5 wt%. To remove the added glycerol and unreacted BDDE, the prepared CPI solutions were dialyzed using a Pur-A-Lyzer™ midi dialysis kit (Sigma Aldrich) with a pore size of 1000 Da molecular weight for 24 h with distilled water with frequent distilled water changes. The purified CPI solutions were dried in a vacuum at 60 °C for 72 h to obtain the samples for FTIR tests. 2.5. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) SDS-PAGE was performed according to the method of Laemmli [46]. Unmodified and cross-linked CPI samples were mixed with sample buffer in the presence of β-mercaptoethanol and SDS, heated for 10 min at 95 °C, and loaded onto a 4–10% polyacrylamide gel with SDS. Gels were run at a constant voltage (200 V) for 38 min in a Mini-Protean II electrophoresis cell (Bio-Rad, Hercules, CA, USA). The gels were developed with Bio-Safe™ Coomassie Stain. 2.6. High performance size exclusion chromatography (HPSEC) The high performance size exclusion chromatographic (HPSEC) system consisted of a Shimadzu SIL-20A HT autosampler, LC-10A T pump and DGU-14A degasser (Shimadzu Corp., Kyoto, Japan), and a Viscotek TDA 305 Triple Detector Array and UV detector 2600 (Malvern Instruments Ltd., Malvern, UK). Two AquaGel columns (PAA-M and PAA-203) were connected in series (PolyAnalytik, London, Canada). The mobile phase (100 mM sodium nitrate and 5 mM sodium azide) was filtered through a 0.22 μm filter. The flow rate was 0.5 mL/min and the columns and detectors were maintained at 40 °C. CPI solutions were prepared at a concentration of 2 mg/mL and 100 μl were injected. 2.7. Thermogravimetric analysis (TGA) The onset decomposition temperature of all CPI samples was determined by TGA using a SDT Q600 (TA Instruments, New Castle, DE, USA). The decomposition measurements were obtained between 50 and 600 °C at a scanning rate of 20 °C/min under a helium atmosphere. 2.8. Dynamic mechanical analysis (DMA) Dynamic mechanical properties of the biopolymer samples were measured using a dynamic mechanical analyzer (DMA) Q800 (TA Instruments, New Castle DE, USA) operated in a multi-frequency strain mode. The specimens were cut into a nominal dimension of 55 mm × 5 mm × 0.25 mm and were conditioned in a laboratory atmosphere (22 °C and 55% relative humidity) for 48 h prior to tensile testing. Each specimen was mounted on the dual cantilever beam clamp, and an amplitude of 250 μm was applied to the specimen at a frequency of 1 Hz. The storage modulus, loss modulus, and tan delta (tanδ ) were recorded as the samples were heated from −70 °C to 150 °C at a heating rate of 5 °C/min. Each sample had two replicates. 2.9. Mechanical properties Mechanical properties were measured using a Com-Ten testing system (model 701SN, FL, USA) with a 0.5 kN load cell according to the Standard Test Method for Tensile Properties of Plastics (ASTM D638). Type-IV tensile specimens were prepared by die cutting 421
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 1. Differential scanning calorimetric thermograms of (a) four CPI+BDDE films prepared by casting without heat compression, (b) CP-BDDE-3 films cured by heat compression at 3, 5, and 10 min at 90 °C.
and were conditioned in a laboratory atmosphere (22 °C and 55% relative humidity) and in a desiccator with saturated potassium sulfate aqueous solution (22 °C and 98% relative humidity) for 48 h prior to tensile testing. A crosshead speed of 5 mm/min was used. The tensile testing was conducted at ambient conditions (23.8 °C, 50% RH). The stress, strain, tensile strength, and elongation-atbreak were calculated from at least seven replicates. 3. Results and discussion 3.1. DSC To investigate the cross-linking reaction taking place in the CPI matrix, DSC thermograms for CP+BDDE-0, CP+BDDE-1, CP +BDDE-2, and CP+BDDE-3 (obtained by casting without heat compression) are shown in Fig. 1(a). A glass transition, Tg , was observed in the thermogram of CP+BDDE-0 film at 48.3 °C as shown in Fig. 1(a). A similar DSC pattern of CPI was reported by Manamperi et al. [6]. However, the Tg at 48.3 °C disappeared in the thermogram of CP+BDDE-1 because BDDE can act as a plasticizer in the matrix before the cross-linking reaction, and a new broad peak was observed at Tp = 115.8 °C, which was an exothermic transition corresponding to the reaction heat of cross-linking reaction [47]. Data obtained from the DSC thermograms, such as the peak temperature (Tp ), the onset temperature (To ), the terminal temperature (Tend ), the curing range, and the cumulative reaction heat 422
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Table 1 Curing characteristics of the cross-linking reaction systems. Sample
To (°C)
Tp (°C)
Tend (°C)
Curing range (°C)
ΔHr (J/g)
CP+BDDE-1 CP+BDDE-2 CP+BDDE-3
86.3 87.7 82.2
115.8 118.7 115.2
153.6 153.8 155.4
67.3 66.1 73.2
4.74 6.78 9.68
(ΔHr ) are listed in Table 1. It was found that the reaction heat increased with the increasing amount of BDDE, indicating that more cross-linking reactions took place in the system with more BDDE. On the other hand, the cross-linking reaction in CP+BDDE-1, CP +BDDE-2, and CP+BDDE-3 had similar To and started before 90 °C, pointing to the possibility of modifying the CPI films with competition by thermal degradation. To determine the kinetics of the present reaction system, CP+BDDE-3 films were cured at 90 °C for 3 min, 5 min, and 10 min and then analyzed by DSC. The resulting thermograms are shown in Fig. 1(b). It was found that the cumulative reaction heat decreased with increase of the reaction time until no reaction heat was detected for the films cured for 10 min. Reproducibly it was found that the cross-linking reaction could be finished completely in 10 min at 90 °C. Consequently, to decrease the effect of degradation on the film properties, the reaction time of 10 min was chosen as suitable for the present reaction systems. 3.2. FTIR The FTIR spectra of unmodified and cross-linked CPI films are shown in Fig. 2(a) and (b). In the FTIR spectrum of unmodified CPI film shown in Fig. 2(a), the absorption peaks between 3500 and 3000 cm−1 were ascribed to the water region, which includes the stretching vibration of NeH and OeH [48]. The peak at 3282 cm−1 in the spectrum of unmodified CPI was assigned to amide A band,
Fig. 2. FTIR spectra at (a) 3700–2700 cm−1 and (b) 1210–1060 cm−1 of unmodified and cross-linked CPI films prepared by solvent casting and heat compression. The unreacted BDDE and added glycerol were removed from the CPI matrix in preparation for the analysis.
423
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 3. The preparation of cross-linked CPI sample.
which is exclusively located on the NH group and for that reason is not sensitive to the conformation of the polypeptide backbone in protein [49]. The peaks at 2851 and 2919 cm−1 in the spectrum of unmodified CPI were assigned to symmetric and asymmetric stretching vibrations of eCH2e groups on proteins. The peak at 3282 cm−1 on the spectra of cross-linked CPI became narrow and a new peak appeared at 3180 cm−1 which is assigned to OeH stretching vibration [48], indicating that new hydrogen groups were generated by the ring-opening reaction between amide groups on protein chains and epoxy groups on BDDE as shown in Fig. 3. In addition, the intensity of peaks at 2851 and 2919 cm−1, ascribed to stretching vibrations of eCH2e, significantly increased with the introduction of BDDE corresponding to its ethyl groups. The unreacted BDDE and added glycerol had been removed from the CPI matrix in preparation for the analysis; therefore, the increase of peak intensity at 2851 and 2919 cm−1 was attributed to the introduction of more BDDE into CPI chains by covalent bonding. In addition, in Fig. 2(b) new peaks assigned to the CeOeC band for BDDE appeared at 1108 cm−1 on the spectra of cross-linked CPI films, although the peaks were weak. Therefore, by the FTIR results it could be confirmed that BDDE reacted with CPI chains to form covalent bonds. 3.3. SDS-PAGE Fig. 4 shows the SDS-PAGE of unmodified and three cross-linked CPI solutions. Although other bands below 250 kDa were not separated in Fig. 4, unmodified CPI (lane 5) showed one band above 250 kDa, which was assigned to cruciferin protein [50]. CPIBDDE-1 (lane 2) showed an additional band with a higher molecular weight than that of unmodified CPI that was attributed to the introduction of cross-linked chains. With increasing amounts of BDDE in the reaction, the additional band was much clearer for CPIBDDE-2 (lane 3) and CPI-BDDE-3 (lane 4), indicating more cross-linked protein chains with similar molecular weight to that of CPBDDE-1 were generated in these two cross-linked CPI films. Furthermore, a third band above the two bands that appeared in CPBDDE-1 could be observed for CP-BDDE-2 and CP-BDDE-3, which suggested that cross-linked protein chains with an even higher molecular weight than in CP-BDDE-1 were being identified. Although the SDS-PAGE could not distinguish the fractions below 250 kDa, the results could qualitatively illuminate the changes in the architecture because of the cross-linking reaction. 3.4. SEC CPI consists of two major families, napin (12.5–14.5 kDa) and cruciferin (300–310 kDa), and other minor protein fractions
Fig. 4. SDS-PAGE of unmodified and cross-linked CPI film solutions. Lanes 1 and 6, molecular weight marker; lane 2, CPI-BDDE-1; lane 3, CPI-BDDE-2; lane 4, CPIBDDE-3; lane 5, unmodified CPI.
