Preparation and Characterization of Biodegradable Composited Films Based on Potato Starch/Glycerol/Gelatin

The use of plastics is resisted worldwide. Therefore, finding alternatives to plastic packaging products is an urgent issue. This work was dedicated to the preparation of biodegradable composited films with potato starch, glycerol, and gelatin. The formulation of the biodegradable film was first optimized via response surface methodology combined with the multi-index comprehensive evaluation method that considered physical properties (thickness, water solution (WS), tensile strength (TS) and elongation at break (E%)) and barrier property (light transmittance (T%)). Results indicated that the optimal conditions were 2.5% starch, 2.0% glycerol, and 1.5% gelatin (based on water). The optimized film presented excellent properties with TS of 4.47 MPa, E% of 109.91%, WS of 43.64%, and T% of 41.21% at 500 nm, and the comprehensive evaluation score of the composite film was 28.68. Moreover, a model verification experiment was further conducted, which proved that the predicted value highly matched experimental values, indicting the credibility and accuracy of the model. The resulting films were further characterized on the basis of rheological measurements, Fourier transform infrared spectroscopy (FTIR), and scanning electron microscopy (SEM). The rheological measurements proved that the film-forming solution exhibited low shear viscosity and non-Newtonian fluid behavior. FTIR and SEM revealed excellent compatibility among starch, glycerol, and gelatin. Hence, the resulting optimized film may be expected to provide theoretical basis and technical support for the food packing industry.


Introduction
In the food industry, food packaging is required to enhance food protection and guarantee shelf life on the basis of their appropriate physicochemical features [1,2]. Petroleum-based plastic packaging is still widely used due to its cost-effectiveness. However, with the worsening of environmental issues caused by nonrenewable and nondegradable plastic packaging and the serious harm to human health due to the accumulation of plastic waste [3,4], some prohibitions about petroleum-based plastic packaging have been enforced. Internationally, the European Chemicals Agency has limited the use of conventional synthetic plastic polymers for food packaging [5]. In the domestic setting, plastic waste materials have been prohibited from importation due to their toxicity [6,7]. erefore, to overcome these issues, replacing traditional plastic packaging with renewable, biodegradable materials has attracted considerable attention in recent decades. Generally, biodegradable packaging films were produced by natural biodegradable materials included polysaccharides, proteins, lipids, and their complexes, which make up for the drawbacks of plastic and solve those root environmental issues caused by plastic waste. Among these natural biodegradable films, starchbased films have incomparable advantages included abundant, low-cost, high film transparency, over other food packaging materials, which has been proved that starch-based film has exceed 60% of the total biodegradable film [8].
Starch is a highly annually renewable and high-molecular-weight polymer material that is composed of two types of α-glucan, namely amylose and amylopectin [9,10]. Amylose is a linear polysaccharide that is connected by α-1,4-glycosidic bonds, and that presents excellent filmforming property similar to fibers [11]. Moreover, the α-1,4glycosidic bonds and α-1,6-glycosidic bonds present in amylopectin are conducive to the thickening effect and freeze-thaw stability [12]. e proportion of amylose and amylopectin in various sources of starch is also different and directly affects film-forming properties [13,14]. erefore, increasing efforts have been devoted to the research of various resources of starch-based films [9,15,16]. e maize starch matrix film was produced and it was found that the films possessed low tensile strength (TS) and water vapor permeability (WVP), which could be enhanced by zein blended, and great compatibility with maize starch matrix [10]. Chollakup et al. [17] investigated cassava starch film, who proved the hydrophobic interactions and hydrogen bonding present in cassava starch and whey protein blended films to explain their barrier mechanisms at the molecular level. Lin et al. [18] utilized cellulose nanocrystals and PVA (polyvinyl alcohol)/glycerol to modify the formability of oxidized starch films with excellent WVP, surface free energy, and oxygen resistance. e characteristic properties of starch-PLA (polylactic acid) composite films [19] significantly improved with the use of cellulosic strengthening agents (cellulosic fibers and cellulose nanocrystals) and coffee husk extracts to enhance tensile properties, reduced barrier permeability, and endowed additional antioxidant capacity to food packaging systems. ese results indicated that the research on starch-based film from various resources had been continued, and the contribution of physicochemical and mechanical properties by a single starch is limited and needed to blend with other components to acquire desirable performances for improving product texture and resistance to various physical changes.
Furthermore, as potato staple food strategy initiated, potatos have been the largest noncereal food crop worldwide [20], and potato planting area expanded 6.7 million hectares (2020) [21]. e yield of potato starch also increased up to 96 million tonnes (2016) [22], which has been a promising natural carbohydrate for food packing material. e size of potato starch granules is relatively large (25-100 μm), and exists in typile B-type crystallinity [23]. Also, in the presence of plasticizers and under high temperature and shearing conditions, potato starch would acquire thermoplastic property which was suitable for preparing biodegradable film [24]. However, the performance of potato starch-based films still needs to be improved to extend the application in food package because of their poor water resistance. To address such issue, plasticizing agents and other active ingredients were incorporated to endow various advantages to films. Glycerol has been reported as the most starch-compatible plasticizing agent that can endow polymers with flexibility and resilience [25,26]. Moreover, gelatin, a protein of animal origin, exhibits excellent thermoreversible capability and thermal stability, presenting possible barrier performance [27,28].
Although a range of various composite films has been investigated, few pieces of research were conducted about the optimized formulation of the composite film via response surface methodology (RSM) combined with the multi-index comprehensive evaluation method that considered physical properties (thickness, water solution (WS), tensile strength (TS), and elongation at break (E%)) and barrier property (light transmittance (T%)). us, this study aimed to prepare a durable biodegradable film with excellent comprehensive performance based on potato starch, glycerol, and gelatin. e optimized films were further characterized by scanning electron microscopy (SEM), rheological measurements, and Fourier transform infrared spectroscopy (FTIR).

