Effect of chitosan and carrageenan-based edible coatings on post-harvested longan (Dimocarpus longan) fruits

ABSTRACT

The effect of chitosan/carrageenan and glycerol as edible coating materials in preserving the fresh longan fruits stored at ambient temperature was evaluated. The concentration of coating components played an important role in the process of controlling quality changes and quantity losses. Changes in fruits’ weight loss, respiration rate and color were used as a measure of the coating’s effectiveness. Results have shown that increase in the chitosan or carrageenan concentration led to significant (p < 0.05) decreases in water loss, weight loss and respiratory rate in coated fruits. However, in carrageenan-coated fruits, high increase in concentration (> 1.19%) of the carrageenan resulted in slight increases in water and weight losses. From the multiple response optimization analysis, a combination of 1.29% (w/v) chitosan with 0.42% glycerol and 1.49% (w/v) carrageenan with 0.03% glycerol were predicted to give the desired coating because they were able to preserve the longan by showing minimal quality changes and quantity losses.

RESUMEN

El presente estudio se propuso evaluar el efecto de materiales de recubrimiento comestibles a base de chitosán/carragenano en la conservación de frutos de longan frescos, almacenados a temperatura ambiente. La concentración de los componentes de recubrimiento desempeñó un rol importante en el proceso de control de los cambios de calidad y las pérdidas de cantidad. Los cambios en el peso, la tasa de respiración y el color fueron utilizados como medida de la eficacia del recubrimiento. Los resultados mostraron que al aumentar la concentración de quitosano o carragenano se redujo significativamente (p < 0.05) la pérdida de agua, la pérdida de peso y la tasa respiratoria en los frutos con recubrimiento. Sin embargo, en los frutos cubiertos con carragenano, el elevado aumento en su concentración (> 1.19%) produjo un ligero aumento del agua y pérdida de peso. El análisis de optimización de respuesta múltiple permitió pronosticar que la combinación de 1.29% (w/v) quitosano con 0.42% de glicerol y 1.49% (w/v) de carragenano con 0.03% glicerol producirían el recubrimiento deseado, pues pudieron conservar el longan con mínimos cambios en la calidad y mínimas pérdidas de cantidad.

KEYWORDS:

• Fresh fruit
• preservation
• weight loss
• respiration rate
• response surface methodology

PALABRAS CLAVE:

• Fruta fresca
• conservación
• pérdida de peso
• tasa de respiración
• metodología de superficies de respuesta

Abbreviations

CO2	=	Carbon dioxide
°C	=	Celsius
CCD	=	Central composite design
Da	=	Daltons
ppm	=	Part per million
Pa	=	Pascal
R2	=	Regression
RSM	=	Response surface methodology
rpm	=	Revolutions per minute

Introduction

Longan fruit (Dimocarpus longan) is a non-climacteric tropical and subtropical fruit. It has high commercial value due to its flavor and nutritional content in vitamin C, minerals and bioactive polyphenols (Wall, 2006; Yang, Jiang, Shi, Chen, & Ashraf, 2011). However, harvested longan fruit has a very short shelf life of only 3–4 days under ambient temperature (Chen, Chen, Hong, & Lin, 2000). Pericarp browning and microbial decay are the major problems in longan fruit. The most popular commercial method for preventing fruit browning and fruit decay is by sulfur dioxide fumigation (Jiang, Zhang, Joyce, & Ketsa, 2002). However, sulfur dioxide fumigation leaves sulfite residues and toxic ingredients that are harmful to human (Jiang et al., 2002; Sodchit, Kongbangkerd, & Phun, 2008). Thus, alternative methods with no toxic substances are needed for food safety issue. In recent years, edible coatings appear to be one of the most innovative ways to improve the commercial shelf life of fruits.

Edible coatings are materials that forms thin layer on the surface of food. Edible coatings can be protein, polysaccharide or lipid based (Dhall, 2013). By using edible coatings, the storability of perishable crops can be prolonged (Dhall, 2013). Edible coating applied on fresh fruits is able to reduce quality changes and slow down quantity losses, e.g., moisture loss by controlling and modifying the internal atmosphere of the individual fruits (Turhan, 2009). It functions by giving a selective barrier to moisture, oxygen and carbon dioxide, preserving fruits and vegetables, improving texture, mechanical properties and preventing flavor loss (Dhall, 2013).