424
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 5. Refractive index chromatograms of unmodified and cross-linked CPI films by SEC.
including oleosins, thionins, and lipid transfer protein (LTP). The composition of unmodified CPI as determined by SEC (Fig. 5) revealed five peaks. The two major canola proteins were well separated; peak 2 was identified as cruciferin and comprised 71.7% of the total peak area while peak 4 was identified as napin and comprised 16.9% of the total. Peak 3 was not pure (mixture of cruciferin and napin) [51] and comprised 7.7% of the total. The number-average molecular weight (Mn) of peak 1 (2.4% of the total) was approximately 800 kDa, which contained aggregates of cruciferin (peak 2, Mw 300 kDa). Peak 5 (8.4 kDa) probably consisted of other minor proteins like LTP or some small polypeptides. The SEC refractive index chromatogram of CP-BDDE-1 showed that the cross-linking reaction led to a remarkable decrease in CPI components corresponding to peak 2 (57.2% of the total) and peak 4 (10.5% of the total), while peak 1 representing aggregates increased to 26.3%, indicating the formation of cross-linked protein chains. The addition of BDDE resulted in a further decrease in peak 2 (38.7% of the total) and peak 4 (5.8% of the total) and increases in peak 1 (52.2% of the total) except for CP-BDDE-3. The decrease of peak 2 and 4 with the addition of BDDE indicated that BDDE had broad reaction specificity and could react with the two major protein fractions in this CPI matrix. In fact, approximately 5.6% of insoluble solid was found in CP-BDDE-3, which indicated some protein chains were further cross-linked and formed insoluble structures. Consequently, the intensity of peak 1 of CP-BDDE-3 in Fig. 5 decreased, attributed to the removal of the insoluble crosslinked proteins during the SEC measurement. Zárate-Ramírez et al. [41] used glyoxal to react with soy protein isolate and found the covalent bonds were predominantly introduced within the polypeptide chains instead of between them (intramolecular crosslinking). However, from the SEC results it could be confirmed that the cross-linking reaction took place between two adjacent CPI chains and a cross-linked architecture was formed in the present reaction system, which increased the weight-average molecular weight from 337 kDa for unmodified CPI to 512 kDa for CP-BDDE-1, 700 kDa for CP-BDDE-2, and 820 kDa for CP-BDDE-3.
Fig. 6. TGA and DTG curves of unmodified and cross-linked CPI films.
425
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Table 2 DMA and TGA data of unmodified and cross-linked CPI. Sample
Unmodified CPI CP-BDDE-1 CP-BDDE-2 CP-BDDE-3
α relaxation
β relaxation
Breakage temperature in TGA curve (°C)
Peak temperature (°C)
tanδ at peak temperature
Peak temperature (°C)
tanδ at peak temperature
42.0 32.6 22.8 17.6
0.44 0.46 0.44 0.42
−50.9 −50.5 −48.0 −47.8
0.24 0.24 0.25 0.25
281.7 294.3 304.5 309.2
3.5. TG analysis The effect of cross-linking architecture on thermostability of CPI films was investigated by TGA. Fig. 6 shows the DTG curve of unmodified CPI film and thermogravimetric decomposition curves of unmodified and cross-linked CPI films. There were four weight loss segments on the DTG curve of unmodified CPI as shown in Fig. 6. The first weight loss of 7.8% below ∼120 °C was attributed to the elimination of free moisture in the protein matrix. The weight loss of approximately 33% appearing between 120 and 280 °C was mainly attributed to evaporation of glycerol and any bonded water with the polar groups on protein molecules; the loss corresponded to the amount of added glycerol as plasticizer in unmodified CPI film. The scission of covalent peptide bonds occurred between 280 and 500 °C, which continued over a long period of decomposition because of the presence of various protein fractions with a broad molecular weight distribution. The last weight loss after 500 °C corresponded to the cleavage of SeS, OeN, and OeO linkages in protein macromolecules, which resulted in the decomposition of protein molecules [52]. There was no remarkable difference between unmodified and cross-linked CPI films in the weight loss before 280 °C corresponding to the loss of water and glycerol; however, the scission temperature of cross-linked CPI molecules increased with the introduction of cross-linked architecture, as summarized in Table 2. Furthermore, the greater the amount of BDDE, the higher the onset of scission. Consequently, the results indicated the introduction of cross-linked architecture increased the thermostability of CPI films. 3.6. Dynamic mechanical analysis Fig. 7 shows the storage modulus E′ and loss factor tanδ as a function of temperature for unmodified and cross-linked CPI films. With an increase of temperature, unmodified CPI clearly showed the transition region (tanδ peak and modulus drop) and the rubbery plateau in sequence. A decrease of three orders of magnitude in E′ was observed at the transition from glass to rubber region. The broad rubbery plateau located at the temperature interval from 85 to 140 °C resembles reported results for intermolecular covalent crosslinking through disulfide bonding among proteins [53]. CP-BDDE-1 showed a slight increase in E′ at the rubbery plateau attributed to the formation of some cross-linked chains. With the increase of BDDE, the E′ at the plateau increased for CP-BDDE-2, indicating that more cross-linking reaction took place and thus the cross-linking density increased. However, CP-BDDE-3 exhibited a thermosetting potential region in Fig. 7(a) because the addition of more BDDE gave rise to the presence of cross-linked architecture, which is in agreement with the presence of some insoluble solid resulting from cross-linking reaction in the results of molecular weight analysis. In Fig. 7(b), two obvious mechanical loss peaks could be observed, one broad peak located between 40 and 78 °C and the other one at −50 °C, in the tanδ curve of unmodified CPI. The peak appearing at higher temperature was associated with the glass transition temperature (Tg ) or the α-transition of plasticized unmodified CPI (as shown in Table 2). This is in accordance with the DSC result of unmodified CPI film, in which a Tg transition at 48.3 °C was observed. It may be noted that unmodified CPI film showed a broad Tg peak, which indicated the presence of phase separation or simply reflected the high heterogeneity of two major protein fractions in molecular size, cruciferin and napin with a molecular weight ranging from 9 to 350 kDa [54]. The low temperature transition was probably associated with Tg of a glycerol-rich phase, which was higher than Tg of glycerol at −93 °C [55]. The appearance of the low temperature transition was ascribed to the partial miscibility of glycerol and CPI chains [56]. It is evident in Fig. 7(b) that the broad Tg peaks of all of cross-linked CPI were becoming narrower with the increase of BDDE. The change is more apparent in CP-BDDE-3 which exhibited a very narrow peak at Tg . BDDE had broad reaction specificity and could react with any fraction in unmodified CPI as discussed in the results of molecular weight distribution; therefore, cross-linking reaction may take place between cruciferin and napin in unmodified CPI matrix so that the compatibility of the two protein fractions was increased in cross-linked CPI. Consequently, the cross-linked protein chains formed the matrix network in CP-BDDE-3 and showed a narrower Tg . The Tg of cross-linked CPI was much lower than that of unmodified protein, which was proposed to be related to the improved hydrophobicity because of decreased polar amine groups and the increase in compatibility of the two main protein fractions [53]. In addition, in the present case, the introduction of BDDE in cross-linked proteins may have increased the mobility of the chain fraction because BDDE is an ether and has a very soft chain, which could have decreased the Tg of the cross-linked CPI. More BDDE was introduced into CP-BDDE-3 and thus a lower Tg was seen in Table 2. Mizuno, Mitsuiki, and Motoki [57] treated 7S and 11S fractions isolated from soy protein using microbial transglutaminase (MTG) and found that cross-linking after MTG treatment rendered Tg lower than that of non-treated soy protein at the same water content. They argued that the MTG treatment caused an increase in immobilized water that contributed to the decrease of Tg . 426
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 7. Temperature dependence of storage modulus E′ (a) and loss tangent tanδ (b) for unmodified and cross-linked CPI films.
3.7. Mechanical properties Fig. 8 shows the strain-stress relationship of unmodified and cross-linked CPI films. As shown in the figure, the deformation nature of the unmodified CPI film at room temperature is typical of an elastomer in terms of the stress and strain. Generally, there were two
Fig. 8. Stress-strain curves of unmodified and cross-linked CPI films at relative humidity 55%.
427
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 9. Tensile strength (a) and elongation-at-break (b) of unmodified and cross-linked CPI films at conditioning humidities of 55% and 98%.
characteristic regions of deformation behavior in the tensile stress-strain curve of unmodified CPI film. At low strains (< 5%) the stress increased rapidly with increasing strain, in which the initial slope corresponds to the elastic moduli of the unmodified CPI film. A yield point appeared at the strain of 5%; beyond the yield point, the unmodified CPI film showed a plastic deformation, i.e., the stress slowly increases with strain (strain hardening) until failure occurred. Furthermore, “necking” phenomenon could be observed during the test, which is a typical thermoplastic behavior. For the cross-linked CPI samples, the macromolecules moved and rearranged during plastic deformation; however, entanglement of cross-linked chains restrained the further movement of chains and thus enhanced the strain hardening until it broke finally at the maximum strain. Therefore, the strain-stress relationship of modified CPI was suggestive of branched linear chains. Effects of cross-linked architecture on tensile strength and elongation at break are shown in Fig. 9. The addition of glycerol as a plasticizer to the unmodified CPI film reduced the interactions between protein chains and even disrupted intermolecular linkages of protein [26], which increased the mobility of protein chains and thus resulted in an increase in elongation-at-break and a decrease in tensile strength at 55% humidity. The entanglement of cross-linked chains limited the large deformation of cross-linked CPI films and led to a decrease in elongation-at-break while the tensile strength increased at 55% humidity, as shown in Fig. 9. Furthermore, with the increase in the amount of BDDE, elongation at break decreased and tensile strength increased further. This is consistent with the general findings using aldehydes as cross-linkers to react with various protein chains [19,36,37,58]. At high humidity of 98%, the tensile strength of unmodified CPI film sharply decreased compared to that at a humidity of 55% because of the poor water resistance property. Although the tensile strength of cross-linked CPI films also decreased at the humidity of 98% in comparison with that at 55%, the decrease was reduced with increasing BDDE. Cross-linking in the modified CPI films led to a decrease in the concentration of hydrophilic amino acid residues and thus improved the hydrophobic interaction of protein molecules. Therefore, the water resistance property of cross-linked CPI films was increased, which potentially diminished the effect of water on tensile strength. The elongation-at-break of unmodified CPI film at 98% humidity increased compared to that at 55% because absorbed water acted as a plasticizer to further increase the flexibility of protein chains. However, the elongation-at-break of cross-linked CPI films at the humidity of 98% showed a slight effect compared to that at 55% because cross-linking decreased the length of segments available for stretching and restricted the slippage of chains [59]. As a result, the elongation in cross-linked CPI was no longer determined by the slippage of molecules; instead, it was a function of the stretch of network-like molecular structure. Therefore, the effect of the absorbed water on the elongation of cross-linked CPI was weakened.
