Sustainable Durio zibethinus-Derived Biosorbents for Congo Red Removal from Aqueous Solution: Statistical Optimization, Isotherms and Mechanism Studies

Abstract

This investigation reports on the biosorption mechanism of Congo Red dyes (CR) in aqueous solution using acid-treated durian peels, prepared for this study. The biosorbent nature was characterized using the Scanning Electron Microscopy (SEM), Fourier Transform infrared spectroscopy (FT-IR) and Brunaure-Emmet-Teller (BET). The effect of process parameters within operational range of pH (2–9), contact time (10–200 min), initial concentration (25–400 mg g⁻¹) and temperature (25–65 °C) for the optimum removal of CR dyes was investigated using central composite design (CCD) under response surface methodology (RSM), and revealed that the optimum condition of biosorption was achieved around a pH of 5.5, contact time of 105 min at initial concentration of 212.5 mg L⁻¹ within 45 °C temperature, which corresponds to 95.2% percent removal of CR. The experimental data fitted better to the second order polynomial model, with a correlation coefficient R² value of 0.9917 and the Langmuir isotherm model with biosorption capacity of 107.52 mg g⁻¹. Gibbs free energy indicated that the adsorption of CR dyes was spontaneous. The mechanism of the adsorption of CR dyes revealed that the biosorption of CR dyes investigated under different operational conditions show that under acidic pH, the adsorption efficiency of the acid treated durian peels is enhanced for the adsorption of CR dye molecules.

1. Introduction

In the last few decades, a large amount of colour effluents have been generated from various industries such as textiles, leather and pharmaceuticals. According to some recent estimates, over half a million tons of Congo red (CR) anionic diazo dyes are produced annually [1]. CR dyes are widely applied in many manufacturing industries such as plastic, paper, leather, silk, cotton and hemp [2]. The residual dyes are the major contributor to the colour in the effluents generated by these industries. These residual dyes contain complex chemical structures, several aromatic rings, and functional groups that are very difficult to biodegrade [3,4]. They can be persistent for a long time in the environment, thereby causing pollution. Furthermore, most of these colours are mutagenic and carcinogenic in nature and can cause many diseases, resulting in adverse effects on the aquatic ecosystem and living beings, including humans [4]. In the same context, these dyes are contained in most of the industrial wastewater all over the world and can easily be released into the surface and ground water [3]. Therefore, studies on the removal of recalcitrant dyes such as CR synthetic dyes is urgent and environmentally significant [5]. Due to the complex chemical structure, it is very difficult to remove them from wastewater and water reservoirs by using commonly used decontamination techniques, including chemical precipitation [6], reverse osmosis [7], solvent extraction [8], photocatalytic processes [9], ion exchange [10], ultrafiltration [11], coagulation, flocculation [12] and adsorption [13,14,15]. In the past few decades, adsorption has been the preferential treatment for reactive dye adsorption due to its simple operation, effective treatment and widespread applicability [16] Many studies have reported on CR dye biosorption from aqueous solution [17,18,19,20,21,22,23,24,25,26,27]. However, none have investigated CR dyes using durian peels. Malaysia and Indonesia are the world’s two major durian (Durio zibethinus) producing countries. Durian is considered the “King of Fruits” in the Southeast Asia peninsula. It is a seasonal dicotyledonous plant with large size fruits (weight: 1 to 3 kg, length: 2 to 7 lb. and dark green in colour) and dreadful thorn-covered husk. Overall durian production in Malaysia is more than 300,470 million tons and 255,353 million tons of durian shells ends as by-products per year, according to reported data by the Agricultural and Agro-based Industry Malaysia [28]. During durian season, agro-waste is the predominant waste all over the Malaysian peninsula, and this could potentially become a future environmental challenge [29].

Motivated by the environmental impacts of durian wastes, which, in turn, can be used as a potential biosorbent for the removal of dye containing effluents, the main objective of this research was to prepare and characterize chemically treated durian peels as an alternative biosorbents for the removal of CR dye from aqueous solution. This study is aimed to optimize the effect of the processing parameters of CR dye removal on the biosorbent in a batch experiment using statistical model. Experimental biosorption data were fitted to Langmuir, Freundlich and Temkin isotherm models. Kinetic studies were investigated using Pseudo first order, Pseudo second order and the intraparticle diffusion equations. Vant hoff’s thermodynamic parameter was used to study the effect of temperature on the sorption capacity of the biosorbent. Moreover, the mechanism of the biosorbent efficiency for CR dye removal was also included in the study.