Chitosan is a high molecular weight cationic polysaccharide (Hirano et al., 1990). Chitosan is obtained through de-acetylation reaction of chitin. Commercially, chitin is available in large amounts as it can easily be extracted from shells of prawns, crabs, etc., which are by-products of shellfish industry (Kumari & Rath, 2014). Chitosan is popular among the edible coating materials as it is biodegradable, biocompatible and highly resistant to microbial attack (Aider, 2010). Moreover, it is non-toxic, safe (Hirano et al., 1990) and increases immunoglobulin M production in human hybridoma cells (Maeda, Murakami, Ohta, & Tajima, 1992). Meanwhile, carrageenan is an anionic linear polysaccharide that is derived from red seaweed. There are three kinds of carrageenan such as kappa, iota and lambda with different numbers and positions of sulfate groups on the galactose dimer. Carrageenan is formed by gelation through a process of moderate drying. After evaporation of solvent, the polysaccharide double helices will form a three-dimensional network, which subsequently forms a solid film (Karbowiak, Debeaufort, & Voilley, 2007).

Plasticizer is a major component in biodegradable films or polymer. Plasticizers are generally small molecules, such as glycerol, that intersperse among the polymer chains. The application of plasticized biodegradable film is necessary as the biodegradable film is easily brittle. The brittleness in biodegradable film is due to excessive cohesive force (Azeredo, 2012). Cohesion is the attractive force between molecules of film components, influencing the mechanical strength of films. During film formation, the incompatibility of fruits with the main biopolymer in the film-forming dispersion increases the cohesion and consequently decreases film strength (Azeredo, 2012). This can be overcome by adding in plasticizers. Plasticizer is able to disrupt hydrogen bonding and increase flexibility of the film (Baldwin & Baker, 2002; Sorbal, Menegalli, Hubinger, & Roques, 2001).

Response surface methodology (RSM) is a statistical and mathematical method used to determine the optimum levels of two or more treatment variables (Lasekan & Abbas, 2011). In the design of experiments, the responses that are influenced by several independent variables are optimized and the relationship between the variables and responses is evaluated and modeled. Numerical and graphical multiple responses optimization procedures are used to predict the best compromise based on the fitted quadratic response surface models constructed in RSM. Numerical optimization is used to determine one point or more that will maximize the function, while, in graphical optimization with multiple responses, it determines the optimum region where requirements simultaneously meet the proposed criteria. In the optimization procedures, different desirability functions can be applied to maximize, minimize or assign a target value, depending on each particular response. D, which represents overall desirability, is the overall objective to be optimized based on all responses. The scale of desirability ranges from 0 to 1; higher values indicate higher degree of satisfactory of responses.

It is important to derive coating formulation with optimum concentration of chitosan/carrageenan and glycerol to effectively increase stability of fresh fruits in terms of weight loss, respiration rate and surface color. Therefore, this study aimed at optimizing chitosan and carrageenan-based edible coating formulations for postharvest longan preservation.

Materials and methods

Materials

Mature longan fruits (D. longan Lour.) cv. Diamond was harvested from an orchard located in Sepang, Selangor, Malaysia, and immediately transported to University Putra Malaysia within 2 h. Fruits were selected for uniformity of size, shape and color in which the blemished and diseased fruits were discarded.

Low molecular weight chitosan (Sigma-Aldrich, USA) and kappa carrageenan (Sigma-Aldrich, USA) were used as polysaccharide-based edible coatings. The molecular weight of chitosan was 190,000–310,000 Da, and the de-acetylation degree was 75–85%. Reagent grade glycerol (Sigma-Aldrich, USA) was used as plasticizer. Pure organic sunflower oil (Melrose; purchased at LOHAS) was added into the coating formulation as an emulsifier and lipid source.

Samples and edible coating solutions preparation

Chitosan solution was prepared according to the method of Jiang, Li, and Jiang (2005) with minor modification. Chitosan (0–2% w/v) powder was dissolved in aqueous solution of glacial acetic acid (0.5% v/v) (Reagent grade, Merck, USA) at room temperature and stirred vigorously using magnetic stirrer (Bibby HB502, UK). The solution was then adjusted to pH 5.6 with 0.1 M sodium hydroxide (AR grade, Friendemann Schmidt, Germany). Glycerol (0–2% v/v) was added to the solution. The solution was made up to 500 ml. Sunflower oil (0.025% v/v) was lastly added as lipid source and the solution was homogenized at 26,000 rpm using a homogenizer (Heidolph DIAX 900, Germany) for 5 min and degassed ultrasonically in an ultrasonic bath (Delta DC150H, Taiwan).