428
European Polymer Journal 89 (2017) 419–430
S. Li et al.
4. Conclusions In this research, the cast method was employed to denature canola protein and unfold the protein chains to an extent in water to form a protein chain network. The unfolded protein chains could increase the possibility of a cross-linking reaction taking place between two adjacent protein chains instead of intra-molecularly. The cross-linking reaction was followed by DSC and FTIR analysis. The increase in molecular weight verified that large protein molecules were formed in the cross-linked CPI films by the reaction. Furthermore, because of the broad reaction specificity of BDDE toward different types of proteins, cross-linking took place between cruciferin and napin in the CPI matrix so that the compatibility of these two major protein fractions was increased in the cross-linked CPI. The cross-linked CPI films showed higher thermostability compared with unmodified CPI films because of the introduction of cross-linked chains. In addition, the tensile strength of cross-linked CPI films was increased at low humidity. At high humidity, the increase of hydrophobicity of cross-linked CPI films decreased the absorption of water and thus restrained the deterioration of mechanical properties. Acknowledgements We thank the Canola Council of Canada, Saskatchewan Pulse Growers and Agriculture and Agri-Food Canada through Growing Forward 2 Canola/Flax and Pulse Science Cluster Programs for financial funding. References [1] S. Ou, Y. Wang, S. Tang, C. Huang, M.G. Jackson, Role of ferulic acid in preparing edible films from soy protein isolate, J. Food Eng. 70 (2) (2005) 205–210. [2] J. Burgain, R. El Zein, J. Scher, J. Petit, E.-A. Norwood, G. Francius, et al., Local modifications of whey protein isolate powder surface during high temperature storage, J. Food Eng. 178 (2016) 39–46. [3] B.E. Dybowska, Whey protein-stabilized emulsion properties in relation to thermal modification of the continuous phase, J. Food Eng. 104 (1) (2011) 81–88. [4] M. da Rocha, M.R. Loiko, G.V. Gautério, E.C. Tondo, C. Prentice, Influence of heating, protein and glycerol concentrations of film-forming solution on the film properties of Argentine anchovy (Engraulis anchoita) protein isolate, J. Food Eng. 116 (3) (2013) 666–673. [5] M.R.G. Tandang-Silvas, T. Fukuda, C. Fukuda, K. Prak, C. Cabanos, A. Kimura, et al., Conservation and divergence on plant seed 11S globulins based on crystal structures, Biochim. Biophys. Acta (BBA)-Proteins Proteom. 1804 (7) (2010) 1432–1442. [6] W.A. Manamperi, S.K. Chang, C.A. Ulven, S.W. Pryor, Plastics from an improved canola protein isolate: preparation and properties, J. Am. Oil Chemists' Soc. 87 (8) (2010) 909–915. [7] S.M. McCurdy, Effects of processing on the functional properties of canola/rapeseed protein, J. Am. Oil Chemists’ Soc. 67 (5) (1990) 281–284. [8] W.A. Manamperi, D.P. Wiesenborn, S.K. Chang, S.W. Pryor, Effects of protein separation conditions on the functional and thermal properties of canola protein isolates, J. Food Sci. 76 (3) (2011) E266–E273. [9] S.H. Tan, R.J. Mailer, C.L. Blanchard, S.O. Agboola, Canola proteins for human consumption: extraction, profile, and functional properties, J. Food Sci. 76 (1) (2011) R16–R28. [10] A.H. Tkaczyk, J.U. Otaigbe, K.-L.G. Ho, Bioabsorbable soy protein plastic composites: effect of polyphosphate fillers on biodegradability, J. Polym. Environ. 9 (1) (2001) 19–23. [11] P. Lodha, A.N. Netravali, Effect of soy protein isolate resin modifications on their biodegradation in a compost medium, Polym. Degrad. Stab. 87 (3) (2005) 465–477. [12] M. Pommet, A. Redl, M.-H. Morel, S. Guilbert, Study of wheat gluten plasticization with fatty acids, Polymer 44 (1) (2003) 115–122. [13] R. Lee, M. Pranata, Z. Ustunol, E. Almenar, Influence of glycerol and water activity on the properties of compressed egg white-based bioplastics, J. Food Eng. 118 (1) (2013) 132–140. [14] R. Kumar, V. Choudhary, S. Mishra, I. Varma, B. Mattiason, Adhesives and plastics based on soy protein products, Indust. Crops Prod. 16 (3) (2002) 155–172. [15] P. Lodha, A.N. Netravali, Thermal and mechanical properties of environment-friendly ‘green’plastics from stearic acid modified-soy protein isolate, Indust. Crops Prod. 21 (1) (2005) 49–64. [16] P. Chen, L. Zhang, New evidences of glass transitions and microstructures of soy protein plasticized with glycerol, Macromol. Biosci. 5 (3) (2005) 237–245. [17] P. Chen, L. Zhang, F. Cao, Effects of moisture on glass transition and microstructure of glycerol-plasticized soy protein, Macromol. Biosci. 5 (9) (2005) 872–880. [18] N.-B. Song, W.-S. Jo, H.-Y. Song, K.-S. Chung, M. Won, K.B. Song, Effects of plasticizers and nano-clay content on the physical properties of chicken feather protein composite films, Food Hydrocolloids 31 (2) (2013) 340–345. [19] A.J. Poole, J.S. Church, The effects of physical and chemical treatments on Na 2 S produced feather keratin films, Int. J. Biol. Macromol. 73 (2015) 99–108. [20] G.W. Selling, K.K. Woods, A. Biswas, J. Willett, Reactive extrusion of zein with glyoxal, J. Appl. Polym. Sci. 113 (3) (2009) 1828–1835. [21] K. Patzsch, K. Riedel, M. Pietzsch, Parameter optimization of protein film production using microbial transglutaminase, Biomacromolecules 11 (4) (2010) 896–903. [22] X. Mo, X. Sun, Thermal and mechanical properties of plastics molded from sodium dodecyl sulfate-modified soy protein isolates, J. Polym. Environ. 8 (4) (2000) 161–166. [23] Z. Zhong, X.S. Sun, Properties of soy protein isolate/polycaprolactone blends compatibilized by methylene diphenyl diisocyanate, Polymer 42 (16) (2001) 6961–6969. [24] J. John, M. Bhattacharya, Properties of reactively blended soy protein and modified polyesters, Polym. Int. 48 (11) (1999) 1165–1172. [25] M. Wei, L. Fan, J. Huang, Y. Chen, Role of star-like hydroxylpropyl lignin in soy-protein plastics, Macromol. Mater. Eng. 291 (5) (2006) 524–530. [26] P. Guerrero, A. Retegi, N. Gabilondo, K. De la Caba, Mechanical and thermal properties of soy protein films processed by casting and compression, J. Food Eng. 100 (1) (2010) 145–151. [27] D.L. Woerdeman, W.S. Veraverbeke, R.S. Parnas, D. Johnson, J.A. Delcour, I. Verpoest, et al., Designing new materials from wheat protein, Biomacromolecules 5 (4) (2004) 1262–1269. [28] Y. Zhang, W. Zhu, Y. Lu, Z. Gao, J. Gu, Nano-scale blocking mechanism of MMT and its effects on the properties of polyisocyanate-modified soybean protein adhesive, Indust. Crops Prod. 57 (2014) 35–42. [29] Y. Chen, L. Zhang, R. Deng, Y. Cui, A new network composite material based on soy dreg modified with polyurethane prepolymer, Macromol. Mater. Eng. 292 (4) (2007) 484–494. [30] H. Tian, Y. Wang, L. Zhang, C. Quan, X. Zhang, Improved flexibility and water resistance of soy protein thermoplastics containing waterborne polyurethane, Indust. Crops Prod. 32 (1) (2010) 13–20. [31] D.J. Sessa, G.W. Selling, A. Biswas, Reaction of zein with methylenediphenyl diisocyanate in the melt state: Thermal, mechanical, and physical properties, Indust. Eng. Chem. Res. 51 (27) (2012) 9199–9203. [32] V. Tropini, J.-P. Lens, W. Mulder, F. Silvestre, Wheat gluten films cross-linked with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide and N-hydroxysuccinimide, Indust. Crops Prod. 20 (3) (2004) 281–289.