Materials.
Potato starch was purchased from a local market in Baoding, China. Glycerol was purchased from Tianjin Kemiou Chemical Reagent Co., Ltd (Tianjin, China). Edible gelatin, extracted from pig skin, was obtained from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China).

Preparation of Complex Films.
Potato starch was dispersed in 100 mL of distilled water and stirred for 30 min at 80°C until completely gelatinized. Glycerol and gelatin were successively added to the film-forming solution. e mixtures were degassed for another 30 min by using an ultrasonic device (SK5200H, Ningbo Scientz Biotechnology Co., Ltd. Zhejiang, China), then casting onto a plate, and dried at 40°C for 36 h in an incubator [29]. After this period, the films were peeled for further analysis.

Comprehensive Score Calculation.
Membership degree and principal component analysis (PCA) were used to evaluate the films comprehensively. In accordance with importance, the indexes of the films were normalized to optimize the production process. Each index weight was determined. Indexes with high membership, such as TS and elongation at break (E%), were calculated using equation (1). WS (water solution) and light transmittance (T%) were calculated with equation (2): where P is the index membership degree; A i is the actual value; A min is the minimum of the same index; A max is the maximum of the same index. e comprehensive score (Y) was calculated with the following equation: where a 1 , a 2 , a 3 , a 4 , and a 5 are the weights of thickness, TS, E %, WS, and T%, respectively, and P 1 , P 2 , P 3 , P 4 , P 5 are membership degrees of thickness, TS, E%, WS, and T%, respectively.

Design of Experiments.
Film concentration ratios were optimized by using RSM to investigate the synergistic effect of the variables. ree-level (starch concentration X 1 [1-6 g/ 100 mL], glycerol concentration X 2 [0.5-3 g/100 mL], and gelatin concentration X 3 [0.5-3 g/100 mL]) factorial design was applied to investigate the effects of the parameters on WVP, TS, T%, and E%. After preliminary tests, the appropriate levels of the independent variables needed to obtain the optimal characteristic properties of the biodegradable composite film were measured. A total of 17 formulas with five replicates at the center point were evaluated to ensure the repeatability of the RSM model. ese formulas are listed in Table 1, which clearly records the coded and uncoded levels of starch, glycerol, and gelatin concentrations. e response for each formulation was independently evaluated with three replicates. A random experimental design was used to eliminate the effects of possible uncontrolled factors.