Carrageenan coating solution was prepared based on the method of Ribeiro, Vicente, Teixeira, and Miranda (2007) with minor modification. Carrageenan powder (0–2% w/v) was dissolved in distilled water, with heating at 80°C for 10 min and stirred using magnetic stirrer. The solution was adjusted to pH 5.6 using 5% w/v anhydrous citric acid (CP grade, R&M Chemicals, UK). Glycerol (0–2% v/v) was added to the solution. The solution was top up to 500 ml. Then, sunflower oil (0.025% v/v) was added and the solution was homogenized at 26,000 rpm using a homogenizer (Heidolph DIAX 900, Germany) for 5 min and degassed ultrasonically in an ultrasonic bath (Delta DC150H, Taiwan).

For sample treatment, the fresh and matured longan samples were initially washed by using 100 ppm sodium hypochlorite (CP grade, R&M Chemicals, UK) and air-dried for 30 min. After that, they were dipped in the coating solution for 2 min. The coated samples were permitted to drain and dried for 3 h at ambient temperature. Samples were packed in polyethylene bags and stored at 28°C for 4 days until analysis.

Analytical methods

For each treatment, the measurement of weight loss was replicated thrice using 10 fruits in each replication. The weight loss was determined gravimetrically using analytical balance (Sartorius TE214S, Switzerland). The weight differences after storage compared to initials weight were expressed in percentage (Azarakhsh, Osman, Tan, Mohd Ghazali, & Mohd Adzahan, 2012).

The respiratory rate was determined by headspace oxygen/carbon dioxide analyzer (Illinois Instruments Model 6600, USA). Ten coated longan fruits (approximately 100 g) were placed in glass jar and incubated at room temperature of 28°C for 1 h. The glass jar had air-tight screw caps and rubber septum to allow headspace sampling. The results were calculated and expressed based on carbon dioxide released (ml CO₂/kg h) (Azarakhsh et al., 2012). For each treatment, three replicates were tested.

The surface lightness was determined using Hunter Lab UltraScan Pro spectrophotometer attached with EasyMatch QC software (Hunter Associate Laboratory Inc., USA). The spectrophotometer has CIE LAB color space, pulsed xenon lamps as light sources. Observer degree of 10 and reflectance specular component were included. The spectrophotometer was calibrated with black calibration light trap and a white tile. For each treatment, peel color of 10 individual fruits per replicates (total of three replicates) were measured. Readings were taken from stem end, mid region and blossom end of each fruits. Readings of L value were taken to represent the lightness of the fruit (0 = black, 100 = white) (Hamzah, Osman, Tan, & Ghazali, 2013).

Experimental design and statistical analysis

Firstly, RSM central composite design (CCD) scheme has been selected to design the experiment with two design variables k = 2, polysaccharide coating (chitosan or carrageenan) concentration and plasticizer (glycerol) concentration. The CCD scheme has 14 runs with 6 center points (Table 2), and the levels of variables are stated in Table 1. This design was done by using Minitab 16 software (Minitab, Inc., USA), and all of the following statistical analyses were performed at a significance level of 0.05.

Table 2. The matrix of central composite design (CCD) and experimental data obtained for response variables studied in chitosan and carrageenan-based edible coating.

Tabla 2. Matriz del diseño compuesto central (CCD) y los datos experimentales obtenidos para las variables de respuesta estudiadas en los recubrimientos comestibles a base de quitosano y carragenano

				Response variables
		Independent variables		Chitosan-based			Carrageenan-based
Run	Block	Chitosan/carrageenan (%, w/v)	Glycerol (%, w/v)	Weight loss (%)	Respiratory rate(ml CO2/kg h)	Surface lightness (L value)	Weight loss (%)	Respiratory rate(ml CO2/kg h)	Surface lightness (L value)
1	1	2	2	2.15	59.74	44.15	3.26	128.44	50.00
2	1	1	1	1.72	96.44	45.82	2.83	102.98	50.16
3	1	2	0	1.74	62.40	44.73	2.57	84.59	51.27
4	1	0	0	2.05	116.20	46.87	2.66	112.25	47.82
5	1	0	2	2.81	124.56	46.23	3.79	85.73	49.66
6	1	1	1	2.03	92.26	46.23	2.43	102.37	50.35
7	1	1	1	1.70	92.72	46.29	2.50	100.32	49.90
8	2	2	1	1.82	64.83	44.44	3.49	119.78	50.44
9	2	1	1	1.73	92.80	46.36	2.83	100.47	50.94
10	2	1	1	1.85	97.58	46.03	2.80	87.55	50.66
11	2	1	2	2.50	90.36	45.61	3.70	100.70	49.98
12	2	0	1	1.75	127.76	47.14	4.61	108.15	46.78
13	2	1	1	2.04	95.61	46.42	2.37	102.45	49.98
14	2	1	0	1.80	93.02	46.20	3.20	67.34	49.99