429
European Polymer Journal 89 (2017) 419–430
S. Li et al.
[33] H.W. Sung, D.M. Huang, W.H. Chang, R.N. Huang, J.C. Hsu, Evaluation of gelatin hydrogel crosslinked with various crosslinking agents as bioadhesives: in vitro study, J. Biomed. Mater. Res. 46 (4) (1999) 520–530. [34] Y. Wang, X. Mo, X.S. Sun, D. Wang, Soy protein adhesion enhanced by glutaraldehyde crosslink, J. Appl. Polym. Sci. 104 (1) (2007) 130–136. [35] W. Teng, J. Cappello, X. Wu, Recombinant silk-elastinlike protein polymer displays elasticity comparable to elastin, Biomacromolecules 10 (11) (2009) 3028–3036. [36] X. Zhang, P. Hoobin, I. Burgar, M.D. Do, Chemical modification of wheat protein-based natural polymers: cross-linking effect on mechanical properties and phase structures, J. Agric. Food Chem. 54 (26) (2006) 9858–9865. [37] R. Spanneberg, F. Osswald, I. Kolesov, W. Anton, H.J. Radusch, M.A. Glomb, Model studies on chemical and textural modifications in gelatin films by reaction with glyoxal and glycolaldehyde, J. Agric. Food Chem. 58 (6) (2010) 3580–3585. [38] O. Suwantong, P. Pavasant, P. Supaphol, Electrospun zein fibrous membranes using glyoxal as cross-linking agent: preparation, characterization and potential for use in biomedical applications, Chiang Mai J. Sci. 38 (1) (2011) 56–70. [39] M.P. Balaguer, J.P. Cerisuelo, R. Gavara, P. Hernandez-Muñoz, Mass transport properties of gliadin films: effect of cross-linking degree, relative humidity, and temperature, J. Membr. Sci. 428 (2013) 380–392. [40] L. Zárate-Ramírez, A. Romero, I. Martínez, C. Bengoechea, P. Partal, A. Guerrero, Effect of aldehydes on thermomechanical properties of gluten-based bioplastics, Food Bioprod. Process. 92 (1) (2014) 20–29. [41] C.M. Vaz, P.F. van Doeveren, G. Yilmaz, L.A. De Graaf, R.L. Reis, A.M. Cunha, Processing and characterization of biodegradable soy plastics: effects of crosslinking with glyoxal and thermal treatment, J. Appl. Polym. Sci. 97 (2) (2005) 604–610. [42] C. Chang, M.T. Nickerson, Effect of plasticizer-type and genipin on the mechanical, optical, and water vapor barrier properties of canola protein isolate-based edible films, Eur. Food Res. Technol. 238 (1) (2014) 35–46. [43] M. Aider, C. Barbana, Canola proteins: composition, extraction, functional properties, bioactivity, applications as a food ingredient and allergenicity–a practical and critical review, Trends Food Sci. Technol. 22 (1) (2011) 21–39. [44] P. Van Wachem, R. Zeeman, P. Dijkstra, J. Feijen, M. Hendriks, M. Cahalan, et al., Characterization and biocompatibility of epoxy cross-linked dermal sheep collagens, J. Biomed. Mater. Res. 47 (2) (1999) 270–277. [45] K. Boulle, R. Glogau, T. Kono, M. Nathan, A. Tezel, J.X. Roca-Martinez, et al., A review of the metabolism of 1, 4-butanediol diglycidyl ether–crosslinked hyaluronic acid dermal fillers, Dermatol. Surgery 39 (12) (2013) 1758–1766. [46] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. [47] H. Cai, P. Li, G. Sui, Y. Yu, G. Li, X. Yang, et al., Curing kinetics study of epoxy resin/flexible amine toughness systems by dynamic and isothermal DSC, Thermochim. Acta 473 (1) (2008) 101–105. [48] N. Perisic, N.K. Afseth, R. Ofstad, A. Kohler, Monitoring protein structural changes and hydration in bovine meat tissue due to salt substitutes by Fourier transform infrared (FTIR) microspectroscopy, J. Agric. Food Chem. 59 (18) (2011) 10052–10061. [49] A. Barth, Infrared spectroscopy of proteins, Biochim. Biophys. Acta (BBA)-Bioenergetics 1767 (9) (2007) 1073–1101. [50] M. Dalgalarrondo, J.M. Robin, J.-L. Azanza, Subunit composition of the globulin fraction of rapeseed (Brassica napus L.), Plant Sci. 43 (2) (1986) 115–124. [51] S. Bérot, J. Compoint, C. Larré, C. Malabat, J. Guéguen, Large scale purification of rapeseed proteins (Brassica napus L.), J. Chromatogr. B 818 (1) (2005) 35–42. [52] C. Marquié, Chemical reactions in cottonseed protein cross-linking by formaldehyde, glutaraldehyde, and glyoxal for the formation of protein films with enhanced mechanical properties, J. Agric. Food Chem. 49 (10) (2001) 4676–4681. [53] S. Sun, Y. Song, Q. Zheng, Morphologies and properties of thermo-molded biodegradable plastics based on glycerol-plasticized wheat gluten, Food Hydrocolloids 21 (7) (2007) 1005–1013. [54] J. Wu, A. Muir, Comparative structural, emulsifying, and biological properties of 2 major canola proteins, cruciferin and napin, J. Food Sci. 73 (3) (2008) C210–C216. [55] A.A. Ogale, P. Cunningham, P. Dawson, J. Acton, Viscoelastic, thermal, and microstructural characterization of soy protein isolate films, J. Food Sci. 65 (4) (2000) 672–679. [56] A. Jerez, P. Partal, I. Martínez, C. Gallegos, A. Guerrero, Protein-based bioplastics: effect of thermo-mechanical processing, Rheol. Acta 46 (5) (2007) 711–720. [57] A. Mizuno, M. Mitsuiki, M. Motoki, Effect of transglutaminase treatment on the glass transition of soy protein, J. Agric. Food Chem. 48 (8) (2000) 3286–3291. [58] G.W. Selling, K.K. Woods, A. Biswas, Electrospun zein fibers using glyoxal as the crosslinking reagent, J. Appl. Polym. Sci. 123 (5) (2012) 2651–2661. [59] H. Khonakdar, J. Morshedian, U. Wagenknecht, S. Jafari, An investigation of chemical crosslinking effect on properties of high-density polyethylene, Polymer 44 (15) (2003) 4301–4309.
430
Contents lists available at ScienceDirect
European Polymer Journal journal homepage: www.elsevier.com/locate/europolj
Preparation and characterization of cross-linked canola protein isolate films
MARK
Shuzhao Lia, Elizabeth Donnera, Michael Thompsonb, Yachuan Zhangc, ⁎ Curtis Rempeld, Qiang Liua, a
Guelph Research and Development Centre, Agriculture and Agri-Food Canada, 93 Stone Road West, Guelph, ON N1G 5C9, Canada Department of Chemical Engineering, McMaster University, Hamilton, ON L8S 4L8, Canada c Lethbridge Research and Development Centre, Agriculture and Agri-Food Canada, Lethbridge, AB T1J 4B1, Canada d Canola Council of Canada, Winnipeg, MB R3B 0T6, Canada b
AR TI CLE I NF O
AB S T R A CT
Keywords: Canola protein Chemical modification Cross-linking reaction Mechanical property Curing effect
Cross-linked canola protein isolate (CPI) films were prepared using a bisepoxide as a chain extender by wet cast followed by heat compression. By DSC measurements, it was verified that the cross-linking reaction took place in the CPI matrix and was completed in 10 min at 90 °C. FTIR analysis confirmed that the chain extenders were covalently bonded with CPI chains. The reaction led to the formation of a cross-linked architecture which was revealed by an increase in molecular weight of the modified CPI films. The newly formed chain architecture increased the thermostability of CPI as evaluated by TGA; furthermore, the thermostability increased with the increase in cross-linking degree of the modified CPI films. DMA results of the unmodified CPI film showed that microcosmic phase separation occurred between two major proteins, cruciferin and napin in CPI; however, the extent of phase separation decreased with the increase of chain extenders and finally disappeared because of the increased compatibility between cruciferin and napin. Elastomeric mechanical behavior was observed with the introduction of cross-linked architecture in the modified CPI films. At low humidity of 55%, the cross-linked CPI films showed an increase in tensile strength and a decrease in elongation at break. At high humidity of 98%, the increase in hydrophobicity of cross-linked CPI matrix resisted the decreasing tensile strength seen in unmodified films.
1. Introduction With a growing interest in green products, the demand for protein-based industrial products is increasing worldwide in the fields of food, gels, adhesives and packaging [1–4]. Canola protein is obtained from the second largest oilseed crop after soy [5,6]; however, the main source of canola protein, canola meal, is not used in human food applications because of the presence of some antinutritional compounds, such as glucosinolates, erucic acid, phytates, and phenolics [7,8], making it an ideal candidate for industrial purposes lest its use be relegated to low value animal feed [9]. However, the poor processability, low mechanical properties, and high moisture sensitivity of protein-based biopolymers have limited their applications [10–12]. Accordingly, to increase their mechanical properties and processability, various efforts have been made, such as plasticizing with glycerol [13], propanediol [14–17], polyethylene glycol, or sorbitol [18]; modification of the protein with chemicals [19–21]; or mixing with hydrophobic materials [22–
⁎
Corresponding author. E-mail addresses: [email protected] (S. Li), [email protected] (E. Donner), [email protected] (M. Thompson), [email protected] (Y. Zhang), [email protected] (C. Rempel), [email protected] (Q. Liu). http://dx.doi.org/10.1016/j.eurpolymj.2017.03.001 Received 17 November 2016; Received in revised form 16 January 2017; Accepted 1 March 2017 Available online 02 March 2017 0014-3057/ Crown Copyright © 2017 Published by Elsevier Ltd. All rights reserved.