Verification Test.
Verification procedures were applied to determine the optimal level of composite films with excellent properties. e experimental data and fitted value predicted by the model were compared to verify the validity and adequacy of the RSM model.

Acquisition of Light Transmission
Curves. UV-vis spectrophotometer (α-1500, Shanghai, China) was used to acquire the light transmittance spectra of the complex films. Briefly, the film was cut into strips (1 × 4 cm) and placed closely to the inner surface of the cuvette to record the T% with three replicates ranging from 400 nm to 800 nm.

Rheology of Film-Forming Solutions.
e static rheological properties of the film-forming solutions were determined by using a rotational rheometer (RheolabQC, Anton Paar Instruments Ltd., Japan) equipped with a concentric cylinder (CC 39, 40 mm diameter, Austria) and temperature control system (TEZ 150P-C). e shear rate was varied from 1 s −1 to 300 s −1 with the Ramp log procedure at 25°C and the flow behavior was recorded every 30 s. e rheological properties of films were further stimulated by the Cross model [30]: where η, η 0, and η ∞ represented viscosity, the initial shear viscosity, and infinite shear viscosity, respectively. c indicated shear rate (s −1 ), and K and Þ were time constant and dimensionless exponent, respectively.

Physical
Properties. e thicknesses of the films was recorded by using an electronic thickness gauge (EC770, Shanghai, China) at 10 random positions. An autotensile tester (PARAM XLW-PC, Jinan, China) was used to measure the TS and elongation at break (E%) of the films at a cross-head speed of 300 mm/s [31]. T% was measured with a UV-vis spectrophotometer at 500 nm.
WVP of the film was measured by the gravimetric method [32]. e film (8 cm × 8 cm) was sealed in a circular cup with 20 g anhydrous CaCl 2 (0% RH (relative humidity)). e cups were placed inside of desiccator with distilled water (100% RH). e difference in RH promoted the transport of water vapor, which caused weight changes.
WS was tested according to the modified weight loss method [33] which was performed with three replicates to determine the WS of films. Briefly, at room temperature, a film (2 cm × 2 cm) was weighed prior to dissolution in distilled water for 24 h and then removed and dried at 90°C to a constant weight. WS was calculated by the following equation: where M 1 and M 2 are the initial and terminal weights of the films (dry basis), respectively.

Morphology of Films.
e microstructures of crosssections and surfaces were measured by using a scanning electron microscope (JSM-7900F, Japan) with an accelerating voltage of 12.5 kV. Before the observation, the films were coated with a thin layer of gold.
2.10. FTIR Spectroscopy. FTIR spectra were acquired over the range of 500-4000 cm −1 in attenuated total reflection mode with a resolution of 4 cm −1 by using an FTIR spectrometer (Shimadzu Instrument Co. Ltd., Japan).

Light Transmission Analysis.
e light transmittance of the film is an important index to directly affect consumer acceptance. And the light transmission curve of a resulting optimized film (starch + glycerol + gelatin) is shown in Figure 1(a). e remarkable UV barrier capability shown by the film over the UV range of 200-280 nm would protect food from oxidative deterioration [32]. is characteristic was due to the intermolecular hydrogen bond between starch and glycerol during the film-forming process, resulting in good miscibility [34]. e optical property of film is significantly reinforced with wavelength increasing. Moreover, over the visible range (at 500 nm), the T% of films reached the maximum value then remained constant at approximately 90%, indicating excellent transparency. Such high transmittance indicated the uniform of film. Perfectly clear food packaging would not adversely affect the consumer experience. e picture covered by these films could be clearly observed (Figure 1(b)), and further proved good transmittance.