2. Materials and Methods

2.1. Adsorbate Preparation

The Congo red dye, namely (3,3′-(Biphenyl-4,4′-diyldidiazene-2,1-diyl)bis(4-aminonaphthalene-1-sulfonic acid)); molecular formula: C₃₂H₂₂N₆Na₂O₆S₂; molar mass: 696.7 g/mol, was procured from QReC^TM chemicals Co Ltd. The chemical structure is illustrated in Figure 1. First, 1000 mg/L stock solution was prepared by dissolving an appropriate amount of dye molecules in demineralized water in a 1 L standard flask. From stock solution, working solution was prepared by diluting the mother solution with appropriate distilled water.

2.2. Adsorbent Preparation

The durian fruit peels were collected from the local area of Johor city, Malaysia. Before being processed, the durian fruit peels were cut into small pieces and washed with distilled water to remove the adhering dirt. The washed materials were oven-dried at 80 °C for 48 h. Thereafter, dried peels were grinded and sieved to particle size ranges from 150–200 µm. About 25 g powered material was soaked overnight in a solution prepared from 1 M H₂SO₄ and 1 M HNO₃ in the ratio of 1:1. It was then filtered and washed with distilled water to remove the residual acid content. It was further dried in an oven for 24 h at 80 °C to obtain chemically activated durian peels powder, which was used for the adsorption studies.

2.3. Characterization of Adsorbents

The functional properties of treated and untreated durian peels powder were obtained from the Fourier Transform Infrared spectroscopy (FTIR) spectra using a Thermo Scientific NICOLET 6700 apparatus in the region of 500–4000 cm⁻¹, at a resolution of± 4.0 cm⁻¹. The surface morphology of the treated and untreated durian peels powder was derived from the scanning electron microscopy (SEM) images by Zeiss SUPRA 55 SEM model. The scanning electron micrographs were obtained at 3 kV. The acid-treated durian powders used for the study was examined from adsorption–desorption test conducted at 77 K to determine the average pore diameter using micrometric model ASAP 2010 sorptometer. The pore size distribution and the surface area of the biosorbent was obtained using the Barrett-Joyner-Halenda (BJH) and Brunaure-Emmet-Teller (BET) method, respectively [30]. The Pore surface characteristics has been compared to previous studies [31,32,33,34]. The surface properties of the biosorbents are shown in

Table 1. BET results of acid-treated durian peel compared with other adsorbents.

Material	BET Surface Area (m2/g)	Pore Volume (cm3/g)	Pore Size (Å)	Reference
Acid-treated durian fruit peel	308.74	0.285	112.17	Present work
sonoenzymolysis of chitin suspension	1.22	8.53 ∗ 10−3	9.33	[31]
coffee waste (CW)	219.69	0.125	40.04	[32]
treated shrimp shell powder (TSSP)	66.35	0.0968	5.61	[33]
Triarrhena Biochar with TiO2	161.5	-	60	[34]

.

2.4. Experimental Design and Modelling

The regression model equations and operational conditions of the qualitative data obtained from the adsorption study was determined using statistical analysis of the response surface methodology. Central Composite Design (CCD) design expert version 7 was used for this study to obtain the optimum process variables for the maximum adsorption of CR dyes onto modified durian peels powder. This method is most appropriate to reduce the time and variability of experimental outcomes, with improved optimized output compared to fixed optimization processes. Thirty (30) experiments were conducted using the full factorial level of the response surface consisting of six replicates number of center points on four (4) independent variables, which were pH ($x_1$), contact time $\left(x_2\right),$ initial concentration ($x_3$) and temperature ($x_4$). The experimental data were designed and analysed using the statistical regression model (Design Expert 6.0.10), central Composite design (CCD) (Stat-Ease, Inc., Minneapolis, MN, USA) software of the response surface methodology to investigate the effect of the interaction of the independent variables on the adsorption of CR dye on the synthesized durian peel powder. A regression model equation was developed which correlated to the response of CR adsorption as a function of the independent variables, using a 2nd degree polynomial equation as illustrated in Equation (1).

$$Y=b_0+\sum b_I{x}_I+\sum b_{ii}{x}^2+\sum b_{ij}{x}_I{x}_j+e$$

(1)

where Y denotes the predicted response, $b_0$ represents the constant coefficient, $b_I$ is the linear coefficients, $b_{ii}$ is the quadratic coefficients $b_{ij}$ depicts the interaction coefficients, and $x_{I,} x_j$ are the coded values of the CR dye adsorption variables.

