Effect of select parameters on the properties of edible film from water-soluble fish proteins in surimi wash-water

Transcription

1 Effect of select parameters on the properties of edible film from water-soluble fish proteins in surimi wash-water Thawien Bourtoom a, Manjeet S. Chinnan b,, Pantipa Jantawat a, Romanee Sanguandeekul a a Department of Food Technology, Faculty of Science, Chulalongkorn University, Bangkok, Thailand b Department of Food Science and Technology, 1109 Experiment St., The University of Georgia, Griffin, GA , USA Abstract Edible film from water-soluble fish proteins were developed by casting film solution on leveled trays and effects of (9.5, 10.0 and 10.5), heating temperature (60, and 80 1C), and heating time (10, 20 and 30 min) of the film solution on various film properties were determined using Response Surface Methodology (RSM). The impact of and heating temperature of film solution was more significant, overall, on the film s properties than heating time. Contour plots of tensile strength and elongation at break was highest at of 10.0 at 1C ( MPa) but low in elongation at break ( %), while water vapor permeability and oxygen permeability were at their lowest ( g mm/m 2 d kpa and cm 3 mm/m 2 d kpa). There was a direct correlation between the films and proteins solubility on one hand, and heating temperature of film solution on the other, which reversed with change in of film solution. Film color was darker and more yellowish with increase in the of film solution. Keywords: Edible film; Water-soluble fish protein; Response surface methodology; Permeability 1. Introduction Food packaging provides important information to the consumer and enables convenient dispensing of food. However, lately food packaging has become the central focus of waste reduction efforts. Packaging represents approximately 30 g/100 g of municipal solid waste (Rawatt, 1993). The public perceives plastic packaging in particular, as using valuable nonrenewable natural resources and is concerned about the environmental effects of food service items. Edible and biodegradable polymer films offer an alternative to traditional packaging without the adverse effect on the environmental costs (Kester & Fennema, 1986; Nelson & Fennema, 1991). Components used for the preparation of edible films can be classified into three categories: Corresponding author. Tel.: ; fax: address: chinnan@uga.edu (M.S. Chinnan). hydrocolloids (such as protein, polysaccharide, and alginate), lipids (such as fatty acids, acyglycerol, waxes) and composites (Donhowe & Fennema, 1993). Hydrocolloid films have good barrier properties to oxygen, carbon dioxide, and lipids but not to water vapor. Most hydrocolloid film also possesses superb mechanical properties, which are quite useful for fragile food products. In industrial surimi manufacturing process, flesh mince is repeatedly washed with chilled water to remove sarcoplasmic proteins to produce a tasteless and odorless product. As a result of washing, approximately g/100 ml of minced fish solids (containing primarily water soluble proteins) are lost in the process (Pacheo-Aguilar, Creawford, & Lampila, 1989; Lin, Park, & Morrissey, 1994). Wash-water from surimi processing industry contains 1 2 water-soluble proteins, which are nutritional and highly functional. Using fish protein from surimi wash-water cannot only reduce the negative environmental impacts and costs of waste