Table 1. Factor level and experimental domain applied to optimized the chitosan/carrageenan coating experimental conditions.

Tabla 1. Nivel de factores y dominio experimental aplicados para optimizar las condiciones experimentales de recubrimiento con quitosano/carragenano.

Factor	Experimental domain
	−1(−α)	0	1(α)
Chitosan/carrageenan (w/v)	0	1	2
Glycerol (w/v)	0	1	2

Secondly, the relationship between the variables and responses was evaluated using a second-order polynomial regression response model as follows:

(1) $Y={\beta }_0+\sum {\beta }_{\mathrm{i}}{\mathrm{x}}_{\mathrm{i}}+\sum {\beta }_{\mathrm{i}\mathrm{i}}{{\mathrm{x}}_{\mathrm{i}}}^2+\sum {\beta }_{\mathrm{i}\mathrm{j}}{\mathrm{x}}_{\mathrm{i}}{\mathrm{x}}_{\mathrm{j}}$(1)

In the model, Y is the response; β₀, β_i, β_ii, β_ij are intercept, linear, quadratic and interaction coefficients, respectively.

Thirdly, numerical and graphical multiple response optimization procedures were used to predict the best compromise based on the fitted quadratic response surface models constructed in RSM. One-sided linear desirability function was used that was weight loss and respiration rate minimization and surface lightness maximization. The optimum coating formulation was based on all responses. In the final verification procedures, the fitted values predicted by the model were compared with experimental data using t-test to verify the adequacy of the regression model.

Results and discussion

Effect of chitosan, carrageenan and glycerol contents on weight loss

Weight loss is just like water loss because other components in fruits like aroma, flavor or gas products of respiration are practically undetectable in terms of weight (Olivas & Barbosa-Cánovas, 2005). The loss of water in fruits is mainly caused by transpiration and respiration processes (Zhu, Wang, Cao, & Jiang, 2008) and thus brings wrinkling, tissue softening, loss of brightness and making fruits more susceptible to deterioration (Nunes & Emond, 2007). From the obtained results presented in Table 2, weight loss of chitosan-based-coated longan varied from 1.70% to 2.81% while for the carrageenan-based-coated longan fruit, the weight loss varied from 2.37% to 4.61%. To aid visualization, the contour plots for the weight loss as a function of coating agents (chitosan and carrageenan) are shown in Figures 1(a) and 2(a). Figure 1(a) shows the contour plot, where the water loss decreased as the chitosan concentration was increased. Conversely, increase in glycerol led to an increase in percentage water loss. A similar trend was also observed in fruits coated with carrageenan (Figure 2(a)). However, in carrageenan, high increase in concentration >1.19% led to slight increase in water loss (Figure 2(a)). The impacts of chitosan and carrageenan edible coatings in preventing water loss in fruits have been reported (Bico, Raposo, Morais, & Morais, 2009; Moraes, Fagundes, Melo, Andreani, & Monteiro, 2012; Petriccione et al., 2015). The current study has shown that chitosan-based films were able to reduce water loss than carrageenan-based films. This observation was expected because chitosan tends to have relatively low water vapor permeability (12.4 × 10¹⁰ g m⁻¹ s⁻¹ Pa⁻¹) compared to carrageenan (23.5 × 10¹⁰ g m⁻¹ s⁻¹ Pa⁻¹) (Desobry & Debeaufort, 2011). For the variability in glycerol concentration, its increment led to increase in weight loss. This result is supported by the findings of Brindle and Krochta (2008), Jongjareonrak, Benjakul, Visessanguan, and Tanaka (2006), and Möller, Grelier, Pardon, and Coma (2004) who reported that the plasticizer generally increased water permeability of the edible coating and therefore the glycerol was added at levels that were able to achieve the desired characteristic without significantly decreasing the barrier properties of mass transfer. Glycerol, as a plasticizer, would decrease the intermolecular attraction between polymer chains, thus, facilitating the penetration of water vapor molecules (Salgado, Ortiz, Musso, Giorgio, & Mauri, 2015).