3. Results

3.1. Characterization of Treated and Untreated Durian Peel

The SEM micrographs of both unsaturated and CR saturated durian fruit peels are shown in Figure 2. From the micrographs, some surface alterations in the unsaturated and dye saturated peel powder was observed. As shown in Figure 2a, the surface of durian fruit peels powder exhibited porous textural surface which was a predominant feature observed. However, after adsorption of CR on durian peels powder, a change in the surface morphology was noticed with the pores completely filled, indicating the adsorption of CR on the surface of the biosorbent (Figure 2b).

The FTIR analysis is a very important and useful tool to describe the main functional groups present on the surface of adsorbent materials. The FTIR spectra observed before and after CR adsorption onto durian fruit peels powder are shown in Figure 3. From the FTIR spectra, a broad band appearance around 3337 cm⁻¹ was mainly attributed to the presence of hydroxyl groups (O–H) of cellulose and lignin. The peak appeared at 2917 cm⁻¹ was assigned to C–H stretching of methylene groups. The strong peak at 1690 and 1647 cm⁻¹ was assigned to the C–O bond of the carboxylic groups present in the hemicellulose and lignocellulosic part of biosorbent. The peaks at 1453 and 1421 cm⁻¹ could be attributed to the ring of aromatic compounds. The band at 1377 cm⁻¹ is assigned for CH group of cellulose and lignin composition on the surface of the durian peels. The peak at 1183 cm⁻¹ corresponds to asymmetric stretch of C-O-C group of ether present in lignin contained in the durian peels. The band appearance around 1037 cm⁻¹ is assigned to C-O stretch of primary alcohol present in lignin. In the FTIR spectra of Figure 3b, some shifts (8 to 22 cm⁻¹) were observed corresponding to the OH and carboxylic groups indicating that these groups significantly enhanced the adsorption of dye onto the durian peels powder.

3.2. Optimization of Dye Adsorption by Response Surface Methodology

The development of second order polynomial regression equation for the analysis of the independent variables and its effect on the removal of CR dyes onto chemically modified durian peels powder was achieved using CCD. Statistical significance of the empirical model expressed in coded factors for the biosorption of CR dyes from aqueous solution is expressed as Equation (2). The biosorption of CR in a batch experiment was performed according to the CCD as presented in

Table 2. Independent Variables and their coded levels for the CCD.

Process Variables	Code	Unit	Coded Variables
			−1	0	+1
pH	$x_1$	-	2	5.5	9
Contact time	$x_2$	min	10	105	200
Initial concentration	$x_3$	Mg g−1	25	212.5	400
Temperature	$x_4$	°C	25	45	65

.

The observed values were compared with the predicted values evaluated from the model in

Table 3. Matrix for the central composite design.

Experiments	Coded Variables				Actual Value	Predicted Value	Residual Value
	$x_1$	$x_2$	$x_3$	$x_4$
1	0	0	0	0	41.60	46.05	−4.45
2	2	0	0	0	63.45	70.41	−6.96
3	1	1	1	1	42.10	47.88	−5.78
4	1	1	1	−1	64.30	69.87	−5.57
5	−1	1	1	−1	45.30	49.43	−4.13
6	1	−1	1	−1	68.82	73.88	−5.06
7	0	0	0	0	47.80	48.46	−0.66
8	−1	−1	1	−1	62.50	70.55	−8.05
9	0	0	0	0	42.30	48.18	−5.88
10	−1	1	−1	1	62.70	75.38	−12.68
11	−1	1	1	1	42.30	51.48	−8.28
12	1	1	−1	−1	66.50	76.30	−9.80
13	1	1	−1	1	41.30	49.07	−7.77
14	0	0	0	0	68.20	76.35	−8.15
15	0	0	2	0	42.60	49.57	−6.97
16	−2	0	0	0	65.60	74.49	−8.89
17	−1	−1	1	1	18.50	10.18	8.32
18	−1	−1	−1	−1	78.40	59.45	18.92
19	−1	1	−1	−1	89.20	75.29	13.91
20	1	−1	−1	1	88.61	75.25	13.36
21	−1	−1	−1	1	85.60	69.53	16.07
22	0	0	0	−2	82.30	71.10	11.20
23	0	0	0	0	67.60	60.90	6.70
24	0	2	0	0	87.54	66.97	20.57
25	1	−1	−1	−1	86.20	81.20	5.00
26	0	0	0	−2	95.20	94.38	0.82
27	0	0	−2	0	85.60	83.20	2.40
28	0	0	0	0	95.70	94.91	0.79
29	0	−2	0	0	89.20	86.40	2.80
30	1	−1	1	1	68.90	61.20	7.70