2 disposal, but may also generate potential profits especially in the form of edible film from water-soluble proteins. The recovered proteins have hitherto been mostly used as animal feed and fertilizer because of the lack of techniques to use them as foodstuffs. At the moment, there is very limited information on films produced from water-soluble proteins, their mechanical properties and applications. The objective of this study was to investigate the effect of process parameters on the mechanical and barrier properties of edible films prepared from fish water-soluble proteins. 2. Materials and methods 2.1. Preparation of raw material Fish water-soluble proteins were recovered from the first stage of surimi wash-water, which was reconstituted in the laboratory from threadfin bream fish (Nemipterus hexodon). The surimi wash-water was freeze dried for 24 h (Dura- Top TM Model TD2DOTS002, FTS Systems, Inc.) and stored in plastic bags at 20 1C until needed. The proximate composition of protein recovered from surimi wash-water was found to be 80.88, 2.94, 6.26 and other 9.92 g/100 g of crude protein, crude fat, ash, and other, respectively Experimental design General Response Surface Methodology (GRSM) was used to determine the optimum combinations of, heating temperature and time. GRSM are given in terms of coded variable, x i (Cochran & Cox, 1957; Cox, 19; Thompson, 19). Selection of levels for independent variables was based on the results from preliminary tests and observations of Iwata, Ishizaki, Handa, and Tanaka (2000). The levels of input variables in coded (x i )and uncoded (x i ) forms are given in Table 1. The complete design consisted of 15 experimental points, which included three replications of the center point. The 15 films were prepared in random order. Each of the 13 dependent Y variables (responses) was assumed to be affected by the three independent variables. Responses under observation were: tensile strength (TS) (Y 1 ), elongation at break (Y 2 ), water vapor permeability (WVP) (Y 3 ), Oxygen permeability (OP) (Y 4 ), film solubility (Y 5 ), protein solubility (Y 6 ), L value (Y 7 ), a value (Y 8 ), b value (Y 9 ), DE ab (Y 10 ), hue angle (Y 11 ), and chroma (Y 12 ). Each value represented the mean of three determinations. The product thus obtained was analysed and experimental values were compared with model predictions Preparation of fish water-soluble proteins film Freeze-dried water-soluble fish proteins (80.88 g/100 g sample) were dissolved in distilled water (3 g/100 ml) to prepare film solutions. The level (9.5, 10 and 10.5) was adjusted prior to adding plasticizer (sorbitol) on a protein to sorbitol ratio of 2:1. The solutions were heated (60, and 80 1C) on a hot plate with magnetic stirrer for the given time (10, 20 and 30 min) Film casting and drying The film solution was filtered through a polyester screen (mesh no. 140 with mesh opening of 106 mm) to remove any small lumps, cooled to room temperature, followed by vacuum application to remove any dissolved air before pouring onto leveled nonstick trays to set. Once set, the trays were held overnight at 30 1C undisturbed, and then cooled to ambient temperature before peeling the films off the plates. Film samples were stored in plastic bags and held in desiccators at 50% RH for further testing Film testing Conditioning All films were conditioned prior to subjecting them to permeability and mechanical tests according to Standard method, D (ASTM, 1993a). Films used for testing WVP, OP, TS and elongation (E) were conditioned at 50% RH and C by placing them in a desiccator over a saturated solution of Mg (NO 3 ) 2 6H 2 O for 48 h or more. For other tests, film samples were transferred to plastic bags after peeling and placed in desiccators Film thickness Thickness of the films was measured with a precision digital micrometer (Digitrix-Mark II, Cole-Palmer Instrument Company, Niles, IL) to the nearest (75%) at five random locations on the film. Mean thickness values for each sample were calculated and used in WVP, OP and TS calculations Film solubility Modified method of Jangchud and Chinnan (1999) was used to measure film solubility. Film pieces, 20 mm 20 mm, were dried at 1C in a vacuum oven for 24 h, and then weighed to the nearest g for the initial dry mass. Films were immersed into 20 ml of distilled water in 50 ml screw cap centrifuge tubes containing 0.01 g/100 g potassium sorbate. The tubes were capped and placed in a shaking water bath for 24 h at 25 1C. A portion of the solution was removed and set aside for use later in protein solubility tests as described later. The remaining solution and film piece were poured onto (Whatman #1) qualitative filter paper, rinsed with 10 ml distilled water, and dried at 1C in a vacuum oven for 24 h to determine dry mass of film. Triple measurements were taken for each treatment triplicate.

3 Table 1 Experimental data for the three-factor, three level response surface analysis a Independent variables Dependent variables Treatment Temp Time Tensile strength (MPa) Elongation at break (%) Water vapor b permeability (%) Oxygen c permeability (%) Film solubility (%) Protein solubility (%) Color L a b DE ab Hue angle Chroma x1 x2 x3 Y 1 Y 2 Y 3 Y 4 Y 5 Y 6 Y 7 Y 8 Y 9 Y 10 Y 11 Y (10.5) 1(80) 0(20) (10.5) 1(60) 0(20) (9.5) 1(80) 0(20) (9.5) 1(60) 0(20) (10.5) 0() 1(30) (10.5) 0() 1(10) (9.5) 0() 1(30) (9.5) 0() 1(10) (10.0) 1(80) 1(30) (10.0) 1(80) 1(10) (10.0) 1(60) 1(30) (10.0) 1(60) 1(10) (10.0) 0() 0(20) (10.0) 0() 0(20) (10.0) 0() 0(20) Values in parentheses are the uncoded independent variables. a Mean of three replication and the experimental runs were performed in a random order. b Water vapor permeability expressed as g mm/m 2 d kpa. c Oxygen permeability expressed as cm 3 mm/m 2 d kpa.