European Polymer Journal 89 (2017) 419–430
S. Li et al.
25]. The introduction of polar additives can increase mechanical properties of protein-based biopolymer to an extent [26]; on the other hand, this method can also lead to an increase in moisture sensitivity. Blending with hydrophobic materials usually generates immiscible/incompatible phase separation, which deteriorates the mechanical properties of the protein matrix. Chemical crosslinking modification can enhance the formation of covalent bonds in the protein matrix; thus, mechanical properties and barrier properties of the protein-based biopolymer can be further improved. Many cross-linkers, such as thiol [27], polyurethane and polyisocyanate [28–31], carbodiimide [32], genipin [33], or enzymes [21] can be used to modify the protein matrix. Of these chemicals, aldehydes including dialdehydes and monoaldehydes have been extensively used for cross-linking protein macromolecules, usually by the compression molding method [20,34–39]. Although the use of an aldehyde generally improved the mechanical properties and water-resistance properties of protein-based biopolymers, the optimal aldehyde type and cross-linking density should be experimentally determined. Zárate-Ramírez et al. [40] evaluated the mechanical properties of wheat gluten biopolymers with and without treatment using formaldehyde, glutaraldehyde, and glyoxal by compression molding. It was found that tensile strength of the aldehyde-modified samples was not improved compared to that of aldehyde-free samples. Vaz et al. [41] found that both the tensile modulus and strength of extruded soy protein isolate biopolymer decreased with the addition of glyoxal. They believed that the probability of glyoxal cross-linking via two amine groups of soy located on two adjacent chains was smaller than cross-linking along the same chain during extrusion. Accordingly, the covalent bonds were predominantly introduced within the polypeptide chains instead of between them. In fact, the compression molding method is usually conducted under high pressure and temperature in order to melt and compact the protein matrix [26], which is not easy to apply in industry. In canola, cruciferin and napin are the two major families of storage proteins, which constitute 60% and 20% of the total protein content in mature seeds, respectively [42]. Cruciferin is a 12S globulin with a high molecular weight (300–310 kDa) and several subunits. Napin is a 2S albumin with a low molecular weight (12.5–14.5 kDa) [43]. Therefore, canola protein is very suitable for the preparation of protein-based films by the casting method because of its water solubility and the well balanced constitution of components with different molecular weight, in which cruciferin can increase mechanical properties of the matrix and napin can act as a plasticizer to increase processability. However, few studies have been conducted on the chemical modification of CPI until now. The objectives of this research are to use canola protein as an ingredient to develop a novel protein-based bioplastic film and to understand the factors that influence the properties of this newly developed product. In this research, 1,4-butanediol diglycidyl ether (BDDE) was used as a chain extender to prepare cross-linked CPI film for the first time. Dermal sheep collagen cross-linked with BDDE was prepared and evaluated after subcutaneous implantation in rats [44]. The results showed cross-linking of dermal sheep collagen with BDDE resulted in biocompatible materials in terms of non-cytotoxicity and non-antigenicity. Additionally, clinical and biocompatibility data spanning more than 15 years support the favorable clinical safety profile of BDDE-crosslinked hyaluronic acid (HA) and its degradation products [45]. Therefore, the usage of BDDE as a chain extender would be anticipated to offer novel properties to CPI in food and industrial applications. Furthermore, the casting method was employed to denature canola protein and unfold the protein chains to an extent in solvent by heating to form a cross-linked protein chain network. The preparation method and the investigation of properties help to increase the knowledge of protein chemical modification and broaden the application fields of canola protein. 2. Materials and methods 2.1. Materials and extraction of CPI from canola meal Glycerol and 1,4-butanediol diglycidyl ether were purchased from Sigma Aldrich and used as received. The protein extraction procedure was based on the work reported by Manamperi et al. [6] with some minor modifications. Conventional dark-seeded (Brassica napus) defatted canola meal was provided by Bunge Canada (Altona, Manitoba). The canola meal was tested for composition which showed 36.5% protein, 11.7% crude fibre, 4.1% crude oil (d.b), 9.8% moisture, and 7.3% ash, in the labs of the Department of Food Science at the University of Manitoba, Canada. The meal was ground into powder using a mill (A10 analytical mill, Ika Works, Wilmington, NC) to pass through a 250 μm sieve and then suspended in distilled water at a ratio of 1/10 (g/ml). The pH of the suspension was adjusted to 12 with NaOH (4 N) with vigorous stirring at 55 °C for 3 h. The first centrifugation took place at 10,000 rpm for 20 min in a Sorvall LYNX 4000 centrifuge (Thermo Scientific, Germany) and the supernatant was collected. The pH of the obtained solution was adjusted to 5 by drop-wise addition of HCl solution (2 N) to allow the canola protein to precipitate. The second centrifugation was conducted and the sediment (protein) was collected and then washed three times with distilled water. The extracted proteins were then dialyzed for 12 h with distilled water with frequent distilled water changes. The protein was recovered after another centrifugation, freeze dried and stored at 5 °C until further use. 2.2. Preparation of cross-linked CPI samples Cross-linked CPI samples were prepared as follows, using the solvent casting method followed by heat compression. A protein suspension was prepared by mixing 32 g of protein powder with 1200 mL of distilled water. After adding 13.6 g of glycerol, the protein suspension was homogenized via a magnetic stirrer for 30 min. The suspension was heated at 65 °C for 30 min and then cooled down to ambient temperature to obtain a protein solution. The solution was then separated equally into four parts, to three of which 16 mg, 32 mg, and 80 mg of BDDE were added to prepare three cross-linked CPI samples. The solution without added BDDE was used as the control. All of the solutions were stirred for 1 h, and then degassed in a vacuum oven at ambient temperature for 420
European Polymer Journal 89 (2017) 419–430
S. Li et al.
45 min, cast onto a Teflon-coated plate with a diameter of 20 cm, and finally dried at ambient temperature for 72 h in a fume hood. The films peeled off of the plates with 0, 16, 32 and 80 mg BDDE were named CP+BDDE-0, CP+BDDE-1, CP+BDDE-2, and CP +BDDE-3, respectively. All of the dried CPI samples were conditioned at ambient conditions (23.8 °C, 50% RH) for 48 h and then heat compression cured with a Carver hydraulic hot press (model 3925) at 90 °C for 10 min under pressure of 1.5 MPa. The thickness of the prepared protein film varied between 0.25 mm and 0.30 mm, and the prepared films containing 0, 16, 32 and 80 mg BDDE after heat compression were denoted as unmodified CPI, CP-BDDE-1, CP-BDDE-2, and CP-BDDE-3. 2.3. Differential scanning calorimetry (DSC) DSC analysis was performed using a differential scanning calorimeter (Q20 DSC, TA Instruments, New Castle, DE, USA) equipped with a refrigerated cooling system. A high-volume pan was loaded with a film sample and then hermetically sealed. The temperature scan range was from 20 to 180 °C at a heating rate of 10 °C/min with a nitrogen purge. The instrument was calibrated with indium and an empty pan was used as a reference. 2.4. Fourier transform infrared spectroscopy (FTIR) Infrared spectra of unmodified and cross-linked CPI films were recorded with a Digilab FTS 7000 spectrometer (Digilab USA, Randolph, MA) equipped with a thermoelectrically cooled deuterated tri-glycine sulfate (DTGS) detector using an attenuated total reflectance accessory at a resolution of 4 cm−1 by 64 scans. A half-band width of 15 cm−1 and a resolution enhancement factor of 1.5 with Bessel apodization were employed. Spectra were subjected to ATR correction (Algorithm 1, correction 1.000), deconvolution (k factor 1.9, half width 24, Bessel apodization), and a multipoint linear baseline correction using Win-IR Pro software. The measured unmodified and cross-linked CPI films were dissolved in distilled water to prepare solutions with a concentration of 5 wt%. To remove the added glycerol and unreacted BDDE, the prepared CPI solutions were dialyzed using a Pur-A-Lyzer™ midi dialysis kit (Sigma Aldrich) with a pore size of 1000 Da molecular weight for 24 h with distilled water with frequent distilled water changes. The purified CPI solutions were dried in a vacuum at 60 °C for 72 h to obtain the samples for FTIR tests. 2.5. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) SDS-PAGE was performed according to the method of Laemmli [46]. Unmodified and cross-linked CPI samples were mixed with sample buffer in the presence of β-mercaptoethanol and SDS, heated for 10 min at 95 °C, and loaded onto a 4–10% polyacrylamide gel with SDS. Gels were run at a constant voltage (200 V) for 38 min in a Mini-Protean II electrophoresis cell (Bio-Rad, Hercules, CA, USA). The gels were developed with Bio-Safe™ Coomassie Stain. 2.6. High performance size exclusion chromatography (HPSEC) The high performance size exclusion chromatographic (HPSEC) system consisted of a Shimadzu SIL-20A HT autosampler, LC-10A T pump and DGU-14A degasser (Shimadzu Corp., Kyoto, Japan), and a Viscotek TDA 305 Triple Detector Array and UV detector 2600 (Malvern Instruments Ltd., Malvern, UK). Two AquaGel columns (PAA-M and PAA-203) were connected in series (PolyAnalytik, London, Canada). The mobile phase (100 mM sodium nitrate and 5 mM sodium azide) was filtered through a 0.22 μm filter. The flow rate was 0.5 mL/min and the columns and detectors were maintained at 40 °C. CPI solutions were prepared at a concentration of 2 mg/mL and 100 μl were injected. 2.7. Thermogravimetric analysis (TGA) The onset decomposition temperature of all CPI samples was determined by TGA using a SDT Q600 (TA Instruments, New Castle, DE, USA). The decomposition measurements were obtained between 50 and 600 °C at a scanning rate of 20 °C/min under a helium atmosphere. 2.8. Dynamic mechanical analysis (DMA) Dynamic mechanical properties of the biopolymer samples were measured using a dynamic mechanical analyzer (DMA) Q800 (TA Instruments, New Castle DE, USA) operated in a multi-frequency strain mode. The specimens were cut into a nominal dimension of 55 mm × 5 mm × 0.25 mm and were conditioned in a laboratory atmosphere (22 °C and 55% relative humidity) for 48 h prior to tensile testing. Each specimen was mounted on the dual cantilever beam clamp, and an amplitude of 250 μm was applied to the specimen at a frequency of 1 Hz. The storage modulus, loss modulus, and tan delta (tanδ ) were recorded as the samples were heated from −70 °C to 150 °C at a heating rate of 5 °C/min. Each sample had two replicates. 2.9. Mechanical properties Mechanical properties were measured using a Com-Ten testing system (model 701SN, FL, USA) with a 0.5 kN load cell according to the Standard Test Method for Tensile Properties of Plastics (ASTM D638). Type-IV tensile specimens were prepared by die cutting 421
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 1. Differential scanning calorimetric thermograms of (a) four CPI+BDDE films prepared by casting without heat compression, (b) CP-BDDE-3 films cured by heat compression at 3, 5, and 10 min at 90 °C.