.

The plot of the normal % probability and studentized residual for the adsorption of CR on modified durian peels powder is shown in Figure 4.

The plot of normal probability is indicated through a straight line, determining whether the residuals follow a normal distribution. It was revealed in Figure 4 that the residuals fall within a straight line, in which case, response transformation is not required [35]. The coefficient of determination of the (R² = 0.8183) obtained in the study was greater than 0.80 [36], suggesting that R² should be at least 0.80 for a good fit of a model. The adequate precision (AP) ratio of the model (9.302) for the biosorption of CR indicated adequacy of the signal for the model. The value of AP higher than 4 is desirable, and confirms that the predicted model can sufficiently navigate the design space as defined by the CCD [37]. The value of AP indicates better precision of the experiment carried out and the reliability of prediction. The determination coefficients R² (0.8183) and adjusted R² R²_adj (0.6487) of the quadratic model indicated good agreement the experimental values and the predicted values of the model as illustrated in Equation (2). The analysis indicated that the independent process variables consisting of pH, contact time, initial concentration and temperature were significant factors with either a positive or negative effect on the percentage removal of dye (

Table 4. ANOVA of response surface quadratic model for biosorption of CR.

Source	Sum of Squares	Degree of Freedom	Mean Square	F-Value	p-Value
Model	11,287.19	14	806.23	4.82	0.0023
$x_1$	3642.53	1	3642.53	21.80	0.0003
$x_2$	2.604	1	2.604	1.56	0.9969
$x_3$	3.66	1	3.66	0.022	0.8843
$x_4$	55.24	1	55.24	0.33	0.5739
$x_1{}^2$	6250.80	1	6250.80	37.41	<0.0001
$x_2{}^2$	680.89	1	680.89	4.07	0.0618
$x_{3 }{}^2$	1061.56	1	1061.56	6.35	0.0235
$x_{4 }{}^2$	1675.65	1	1675.65	10.03	0.0064
$x_1{x}_2$	5.61	1	5.61	0.034	0.0085
$x_1{x}_3$	8.556	1	8.556	5.12	0.0499
$x_1{x}_4$	8.02	1	8.02	0.048	0.0082
$x_2{x}_3$	2.15	1	2.15	0.013	0.9111
Residual	2506.57	15	167.10

).

$$\begin{gathered} Y=+95.20+12.32x_1-0.01x_2+0.39x_3+1.52x_4-15.1x_1{}^2-4.98x_2{}^2-6.22x_3{}^2-7.84x_4{}^2-0.59 x_1{x}_2 \\ +0.023x_1{x}_3+0.71x_1{x}_4 - 0.70x_2{x}_3+0.37x_2{x}_4 - 0.62x_3{x}_4 \end{gathered}$$

(2)

The effectiveness of the predictive ability of the model was found within the desirable correlation coefficient of R² value of (0.9999), which indicated qualitative agreement between the predicted values and experimental values (

Table 5. Model validation of optimum operational parameters and response.

Desirability	$x_1$	$x_2$	$x_3$	$x_4$	%CR Removal		Error
					Actual	Predicted
0.9999	5.5	105	212.50	45	95.70	94.91	0.79

).