Figure 1. Contour plots as a function of chitosan and glycerol concentration on the (a) weight loss, (b) respiratory rate and (c) surface lightness of coated longan fruits.

Figura 1. Gráfico de contorno como función de la concentración de quitosano y glicerol en la (a) pérdida de peso, (b) tasa respiratoria y (c) claridad de la superficie de frutos de longan recubiertos.

Figure 2. Contour plots as a function of carrageenan and glycerol concentration on the (a) weight loss, (b) respiratory rate and (c) surface lightness of coated longan fruits.

Figura 2. Gráfico de contorno como función de la concentración de carragenano y glicerol en la (a) pérdida de peso, (b) tasa respiratoria y (c) claridad de la superficie de frutos de longan recubiertos.

Effect of chitosan, carrageenan and glycerol contents on respiratory rate

Rate of respiration is an important factor in the determination of the quality of postharvest fruits. Therefore, when the rate of respiration of fruits is increased, the postharvest lives of the fruits become shortened (Aked & Jongen, 2002). From the obtained results shown in Table 2, respiratory rate of chitosan-based-coated longan varied from 59.74 to 127.76 ml CO₂/kg h. Meanwhile for the carrageenan-based-coated longan fruits, the respiratory rate varied from 67.34 to 128.44 ml CO₂/kg h (Table 2). Contour plots for respiratory rate are illustrated in Figures 1(b) and 2(b) for better visualization. From Figure 1(b), respiratory rate decreased when chitosan concentration was increased. However, different levels of glycerol concentrations maintained the respiratory rate at a low level. For carrageenan-based coating, the respiratory rate showed a similar trend (Figure 2(b)). However, when the glycerol concentration was increased, the respiratory rate also increased. The observed decrease in respiratory rate by chitosan or carrageenan has been reported in previous studies (Ghaouth, Arul, Ponnampalam, & Boulet, 1991) in which semi-permeable-coated fruits modified the endogenous carbon dioxide and oxygen and subsequently lowered the respiratory rate of fresh produces. Both chitosan and carrageenan were able to produce an internally modified atmosphere with its selective oxygen and carbon dioxide permeability, thus isolating the coated products from the environment. As a result, the respiratory rate was reduced, and this also slowed down the postharvest metabolism of the fruits, thereby increasing its shelf life (Fernández-Pan, Ignacio, & Caballero, 2011). In addition, the combination of chitosan/carrageenan and glycerol produced different oxygen permeability as chitosan plus glycerol’s oxygen permeability (0.009 × 10⁻¹⁵ g m⁻¹ s⁻¹ Pa⁻¹) was relatively lower than carrageenan plus glycerol (0.72 × 10⁻¹⁵ g m⁻¹ s⁻¹ Pa⁻¹) (Desobry & Debeaufort, 2015).

Effect of chitosan, carrageenan and glycerol contents on surface lightness

Color and appearance are quality attributes that are commonly used by consumers to judge a product’s freshness. In longan fruits, pericarp darkening limits its shelf life by reducing its visual appearances and qualities, and consequently decreasing its commercial values. From the obtained results shown in Table 2, surface lightness of chitosan-based-coated longan varied from L value 44.15 to 47.14. Meanwhile for the carrageenan-based-coated longan fruit, L value varied from 46.78 to 51.27. The contour plots are shown in Figures 1(c) and 2(c). In Figures 1(c), it can be seen that the increase of chitosan or glycerol decreased the surface lightness. However, there was a difference in the case of carrageenan-based coating. The increase of carrageenan increased the lightness while the increase in glycerol decreased the lightness (Figure 2(c)). This result was most probably due to the differences in the natural optical properties of chitosan and carrageenan. The result was supported by the findings of Casariego et al. (2009), who reported a decrease in lightness of film surface from 94% to 91% when the concentration of chitosan was increased from 1% to 2%. Conversely there was an increase in lightness when carrageenan was increased. It may be due to its significant difference in opacity level as compared to the chitosan film. Opacity level is an indication of the amount of light that passes through the film. Carrageenan film of 2% was reported to have opacity level of 2.65% while chitosan film (2%) opacity level was 10.79% (Casariego et al., 2009). A smaller value of opacity means a greater transparency (Cuq, Gontard, Cuq, & Guilbert, 1996).