and were conditioned in a laboratory atmosphere (22 °C and 55% relative humidity) and in a desiccator with saturated potassium sulfate aqueous solution (22 °C and 98% relative humidity) for 48 h prior to tensile testing. A crosshead speed of 5 mm/min was used. The tensile testing was conducted at ambient conditions (23.8 °C, 50% RH). The stress, strain, tensile strength, and elongation-atbreak were calculated from at least seven replicates. 3. Results and discussion 3.1. DSC To investigate the cross-linking reaction taking place in the CPI matrix, DSC thermograms for CP+BDDE-0, CP+BDDE-1, CP +BDDE-2, and CP+BDDE-3 (obtained by casting without heat compression) are shown in Fig. 1(a). A glass transition, Tg , was observed in the thermogram of CP+BDDE-0 film at 48.3 °C as shown in Fig. 1(a). A similar DSC pattern of CPI was reported by Manamperi et al. [6]. However, the Tg at 48.3 °C disappeared in the thermogram of CP+BDDE-1 because BDDE can act as a plasticizer in the matrix before the cross-linking reaction, and a new broad peak was observed at Tp = 115.8 °C, which was an exothermic transition corresponding to the reaction heat of cross-linking reaction [47]. Data obtained from the DSC thermograms, such as the peak temperature (Tp ), the onset temperature (To ), the terminal temperature (Tend ), the curing range, and the cumulative reaction heat 422
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Table 1 Curing characteristics of the cross-linking reaction systems. Sample
To (°C)
Tp (°C)
Tend (°C)
Curing range (°C)
ΔHr (J/g)
CP+BDDE-1 CP+BDDE-2 CP+BDDE-3
86.3 87.7 82.2
115.8 118.7 115.2
153.6 153.8 155.4
67.3 66.1 73.2
4.74 6.78 9.68
(ΔHr ) are listed in Table 1. It was found that the reaction heat increased with the increasing amount of BDDE, indicating that more cross-linking reactions took place in the system with more BDDE. On the other hand, the cross-linking reaction in CP+BDDE-1, CP +BDDE-2, and CP+BDDE-3 had similar To and started before 90 °C, pointing to the possibility of modifying the CPI films with competition by thermal degradation. To determine the kinetics of the present reaction system, CP+BDDE-3 films were cured at 90 °C for 3 min, 5 min, and 10 min and then analyzed by DSC. The resulting thermograms are shown in Fig. 1(b). It was found that the cumulative reaction heat decreased with increase of the reaction time until no reaction heat was detected for the films cured for 10 min. Reproducibly it was found that the cross-linking reaction could be finished completely in 10 min at 90 °C. Consequently, to decrease the effect of degradation on the film properties, the reaction time of 10 min was chosen as suitable for the present reaction systems. 3.2. FTIR The FTIR spectra of unmodified and cross-linked CPI films are shown in Fig. 2(a) and (b). In the FTIR spectrum of unmodified CPI film shown in Fig. 2(a), the absorption peaks between 3500 and 3000 cm−1 were ascribed to the water region, which includes the stretching vibration of NeH and OeH [48]. The peak at 3282 cm−1 in the spectrum of unmodified CPI was assigned to amide A band,
Fig. 2. FTIR spectra at (a) 3700–2700 cm−1 and (b) 1210–1060 cm−1 of unmodified and cross-linked CPI films prepared by solvent casting and heat compression. The unreacted BDDE and added glycerol were removed from the CPI matrix in preparation for the analysis.
423
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 3. The preparation of cross-linked CPI sample.
which is exclusively located on the NH group and for that reason is not sensitive to the conformation of the polypeptide backbone in protein [49]. The peaks at 2851 and 2919 cm−1 in the spectrum of unmodified CPI were assigned to symmetric and asymmetric stretching vibrations of eCH2e groups on proteins. The peak at 3282 cm−1 on the spectra of cross-linked CPI became narrow and a new peak appeared at 3180 cm−1 which is assigned to OeH stretching vibration [48], indicating that new hydrogen groups were generated by the ring-opening reaction between amide groups on protein chains and epoxy groups on BDDE as shown in Fig. 3. In addition, the intensity of peaks at 2851 and 2919 cm−1, ascribed to stretching vibrations of eCH2e, significantly increased with the introduction of BDDE corresponding to its ethyl groups. The unreacted BDDE and added glycerol had been removed from the CPI matrix in preparation for the analysis; therefore, the increase of peak intensity at 2851 and 2919 cm−1 was attributed to the introduction of more BDDE into CPI chains by covalent bonding. In addition, in Fig. 2(b) new peaks assigned to the CeOeC band for BDDE appeared at 1108 cm−1 on the spectra of cross-linked CPI films, although the peaks were weak. Therefore, by the FTIR results it could be confirmed that BDDE reacted with CPI chains to form covalent bonds. 3.3. SDS-PAGE Fig. 4 shows the SDS-PAGE of unmodified and three cross-linked CPI solutions. Although other bands below 250 kDa were not separated in Fig. 4, unmodified CPI (lane 5) showed one band above 250 kDa, which was assigned to cruciferin protein [50]. CPIBDDE-1 (lane 2) showed an additional band with a higher molecular weight than that of unmodified CPI that was attributed to the introduction of cross-linked chains. With increasing amounts of BDDE in the reaction, the additional band was much clearer for CPIBDDE-2 (lane 3) and CPI-BDDE-3 (lane 4), indicating more cross-linked protein chains with similar molecular weight to that of CPBDDE-1 were generated in these two cross-linked CPI films. Furthermore, a third band above the two bands that appeared in CPBDDE-1 could be observed for CP-BDDE-2 and CP-BDDE-3, which suggested that cross-linked protein chains with an even higher molecular weight than in CP-BDDE-1 were being identified. Although the SDS-PAGE could not distinguish the fractions below 250 kDa, the results could qualitatively illuminate the changes in the architecture because of the cross-linking reaction. 3.4. SEC CPI consists of two major families, napin (12.5–14.5 kDa) and cruciferin (300–310 kDa), and other minor protein fractions
Fig. 4. SDS-PAGE of unmodified and cross-linked CPI film solutions. Lanes 1 and 6, molecular weight marker; lane 2, CPI-BDDE-1; lane 3, CPI-BDDE-2; lane 4, CPIBDDE-3; lane 5, unmodified CPI.
424
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 5. Refractive index chromatograms of unmodified and cross-linked CPI films by SEC.
including oleosins, thionins, and lipid transfer protein (LTP). The composition of unmodified CPI as determined by SEC (Fig. 5) revealed five peaks. The two major canola proteins were well separated; peak 2 was identified as cruciferin and comprised 71.7% of the total peak area while peak 4 was identified as napin and comprised 16.9% of the total. Peak 3 was not pure (mixture of cruciferin and napin) [51] and comprised 7.7% of the total. The number-average molecular weight (Mn) of peak 1 (2.4% of the total) was approximately 800 kDa, which contained aggregates of cruciferin (peak 2, Mw 300 kDa). Peak 5 (8.4 kDa) probably consisted of other minor proteins like LTP or some small polypeptides. The SEC refractive index chromatogram of CP-BDDE-1 showed that the cross-linking reaction led to a remarkable decrease in CPI components corresponding to peak 2 (57.2% of the total) and peak 4 (10.5% of the total), while peak 1 representing aggregates increased to 26.3%, indicating the formation of cross-linked protein chains. The addition of BDDE resulted in a further decrease in peak 2 (38.7% of the total) and peak 4 (5.8% of the total) and increases in peak 1 (52.2% of the total) except for CP-BDDE-3. The decrease of peak 2 and 4 with the addition of BDDE indicated that BDDE had broad reaction specificity and could react with the two major protein fractions in this CPI matrix. In fact, approximately 5.6% of insoluble solid was found in CP-BDDE-3, which indicated some protein chains were further cross-linked and formed insoluble structures. Consequently, the intensity of peak 1 of CP-BDDE-3 in Fig. 5 decreased, attributed to the removal of the insoluble crosslinked proteins during the SEC measurement. Zárate-Ramírez et al. [41] used glyoxal to react with soy protein isolate and found the covalent bonds were predominantly introduced within the polypeptide chains instead of between them (intramolecular crosslinking). However, from the SEC results it could be confirmed that the cross-linking reaction took place between two adjacent CPI chains and a cross-linked architecture was formed in the present reaction system, which increased the weight-average molecular weight from 337 kDa for unmodified CPI to 512 kDa for CP-BDDE-1, 700 kDa for CP-BDDE-2, and 820 kDa for CP-BDDE-3.
Fig. 6. TGA and DTG curves of unmodified and cross-linked CPI films.
425
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Table 2 DMA and TGA data of unmodified and cross-linked CPI. Sample
Unmodified CPI CP-BDDE-1 CP-BDDE-2 CP-BDDE-3
α relaxation
β relaxation
Breakage temperature in TGA curve (°C)
Peak temperature (°C)
tanδ at peak temperature
Peak temperature (°C)
tanδ at peak temperature
42.0 32.6 22.8 17.6
0.44 0.46 0.44 0.42
−50.9 −50.5 −48.0 −47.8
0.24 0.24 0.25 0.25
281.7 294.3 304.5 309.2
3.5. TG analysis The effect of cross-linking architecture on thermostability of CPI films was investigated by TGA. Fig. 6 shows the DTG curve of unmodified CPI film and thermogravimetric decomposition curves of unmodified and cross-linked CPI films. There were four weight loss segments on the DTG curve of unmodified CPI as shown in Fig. 6. The first weight loss of 7.8% below ∼120 °C was attributed to the elimination of free moisture in the protein matrix. The weight loss of approximately 33% appearing between 120 and 280 °C was mainly attributed to evaporation of glycerol and any bonded water with the polar groups on protein molecules; the loss corresponded to the amount of added glycerol as plasticizer in unmodified CPI film. The scission of covalent peptide bonds occurred between 280 and 500 °C, which continued over a long period of decomposition because of the presence of various protein fractions with a broad molecular weight distribution. The last weight loss after 500 °C corresponded to the cleavage of SeS, OeN, and OeO linkages in protein macromolecules, which resulted in the decomposition of protein molecules [52]. There was no remarkable difference between unmodified and cross-linked CPI films in the weight loss before 280 °C corresponding to the loss of water and glycerol; however, the scission temperature of cross-linked CPI molecules increased with the introduction of cross-linked architecture, as summarized in Table 2. Furthermore, the greater the amount of BDDE, the higher the onset of scission. Consequently, the results indicated the introduction of cross-linked architecture increased the thermostability of CPI films. 3.6. Dynamic mechanical analysis Fig. 7 shows the storage modulus E′ and loss factor tanδ as a function of temperature for unmodified and cross-linked CPI films. With an increase of temperature, unmodified CPI clearly showed the transition region (tanδ peak and modulus drop) and the rubbery plateau in sequence. A decrease of three orders of magnitude in E′ was observed at the transition from glass to rubber region. The broad rubbery plateau located at the temperature interval from 85 to 140 °C resembles reported results for intermolecular covalent crosslinking through disulfide bonding among proteins [53]. CP-BDDE-1 showed a slight increase in E′ at the rubbery plateau attributed to the formation of some cross-linked chains. With the increase of BDDE, the E′ at the plateau increased for CP-BDDE-2, indicating that more cross-linking reaction took place and thus the cross-linking density increased. However, CP-BDDE-3 exhibited a thermosetting potential region in Fig. 7(a) because the addition of more BDDE gave rise to the presence of cross-linked architecture, which is in agreement with the presence of some insoluble solid resulting from cross-linking reaction in the results of molecular weight analysis. In Fig. 7(b), two obvious mechanical loss peaks could be observed, one broad peak located between 40 and 78 °C and the other one at −50 °C, in the tanδ curve of unmodified CPI. The peak appearing at higher temperature was associated with the glass transition temperature (Tg ) or the α-transition of plasticized unmodified CPI (as shown in Table 2). This is in accordance with the DSC result of unmodified CPI film, in which a Tg transition at 48.3 °C was observed. It may be noted that unmodified CPI film showed a broad Tg peak, which indicated the presence of phase separation or simply reflected the high heterogeneity of two major protein fractions in molecular size, cruciferin and napin with a molecular weight ranging from 9 to 350 kDa [54]. The low temperature transition was probably associated with Tg of a glycerol-rich phase, which was higher than Tg of glycerol at −93 °C [55]. The appearance of the low temperature transition was ascribed to the partial miscibility of glycerol and CPI chains [56]. It is evident in Fig. 7(b) that the broad Tg peaks of all of cross-linked CPI were becoming narrower with the increase of BDDE. The change is more apparent in CP-BDDE-3 which exhibited a very narrow peak at Tg . BDDE had broad reaction specificity and could react with any fraction in unmodified CPI as discussed in the results of molecular weight distribution; therefore, cross-linking reaction may take place between cruciferin and napin in unmodified CPI matrix so that the compatibility of the two protein fractions was increased in cross-linked CPI. Consequently, the cross-linked protein chains formed the matrix network in CP-BDDE-3 and showed a narrower Tg . The Tg of cross-linked CPI was much lower than that of unmodified protein, which was proposed to be related to the improved hydrophobicity because of decreased polar amine groups and the increase in compatibility of the two main protein fractions [53]. In addition, in the present case, the introduction of BDDE in cross-linked proteins may have increased the mobility of the chain fraction because BDDE is an ether and has a very soft chain, which could have decreased the Tg of the cross-linked CPI. More BDDE was introduced into CP-BDDE-3 and thus a lower Tg was seen in Table 2. Mizuno, Mitsuiki, and Motoki [57] treated 7S and 11S fractions isolated from soy protein using microbial transglutaminase (MTG) and found that cross-linking after MTG treatment rendered Tg lower than that of non-treated soy protein at the same water content. They argued that the MTG treatment caused an increase in immobilized water that contributed to the decrease of Tg . 426
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 7. Temperature dependence of storage modulus E′ (a) and loss tangent tanδ (b) for unmodified and cross-linked CPI films.