Interactive Effect of Two Independent Variables

The 3D response surface indicating percentage removal of CR dyes versus the interactive effect of the independent variables is shown in Figure 5. The 3D plots represent the effect of the interaction of two operational factors with the other variables fixed at zero levels. Figure 5a shows the interaction between pH and time (initial concentration and temperature was fixed at zero level). As can be observed, the maximum dye biosorption was achieved around pH 5, and there was rapid removal of dye within 105 min contact time and a low increase was achieved after this period. This can be attributed to a decrease in available sites as they become saturated with dye molecules [38]. Figure 5b denotes the interaction between pH and initial concentration (contact time and temperature was fixed at zero level). The maximum biosorption of dye was in the pH range of 4.0–6.0. It was indicated that percent dye removal was maximum in acidic conditions and reached maximum at five, while in alkaline pH, a decrease in the adsorption of CR dye onto durian surface was achieved [39]. Biosorption increased when the initial concentration was increased in acidic pH due to presence of active functional groups such as carboxyl, which was positively charged due to protonation on the exposed biosorbent surface, and subsequently attracted to the negatively charged CR dye molecules. As a result, the binding efficiency increased due to the electrostatic interaction between positively charged surface binding sites and the negatively charged dye molecules [40]. Figure 5c depicts the effect of the interaction of pH and temperature (contact time and initial concentration was fixed at zero level). As per the result obtained in Figure 5b, maximum biosorption of dye was obtained as the temperature increased from 45 °C. The increase in biosorption may be attributed to the increased penetration of dyes within the micropores. Figure 5d revealed the effect of the interaction of contact time and initial concentration (pH and temperature was fixed at zero level). The percent removal of CR was rapid in the first 30 min, after which the rate of biosorption became slow. This may be explained as a result of rapid adsorption on the outer surface of the biosorbent, followed by slower adsorption within the micropores [35]. Maximum biosorption, according to the 3D response plot, was obtained at a lower pH and higher contact time and initial concentration and temperature. The optimum adsorption of CR (95.2%) was achieved at the operational conditions of pH, contact time, initial concentration and temperature at corresponding values of 5.5, 105 min, 212.5 mg g⁻¹ and 45 °C, respectively.

3.3. Adsorption Kinetic Studies

Kinetic studies provide information about the prediction of rate of adsorption and mechanism. Accordingly, to determine the rate of loading of dye on the biosorbent, batch experiments were carried out. The treated durian peel materials (0.1 g) were stirred with 50 mL of solution containing the CR dye (100 µg mL⁻¹) at temperature intervals of 10, 20, 30, 40, 60, 90, 120, 180 and 200 min. The dynamics of the adsorption process in terms of the order and the rate constant was evaluated using the kinetic adsorption data. The process of the dye removal from an aqueous phase by any adsorbent can be explained by using kinetic models in order to examine the rate-controlling mechanism of the adsorption process, such as diffusion control, chemical reaction, and mass transfer [41]. The kinetic parameters are useful in predicting the adsorption rate, which can be used as important information in the designing and modeling of the adsorption operation. To investigate the kinetics of the biosorption of CR dye, pseudo first-order, second-order kinetic models were used. The kinetic Equations are as follows (3) and (4).

Pseudo-first-order model:

$$\log \left(q_{e-}{q}_t\right)=\log q_{e-}\frac{K_{1}t}{2.303}$$

(3)

Pseudo-second-order model:

$$\frac{t}{q_t}=\frac{1}{K_{2 }{q}_e{}^2}+\frac{t}{q_e}$$

(4)

where k₁ and k₂ are the rate constants of the pseudo-first-order and pseudo-second- order model, respectively. $q_e$ and $q_t$ represent the adsorption capacity of biosorbent at equilibrium (mg g⁻¹) and at time t, respectively. The pseudo-first-order model indicates reversible adsorption phenomenon, the model signifies that the rate of adsorption is directly proportional to the difference in the saturation of concentration and the number of unoccupied sites of the adsorbate [42]. The pseudo-second-order model defines the rate of adsorption, which is directly proportional to the square of unoccupied sites. The model describes the kinetics of pollutant binding reaction of the biosorbent [41]. The parameters obtained from the kinetics models are shown in

Table 6. Kinetic model for CR adsorption parameters onto acid-treated durian fruit peel.

Pseudo-First–Order Model		Pseudo-Second–Order Model		Intraparticle Diffusion Model
qe(cal) (mg g−1)	26.45	qe(cal) (mg g−1)	111.11	Kid (g mg−1 min−1/2)	11.814
K1 (min−1)	0.0274	K2 (g mg−1 min−1)	0.0068	C	6.950
R2	0.9701	R2	0.9917	R2	0.8241

. It was indicated that a more precise fit of kinetics data was shown by the pseudo-second order model. The calculated qe values are closer to the experimental data than the calculated values of pseudo-first order model and the values of regression coefficients (R²) are higher (0.9917) than pseudo-first order kinetic model.

