Measurement of pH of solutions was done with the aid of a pH meter (Sartorius PB-10, Göttingen, Germany).

DPPH-RS-A was examined [10]. A standard curve of Trolox (10–60 µM) was prepared. The activity was computed and reported as µmol Trolox equivalents (TE)/mL sample.

UV-absorbance and browning index of MRPs were measured at 294 and 420 nm, respectively [13].

A micrometer (Mitutoyo, Model ID-C112PM, Mitutoyo Corp., Kawasaki-shi, Japan) was employed to determine the thickness of film. The thickness of 12 randomly chosen areas in film samples was obtained to calculate the average thickness.

Before measuring mechanical properties, the films were conditioned (48 h, 25°C, 50% ± 5% RH). Tensile strength (TS) and elongation at break (EAB) were assessed [18] using the Universal Testing Machine (Lloyd Instrument, Hampshire, UK) equipped with 100 N tensile load cell using cross-head speed of 30 mm/min. Ten samples (20 mm × 50 mm) with 30 mm of initial length were used for testing.

WVP of films was measured [18]. The films were placed to cover the aluminium cup, and a rubber gasket was used to fix the films tightly. The aluminium cups were then transferred into an environmental chamber (model H110K-30DM; Seiwa Riko Co., Tokyo, Japan) (25°C ± 0.5°C, 50% ± 5% RH). Over 10 h period, the cups were weighed every 1 h. WVP was determined by the following equation:

(1) WVP (g mm2 s Pa)=wlAt(P2−P1) where w is the weight gain of the cup (g); l is the film thickness (m); A is the exposed area of film (m²); t is the time of gain (s); P₂−P₁ is the vapor pressure difference across the film (Pa).

A CIE colorimeter (Hunter associates Laboratory, Inc., Reston, VA, USA) was used to measure the color of the films using D₆₅ (day light). A measuring cell having 1.25 inch opening was used. L* (Lightness/darkness), a* (redness/greenness) and b* (yellowness/blueness) were recorded. The total difference in color (ΔE*) was computed with the equation shown below [5]:

(2) ΔE∗=(ΔL∗)2+(Δa∗)2+(Δb∗)2 where ΔL*, Δa* and Δb* are the differences between the color parameter of the samples and those of the white standard (L* = 92.82, a* = −1.24, b* = 0.46).

Light transmission of films was determined at UV and visible range (from 200–800 nm) using UV–vis spectrophotometer (Shimadzu, Kyoto, Japan) [5]. Opaqueness value of film was computed:

(3) Opaqueness value= −logT600x where T₆₀₀ is fractional transmission at 600 nm, and x is the thickness of film (mm).

Protein patterns of solubilized films were determined using sodium dodecyl sulphate–polyacrylamide gel electrophoresis. The prepared samples (15 mg protein) were loaded onto the gel (4% stacking gel and 10% separating gel). Constant current (15 mA/gel) was applied using Mini Protean Tetra Cell unit (Bio-Rad Laboratories, Richmond, CA, USA). Thereafter, the proteins separated in gel were stained and destained [7].

Film solubility was examined with a slight modification [18]. Conditioned films (width: 20 mm; length: 50 mm) were weighed and transferred into Erlenmeyer flask containing 10 mL of distilled water and 0.1% (w/v) sodium azide. The mixture was then agitated using a shaker (Heidolth Inkubator 10000, Schwabach, Germany) (250 rpm, 30°C, 24 h). After 24 h, the mixture was filtered using Whatman filter paper No. 1 to obtain the retentate. The retentate was dried (105°C, 24 h) and the dried insolubilized matter was weighed. Film solubility value was computed and reported as a percentage of total soluble matter weight relative to the weight of initial film.

SEM (Quanta 400, FEI, Eindhoven, the Netherlands) was used to visualize the surface and cross-section of the selected films [18]. Film samples were placed on bronze stubs and sputtered with gold (Sputter coater SPI-Module, PA, USA) to make them conductive before the morphological visualization. An acceleration voltage of 20 kV was employed. The magnification of 500× and 1000× were used for visualization of film surface and cross-section, respectively.

The selected films containing glycerol and MRPs at a ratio of 10:20 as well as LDPE film were used to prepared 3-side sealed bags. An impulse sealer with a magnet (Model ME-300HIM, S.N.MARK Ltd., Park, Nonthaburi, Thailand) was used to seal the films (90 mm × 90 mm) at 150°C ± 0.5°C for 2.5 s. The sealed films were then cooled for 3 s.