3.7. Mechanical properties Fig. 8 shows the strain-stress relationship of unmodified and cross-linked CPI films. As shown in the figure, the deformation nature of the unmodified CPI film at room temperature is typical of an elastomer in terms of the stress and strain. Generally, there were two
Fig. 8. Stress-strain curves of unmodified and cross-linked CPI films at relative humidity 55%.
427
European Polymer Journal 89 (2017) 419–430
S. Li et al.
Fig. 9. Tensile strength (a) and elongation-at-break (b) of unmodified and cross-linked CPI films at conditioning humidities of 55% and 98%.
characteristic regions of deformation behavior in the tensile stress-strain curve of unmodified CPI film. At low strains (< 5%) the stress increased rapidly with increasing strain, in which the initial slope corresponds to the elastic moduli of the unmodified CPI film. A yield point appeared at the strain of 5%; beyond the yield point, the unmodified CPI film showed a plastic deformation, i.e., the stress slowly increases with strain (strain hardening) until failure occurred. Furthermore, “necking” phenomenon could be observed during the test, which is a typical thermoplastic behavior. For the cross-linked CPI samples, the macromolecules moved and rearranged during plastic deformation; however, entanglement of cross-linked chains restrained the further movement of chains and thus enhanced the strain hardening until it broke finally at the maximum strain. Therefore, the strain-stress relationship of modified CPI was suggestive of branched linear chains. Effects of cross-linked architecture on tensile strength and elongation at break are shown in Fig. 9. The addition of glycerol as a plasticizer to the unmodified CPI film reduced the interactions between protein chains and even disrupted intermolecular linkages of protein [26], which increased the mobility of protein chains and thus resulted in an increase in elongation-at-break and a decrease in tensile strength at 55% humidity. The entanglement of cross-linked chains limited the large deformation of cross-linked CPI films and led to a decrease in elongation-at-break while the tensile strength increased at 55% humidity, as shown in Fig. 9. Furthermore, with the increase in the amount of BDDE, elongation at break decreased and tensile strength increased further. This is consistent with the general findings using aldehydes as cross-linkers to react with various protein chains [19,36,37,58]. At high humidity of 98%, the tensile strength of unmodified CPI film sharply decreased compared to that at a humidity of 55% because of the poor water resistance property. Although the tensile strength of cross-linked CPI films also decreased at the humidity of 98% in comparison with that at 55%, the decrease was reduced with increasing BDDE. Cross-linking in the modified CPI films led to a decrease in the concentration of hydrophilic amino acid residues and thus improved the hydrophobic interaction of protein molecules. Therefore, the water resistance property of cross-linked CPI films was increased, which potentially diminished the effect of water on tensile strength. The elongation-at-break of unmodified CPI film at 98% humidity increased compared to that at 55% because absorbed water acted as a plasticizer to further increase the flexibility of protein chains. However, the elongation-at-break of cross-linked CPI films at the humidity of 98% showed a slight effect compared to that at 55% because cross-linking decreased the length of segments available for stretching and restricted the slippage of chains [59]. As a result, the elongation in cross-linked CPI was no longer determined by the slippage of molecules; instead, it was a function of the stretch of network-like molecular structure. Therefore, the effect of the absorbed water on the elongation of cross-linked CPI was weakened.
428
European Polymer Journal 89 (2017) 419–430
S. Li et al.
4. Conclusions In this research, the cast method was employed to denature canola protein and unfold the protein chains to an extent in water to form a protein chain network. The unfolded protein chains could increase the possibility of a cross-linking reaction taking place between two adjacent protein chains instead of intra-molecularly. The cross-linking reaction was followed by DSC and FTIR analysis. The increase in molecular weight verified that large protein molecules were formed in the cross-linked CPI films by the reaction. Furthermore, because of the broad reaction specificity of BDDE toward different types of proteins, cross-linking took place between cruciferin and napin in the CPI matrix so that the compatibility of these two major protein fractions was increased in the cross-linked CPI. The cross-linked CPI films showed higher thermostability compared with unmodified CPI films because of the introduction of cross-linked chains. In addition, the tensile strength of cross-linked CPI films was increased at low humidity. At high humidity, the increase of hydrophobicity of cross-linked CPI films decreased the absorption of water and thus restrained the deterioration of mechanical properties. Acknowledgements We thank the Canola Council of Canada, Saskatchewan Pulse Growers and Agriculture and Agri-Food Canada through Growing Forward 2 Canola/Flax and Pulse Science Cluster Programs for financial funding. References [1] S. Ou, Y. Wang, S. Tang, C. Huang, M.G. Jackson, Role of ferulic acid in preparing edible films from soy protein isolate, J. Food Eng. 70 (2) (2005) 205–210. [2] J. Burgain, R. El Zein, J. Scher, J. Petit, E.-A. Norwood, G. Francius, et al., Local modifications of whey protein isolate powder surface during high temperature storage, J. Food Eng. 178 (2016) 39–46. [3] B.E. Dybowska, Whey protein-stabilized emulsion properties in relation to thermal modification of the continuous phase, J. Food Eng. 104 (1) (2011) 81–88. [4] M. da Rocha, M.R. Loiko, G.V. Gautério, E.C. Tondo, C. Prentice, Influence of heating, protein and glycerol concentrations of film-forming solution on the film properties of Argentine anchovy (Engraulis anchoita) protein isolate, J. Food Eng. 116 (3) (2013) 666–673. [5] M.R.G. Tandang-Silvas, T. Fukuda, C. Fukuda, K. Prak, C. Cabanos, A. Kimura, et al., Conservation and divergence on plant seed 11S globulins based on crystal structures, Biochim. Biophys. Acta (BBA)-Proteins Proteom. 1804 (7) (2010) 1432–1442. [6] W.A. Manamperi, S.K. Chang, C.A. Ulven, S.W. Pryor, Plastics from an improved canola protein isolate: preparation and properties, J. Am. Oil Chemists' Soc. 87 (8) (2010) 909–915. [7] S.M. McCurdy, Effects of processing on the functional properties of canola/rapeseed protein, J. Am. Oil Chemists’ Soc. 67 (5) (1990) 281–284. [8] W.A. Manamperi, D.P. Wiesenborn, S.K. Chang, S.W. Pryor, Effects of protein separation conditions on the functional and thermal properties of canola protein isolates, J. Food Sci. 76 (3) (2011) E266–E273. [9] S.H. Tan, R.J. Mailer, C.L. Blanchard, S.O. Agboola, Canola proteins for human consumption: extraction, profile, and functional properties, J. Food Sci. 76 (1) (2011) R16–R28. [10] A.H. Tkaczyk, J.U. Otaigbe, K.-L.G. Ho, Bioabsorbable soy protein plastic composites: effect of polyphosphate fillers on biodegradability, J. Polym. Environ. 9 (1) (2001) 19–23. [11] P. Lodha, A.N. Netravali, Effect of soy protein isolate resin modifications on their biodegradation in a compost medium, Polym. Degrad. Stab. 87 (3) (2005) 465–477. [12] M. Pommet, A. Redl, M.-H. Morel, S. Guilbert, Study of wheat gluten plasticization with fatty acids, Polymer 44 (1) (2003) 115–122. [13] R. Lee, M. Pranata, Z. Ustunol, E. Almenar, Influence of glycerol and water activity on the properties of compressed egg white-based bioplastics, J. Food Eng. 118 (1) (2013) 132–140. [14] R. Kumar, V. Choudhary, S. Mishra, I. Varma, B. Mattiason, Adhesives and plastics based on soy protein products, Indust. Crops Prod. 16 (3) (2002) 155–172. [15] P. Lodha, A.N. Netravali, Thermal and mechanical properties of environment-friendly ‘green’plastics from stearic acid modified-soy protein isolate, Indust. Crops Prod. 21 (1) (2005) 49–64. [16] P. Chen, L. Zhang, New evidences of glass transitions and microstructures of soy protein plasticized with glycerol, Macromol. Biosci. 