Formulation optimization of bionanocomposite film based on polyvinyl alcohol/glycerol/cellulose nanocrystal from pineapple crown leave fibers using response surface methodology

The current research is intended to optimize the formulation of bionanocomposite films produced from polyvinyl alcohol by adding glycerol (4–8%) and cellulose nanocrystals extracted from pineapple crown leaf (3–7%). The interaction between the bionanocomposite films characteristic, such as thickness, tensile strength, and elongation are determined using a response surface methodology with a central composite design. To show the practical relevance of the prediction model, the optimized bionanocomposite film composition required additional validation. Adding glycerol at 4% and cellulose nanocrystal at 3.52% was determined to be the film’s optimum parameters. Improved mechanical properties with a low thickness value and high elongation were found in the enhanced film formulation. The findings showed that an optimized bionanocomposite film formulation based on polyvinyl alcohol with glycerol and cellulose nanocrystals from pineapple crown leaves provided good model validation and could be utilized optimally in the food packaging application.


Introduction
The growing of bio-degradable alternatives for food packaging has developed significantly because of several concern of the environment related to synthetic materials based on petroleum [1].In the past two decades, research has focused heavily on the use of biopolymers as biodegradable materials as a sustainable development strategy that contributes to secure the environment.mainly because there is a number of benefits, particularly when compared to synthetic materials, such as non-toxicity, biodegradability, biocompatibility, and their broad availability [2].Many different resources from nature, such starch, protein, chitosan, cellulose and other source from plants and animals, are used to produce biopolymers, which have been seen as attractive alternatives.Polyvinyl alcohol has become one of the most often utilized materials in the production of bioplastics.It is a transparent biopolymer that is non-toxic and has a high level of biocompatibility and biodegradability.Additionally, this substance is water soluble, exhibits excellent film-forming abilities, extremely polar, generates a number of beneficial interaction through its hydroxyl group, and exhibits the capacity to interact and combine with other polar polymers [3].Films produced from polyvinyl alcohol also have useful qualities that make them desirable for packaging applications, including transparency, flexibility, and a great barrier against oxygen absorption.Polyvinyl alcohol was successfully used as a primary matrix based film in previous studies [4].However, there were still some problems with prior works.The hydrophilic characteristics polyvinyl alcohol-based film and their low mechanical qualities as biodegradable polymer limited their potential to find significant commercial applications.The addition of nanomaterials, biomaterials and other active agents has been one method used to enhance the polyvinyl alcohol-based films properties.According to prior studies, the inclusion of binding agents and nanoparticles can enhance the characteristics of polyvinyl alcohol-based film [2].
Numerous research on bionanocomposite films based on polyvinyl alcohol in addition to incorporate plasticizing agents and nanoparticles like glycerol and cellulose nanocrystals have demonstrated considerable improvement in the characteristics of the resulting films [5].Studies have additionally shown the positive effects of using cellulose nanocrystals as an excellent filler when forming bionanocomposite films [6].Cellulose nanocrystals have gained a lot of attention as prospective nanofiller for polyvinyl alcohol-based films because of their availability, sustainability, large surface area, light-weight structure, and superior mechanical strength.Cellulose nanocrystals are stable aqueous colloidal suspensions made of nanometer-sized crystalline particles that are often obtained from cellulose sources by acid hydrolysis [7].Various materials, including bagasse, kenaf fiber, rice husk, walnut shell, conorcapus fiber, and pineapple crown leaves, were used to extract cellulose nanocrystals.The demand for pineapples and related products is rising significantly on the international market.3 billion tons a year of byproducts from the processing of pineapples, including pineapple crown leaves, are produced and damaging to the environment.Sometimes pineapple crown leaves are thrown into landfills or burned without considering the implications for the environment.A proper method of processing pineapple crown leaves should be used in order to produce a superior product for industry application.Fibers from pineapple crown leaves have been found to be a superior source of cellulose in the range of 79-83% [8].Therefore, as shown by improvements in physical and mechanical properties in biopolymer-based films, incorporating cellulose nanocrystals from pineapple crown leaf waste offers interesting qualities that possible utilized for packaging application.The purpose of the studies is to maximize the inclusion of glycerol and cellulose nanocrystals produced from pineapple crown leaves as natural fillers to enhance the physical and mechanical qualities of polyvinyl alcohol-based film bionanocomposites.
This study also produced an optimum concentration of glycerol and cellulose nanocrystals additions and the stage of making bionanocomposite films by utilizing polyvinyl alcohol as an adequate biopolymer.In addition, the proposed invention is intended to overcome environmental problems by replacing petroleum-based films with more biodegradable materials and feasible to decompose in the environment.

