KINETICS STUDY OF CELLULOSE NANOCRYSTALS MODIFICATION USING RARASAPONINS BY ELOVICH EQUATION

The modiﬁcation of cellulose nanocrystals (CNCs) using rarasaponins (RSs) was carried out to enhancing the hydrophobicity of the CNCs. The RSs are a natural surfactant that has hydrophilic and hydrophobic sides. The linked RSs on the CNCs surface can be used to bond the hydrophobic drugs so that the modiﬁed CNCs can be applied as the hydrophobic drugs carrier in the medical ﬁeld. The kinetics study was successfully carried out using the Elovich equation as the modeling equation. The Elovich equation ﬁts the modiﬁcation results well based on two parameters, i.e., the RSs/CNCs ratios and the times. The dispersion characteristics analysis was carried out to ﬁgure the enhancement of the hydrophobicity on the modiﬁed CNCs compared to the unmodiﬁed CNCs. According to the kinetics study and the dispersion characteristics analysis, the modiﬁcation of CNCs using RSs could enhance CNCs utilization in the hydrophobic drug delivery system.


INTRODUCTION
Cellulose nanocrystals (CNCs) are unique and very useful materials for various applications, such as pharmaceutical, food, and chemical industries. CNCs can be used for drug delivery systems, tissue engineering, composite material, other nanomaterial syntheses, protein and enzyme immobilization, and emulsion stabilizers [1][2][3] . Due to their fabulous physical and chemical characteristics, CNCs are suitable as a drug carrier in a drug delivery system. This utilization of CNCs is very advantageous for the medical field. CNCs have several characteristics, such as biodegradable, biocompatible, non-toxic, rich in reactive hydroxyl groups on the CNCs surface, good hydrophilicity, and easily modified material [4,5] .
Based on previous studies, CNCs have been isolated from various lignocellulose materials, such as pinewood, corncob [6] , sugar palm fibers [7] , discarded cigarette filters [8] , sago seed shells [9,10] , and passion fruit peels [11] . Many lignocellulose materials can be used for CNCs production due to their high cellulose content. Besides that, several natural resources have been studied to produce CNCs for enhancing their economic value. In this study, bamboo shoots (BSs) were used as the raw material for CNCs production. Dendrocalamus asper is one of BSs kinds that is abundant in Southeast Asia, especially Indonesia. This kind of BSs is very popular among many people due to its large size and sweet taste. Because of that, these BSs are widely used in food applications. The utilization of BSs would increase the economic value in the form of CNCs. BSs have high cellulose content (22.8-34.8 wt%) and low lignin content (1.8-11 wt%) [12] . The high enough cellulose content is an important consideration in this BSs utilization. Another advantage of using BSs is the low lignin content that can facilitate the pretreatment process of BSs. The pretreatment process removes the hemicellulose and lignin contents and obtains the cellulose content as pure as possible. In obtaining pure cellulose, the pretreatment process is needed to break down the lignocellulose material's complex structure, where a strong lignin structure covers the cellulose. The pretreatment process can be done by physical, chemical, biological, and combined methods. The low lignin content can reduce the pretreatment process's cost, where the simple pretreatment process can be used to isolate the cellulose content. Based on these considerations, BSs are assessed as a potential raw material for producing CNCs.
In the drug delivery system, CNCs are easily used as a drug carrier for hydrophilic drugs but not hydrophobic drugs. The abundant hydroxyl groups on the CNCs surface make CNCs have hydrophilic characteristics. In the medical field, drug carriers are needed so that drugs injected into the human body can be taken orally. This makes it easier for the recipient to consume these drugs. The utilization of CNCs as a hydrophobic drug carrier is a challenge because CNCs have a low drug loading capacity for hydrophobic drugs. Although CNCs have a weakness as a hydrophobic drugs carrier, it is still worth to be used due to its safety for humans. The modification process of CNCs is needed to overcome this challenge. Previous studies have been carried out to enhance the hydrophobicity of CNCs. Several modification agents have been used such as castor oil [5], cyclodextrin [13] , chitosan [14] , hexadecyltrimethylammonium (HDTMA) [15] , and cetyltrimethylammonium bromide (CTAB) [1,16] . This study develops the modification of CNCs using rarasaponins (RSs). RSs are a natural surfactant with a hydrophobic side (aglycone chain) and a hydrophilic side (oligosaccharides chain). This characteristic is suitable for CNCs modification. The hydrophilic side of RSs is used to bind to CNCs, while the hydrophobic side can be used to bind the hydrophobic drugs. RSs is chosen because it is safe for human health and the environment. RSs can be isolated from Sapindus rarak De Candole that is abundant in Indonesia.
In general, the Elovich equation has been usually used for describing the adsorption kinetic that indicates a chemical reaction. The Elovich equation can figure out the type of chemical interaction in the adsorption phenomenon. Here, the Elovich equation was tried to be used as models for this modification process's kinetics study. This work aimed to study the modification process of CNCs kinetically using RSs. To prove the hydrophobicity enhancement, the characteristic dispersion analysis was carried out towards the modification product.