Chicken skin containing depot fat was used [19]. Chicken skins (20 mm × 30 mm) were rendered using a pan at 130°C–150°C for 10 min. The obtained oil namely CSO was filtered through two layers of cheesecloth and centrifuged at 3000 × g for 20 min.

CSO (4 g) was filled into the prepared bags from different films and headspace air was removed. Finally, the bags were heat-sealed. All samples were kept at 30°C ± 0.5°C with 52% ± 5% RH mediated by Mg(NO₃)₂ saturated salt [20] and kept away from direct light exposure. CSO samples were collected every 3 days for totally 15 days for quality assessment.

The pHs of MRPs solution produced under varying conditions are presented in Table 1. The pH of MRPs solution decreased as the Maillard reaction took place [7]. As reaction time was augmented, the continuous decrease of pH values was found for all MRPs solutions, regardless of temperature and gelatin-fructose (GE/FR) ratio used. The drastic decreases in pH were more pronounced for MRPs solutions after 12 h of reaction (p < 0.05). The lowest pH of MRPs solution was obviously obtained in all MRPs solutions after 24 h of reaction, compared to those of MRPs solution at 0 h. This result indicated that the Maillard reaction more proceeded with a rising reaction time [7]. Despite different GE/FR ratios, pH values between all MRPs solutions at the initial reaction time (0 h) were similar (p > 0.05). Differences in pH were attained after the heating was applied. At the same temperature and reaction time (except at 0 h), the lower pH (p < 0.05) was observed for MRPs solutions with GE/FR ratio of 1:1 (w/w), compared to those of MRPs solutions with other GE/FR ratios (4:1 and 9:1) (p < 0.05). The equal ratio between gelatin and fructose could provide the appropriate number of functional groups, which were mainly involved in the Maillard reaction as witnessed by the lowering of pH in resulting MRPs solution. Nevertheless, lower pH value was obtained for MRPs solutions prepared at 90°C, compared to those of MRPs solutions heated at 60°C, irrespective of GE/FR ratio and reaction time used (except at 0 h). This was suggested that a higher temperature used could accelerate Maillard reaction [10], which might be associated with high energy for glycation between fructose and α- or ε-NH₂ of amino acids in gelatin. Lowering of pH of MRPs might be associated with the formation of organic acids such as formic acid, acetic acid, etc., mainly degraded from reducing sugars during Maillard reaction [23]. Additionally, the amine-amine reaction could be another important mechanism to form acidic products, resulting in the decreased pH of MRPs [24]. Therefore, the temperature, GE/FR ratio and reaction time had the profound impact on pH of MRPs solutions.

The impact of heating temperature, GE/FR ratio and reaction time on the changes of UV-absorbance (A₂₉₄) and browning index (A₄₂₀) of MRPs solutions is shown in Table 1. Typically, A₂₉₄ has been used for the measurement of the production of uncolored intermediate MRPs [10], whereas A₄₂₀ represents the formation of the final browning product [13]. At the same temperature and GE/FR ratio, higher A₂₉₄ were observed for MRPs solutions after 12 h of reaction. This result suggested that the formation of uncolored intermediate compounds was increased as reaction time upsurged, in which those intermediate compounds could serve as Maillard reaction’s precursor [10]. Regardless of GE/FR ratio and reaction time used, no difference in A₂₉₄ (p > 0.05) between MRPs solutions produced at 60°C and 90°C was detected, except MRPs solutions with GE/FR ratio of 9:1 after 12 and 24 h of reaction. Lower A₂₉₄ was found for those aforementioned MRPs solutions prepared at 60°C, compared to that of MRPs solutions prepared at 90°C, suggesting that the Maillard reaction occurred at a lower extent when lower level of sugar and temperature were used. Additionally, at the same temperature and reaction time used, no differences in A₂₉₄ between MRPs solutions with GE/FR ratios of 1:1, 4:1 and 9:1 were observed, except A₂₉₄ of MRPs solutions with GE/FR ratio of 9:1 after heating for 12 and 24 h. For the MRPs solutions after 12 h of reaction, higher values of A₄₂₀ were obviously observed for MRPs solution heated at 90°C, compared to those of MRPs solutions prepared at 60°C (p < 0.05). Higher A₄₂₀ values were also obtained when MRPs solutions were heated for longer time. A₄₂₀ of MRPs solutions heated for 24 h were higher than those of MRPs solutions prepared for 0 and 12 h, irrespective of temperature and GE/FR ratio used, except A₄₂₀ of MRPs solutions with GE/FR ratio of 1:1 heated for 12 h. In particular, the highest A₄₂₀ value was observed for MRPs solution with GE/FR ratio of 1:1 heated at 90°C for 12 and 24 h. Therefore, temperature, GE/FR ratio, and reaction time had a direct influence on the formation of intermediate and final browning products in Maillard reaction.