4 Total soluble matter was calculated from the initial gross mass and final dry mass using the following equation: %FSðdbÞ ðfilm mass before test film mass after testþ100 ¼. Film mass before test Protein solubility Solution samples set aside from film solubility tests were analysed for protein content by the Lowry method (Lowry, Rosebrough, Farr, & Randall, 1951). followed (ASTM, 1993b). Film samples were double masked in aluminum foil masks with an effective film test area of 5 cm 2. Tests were performed at C and 50% RH. WVP was calculated by multiplying water vapor transmission rate (WVTR) with the film thickness and dividing it by water vapor pressure gradient across the exposed film. %PS¼ Mass of protein in 20 ml solution 100 Initial mass of film ð% protein in filmþð% dry matter of filmþ Color A portable colorimeter (MiniScan XE, Associate Laboratory, Inc., Reston, VA) was used to determine the film L ; a and b color value (L ¼ 0 (black) to 100 (white); a ¼ 60 (green) to +60 (red); and b ¼ 60 (blue) to +60 (yellow). Yellow standard plate (calibration plate CR-A47, L ¼ 85:45; a ¼ 0:15 and b ¼ 54:55) was used as a standard. Color (means of five measurements at different locations on each specimen) was measured on 10 cm 10 cm segment of film. Film specimens were placed on a black plate when measurements were performed. Total color difference (DE ab ), hue angle and chroma were calculated using the following equation: DL ¼ L sample L standard, Da ¼ a sample a standard, Db ¼ b sample b standard, DE ab ¼½ðDL Þ 2 þðda Þ 2 þðdb Þ 2 Š 0:5, C ¼½ða Þ 2 þðb Þ 2 Š 0:5 and H ¼ tan 1 ðb =a Þ when a 40andb 40, H ¼ 180 þ tan 1 ðb =a Þ when a o0, H ¼ 360 þ tan 1 ðb =a Þ when a 40 and b o0. Prior to taking color measurements, film specimens were pre conditioned at 50% RH and C for 3 days Water vapor permeability (WVP) The WVP values were determined using the water transmission rate instrument (Permatran-W1A, Modern Controls, Inc. Minneapolis, MN). Testing method as described by ASTM F Standard Method was Oxygen permeability (OP) The OP values were determined using a MOCON unit (Ox-Tran 100A, Modern Control, Inc., Minneapolis, MN) according to ASTM D Standard Method (ASTM, 1993c). Film samples were masked by aluminum foil mask with effective film test area of 5 cm 2. Tests were performed at C and 50% RH. OP was calculated by multiplying oxygen gas transmission rate (OGTR) with the thickness and dividing by partial pressure difference of oxygen across the films surface Tensile strength and elongation at break (TS and E) TS was performed using an Instron universal testing instrument (Model 1122, Instron Corp., Canton, MA) as per ASTM D8-91 Standard Method (ASTM, 1995). Fifteen samples, 3 cm 10 cm each, were cut from each film. Initial gap separation and cross-head speed were set at 50 mm and 50 mm/min, respectively. TS was calculated by dividing the maximum force at break by initial specimen cross-sectional area, and percent elongation at break was calculated as follows: E ¼ 100 ðd after d before Þ=d before, where d was the distance between grips holding the specimen before or after the break of the specimen Statistical analysis Data were analysed to fit the following second-order polynomial equation to all dependent Y variables: Y ¼ b ko þ X3 i¼1 b ki x i þ X3 i¼1 b kii x 2 i þ X2 X 3 i¼1 j¼iþ1 b kij x i x j, (1) where: b ko ; b ki ; b kij are constant coefficients and x i is the coded independent variable. The SAS version 6.12 program (SAS Institute Inc., 1996) was used for analysis of variance and regression coefficient calculation. Contour plot of responses for these models were also drawn using the Statistica for Windows, Version 5.0 by plotting as a function of two variables, while keeping other variables at the constant value.