Response surface analysis for chitosan and carrageenan-based edible coating

The results of the studied responses such as weight loss, respiratory rate and surface lightness for chitosan-based and carrageenan-based coating formulation are shown in Table 2. Each response was assessed as the function of main, quadratic and interaction effects of the two edible coating components, chitosan or carrageenan (X₁) and glycerol (X₂). Chitosan/carrageenan and glycerol significantly (p < 0.05) affected weight loss, respiratory rate and surface lightness. Weight loss of chitosan-based edible coated longan fruit was significantly (p < 0.05) affected by chitosan concentration in terms of main effect and by glycerol concentration in terms of quadratic effect. A fitted model for predicting the response variable is as follows:

(2) $\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{s},Y_1=2.0133-0.1500X_1+0.3450{X_2}^2$(2)

Meanwhile, weight loss of carrageenan-based edible-coated longan fruit was significantly (p < 0.05) affected by carrageenan concentration in terms of main and quadratic effect and by glycerol concentration in terms of main effect, respectively:

(3) $\mathrm{W}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{s},Y_2=3.4062-1.7900X_1+0.3867X_2+0.7500{X_1}^2$(3)

The main effect reveals that individual factor either gave positive or negative effect with variation in the response. When a quadratic effect was determined, the optimal levels of the relevant factors were not in the extremes of the experimental region but within the experimental range. The respiratory rate of chitosan-based edible coated fruit was significantly affected by chitosan concentration in terms of main and interaction effects, and by glycerol concentration in terms of main, quadratic and interaction effects, respectively (Equation (4)) while for carrageenan-based edible-coated fruit, it was significantly (p < 0.05) affected by carrageenan concentration in terms of main, quadratic and interaction effects, and by glycerol concentration in terms of quadratic and interaction effects, respectively (Equation (5)):

(4) $\mathrm{R}\mathrm{e}\mathrm{s}\mathrm{p}\mathrm{i}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{o}\mathrm{r}\mathrm{y}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e},Y_3=118.047-27.506X_1+11.166X_2-3.952{X_2}^2-2.755X_1{X}_2$(4)

(5) $\mathrm{R}\mathrm{e}\mathrm{s}\mathrm{p}\mathrm{i}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{o}\mathrm{r}\mathrm{y}\mathrm{r}\mathrm{a}\mathrm{t}\mathrm{e},Y_4=106.72-46.24X_1+16.55{X_1}^2-13.40{X_2}^2+17.59X_1{X}_2$(5)

Interaction effect represents the combined effect of factors on the dependent measure. When an interaction effect occurred, the impact of one factor (e.g., chitosan concentration) depended on the level of the other factor (e.g., glycerol concentration). For response of surface lightness, chitosan-based-coated fruit was significantly (p < 0.05) affected by chitosan concentration and glycerol concentration in terms of quadratic effect (Equation (6)) while carrageenan-based-coated fruit was significantly (p < 0.05) affected by carrageenan concentration in terms of main and quadratic effect and also by interaction effect together with glycerol concentration (Equation (7)):

$\begin{array}{rl}& \mathrm{S}\mathrm{u}\mathrm{r}\mathrm{f}\mathrm{a}\mathrm{c}\mathrm{e}\mathrm{b}\mathrm{r}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{n}\mathrm{e}\mathrm{s}\mathrm{s},Y_5=46.9515-0.4056{X_1}^2\\ & -0.2906{X_2}^2(6)\end{array}$

$\begin{array}{rl}& \mathrm{S}\mathrm{u}\mathrm{r}\mathrm{f}\mathrm{a}\mathrm{c}\mathrm{e}\mathrm{b}\mathrm{r}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}\mathrm{n}\mathrm{e}\mathrm{s}\mathrm{s},Y_6=47.2158+3.8525X_1\\ & -0.9167{X_1}^2-0.7775X_1{X}_2(7)\end{array}$

The adequacy of models was determined using model analysis, coefficient of determination and lack of fit test. Also, relatively high R² values, which ranges from 0.7098 to 0.9919, with no significant (p > 0.05) lack of fit were obtained (Table 3). This indicated that the response surface models were significantly fitted for the studied responses.

Table 3. Regression coefficients and analysis of variance (ANOVA) for chitosan and carrageenan-based edible coating.