Isolation of Cellulose Nanocrystals
BSs were dried using an oven at 50°C and powdered into small size. In the pretreatment process, BSs were refluxed using sodium hydroxide solution (20 wt%) at 100°C for four h. The solid residues were dried and continued for the next steps. The solid residues were mixed with sodium hydroxide solution (2 M) and hydrogen peroxide solution (50%) with a ratio of 1:1. The solid residues contained high cellulose content.
Pretreated bamboo shoots (PBS) were hydrolyzed using a sulfuric acid solution (55 wt%) at 45°C. 10-folded cold distilled water was added to the suspension to stop the acid hydrolysis process. The acid solution was separated, and then the solid residues were washed until neutral. The supernatant was obtained and dried using a freeze dryer at a pressure of 0.08 mbar and a temperature of -42°C.

Extraction of RSs
The extraction of the powdered and dried Sapindus rarak De Candole was conducted using 150 mL of distilled water at 80°C for 60 min with a solid/solvent ratio of 1:10 g/mL. After the process, the solids residue was separated, and the liquid extract was dried using a freeze dryer at a pressure of 0.08 mbar and a temperature of -42°C. This process was carried out by following the previous method [17] .

Modification of Cellulose Nanocrystals
RS solution was added CNCs suspension with RS/CNC mass ratios of 0.5, 1.0, and 1.5. The mixture was stirred for a certain time (kinetics study). The time parameter was varied from 10 to 600 min (for 10 data). The combined parameters were used to study the kinetics behavior of this modification process.

Analyses
There are two analysis method were carried out in this study, i.e. determination of linked RSs on the CNCs and dispersion characteristic analysis. In determination first analysis, the total saponins content analyzed the RS solution's residual concentration to calculate the linked RSs on the CNCs. The calibration curve was figured out using diosgenin as the standard solution. The diosgenin solutions were varied by 100-700 mg/L in methanol-water solvent (a ratio of 4:1). 0.5 mL of the diosgenin solution was added by 0.5 mL vanillin reagent (8% in ethanol) and 5 mL of sulfuric acid solution 72% volume. It was heated at 60°C until the color changed, and then it was cooled. The absorbance measurement was done at 544 nm using UV/Vis Spectrophotometer. In the residual RSs solution, the concentration was measured in the same way by replacing the diosgenin solution with the residual RSs solution. The amount of linked RSs on the CNCs surface ( ) in mg/g was calculated using the following equation: where and are the initial and residual RSs concentration in mg/L. V is the volume of the modification process in Liter, and m is the mass of CNCs in gram.
In the second analysis, i.e. dispersion characteristic analysis, the unmodified and modified CNCs samples were dispersed in two kinds of solvents, such as water and ethanol solvents (0.1 wt%). The mixtures were sonicated for one hour. The investigation was done after three weeks of resting at room temperature.

Elovich Equation
Elovich equation is widely used for modeling the chemical adsorption kinetics. This equation can model the kinetics interaction in this modification process with similar behavior with the chemical adsorption. In this modification, the CNCs and RSs would act as the adsorbent and the adsorbate based on their phases during the process. This equation's basic assumption is the heterogenous adsorbing surfaces where it figures out the CNCs surface as the heterogeneous surface. The mildly rising kinetics behavior of the process fits this equation well [18] . The basic equation of the Elovich equation is given Aharoni and Tompkins [19] by Eq. 2.
A and b are the initial rate constant (mg/g min) and the Elovich constant (g/mg). Using the boundaries of = 0 at = 0 and = at = the integrated form of Eq. 2 becomes: where = 1 . If the assumption of ≫ is applied, Eq. 3 becomes: Eq. 4 is the final Elovich equation for investigating the constants.