5 3. Results and discussion 3.1. Statistical analysis and contour plots The RSREG procedure of Statistical Analysis System (Box & Draper, 1987) which was used to fit the secondorder polynomial Eq. (1) to the data on films properties is shown in Table 1. The regression coefficients (b ki )and analysis of variance variables are presented in Table 2. The results (not presented) showed that the model developed for the TS, elongation at break, OP, film solubility, protein solubility, color (L ; a ; b ; hue angle, DE ab and chroma) was adequate, and did not show a significant lack of fit. However, with regards to the WVP and film solubility values, the highly significant lack of fit suggests that the chosen model did not represent the system appropriately (Thompson, 19). In such a case, it would be desirable to perform some kind of mathematical transformation on the dependent or independent variables, to obtain an acceptable model with nonsignificant lack of fit. Several such transformations of the experimental data were tried. The model obtained by the logarithmic transformation of water vapor and OP data yielded the best results and are given below: A ¼ ln ðwvpþ ¼4: þ 0:014X 1 0:06X 2 þ 0:001X 3 þ 0:235X 1 X 1 þ 0:022X 1 X 2 þ 0:208X 2 X 2 0:035X 1 X 3 0:013X 2 X 3 þ 0:115X 3 X 3, B ¼ ln ðfsþ ¼4:047 þ 0:073X 1 0:024X 2 0:014X 3 þ 0:111X 1 X 1 0:003X 1 X 2 þ 0:084X 2 X 2 þ 0:012X 1 X 3 0:031X 3 X 3. These models were the most appropriate for calculating WVP and film solubility, giving a statistically nonsignificant lack of fit and explaining 88.08% and 83.22% of the variability, respectively. Further statistical analysis (not presented) revealed that, heating temperature and heating time had a significant overall effect on all the responses. The and heating temperature of film solution had the most significant effect on TS, elongation at break, WVP and a values, while heating time was the least effect. OP, E ab seemed to be most affected by all three process variables. Film solubility, protein solubility, hue angle, chroma, were most affected by. However,, heating temperature and heating time did not have significant effect on L value Tensile strength and elongation at break The and heating temperature of the film solution were the most important factors affecting the mechanical properties, while heating time had lesser effect. Contour plots of TS and elongation at break as affected by and heating temperature were given in Fig. 1. Depending upon the film conditions, TS and elongation at break showed a high variation between 1. and 3.02 MPa and 8.50% and 14.72%, respectively. Table 2 Regression coefficients and analysis of variance of the second order polynomial for 13 response variables Coefficient Tensile strength Elongation at break Water vapor permeability Oxygen permeability Film solubility Protein solubility Color L a b DE ab Hue angle Chroma Y 1 Y 2 Y 3 Y 4 Y 5 Y 6 Y 7 Y 8 Y 9 Y 10 Y 11 Y 12 b ko Linear b k ** * 5.29** 1.38* 0.21** 2.41** 2.64** 2.87* 2.44* b k ** * ** b k ** * Interaction b k ** * ** 0.11* * 1.75* 0.21 b k ** * ** * * b k ** ** * * Quadratic b k ** ** b k b k * Variability Explained (R 2 ) F Probability of F *Significant at 5% level, **Significant at 1% level.

6 Tensile strength Elongation at break (A) (B) (C) Fig. 1. Contour plots showing response behavior of and heating temperature of film solutions under constant heating time. The numbers inside the contours represent TS (kpa) and elongation at break (%) of films at given heating; (A) ¼ 10 min, (B) ¼ 20 min and (C) ¼ 30 min. Comparing within the same heating temperature of film solutions, the results demonstrated that, TS increased as of film solutions increased. This result implied that higher of film solutions induced formation of resistant films. Banker (19) reported that played an important role in protein films made from watersoluble materials. At alkaline away from the isoelectric point of 3.5 (Bourtoom, Jantawat, Sanguandeekul, & Chinnan, 2002), denaturation of proteins was promoted and resulted in unfolding and solubilizing of the proteins. During solubilization, the cohesive forces between the protein macromolecules were neutralized by complexing with the solvent molecules (Banker, 19). In general, functions of polymers were related to