DPPH-RS-A of MRPs solution produced at various temperatures, GE/FR ratios and times is displayed in Table 1. In general, the highest DPPH-RS-A was found for MRPs solution heated at 90°C using GE/FR ratio of 1:1 for 24 h (p < 0.05). Xu et al. [25] documented that the higher pH of the reaction system provided a higher antioxidant activity of MRPs than the lower pHs used in amino acids-reducing sugars system. In addition, Chen et al. [26] found that higher temperature also increased antioxidant activity of MRP in gelatin hydrolysate-sugars model systems. DPPH-RS-A increased along with the rises of A₂₉₄ and browning intensity (Table 1). This results were also documented by Benjakul et al. [10] and Karnjanapratum et al. [7]. In general, DPPH-RS-A of MRPs was related with the formation of intermediate and final products or browning intensity. According to Chen et al. [26], MRPs can act as hydrogen donors in antiradical activity. Apart from melanoidin, different MRPs such as furans, pyrroles, pyridines, and imidazoles might contribute to antioxidant potential during extended heating [27]. Therefore, DPPH-RS-A of MRPs produced from gelatin-fructose systems could be improved under the optimal MRP production condition.

Film with GLY/MRPs ratio of 10:20 was chosen due to the excellent film properties, including lowest WVP (Fig. 1D) and high UV barrier properties (Fig. 2E). Properties of films with GLY/MRPs ratios of 30:0 and 0:30 were also examined.

The changes in FFA content of CSO as affected by different types of bag are depicted in Fig. 4A. FFA content increased with the upsurge in storage time. FFAs were formed by hydrolysis of triacylglycerol or phospholipid, due to the presence of lipase, water, and excessive heating process [18]. The main factor affecting FFA increment in CSO packed in bags during storage was the water that possibly passed through the film or water present in the CSO during dry rendering. CSO packed in bag from gelatin containing glycerol alone (30:0) generally had higher FFA content compared to other samples (p < 0.05). Poor WV-BP of film was generally obtained when added with glycerol. Nilsuwan et al. [36] also reported the high water absorptivity of gelatin film added with glycerol compared to nylon/LLDPE film. In contrast, the changes in FFA contents between CSO packed in bag from gelatin incorporated with GLY/MRPs ratio of 10:20 and 0:30 as well as LDPE bag were not noticeably different from each other, especially from day 9 to day 15 of storage (p > 0.05). Therefore, packaging material with lower water vapor migration could be achieved when gelatin film was added with MRPs to slow down the hydrolysis of CSO during storage.

Changes in PV of CSO packed in various bags during 15 days of storage are depicted in Fig. 4B. Although PV increased during storage times, the differences between samples were not obvious (p > 0.05), except for CSO packed in gelatin bag containing only glycerol (30:0), which had higher PV than others, especially within the initial 6 days of storage (p < 0.05). The gradual upsurge in PV might be caused by the presence of some oxygen in CSO that could be involved in the oxidation reaction in CSO to form the primary oxidation products, e.g., hydroperoxide. This finding is supported by Nilsuwan et al. [36], who documented that similar PV in shrimp crackers packed with glycerol-plasticized gelatin film and nylon/LLDPE were obtained after storage. For day 3 and 6, the CSO packed in the bag produced from film with GLY/MRPs at 10:20 had the lower PV than other samples packed in bag produced from film containing only glycerol. Furthermore, the presence of natural antioxidant from MRPs added into gelatin films regardless of its proportion in GLY/MRPs mixture did not significantly affect PV of CSO during storage. The water-soluble polymeric fractions of MRPs that have antioxidant potential could not be released effectively in CSO, a hydrophobic system [37]. Additionally, a slight decrease of PV was observed on day 12 of storage. As the temperature and light were controlled during storage, PV still increased in the presence of oxygen. Hydroperoxide formed at the earlier stages of oxidation underwent decomposition at the advanced stages of oxidation [38]. Minor reduction of PV on day 12 indicated the generation of secondary oxidation products, which negatively affected the quality of CSO.