FIGURE 1
The calibration curve of RSs concentration.
For approaching the equilibrium factor of the Elovich equation, the dimensionless Elovich equation (Eq. 7) is developed by determining as the longest time in the process and as the concentration in the solid phase at = . Thus, Eq. 4 is rewritten as: where RE indicates the equilibrium factor of the Elovich equation. The RE value is divided into four zones (I -IV) that vary from sloping to steep curves of versus . The classification of these zones is given in Table 1 [18] . The third column, i.e. The Behavior, is the behavior of approaching the equilibrium condition. Figure 1 shows the calibration curve for determining the total saponins content or concentration. The diosgenin was used as the standard solution because it represents the hydrolysate of the saponins content in the solution. The hydrolysate is the

FIGURE 2
The nonlinear regression of the modification results using the Elovich equation.
aglycone region only that the oligosaccharides region has been hydrolyzed using a sulfuric acid solution. The concentration can be determined using Eq. 8.
The extraction was carried out from the Sapindus rarak De Candole. The RSs concentration of the RSs extract was 68.48 ± 2.05% through the total saponins content determination.
The modification of CNCs using RSs was carried out, where the results are shown in Table 2. The combined parameters (both initial concentrations and times) were used to investigate this process's kinetics behavior. Table 2 shows that the RSs/CNCs ratio of 1.0 gives the highest amount of linked RSs on the CNCs surface. From the RSs/CNCs ratio of 0.5 to 1.0, the q values show the enhancement every time. The increase of the driving force causes it. This process's driving force is the RSs concentration difference between the liquids and solids phases' concentration. The increase of ratio would increase the RSs concentration difference so that the mass transfer of RSs compounds from the liquids phase to the CNCs surface would enhance. However, from the RSs/CNCs ratio of 1.0 to 1.5, the q values show the reduction values. The overmuch RSs compounds in the process make the interaction of RSs compounds and the CNCs decrease due to the reduced free space in the solution. So, the RSs/CNCs ratio of 1.0 is the best ratio among the used ratios.  Based on the Elovich equation, the nonlinear regression was carried out as shown in Figure 2, where the constants are written in Table 3. The Elovich question fits the data of modification results well. It is proven by the values of 2 that show a value of more than 0.99. The values indicate the initial attachment rate of RSs to CNCs. It demonstrates the increased mass transfer statement and the RSs/CNCs ratio (RSs concentration difference). The values increase along with the increase of the RSs/CNCs ratio. values represent the rate constant that is linear with the initial concentrations [20] . The 1 values increase along with the increase of the RSs/CNCs ratios (initial RSs concentrations), as shown in Table 3. The 2 values represent the rate constant, but these have a complex function towards the initial concentration [20] . The 2 values are affected by the initial RSs concentrations and the affinity between the RSs and CNCs.   nonpolar materials. The unmodified CNCs could be dispersed in water but not in ethanol. It means that the unmodified CNCs have very high hydrophilicity. After the modification process, the RSs were linked on the CNCs surface, making the hydrophobicity increase due to the surface's hydrophobic side. It was proven by the good dispersion of the modified CNCs in ethanol.
Moreover, the modified CNCs could not be dispersed in water. This indicates that the modification process changes the hydrophilic characteristics to the hydrophobic characteristics. This characteristic dispersion analysis was successfully carried out to figure out the hydrophilic-hydrophobic characteristics of the CNCs.

CONCLUSION
The kinetics study has been carried out to the modification results by using the Elovich equation. This equation gave more satisfactory modeling results toward the Pseudo-first order and Pseudo-second order equations. In this modification process, the RSs/CNCs ratios affected the mass transfer rate of the RSs from the liquid phase onto the CNCs surface. However, the highest amount of linked RSs on the CNCs surface was obtained at a RSs/CNCs ratio of 1.0. From the dimensionless Elovich equation, RE's values also showed satisfying results ( 2 > 0.99). The importance of RE indicates the mild interaction rate behavior. It confirmed the results of the direct nonlinear regression towards the modification data.