Materials
Polyvinyl Alcohol with 200 kDa was purchased from Sigma Aldrich, Singapore.Pineapple crown leaves with the age of 1 to 1.5 years were obtained and collected from a market in Aceh, Indonesia.Glycerol, sodium hydroxide, and sulfuric acid, hydrogen peroxide were used as active and reaction agents in this study.All reagent is purchased from Merck, Singapore.

Preparation of Polyvinyl Alcohol/Glycerol/Cellulose Nanocrystal Bionanocomposite Films
The approach previously described by Fitriani et al. [7] was used to extract the cellulose nanocrystal (CNC) using pre-chemical treatment and hydrolysis.First, the dried pineapple crown leaf (PCL) was crushed and sieved through a 40 mesh screen until fibers formed.To accomplish the alkali and bleaching process, the ground fibers were heated up in NaOH and H2O2 (1:1) for 1 h at 80°C.The remaining fibers after first treatment was rinsed with distillate water repeatedly until pH neutral.The fiber was subsequently exposed to acid hydrolysis using 1 M H2SO4 throughout the course of a 3 h reaction time at 45°C to create CNC particles through depolymerization.The acidic reaction in the process of hydrolysis was stopped by blending distilled water to the resultant solution and cooling it in a water bath for 24 h at 20°C.In order to neutralize the suspension, it was centrifuged repeatedly for 30 min at 2000 rpm while adding distilled water each time, and the remaining acid was then removed using ultrasonication for 30 min until the suspensions pH was neutral.The final CNC product was then dried, processed into powder, and placed in a room temperature storage container.For the fabrication of bionanocomposite film, 10 gr PVA was added to distilled water (200 ml) and kept on a stirrer on the hot plate at 80°C for 1 h.Then, varied CNC (3-7%) and glycerol (4-8%) were added to the PVA solution.
The film solution was poured on a silicon mold and dried for 24 h at room temperature before peeled off and storage for further analysis.

Mechanical and Physical Analysis
For each sample, the film thickness was tested three times with a micrometer that had a 0.001 mm precision.A Universal Testing Machine (MTS-UTM type E43, China) and ASTM standard method using D638-IV were used to analyze the mechanical characteristics of the sample.

Statistical Analysis and Experimental Design of bionanocomposite films
A central composite design (CCD) from response surface methodology (RSM) were used to create the formulation for fabricating and optimizing bionanocomposite films.The experiment design was produced using Design Expert 11 by Stat-Ease Inc. (United States).Concentrations of glycerol (X1) and CNC (X2) constitute the two independent variables.There were 13 runs total in the approach, including eight factorials (levels ±1), four axials (levels ±α), and four replications in the central point.Table 1 and 2 shows the factors levels and coding of experiment.

Model Development
RSM-CCD was used to assess the glycerol factor levels (X1) and CNC (X2) and the impact of those factors on thickness (Y1), tensile strength (Y2), and elongation (Y3).Table 3 shows the results of 13 tests that were carried out at various factor-level combinations.For all responses, the model shows a significant of F value (p <0.05), indicating that all model definition is essential.While the lack-of-fit offers evidenced on the suitability of the model used to quantify errors resulting from deficiencies in the model [9], it does not constitute perfect.The results demonstrated that the input of p-value for lack-offit was insignificant for any response (p > 0.05).The model is suitable and sufficient for data clarification when the value is insignificant.To fit the whole response surface model, the mathematical model equation is developed using the coefficients of regression in actual values and is written as follows: (1) (2) Additionally, as demonstrated in Tables 4 to Table 6, this model sufficiently describes the fluctuation of the response and has acceptable values for the level of determination (R 2 ), coefficient of variation (C.V.), and adjusted level of determination (Adj-R 2 ) in analysis of variance (ANOVA).The R 2 values in table for Y1, Y2, and Y3 responses were found at 0.895, 0.915 and 0.946, respectively.A greater R 2 value implies a high degree of relation between the experimental value and the predictive value of the variable response.All of the results show that the model is able of accurately predicting the mechanical and physical characteristics of films.However, a high R 2 value does not always indicate that the regression model is adequate.The R 2 value will always rise when more variables are included in the model, regardless of their statistical significance.Therefore, models with high R 2 values may give incorrect predictions of average responses or estimates of new observations.Thus, some regression models tend to consider the R 2 adjustment because its value stays the same as more variables are added to the model.As a result, several regression models frequently take the R 2 adjustment into consideration because its value remains constant as the number of variables in the model increases [10].The R 2 is adjusted by calculating the coefficient of variance based on each independent variable that influences the response.The R 2 value is based the assumption that each independent variable describes any variation in the response variable of bionanocomposite films.The adjusted R 2 values for Y1, Y2, and Y3 were found to be 0.820, 0,854, and 0.907, respectively.The C.V. value for this model, which represents the percentage of the mean as a standard deviation, is stated to range from 5.54 to 16.32.In addition, precision adequacy calculates the signal-to-noise ratio where the ratio is higher than four which is highly desirable and indicates a sufficient model selection.The model optimization ratio is between 8.198 and 14.412, which is expressed as an adequate signal.Therefore, based on the explanation of the result above, the model provided by RSM can be used to manage the design area development in optimization.