Tabla 3. Coeficientes de regresión y análisis de varianza (ANOVA) para los recubrimientos comestibles a base de quitosano y carragenano.

	Chitosan-based			Carrageenan-based
Regression coefficients	Weight loss	Respiratory rate	Surface lightness	Weight loss	Respiratory rate	Surface lightness
b0	2.0133	118.047	46.9515	3.4062	106.72	47.2158
b1	−0.1500	−27.506	−0.3422	−1.7900	−46.24	3.8525
b2	−0.3783	11.166	0.2795	0.3867	17.65	0.8708
b11	–	–	−0.4056	0.7500	16.55	−0.9167
b22	0.3450	−3.952	−0.2906	–	−13.40	–
b12	–	−2.755	–	–	17.59	−0.7775
R^2	0.8082	0.9919	0.9630	0.7098	0.8728	0.8016
R^2 (adj)	0.7507	0.9883	0.9466	0.5809	0.7933	0.7135
Regression (p-value)	0.001*	0.000*	0.000*	0.015*	0.002*	0.003*
Lack of fit (F-value)	1.19	0.99	0.57	4.75	2.13	4.27
Lack of fit (p-value)	0.428**	0.491**	0.696**	0.078**	0.216**	0.071**

b₀: constant.

b₁/b₂: the predicted regression coefficient for the main linear effect.

b₁₁/b₂₂: the predicted regression coefficient for the quadratic effect.

b₁₂: the predicted regression coefficient for the interaction effect.

(1): chitosan/carrageenan.

(2): glycerol.

*Significant (p < 0.05).

**Not significant (p > 0.05).

Tabla 3 nota.

b₀: Constante

b₁/b₂: Coeficiente de regresión pronosticado para el principal efecto lineal

b₁₁/b₂₂: Coeficiente de regresión pronosticado para el efecto cuadrático

b₁₂: Coeficiente de regresión pronosticado para el efecto de interacción

(1): Chitosán/Carragenano

(2): Glicerol

*Significativo (p < 0.05)

**No significativo (p > 0.05)

Optimization and verification of the models

Numerical and graphical multiple responses optimization were used to determine the exact values of the optimum levels of the independent variables that led to the overall optimum condition. The optimum region was predicted based on weight loss and respiration rate minimization, and surface lightness maximization. For chitosan-based coating, the optimized region was predicted to be 1.72% weight loss, 85.00 ml CO₂/kg h respiratory rate and 45.90 L value for surface lightness. This predicted result was achieved by 1.29% (w/v) chitosan and 0.42% (w/v) glycerol, with desirability of 0.7117. For carrageenan-based coating formulation, the optimized region was predicted to be 2.42% weight loss, 75.71 ml CO₂/kg h respiratory rate and 50.92 L value for surface lightness, which was contributed by 1.49% (w/v) carrageenan and 0.03% glycerol, with desirability of 0.9201.

The fitted values predicted by the response regression models were compared to the experimental values to verify the adequacy of the final reduced equation. The values of comparison did not show any significant (p > 0.05) difference in t-test. So, the predicted models were indicated to be able to describe the response variables satisfactory.

Conclusion

The results of the RSM on the optimal concentrations of carrageenan/chitosan and glycerol necessary to maintain the quality parameter relevant to postharvest storage of longan fruit revealed that chitosan-based coating was better than carrageenan-based coating in maintaining reduced respiratory rate, water and weight losses, respectively. On the other hand, carrageenan-based coating produced better surface lightness. Under the optimum conditions, the corresponding predicted response values for the chitosan and carrageenan-based coatings were 1.29% (w/v) chitosan, 0.42% (w/v) glycerol and 0.025% (w/v) sunflower oil; 1.49% (w/v) carrageenan, 0.03% (w/v) glycerol and 0.025% (w/v) sunflower oil. Both optimum formulations obtained had high desirability, which are 0.7117 and 0.9201 for chitosan and carrageenan-based edible coating, respectively. So, they both have the potential to increase the shelf life of postharvest longan fruits. While chitosan has more benefits because of its antifungal and antibacterial properties, it has a major drawback, which is poor solubility in neutral solution.

Acknowledgments

The authors are grateful for the extensive financial support of the Grant Putra IPS (GP-IPS/2016/9478500) at the University Putra Malaysia.

Disclosure statement

The authors confirm no conflicting interest.

Additional information

Funding

The University Putra Malaysia is acknowledged for providing the necessary research grant [number: GP-IPS/2016/9478500].