7 solution properties which further influenced the film characteristics. The same charged groups repelled each other and produced a stretching of the polymer chain when functional groups on a linear polymer ionized during dissolution. This phenomenon, facilitated molecule orientation and fine-stranded network (Banker, 19). The resulting interaction between polymers may have been responsible for this result. Anker, Stading, and Hermansson (2000) reported that, when the of the film solutions from X -lactoglobulin was increased above 8, SH/S-S interchange reactions or thiol/thiol (SH/SH) oxidations could occur upon heating and intermolecular disulfide (S S) bonds formed. The highest TS value was obtained at about 10.0 (Fig. 1). However, increasing of film solutions higher than 10.0 resulted in decrease of TS, by the reason that strongly repulsive force that occurred between negative (extreme ) charges along the protein chains could have decreased the occurrence of molecular associations within the protein matrix (Rhim, Gennadios, Weller, & Hanna, 2002). Gennadios, Brandenburg, Weller, and Testin (1993) studied the effect of on soy protein isolate film and found that highly alkaline condition (412) inhibited soy protein isolate film formations. The weakest film was obtained at the lower and higher end of the studied, most likely due to less protein protein interaction. The TS was enhanced as heating temperature of film solutions increased from 60 to 80 1C. This result demonstrated that, TS increased from 1. to almost 3.0 MPa when heating temperatures of film solutions increased from 60 to 80 1C. This might be due to the fact that higher heating temperature induced protein denaturation and resulted in increase in the number and/or a better localization of bonds between protein chains. According to the contour plots, the experimental condition involving higher of both (10.0) and heating temperature resulted in higher film formations and high TS of the formed films. It may be noted that the protein denaturation depends not only upon temperature but also heating time. However, the results showed that the heating temperature affected TS more than the heating time. The elongation at break value was also most affected by the and heating temperature of film solution. All linear, quadratic and interaction terms for, heating temperature and heating time were significant. The contour plots of elongation at break (Fig. 1) indicate that water-soluble fish protein films exhibit properties of an elastic material with elongation at break values between 8.50% and 14.72% and showed the highest elongation at break when lower and higher heating temperature of film solutions were employed. An increase in elongation at break of heat-induced edible films was suggested to be due to an increased number of intermolecular disulfide (SS bond) bonds (Shimada & Cheftel, 1988). Prolonged heating time, however, resulted in increase in elongation at break. The experiments showed that TS and elongation at break of films are almost inversely related Water vapor permeability The main factor influencing WVP of films made from water-soluble fish proteins film are and heating temperature of film solution. The contour plots (Fig. 2) were characteristics of the effects of these variables and showed that the WVP value was the highest at around 9.5 ( g mm/m 2 d kpa) and tended to decline when of film solutions reached to 10.0 ( g mm/m 2 d kpa). However, the WVP increased again when the was adjusted to 10.5 ( g mm/m 2 d kpa). These results could arise from the fact that, at higher, protein can denature, unfold and solubilize, facilitating favorable molecule orientation. This phenomenon facilitated favorable molecule orientation and formation of intermolecular disulfide bond by thiol disulfide interchange and thiol oxidation reactions. The function of disulfide bonds on protein coagulation during drying of soymilk was studied by Fukushima and Van Buren (19). Thiol disulfide interchanged via thiol oxidation also implicated in whey protein gelation (Shimada & Cheftel, 1988; Donovan & Mulvihill, 19). Extreme (410.0) of film solutions as in this study might inhibit the water-soluble fish protein film formation. Most likely, strong repulsive forces between highly negative charges prevented protein molecules form associating and forming the films. The highest WVP was observed at the lowest and the highest of this study. The WVP of edible films was affected by heating temperature of the film solution as well. Basically, proteins must be denatured in order to form more extended structures that are required for film formations. Once extended, protein chains can associate through hydrogen, ionic, hydrophobic and covalent bondings. The chain-to-chain interaction that produces cohesive films is affected by the degree of chain extension and the nature and sequence of amino acid residues. Uniform distribution of polar, hydrophobic, and/or thiol groups along the polymer chain increased the likelihood of the respective interactions (1). The result of this experiment showed that increasing of heating temperature of film solutions (60 1C) resulted in lower WVP (Fig. 2). The thermal energy might promote a greater cross-link between protein protein chains resulting in a tight and compact protein network and structure. Shimada and Cheftel (1988) reported that the first step of ovalbumin aggregation involved the formation of SS bonds and the exposure of hydrophobic groups, and that, during further heating, ovalbumin was then polymerized and the intermolecular sulfhydryl/ disulfide (SH/SS) exchanged to form a higher protein net work structure. However, extremely heating