5 (3) (2005) 237–245. [17] P. Chen, L. Zhang, F. Cao, Effects of moisture on glass transition and microstructure of glycerol-plasticized soy protein, Macromol. Biosci. 5 (9) (2005) 872–880. [18] N.-B. Song, W.-S. Jo, H.-Y. Song, K.-S. Chung, M. Won, K.B. Song, Effects of plasticizers and nano-clay content on the physical properties of chicken feather protein composite films, Food Hydrocolloids 31 (2) (2013) 340–345. [19] A.J. Poole, J.S. Church, The effects of physical and chemical treatments on Na 2 S produced feather keratin films, Int. J. Biol. Macromol. 73 (2015) 99–108. [20] G.W. Selling, K.K. Woods, A. Biswas, J. Willett, Reactive extrusion of zein with glyoxal, J. Appl. Polym. Sci. 113 (3) (2009) 1828–1835. [21] K. Patzsch, K. Riedel, M. Pietzsch, Parameter optimization of protein film production using microbial transglutaminase, Biomacromolecules 11 (4) (2010) 896–903. [22] X. Mo, X. Sun, Thermal and mechanical properties of plastics molded from sodium dodecyl sulfate-modified soy protein isolates, J. Polym. Environ. 8 (4) (2000) 161–166. [23] Z. Zhong, X.S. Sun, Properties of soy protein isolate/polycaprolactone blends compatibilized by methylene diphenyl diisocyanate, Polymer 42 (16) (2001) 6961–6969. [24] J. John, M. Bhattacharya, Properties of reactively blended soy protein and modified polyesters, Polym. Int. 48 (11) (1999) 1165–1172. [25] M. Wei, L. Fan, J. Huang, Y. Chen, Role of star-like hydroxylpropyl lignin in soy-protein plastics, Macromol. Mater. Eng. 291 (5) (2006) 524–530. [26] P. Guerrero, A. Retegi, N. Gabilondo, K. De la Caba, Mechanical and thermal properties of soy protein films processed by casting and compression, J. Food Eng. 100 (1) (2010) 145–151. [27] D.L. Woerdeman, W.S. Veraverbeke, R.S. Parnas, D. Johnson, J.A. Delcour, I. Verpoest, et al., Designing new materials from wheat protein, Biomacromolecules 5 (4) (2004) 1262–1269. [28] Y. Zhang, W. Zhu, Y. Lu, Z. Gao, J. Gu, Nano-scale blocking mechanism of MMT and its effects on the properties of polyisocyanate-modified soybean protein adhesive, Indust. Crops Prod. 57 (2014) 35–42. [29] Y. Chen, L. Zhang, R. Deng, Y. Cui, A new network composite material based on soy dreg modified with polyurethane prepolymer, Macromol. Mater. Eng. 292 (4) (2007) 484–494. [30] H. Tian, Y. Wang, L. Zhang, C. Quan, X. Zhang, Improved flexibility and water resistance of soy protein thermoplastics containing waterborne polyurethane, Indust. Crops Prod. 32 (1) (2010) 13–20. [31] D.J. Sessa, G.W. Selling, A. Biswas, Reaction of zein with methylenediphenyl diisocyanate in the melt state: Thermal, mechanical, and physical properties, Indust. Eng. Chem. Res. 51 (27) (2012) 9199–9203. [32] V. Tropini, J.-P. Lens, W. Mulder, F. Silvestre, Wheat gluten films cross-linked with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide and N-hydroxysuccinimide, Indust. Crops Prod. 20 (3) (2004) 281–289.
429
European Polymer Journal 89 (2017) 419–430
S. Li et al.
[33] H.W. Sung, D.M. Huang, W.H. Chang, R.N. Huang, J.C. Hsu, Evaluation of gelatin hydrogel crosslinked with various crosslinking agents as bioadhesives: in vitro study, J. Biomed. Mater. Res. 46 (4) (1999) 520–530. [34] Y. Wang, X. Mo, X.S. Sun, D. Wang, Soy protein adhesion enhanced by glutaraldehyde crosslink, J. Appl. Polym. Sci. 104 (1) (2007) 130–136. [35] W. Teng, J. Cappello, X. Wu, Recombinant silk-elastinlike protein polymer displays elasticity comparable to elastin, Biomacromolecules 10 (11) (2009) 3028–3036. [36] X. Zhang, P. Hoobin, I. Burgar, M.D. Do, Chemical modification of wheat protein-based natural polymers: cross-linking effect on mechanical properties and phase structures, J. Agric. Food Chem. 54 (26) (2006) 9858–9865. [37] R. Spanneberg, F. Osswald, I. Kolesov, W. Anton, H.J. Radusch, M.A. Glomb, Model studies on chemical and textural modifications in gelatin films by reaction with glyoxal and glycolaldehyde, J. Agric. Food Chem. 58 (6) (2010) 3580–3585. [38] O. Suwantong, P. Pavasant, P. Supaphol, Electrospun zein fibrous membranes using glyoxal as cross-linking agent: preparation, characterization and potential for use in biomedical applications, Chiang Mai J. Sci. 38 (1) (2011) 56–70. [39] M.P. Balaguer, J.P. Cerisuelo, R. Gavara, P. Hernandez-Muñoz, Mass transport properties of gliadin films: effect of cross-linking degree, relative humidity, and temperature, J. Membr. Sci. 428 (2013) 380–392. [40] L. Zárate-Ramírez, A. Romero, I. Martínez, C. Bengoechea, P. Partal, A. Guerrero, Effect of aldehydes on thermomechanical properties of gluten-based bioplastics, Food Bioprod. Process. 92 (1) (2014) 20–29. [41] C.M. Vaz, P.F. van Doeveren, G. Yilmaz, L.A. De Graaf, R.L. Reis, A.M. Cunha, Processing and characterization of biodegradable soy plastics: effects of crosslinking with glyoxal and thermal treatment, J. Appl. Polym. Sci. 97 (2) (2005) 604–610. [42] C. Chang, M.T. Nickerson, Effect of plasticizer-type and genipin on the mechanical, optical, and water vapor barrier properties of canola protein isolate-based edible films, Eur. Food Res. Technol. 238 (1) (2014) 35–46. [43] M. Aider, C. Barbana, Canola proteins: composition, extraction, functional properties, bioactivity, applications as a food ingredient and allergenicity–a practical and critical review, Trends Food Sci. Technol. 22 (1) (2011) 21–39. [44] P. Van Wachem, R. Zeeman, P. Dijkstra, J. Feijen, M. Hendriks, M. Cahalan, et al., Characterization and biocompatibility of epoxy cross-linked dermal sheep collagens, J. Biomed. Mater. Res. 47 (2) (1999) 270–277. [45] K. Boulle, R. Glogau, T. Kono, M. Nathan, A. Tezel, J.X. Roca-Martinez, et al., A review of the metabolism of 1, 4-butanediol diglycidyl ether–crosslinked hyaluronic acid dermal fillers, Dermatol. Surgery 39 (12) (2013) 1758–1766. [46] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. [47] H. Cai, P. Li, G. Sui, Y. Yu, G. Li, X. Yang, et al., Curing kinetics study of epoxy resin/flexible amine toughness systems by dynamic and isothermal DSC, Thermochim. Acta 473 (1) (2008) 101–105. [48] N. Perisic, N.K. Afseth, R. Ofstad, A. Kohler, Monitoring protein structural changes and hydration in bovine meat tissue due to salt substitutes by Fourier transform infrared (FTIR) microspectroscopy, J. Agric. Food Chem. 59 (18) (2011) 10052–10061. [49] A. Barth, Infrared spectroscopy of proteins, Biochim. Biophys. Acta (BBA)-Bioenergetics 1767 (9) (2007) 1073–1101. [50] M. Dalgalarrondo, J.M. Robin, J.-L. Azanza, Subunit composition of the globulin fraction of rapeseed (Brassica napus L.), Plant Sci. 43 (2) (1986) 115–124. [51] S. Bérot, J. Compoint, C. Larré, C. Malabat, J. Guéguen, Large scale purification of rapeseed proteins (Brassica napus L.), J. Chromatogr. B 818 (1) (2005) 35–42. [52] C. Marquié, Chemical reactions in cottonseed protein cross-linking by formaldehyde, glutaraldehyde, and glyoxal for the formation of protein films with enhanced mechanical properties, J. Agric. Food Chem. 49 (10) (2001) 4676–4681. [53] S. Sun, Y. Song, Q. Zheng, Morphologies and properties of thermo-molded biodegradable plastics based on glycerol-plasticized wheat gluten, Food Hydrocolloids 21 (7) (2007) 1005–1013. [54] J. Wu, A. Muir, Comparative structural, emulsifying, and biological properties of 2 major canola proteins, cruciferin and napin, J. Food Sci. 73 (3) (2008) C210–C216. [55] A.A. Ogale, P. Cunningham, P. Dawson, J. Acton, Viscoelastic, thermal, and microstructural characterization of soy protein isolate films, J. Food Sci. 65 (4) (2000) 672–679. [56] A. Jerez, P. Partal, I. Martínez, C. Gallegos, A. Guerrero, Protein-based bioplastics: effect of thermo-mechanical processing, Rheol. Acta 46 (5) (2007) 711–720. [57] A. Mizuno, M. Mitsuiki, M. Motoki, Effect of transglutaminase treatment on the glass transition of soy protein, J. Agric. Food Chem. 48 (8) (2000) 3286–3291. [58] G.W. Selling, K.K. Woods, A. Biswas, Electrospun zein fibers using glyoxal as the crosslinking reagent, J. Appl. Polym. Sci. 123 (5) (2012) 2651–2661. [59] H. Khonakdar, J. Morshedian, U. Wagenknecht, S. Jafari, An investigation of chemical crosslinking effect on properties of high-density polyethylene, Polymer 44 (15) (2003) 4301–4309.
430