Thickness of Films
Bionanocomposite film thickness analysis is essential for determining a packaging films mechanical and physical characteristic.The ability to retract film bionanocomposite structures, composition, fabrication procedure, and the use of consistent capacity on a specific surface are the main factors influencing film thickness [11].PVA-based film thicknesses added with varying concentrations of glycerol and CNC were found to be between 0.06 and 0.23 mm, as observed in Table 3.The model of surface response for thickness was generated using two variable concentrations of glycerol and CNC on film bionanocomposites.Figure 1 shows that glycerol and CNC concentrations were identified to have a significant result on the thickness of bionanocomposite film (p<0.05),while the lack of fit results implies insignificance (p>0.05)relative to pure error, as reported in Table 4.This can be due to the dependence on the composition of the film formulation, which causes a varying value in the film thickness.Previous study also stated that the thickness of bionanocomposite films with different concentrations of additives and fillers was in the range between 0.14 mm and 0.20 mm [12].A higher amount of glycerol and CNC in the film composition induces the creation of film thickness in a parabolic trend.This increase in thickness also allows the contribution of the properties and characteristics of glycerol and CNC to cause thickening, suspension, and interaction between components.Similar results on increased thickness with higher concentrations of glycerol and CNC were also reported in previous studies with PVA-based films [13,14].

Tensile Strength of Films
Tensile strength is the highest or maximum part tensile stress material can endure before it deforms or breaks down.The tensile strength of PVA-based films blend with varying concentrations of glycerol and CNC were found to be between 5.24 and 11.12 MPa as observed in Table 3. Response surface plots were studied to demonstrate a better knowledge of the different effects of variable glycerol and CNC on tensile strength parameters.Figure 2 shows that higher tensile strength curve values are shifting at lower glycerol concentrations and CNC.In addition, the quadratic concentration of glycerol CNC was significant to the tensile strength value (p <0.05) and was strongly influenced by the interaction of glycerol and CNC.The lack of fit model gave insignificant results (p>0.05) relative to pure error rates, as reported in Table 5.These findings are supported by previous studies results, which reported an increase in the amount of cellulose (1-5%) added to PVA-based films led to a higher in tensile strength variable [15].The increase in tensile strength of film bionanocomposites at low CNC concentrations (3-5%) resulted a strong interaction among the polymer matrix and filler.The bond hydrogen between the group of O-H in the CNC and the PVA matrix is the main reason for improving the adhesion between phases, which results in good interactions between the bionanocomposite substance of the film [16].Similar results were also reported by earlier research using nanocellulose from palm oil waste, cobs, and durian bark fibers as fillers in PVA films [4,17,18].Although, a decrease slightly was seen when the filler concentration increased to 7%, possibly due to poor dispersion.This poor dispersion may be due to aggregate particles, which as well create a stress levels in the polymer matrix and lower the positive input of CNC as filler to the PVA films tensile strength [19].This result was also observed in the research reported by Tan et al. [20] and Kumar et al. [21] using CNC concentrations at 7-9% and 8-10%.
In contrast, the increase in glycerol as a plasticizer has a good effect on reducing the tensile strength of the PVA based film.This behavior was also observed in a previous study by Cazon et al. [22], which state that the use of glycerol ranging from 0 to 5% into the PVA film has reduced the films strength compared to the pure PVA film.Another study also observed similar occurrences in PVA films with different glycerol concentrations (0-5%) found to decrease the tensile strength value of the bionanocomposite film from 20.76 to 2.15 MPa [23].This phenomenon probably cause by the plasticizing effects of glycerol that damage bonds of hydrogen and crystalline regions on the film structure and decrease film strength and hardness [24].The glycerol effect on bionanocomposite film can also be indicated by a hydrogen bond competition among polymers and polymer-plasticizers when plasticizers are incorporated into the polymer matrix [22].Similar results can be observed in other PVA film studies incorporating plasticizers such as sorbitol, propylene glycol, and polyethylene glycol [25][26][27].Therefore, it can be assumed that PVA-based films will show a more dense and rigid form with increasing levels of CNC and less glycerol incorporation.