Comprehensive Evaluation of the Composite Film.
PCA was performed by using SPSS19.0 software on six randomly selected sets of data (Table 2) to evaluate film comprehensive performance. According to the criterion of eigenvalues exceeding 1, the results of PCA (Table 3) showed that the three principal components (PC1, PC2, and PC3) are extracted and the variance contributions of PC1, PC2, and PC3 were 40.425%, 29.065%, and 20.892%, respectively. In addition, the cumulative variance of the three components reached 90.382%, indicating that PC1, PC2, and PC3 could contain most of the information (more than 90%) of all indexes (TS, E%, WS, thickness, and T%) and reflect the comprehensive performance of the films. erefore, PC1, PC2, and PC3 could replace the original complex indexes.   Table 4 lists the factor loading matrix and the contribution rate of thickness, TS, E%, and WS to PC1, PC2, and PC3 (eigenvalue > 1). As shown in Table 4, thickness, E%, and WS mainly affected PC1; TS mainly affected PC2; T% and TS mainly affected PC3. After normalization, the percentages of thickness, TS, E%, WS, and T%, were 13%, 34%, 18%, 30%, and 5%, respectively. e comprehensive score (Y) was calculated by using the following equation in accordance with equations (1)- (3): 3.3. RSM Analysis. RSM was applied to investigate the linear, quadratic, and interaction effects of each independent variable (starch, glycerol, and gelatin concentrations) on response values. All the response experimental data are listed in Table 5. e different evaluation indexes, including the F value, the lack-of-fit value, and the correlation coefficient R 2 , are presented in Table 6 to verify the model's adequacy and accuracy. e predictive model was extremely significant (p < 0.01). A low p value indicates the high significance and applicability of the regression model. Moreover, the p values of the linear (X 1 ), quadratic (X 13 ), and interactive (X 12 , X 13 , and X 23 ) models were all less than 0.05, which indicated that the model fits well with the measured data. R 2 and adj-R 2 were used to prove the adequacy and goodness of fit of the applied model. A high R 2 value (R 2 > 0.9) indicated the high reliability of the models and acceptable error between the calculated and experimental results. e insignificance between R 2 and R 2 -adj indicated the adequate accuracy and applicability of the selected polynomial model. e response comprehensive score (Y) was fitted with the regression equation as shown in the following equation: Y � 50.96 − 7.92 X 1 + 4.12 X 2 + 2.71 X 3 + 0.012 X 1 X 2 − 5.48 X 1 X 3 + 0.95 X 2 X 3 − 8.20 X 12 − 10.53 X 22 − 8.70 X 32 .
3D response surfaces were used ( Figure 2) to further interpret the effects of interactions among independent variables (starch, glycerol, and gelatin concentrations) on response value. e closer the bottom curve, which is the projection of the 3D response surface, to the ellipse, the more obvious the interaction effect and vice versa. In accordance with Table 5, the figure shows significant interaction between X 1 and X 2 and no significant interaction between X 1 and X 2 and between X 2 and X 3 . Such results further indicated that the RSM model was reliable and veritable.