8 Water vapor permeability Oxygen permeability (A) (B) (C) Fig. 2. Contour plots showing response behavior of and heating temperature of film solutions under constant heating time. The numbers inside the contours represent water vapor permeability (g mm/m 2 d kpa) and oxygen permeability (cm 3 mm/m 2 d kpa) of film at given heating time: (A) ¼ 10 min, (B) ¼ 20 min and (C) ¼ 30 min. temperature (4 1C) of film solutions provided an increase in WVP, most likely due to increase in protein denaturation and the protein precipitation that obstructed the film formation. The highest WVP of edible films was found at lowest and highest heating temperature of film solutions. The effect of heating time of film solutions on WVP of edible film showed a similar trend with that of the heating temperature Oxygen permeability The main factors to influence OP of water-soluble fish protein film were, heating temperature and heating time of film solution. The contour plots (Fig. 2) showed that the OP value increased when the of film solution was 9.5 and decreased when of film forming solution was increased to 10.0; however, OP rose again when

9 of film solution was increased above The decrease in OP with increase in of film solution (10.0) may be explained by the incidence of higher protein protein interaction. Upon comparing the heating temperature of the film solution with the effects of, it was interesting to note that the OP decreased as heating temperature was increased from 60 to 1C (Fig. 2). It was believed that higher heating temperature of film solution would form greater cross-links resulting in a tight and compact protein network and structure. However, when the heating temperature of the film solution was increased above 1C, it resulted in increased OP. It was probably due to increased protein denaturation leading to protein precipitation where the film could not be formed well. The highest OP occurred at the lowest and highest temperature. OP was also affected by heating time, the Film solubility Protein solubility Temperature ( C) Temperature ( C) (A) Temperature ( C) Temperature ( C) (B) Temperature ( C) Temperature ( C) (C) Fig. 3. Contour plots showing response behavior of and heating temperature of film solutions under constant heating time. The numbers inside the contours represent film solubility (%) and protein solubility (%) of film at a given heating time: (A) ¼ 10 min, (B) ¼ 20 min and (C) ¼ 30 min.

10 OP contour plots (Fig. 2) indicate that higher heating time (20 min) produced lower OP than lower heating time (10 min); however, when heating time increased to 30 min, OP increased again. It was believed that the higher heating time (30 min) of film solution might have caused higher protein denaturation resulting in precipitation. The contour plots (Fig. 2) were characteristic of the strong interaction between the and heating temperature of the film solution (Po0:05) and showed that the highest OP value resulted with highest and highest heating temperature of the film solution. Additionally, at lowest and lowest heating temperature of the film solution, a similar trend was observed. At 410.0, increase in the heating temperature of the L* a* (A) (B) (C) Fig. 4. Contour plots showing response behavior of and heating temperature of film solutions under constant heating time. The numbers inside the contours represent L and a values of film at given heating time: (A) ¼ 10 min, (B) ¼ 20 min and (C) ¼ 30 min.

11 film solution increased the OP. At low heating temperature of the film solution, the levels o10.0 and both increased OP. Films formed under medium ( 10.0) and medium heating temperature ( 1C) produced very low OP. Thus, it was concluded that to maintain relatively low OP, a of 10.0 and heating temperature of 1C were required Film and protein solubility The of film solution significantly affected the film and protein solubility. The contour plots of film and protein solubility showed that both had increased significantly at each heating temperature when the of the film solution was increased (Fig. 3). It was (A) b* E* ab (B) (C) Fig. 5. Contour plots showing response behavior of and heating temperature of film solutions under constant heating time. The numbers inside the contours represent b and DE ab values of film at given heating time: (A) ¼ 10 min, (B) ¼ 20 min and (C) ¼ 30 min.