TBARS values of CSO were changed as affected by various types of bag during storage as depicted in Fig. 4C. TBARS value represents the amount of secondary oxidation products in CSO. Secondary oxidation products occurred even in the presence of oxygen at a low extent. TBARS value increased in all samples on day 3, suggesting secondary oxidation products were produced. Thereafter, TBARS values were fluctuated in the narrow range up to 15 days of storage. The highest TBARS value was attained in sample packed in LDPE bag. Higher lipid oxidation therefore occurred in the sample packed in LDPE bag. In contrast, lowest TBARS value was observed in CSO packed in gelatin bag containing GLY/MRPs of 10:20 (p < 0.05), especially during 9–15 days of storage. TBARS value of CSO packed in bag added with only MRPs was similar to that found in CSO packed in LDPE bag. Protein-based films are recognized to possess better O-BP than synthetic films [39,40]. The result indicated the necessity to develop a film with proper ratio of additives to improve the film properties. Thus, packaging material or bag that possessed an excellent oxygen barrier and antioxidative properties was important to retard lipid oxidation in CSO samples.

The volatile compounds of CSO packed in various types of bag after 15 days of storage in comparison with fresh CSO (day 0) are presented in Table 2. The fatty acid content in CSO is mainly dominated by unsaturated fatty acid, especially oleic (c18:1 n-9) and linoleic acid (c18:2 n-6) [18]. The unsaturated fatty acid content in CSO accounted up to 70% in 100 g of oil, which influenced the oxidative stability during storage [18]. Autoxidation occurred in the presence of oxygen, leading to the formation of hydroperoxides and subsequent decomposition to the secondary oxidation products such as aldehydes, ketones, and alcohol [41]. The presence of secondary oxidation products are often linked to off-odors and rancidity in oil [42]. Major compounds generated from the oxidation of oleic and linoleic acids, such as pentanal and hexanal were detected at day 0. Hexanal content raised up to 8 times after 15 days of storage in samples packed in all bags. During rendering in air and at high temperature, oxidation could be enhanced. Apart from hexanal, significant increment was also observed for other volatile aldehydes (pentanal, heptenal, 3-methylbutanal, 2,4-heptadienal) after 15 days. The linearity between PV and the amount of 2,4-heptadienal was previously reported [43]. In the present study, similar levels of 2,4-heptadienal in all samples were obtained. Relationship between PV and 2,4-heptadienal was also observed (Fig. 4B). Additionally, the least abundance of aldehydic volatile compounds was mostly observed in CSO packed in gelatin bag added with GLY/MRPs at 10:20 ratio. These results were coincidental with the lowest content of TBARS value in CSO packed in the bag containing GLY/MRPs ratio of 10:20 (Fig. 4C). For alcoholic volatile compounds, significant increases of 1-penten-3-ol and 3-methyl-1-butanol were observed after 15 days of storage. 1-octen-3-ol and 1-pentanol were also detected at low abundance. It demonstrated the oxidation of linoleic acid in the presence of single oxygen [18,43]. Among all samples, CSO packed in the gelatin bag containing only glycerol (30:0) had the highest alcoholic volatiles compounds. Ketonic volatile compounds that increased in most samples after storage were 3-pentanone, 3-hydroxy-2-butanone, and dihydro-2(3H)-furanone. CSO packed in gelatin bag containing the GLY/MRPs at a ratio of 10:20 contained the least abundance of ketonic volatile compounds compared to other samples. Overall, CSO packed in gelatin bag incorporated with GLY/MRPs at 10:20 possessed the lowest abundance of volatile compounds, indicating better oxidative stability compared to other samples after storage. The results were in tandem with the changes in FFA (Fig. 4A) and TBARS values (Fig. 4C). Thus, the gelatin bag with GLY/MRPs ratio of 10:20 was capable of retardation of the formation of secondary oxidation products and maintained the quality of CSO during storage.