Verification Experiments.
Verification experiments on the optimal variable concentrations (2.40% starch, 2.11% glycerol, and 1.68% gelatin) and experimental concentrations (2.5% starch, 2.0% glycerol, and 1.5% glycerin) were performed to validate the reliability and validity of the model equations.
e results of the verification experiments are shown in Table 7. e acceptable percentage error between the predicted value and experimental value further demonstrated the reasonable adequacy of the response surface equations for responses and the reliability of the experiments. Figure 3(a) depicts the variation curve of viscosity with the shear rate of 3 film solutions. Film thickness and uniformity are affected by the fluidity (viscosity) of filmforming solutions. e figure shows that the viscosity of the film-forming solutions decreased with the increase in shear rate, exhibiting non-Newtonian fluid behavior that might be attributed to the rearrangement of entangled molecules under high-speed shear [35,36]. Such a result was in accordance with the finding of Glusac et al. [37], who proved that potato starch and zein proteins exhibited shear-thinning behavior. And the rheological behavior of the three films was successfully fitted by the Cross model (Table 8). e p value of all films was lower than 1, which further proved the pseudoplastic fluids. K-value was related to the destruction of the molecular structure inside the fluid [30]. And the three-component film showed the smallest K-value, which indicated the least time needed to form a new network structure and hydrogen bonding [38].   [30]. In principle, film-forming solutions become viscous with the increase in solute concentration. However, Figure 3(a) indicates that the film solution of the three mixed matrixes (starch + glycerol + gelatin) showed low viscosity. is result was consistent with Hussain et al. [39], who believed that the viscosity decrease could be attributed to the alteration in the macromolecular organization of the blended gelatin at a high shear rate.      the forming of intermolecular hydrogen bonds between glycerol and starch with the matrix, resulting in strong interfacial adhesion [16] and lower WS indicated water resistance enhanced. Meanwhile, the lower TS and E% of the three-component film may probably be due to the agglomerates formed by the incorporation of gelatin, which disrupted the compactness of the hydrogen bonds [40]. Moreover, the T% showed no significance, which may be explained by the great miscibility of three components.
3.6. SEM Analysis. As shown in Figure 4, SEM was applied to observe the microstructures of the surface and cross section of the optimized film. e smoothness and homogeneity of the surfaces presented in Figure 4(a) indicated that the components dispersed well within each other. e appearance of surface wrinkles did not mean that the film was uneven and discontinuous. Such wrinkles might be caused by adhesion between the film and the plastic plate when the film was peeled off. Additionally, as shown in Figure 4(b), distinctive ridged structures were presented in the crosssections of the composite biodegradable film. e high uniformity of the biodegradable film was attributed to the plasticizing capability of glycerol and the thickening capability of gelatin. Furthermore, the numerous intermolecular hydrogen bonds that formed between hydroxyl groups in glycerol with the hydroxyl groups present in starch promoted close bonding in polymers [41] and further enhanced miscibility among starch, gelatin, and glycerol.

FTIR Analysis.
e spectra of the single starch film and optimized biocomposite films are shown in Figure 5. Typically, the broadening of the O-H and the decreasing of wavenumber could be possible due to the characteristic of hydrogen bonding interaction in the film [42]. For the control film (single starch film), the characteristic band around 3267 cm −1 was originally from the O-H stretching vibration, and the band absorption intensity was broadened with incorporation with gelatin and glycerol. Also, the characteristic peak at 1644 cm −1 slightly shifts to 1636 cm −1 .
is phenomenon indicated the formation of hydrogen bonds. e similar characteristic peaks at approximately 2940 and 2875 cm −1 in the three biocomposite film spectra were attributed to C-H vibration and C�C vibration, respectively [43]. e peaks at 1636 and 1023 cm −1 represented the C�O and C-O stretching vibrations of glycoside bonds in the starch branch. e intensity of the peak at 1023 cm −1 increased with the incorporation of glycerol, and the enhancement in the -OH stretching vibration at approximately 3267 cm −1 suggested the occurrence of association [44]. Cross-linking between polymers tightened in the presence of intermolecular hydrogen and hydrogenoxygen bonds ( Figure 6). Such results illustrated the good   compatibility between matrixes and were consistent with the SEM and rheology measurements.

Conclusions
A comprehensive performance film was successfully fabricated with biodegradable materials (starch, glycerol, and gelatin), and its formulation was optimized by using RSM. Responses, including mechanical, physical, and barrier properties, were subjected to a comprehensive multi-index evaluation. e RSM results verified the reliability of the model, and the optimized film exhibited excellent performance. In addition, the verification result indicated that the experimental values were highly consistent with the model-predicted values, and the optimal preparation conditions were 2.5% starch, 2% glycerol, and 1.5% gelatin. Rheological measurements proved that the film-forming solution exhibited non-Newtonian fluid behavior with low shear viscosity. Moreover, FTIR and SEM analysis demonstrated that the optimized films showed homogeneous with smooth microstructures, indicating good compatibility. In this work, the resulting biodegradable starch-based film with great comprehensive performance may be expected to become a green food packaging.

Data Availability
e data used to support the study are included within the article. Raw data can be acquired from the corresponding author upon reasonable request (Causunjf@hebau.edu.cn).

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.