To examine the diffusion process of CR onto durian peel, Weber-Morris intraparticle diffusion model was studied. The linear equation for Weber-Morris diffusion model is expressed as follows in Equation (5).

$$q_t=K_{id}{t}^{\frac{1}{2}}+C$$

(5)

where, q_t = the amount of CR adsorbed (mg g⁻¹) at time t (min), K_id = the diffusion rate constant (mg g⁻¹min^1/2), and C = constant (mg g⁻¹).

The value of K_id, C and R² were calculated from the gradient of plot q_t Vs. t^1/2 as shown, and the related data is presented in Table 6.

The result revealed that there was multi-linearity for CR dye sorption onto durian peel powder, indicating that more than one rate determining steps was involved in the sorption process. Firstly, linear section indicated external diffusion of dye (CR) onto the durian peel surface. Moreover, second linear portion illustrated the diffusion (intra–particle) of CR, as a slowed process. On the other hand, the plot shown in Figure 6c. illustrates that there are few factors playing a significant role in the sorption process due to the multi-linearity correlation. The deviation in line where it does not pass through the origin, indicates that in the adsorption process the film diffusion and intraparticle diffusion were involved [43].

3.4. Adsorption Isotherm Studies

The adsorption data at equilibrium for a wide range of adsorbate concentrations are well described by various models of adsorption isotherm, such as the Langmuir, Freundlich and Temkin models. The Langmuir isotherm model assumes that adsorption occurs on different functional group sites of the adsorbent with no interaction between the adsorbed species, and that each site can accommodate many molecules (homogeneous adsorption) with the same enthalpy sorption, independent of surface coverage. The model is shown in a linear form according to Equation (6):

$$\frac{C_e}{q_e}=\frac{1}{q_{m}b }+\frac{C_e}{q_m}$$

(6)

where $C_e$ denotes equilibrium concentration in (mg/L), $q_e$ represents the amount of biosorption of dye in (mg/g). The biosorption capacity and energy of adsorption are represented by the Langmuir constants, $q_m$ and b, respectively. The value of $q_m$ and b are obtained from the intercept and slope of the linearized plot $C_e/q_e$ against $C_e$, respectively. The nature of the adsorption is expressed in terms of a dimensionless coefficient known as the separation coefficient ($R_L$) according to Equation (7).

$$R_L=\frac{1}{1+b{C}_o}$$

(7)

The R_L values (R_L = 0) indicates adsorption is irreversible, favorable if (0 < R_L < 1), linear if (R_L = 1) or unfavorable if (R_L > 1) [44].

The Freundlich model assumes that the ratio of the amount of adsorbed solute on a mass of biosorbent to the concentration of solute is not constant at varying concentration.

The linearized form of Freundlich is given by Equation (8).

$$ln{q}_e=ln{K}_{F }+\frac{1}{n }ln{C}_e$$

(8)

A plot of lnq_e versus lnC_e describes the Frendlich isotherm, where K_F is the Freundlich constant which denotes amount of dye adsorbed on biosorbent per unit equilibrium concentration. The magnitude of $\frac{1}{n}$ < 1 illustrates Freundlich isotherm describing the favourability of the adsorption condition. The model parameter evaluated from (Figure 7b) indicated high R² value of 0.9831.

The Temkin model expresses that heat of adsorption decreases linearly due to increased surface coverage [45]. The Temkin model is according to Equation (9):

$$q_e=BlnA+BlnA+Bln{C}_e$$

(9)

where q_e denotes the equilibrium amount of adsorbate adsorbed (mg g⁻¹), C_e indicates the equilibrium concentration of adsorbate (mg L⁻¹), B constant relates to the heat of absorption and is expressed in terms of B = RT/b, b represents the Temkin constant (J mol-1), R signifies the gas constant (8.314 J/mol K), A represents the Temkin constant (Lg⁻¹). q_e is the amount adsorbed at equilibrium (mg g-1) and q_m and b is Langmuir constants related to adsorption efficiency and energy of adsorption, respectively. K_F and n are Freundlich constant. A is the equilibrium binding constant (L mg⁻¹). The slope (B) and intercept (B lnA) are derived from the plot of q_e against lnC_e. R² value of 0.9304 was obtained from the Temkin model.