Elongation of Films
Elongation is the ratio between the change of films length before and after the film specimen is broken.The results reported that the film's elongation value was between 50.95 and 89.17%.The ANOVA results in Table 6 show that the variable F values of glycerol concentration and CNC significantly affect film elongation (p <0.05).Meanwhile, lack of fit implies that the model is insignificant (p >0.05) and relative to pure error.A graph plot with three-dimensional surface is created to demonstrate the effect of each variable the response.It shows in Figure 3 that a higher film elongation values were observed depending on the amount of glycerol and CNC concentrations.Based on these results, increasing the amount of glycerol and CNC at higher concentrations tends to lower the elongation value of film bionanocomposites.The previous study by Cazon et al. [22] and Kumar et al. [21] revealed that the glycerol and CNC added to PVA-based films also reduced film elongation, which is in agreement with the results.This may be due to the exhibition of molecular chain by CNC materials that have substantially low in rigidity, mostly small elongation, and fragile structure [16].This CNC structure may have modified the PVA molecular chain.The addition of a higher amount of CNC to the film, the more likely it is to drive lower elongation and a more fragile film structure [18].Similar results were also reported in previous studies using CNC concentrations (0-10%) [16,21], .
In contrast, the decrease in film elongation using high glycerol concentrations may be due to the effect of cellulose components as fillers [22].Intense contact may also occur between plasticizer at higher concentrations and the polymer-based, causing a crosslinking effect that reduces the mobility in the polymer matrix and free volume molecules [28,29].This phenomenon gives higher flexibility to the film at low glycerol concentrations (0-6%).Therefore, film elasticity and a less fragile component with higher resistance to film elongation can be acquired by reducing the amount of glycerol and CNC at low-level concentrations before being added to the film.

Numerical optimization
Numerical optimization is performed to estimate the optimal level of the independent parameters to attain the desired result for all responses.Optimal conditions are obtained at 4% glycerol and 3.52% CNC, as observed in Table 7.The response results for all mechanical and physical properties predicted under optimal conditions are 0.11 mm, 10.28 MPa, and 91.03%, respectively.8, where the thickness and elongation values are higher than the predicted values, whose anticipated values are 0.12 mm and 92.22%, respectively.While the tensile strength parameter showed a lower value (10.15 MPa) than expected.However, all responses validation results were within 95% confidence interval.The parameter's predicted range with a given level of confidence is known as the estimated confidence interval.According to the confidence interval, there is a 95% chance that the average data across all responses will fall within this range of validation.Results for every response were also discovered to be within a 95% prediction range.All replies prediction intervals, which were found to be higher than the confidence intervals, fell inside the 95% prediction interval.Because there is more uncertain and inconsistent data when predicting a single variable response as opposed to an average response, the prediction interval was higher than the confidence interval.This means there is a 95% chance that the findings will include all possible responses in this situation.Therefore, the obtained data suggests that an optimized PVA-based bionanocomposite film is ideal to be applied as one of the packaging materials in food industry.

Conclusions
In this study, the stability of PVA-based film bionanocomposites is increased by the addition of glycerol and CNC.RSM-CCD was used to examine the impact of the interaction of these two factors, glycerol and CNC, on physical and mechanical properties.To show the accuracy of the prediction model, the improved film formulation underwent additional validation.The optimum value of the bionanocomposite film is selected with the addition of 4% glycerol and 3.52% CNC.The validated optimal model yielded values of 0.12 mm (thickness), 10.15 MPa (tensile strength), and 92.22% (elongation).Improved bionanocomposite film compositions have greater characteristics and properties that can be utilized as packaging in food industry.

Figure 1 .
Figure 1.Plot the surface response to thickness.

Figure 2 .
Figure 2. Plot the surface response to tensile strength.

Figure 3 .
Plot the response of the surface to elongation.

Table 1 .
Different variable with coded level a center point; k = 2, and α = 1.414

Table 2 .
The configuration of the CCD in coded and actual variable

Table 3 .
Response Parameters in CCD

Table 4 .
ANOVA for quadratic models of film thickness

Table 5 .
ANOVA for quadratic models of film tensile strength

Table 6 .
ANOVA for quadratic models of film elongation

Table 7 .
Recommended solutions for optimal conditions of PVA bionanocomposite film Validation of Optimized FormulationsValidation tests are performed to determine the actual mechanical and physical properties values at optimum conditions (4% glycerol and 3.52% CNC).The validation results are shown in Table

Table 8 .
The validation for optimal bionanocomposite film formulation