12 observed that water-soluble fish protein films showed higher solubility when the of the film solution was Increased soluble matter may be due to increased protein solubility. Higher of film solution (410.0), with enhanced dispersion capability in water might have resulted in loosing the film structure, causing dissolution of the nonprotein materials (Gnanasambandam, Hettiarachy, & Coleman, 1997). It was observed that film and protein solubility were lowest at around 10.0, most likely due to better film formation. The contour plots of the effect of heating temperature of the film solution on film and protein solubility are shown in Fig. 3. Comparing them at the same of the film solution, demonstrates that an increase in heating temperature of film solution from 60 to 1C resulted in a decrease in film and protein Hue angle Chroma (A) (B) (C) Fig. 6. Contour plots showing response behavior of and heating temperature of film solutions under constant heating time. The numbers inside the contours represent hue angle and chroma values of film at given heating time: (A) ¼ 10 min, (B) ¼ 20 min and (C) ¼ 30 min.

13 solubility. Roy, Weller, Gennadios, Zeece, and Testin (1999) reported that wheat gluten film and protein solubility in water decreased (Po0:05) as heating temperature of film solutions increased. This was attributed to more pronounced heat-induced protein denaturation at higher temperatures. Heating denature (unfolds) protein chains, exposing previously buried groups such as hydrophobic and sulfhydryl (SH) groups (Fukushima & Van Buren, 19; Mine, Noutomi, & Haga, 1990). However, when the heating temperature of the film solution was raised higher than 1C, it yielded an increase in film and protein solubility. Heating time of the film solution in this study did not significantly affect the film and protein solubility Film color The color of films was most affected by of film solution, while heating temperature and heating time had little effect. Film formed at lower and heating temperature was a lighter yellow than the film formed at higher and heating temperature. Instrumental color parameters L showed a little increase with increase in and heating temperature of film solution (Fig. 4); however, value b markedly increased with increase in and heating temperature, and this made the film appear more yellowish (Fig. 5). At alkaline, proteins were observed to form complex polyphenolic compounds. Such complexes might have contributed to discoloration of films prepared at higher (Gnanasambandam et al., 1997). The value a increased as of film solution increased, which reversed with change in heating temperature (Fig. 4) resulting in a reddish yellow film. The main factors to influence DE ab of watersoluble fish protein film were, heating temperature and heating time of film solution (Fig. 5), while hue angle and chroma were most influenced by the of the film solution (Fig. 6). From the model where DE ab was plotted against and heating temperature of film solution at various heating times (Fig. 5), as can be seen that the and heating temperature of film solution were more influential than heating time. The hue angle decreased when the and heating temperature of the film solution were increased (Fig. 6). The contour plots of chroma (Fig. 5) show an increase in the chroma value as a result of an increase in the of the film solution which was reflected in the yellowish film appearance. 4. Conclusions The impact of and heating temperature of the film solution was most significant, overall, on the film s properties than heating time. TS and elongation at break were highest at 10.0 and heating temperature of 1C, while water vapor permeability and oxygen permeability were at their lowest. There was a direct correlation between the film and proteins solubility on one hand, and heating temperature of the film solution on the other, which reversed with change in of the film solution. The film color was darker and more yellowish with increase in. Acknowledgements This research was conducted under the Thai program for International Collaboration with the Department of Food Science and Technology, University of Georgia, USA, funded by the Royal Thai Government. The authors are very grateful to Mr. Glenn Farrell for his technical assistance and Mrs. Vijaya Mantripragada for her editorial suggestions and general laboratory assistance. References Anker, M., Stading, M., & Hermansson, A. M. (2000). Relationship between the microstructure and the mechanical and barrier properties of whey protein films. Journal of Agricultural and Food Chemistry, 48, ASTM (1993a). Standard practice for conditioning plastics and electrical insulating materials for testing: D (Reproved 1990). In: ASTM. Annual book of American standard testing methods (Vol. 8.01, pp ). Philadelphia, PA. ASTM (1993b). Standard test method for water vapor transmission rate through plastic film and sheeting using a modulated infrared sensor: D In: ASTM. Annual book of American standard testing methods (Vol , pp ). Philadelphia, PA. ASTM (1993c). Standard test method for oxygen gas transmission rate through plastic film and sheeting using coulometric sensor: D (Reproved 1988). In ASTM. Annual book of American standard testing methods (Vol , pp ). Philadelphia, PA. ASTM (1995). Standard test methods for tensile properties of thin plastics sheeting: D8-91. In ASTM. Annual book of American standard testing methods (Vol. 8.01, pp ). West Conshohocken, PA. Banker, G. S. (19). Film coating theory and practice. Journal of Pharmacy Pharmacology, 55, Bourtoom, T., Jantawat, P., Sanguandeekul, R., & Chinnan, M. S. (2002). Recovery of proteins from surimi wash-water (abstract). In Institute of food technologists. Annual meeting book of abstracts (abstract no. 76E-39). Anaheim: CA. Box, G. E. P., & Draper, N. R. (1987). A basis for the selection of response surface design. Journal of American Statistical and Association, 54, Cochran, W. G., & Cox, G. M. (1957). Experimental designs. NewYork: Wiley. Cox, D. R. (19). Planning of experiments. New York: Wiley. Donhowe, I. G., & Fennema, O. R. (1993). The effects of plasticizers on crystallinity, permeability, and mechanical properties of methylcellulose films. Journal of Food Processing and Preservation, 17,