The obtained experimental data is presented in

Table 7. Adsorption isotherm limits for CR adsorption onto acid-treated durian fruit peel.

Langmuir Isotherm		Freundlich Isotherm		Temkin Isotherm
$q_m$ (mg g−1)	107.52	KF (mg/g (L/mg)1/n)	73.46	A (L mg−1)	18.620
b (L mg−1)	0.0602	n	1.717	B	28.219
R2	0.9989	R2	0.9831	R2	0.9304

.

From the result evaluated from (7), the value of $R_L$ = 0.076 revealed favourable adsorption of CR on the biosorbent. The Langmuir model (Figure 7a) of the biosorption of CR indicated a closer proximity to unity with correlation coefficient value (R² = 0.9989). The Freundlich model assumes that the ratio of the amount of adsorbed solute on a mass of biosorbent to the concentration of solute is not constant at varying concentration. The linearized form of Freundlich is given by Equation (8).

A plot of lnq_e versus lnC_e describes the Frendlich isotherm, where K_F is the Freundlich constant which denotes amount of dye adsorbed on biosorbent per unit equilibrium concentration. The magnitude of $\frac{1}{n}$ <1 illustrates Freundlich isotherm the favorability of the adsorption condition. The model parameter evaluated from (Figure 7b) indicated high R² value of 0.9831. The Temkin model expresses that heat of adsorption decreases linearly due to increased surface coverage [45]. The Temkin model is illustrated as Equation (9).

From the values of maximum adsorption efficiency q_m, it was revealed that the maximum biosorption of CR corresponds to a monolayer of adsorbate molecules on the surface of the modified durian peel with constant energy and with no transmission of adsorbate in the plane of the adsorbent surface. The maximum monolayer adsorption capacity ($q_m$) of modified durian peels for the uptake of CR dye was compared with ($q_m$) for other biosorbents available in the literature [46,47,48,49,50] (

Table 8. Comparison of maximum adsorption capacity for CR by another adsorbent.

Adsorbent	Maximum Adsorption Capacity (mg/g)	References
Egg shell membrane	117.65	[46]
Eichhornia crassipes biomass	14.49	[47]
Myrtus communis (AC-MC)	19.23	[48]
pomegranate (AC-PG)	10.00	[48]
acid-treated pinecone	40.19	[49]
Pinus radiata cone biomass	0.47	[50]
Acid-treated durian fruit peel	105.92	Present work

).

3.5. Thermodynamic Parameter

Thermodynamic parameters for the adsorption process of CR dye removal, including Gibbs free energy change (ΔG°), entropy change (ΔS°) and enthalpy change (ΔH°), were calculated using the following Equations:

$$\Delta \mathrm{G}°=-RTln{K}_c$$

(10)

$$ln{K}_c=\frac{\Delta \mathrm{S}°}{R}-\frac{\Delta \mathrm{H}°}{RT}$$

(11)

where, T (K) = the absolute temperature, R = (8.314 J ${\mathrm{mol}}^{-1}{\mathrm{K}}^{-1}$) which is universal gas constant and $K_c$ = the distribution coefficient.

The nature of the interaction between adsorbate molecules and adsorbent can be classified according to the magnitude of the enthalpy change (ΔH°) and entropy change (ΔS°). The evaluation of enthalpy (ΔH°) and entropy (ΔS°) were calculated from the slope and intercept of ln$K_c$ vs 1/T plot at different experimental temperatures as shown in Figure 8.

It can be seen from

Table 9. Thermodynamic studies for CR adsorption onto acid-treated durian fruit peel.