14 Donovan, M., & Mulvihill, D. M. (19). Thermal denaturation and aggregation of whey proteins. Journal of Food Science and Technology, 11, Fukushima, D., & Van Buren, J. (19). Mechanisms of protein insolubilization during the drying of soy milk. Role of disulfide and hydrophobic bonds. Cereal Chemistry, 47, Gennadios, A., Brandenburg, A. H., Weller, C. L., & Testin, R. F. (1993). Effect of on properties of wheat gluten and soy protein isolate films. Journal of Agricultural and Food Chemistry, 41, Gnanasambandam, R., Hettiarachy, N. S., & Coleman, M. (1997). Mechanical and barrier properties of rice bran films. Journal of Food Science, (2), Iwata, K., Ishizaki, S., Handa, A., & Tanaka, M. (2000). Preparation and characterization of edible film from fish water-soluble proteins. Fisheries Science,, Jangchud, A., & Chinnan, M. S. (1999). Peanut protein film as affected by drying temperature and of film forming solution. Journal of Food Science, 64(1), Kester, J. J., & Fennema, O. R. (1986). Edible films and coatings: A review. Food Technolology, 40(12), Lin, T. M., Park, J. W., & Morrissey, M. T. (1994). Recovered protein and reconditioned water from surimi wash process waste. Journal of Food Science, 60(1), 4 9. Lowry, O. H., Rosebrough, N. J., Farr, A. L., & Randall, R. J. (1951). Protein measurement with Folin phenol reagent. Journal of Biological and Chemistry, 193(1), Mine, Y., Noutomi, T., & Haga, N. (1990). Thermally induced change in egg white proteins. Journal of Agricultural and Food Chemistry, 38, Nelson, K. L., & Fennema, O. R. (1991). Methylcellulose films to prevent lipid migration in confectionery products. Journal of Food Science, 56, Pacheo-Aguilar, R., Creawford, D. P., & Lampila, L. E. (1989). Procedures for the efficient washing of minced whiting (Merluccius products) flesh for surimi production. Journal of Food Science, 54(2), Rawatt, R. J. (1993). The plastic waste problem. Chemical Technology, 23, Rhim, J. W., Gennadios, A., Weller, C. L., & Hanna, M. A. (2002). Sodium dodecyl sulfate treatment improves properties of cast films from soy protein isolate. Industrial Crops and Products, 15, Roy, S., Weller, A., Gennadios, A., Zeece, M. G., & Testin, R. F. (1999). Physical and molecular properties of wheat gluten films casting from heated film-forming solutions. Journal of Food Science, 64(1), SAS. (1996). SAS/STAT user s guide: Statistic: Version North Carolina: SAS Institute Inc. Shimada, K., & Cheftel, J. C. (1988). Texture characteristics, protein solubility, and sulfhydryl group/disulfide contents of heat-induced gels of whey protein isolate. Journal of Agricultural and Food Chemistry, 36, Thompson, D. R. (19). Response surface experimentation. Journal of Food Processing and Preservation, 6,