Δ°H (Kj/mol)	Δ°S (Kj/molK)	Temperature (K)	Δ°G (Kj/mol)
9.219	38.477	298	−7.531
		308	−7.912
		318	−8.328
		328	−8.699
		338	−9.076

that ΔG° values differ from −7.531 to −9.076 kJ${\mathrm{mol}}^{-1}$ as the temperature increases from 298 to 338 K and the -ve value of ΔG° could be attributed to the feasibility and spontaneity of adsorption process with greater sorption performance of CR on acid-treated durian peels powder. The decreasing value of Gibbs free energy (ΔG°) with increasing temperature indicates that the degree of spontaneity of the adsorption process increases with increasing temperature. Generally, changes in adsorption enthalpy denoting physiosorption ranges from −20 to 40 kJ/mol and chemisorption ranges between −400 and −80 kJ/mol [51]. Experimental data shows that positive ΔH° (9.219 kJ/mol) indicates endothermic process and physical adsorption. Hence, the result revealed that physical forces were involved in adsorption of CR onto durian peels powder. Furthermore, the positive values of ΔS° (38.477 Kj/molK) suggest increased disorderliness of solid-liquid interfacial interaction during the adsorption process of dye (CR) onto durian peels biosorbent [52]. The magnitude and sign of ΔS° gives an indication whether the adsorption process is an associative or dissociative mechanism. A positive ΔS° value indicates that the adsorption procedure comprises a dissociative mechanism.

3.6. Biosorption Mechanism

The mechanism for the adsorption of CR dyes onto modified durian peels is illustrated Figure 9. The removal mechanism of biosorption of CR dyes has essentially been subjected to electrostatic interactions and hydrogen bonding. According to FTIR spectra, different functional groups such as hydroxyl, ester, ether and carboxylic groups were predominant on the surface of the durian fruit peels. From the FTIR spectra, some shift (8 to 22 cm⁻¹) was observed in position of hydroxyl and carboxylic groups after adsorption of CR dye suggesting that these groups were mainly responsible for the adsorption of CR dye onto the surface of durian peels.

The biosorption system depends on the active sites which also depends on the pH of the adsorbate. The surface interaction of dye molecules on active sites of the biosorbent is greatly influenced by the solution pH [53]. In this study, maximum adsorption capacity of CR dyes on modified durian peels were observed under acidic solution (pH 5). In acidic pH, surface charge of durian fruit peels was positively charged and more protonated sites (H⁺) were present on the surface resulting in increased electrostatic interaction between the positively charged surface and negatively charged CR dye molecules [54]. Furthermore, in basic pH, the surface of durian peels contained more hydroxyl ions (OH⁻), which resulted in the deprotonation of carboxylic and hydroxyl groups. This may cause a decrease in electrostatic interaction between surface charge of durian peel and negatively charged CR dye molecules [55]. Aside from electrostatic interaction, H-bonding (between electronegative atom and H-atom) and pi-pi interaction also played a significant role in CR adsorption onto the surface of the modified durian peels powder. According to isotherm studies, experimental data were better fitted to the Langmuir model for the adsorption of CR dyes molecules on the surface of the biosorbent, which implies that monolayer adsorption occurs on the homogeneous surface filled with localized adsorption sites. Same findings reported CR adsorption onto PEI-wheat straw and coffee waste [56,57].The biosorption of CR dyes investigated under different operational conditions indicated that under acidic pH, increased adsorption efficiency was favoured through electrostatic interactions, along with hydrogen bonding which enhanced the adsorption efficiency of acid-treated durian peels for the adsorption of CR dye molecules.

4. Conclusions

In this study, acid-treated durian peels were prepared and investigated as a possible biosorbent for the removal of CR dyes in aqueous solution. The biosorption investigation was carried out as a function of pH ($x_1$), contact time ($x_2$), initial concentration ($x_3$), and temperature ($x_4$). Percentage removal of CR dye molecules increased in acid pH condition as the initial concentration increased, and also as the contact time and temperature increased to a certain level. Therefore, the biosorption of CR was effective within the range of the investigated parameters. The optimum condition for 95.2% removal of CR was achieved at pH 5, contact time of 105 min at initial concentration of 212.5 mg g⁻¹ within 45 °C. The Langmuir isotherm model best described the biosorption process. The kinetic data was better fitted to the pseudo-second-order model. Overall, the thermodynamic parameters revealed that the adsorption of CR dyes on acid-treated durian peels powder was spontaneous, endothermic and physical forces determined the biosorption of CR. Generally, the rapid adsorption of CR at shorter contact time was attributed to film diffusion and intraparticle diffusion at longer adsorption time. The mechanism of biosorption was influenced by electrostatic interaction and hydrogen bonding. The high removal capacity of CR dyes indicted that acid-treated durian peels could be used as an alternative biosorbent for the treatment of effluents from textiles and dyeing industries.