Kinetic, Isothermal evaluation of Jackfruit Activated Carbon to removal Methylene Blue, and optimization by Response Surface Methodology (RSM)

Chia sẻ: _ _ | Ngày: | Loại File: PDF | Số trang:13

Thêm vào BST

Báo xấu

10
lượt xem 2
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

Waste from Jackfruit has little economic value and often causes serious environmental problems. The conversion of jackfruit waste will help reduce waste treatment costs, provide a potential source of raw materials for the activated carbon (AC) and increase economic value. In this study, the ACs from the Jackfruit shell and pulp were synthesized by microwave assistance for removing organic dyes.

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Kinetic, Isothermal evaluation of Jackfruit Activated Carbon to removal Methylene Blue, and optimization by Response Surface Methodology (RSM)

Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 Kinetic, Isothermal evaluation of Jackfruit Activated Carbon to removal Methylene Blue, and optimization by Response Surface Methodology (RSM) Bich Ngoc Hoang1*, Thi Kim Ngan Tran1, Thi Cam Quyen Ngo1 1 Institute of Applied Technology and Sustainable Development, Nguyen Tat Thanh University, Ho Chi Minh City, Vietnam * Corresponding author: bichhn@ntt.edu.vn ARTICLE INFO ABSTRACT Waste from Jackfruit has little economic value and often causes serious environmental problems. The conversion of jackfruit waste will help reduce waste treatment costs, provide a potential source of raw materials for the activated carbon (AC) and increase economic value. In this study, the ACs from the Jackfruit shell and pulp were synthesized by microwave assistance for removing organic dyes. The removal process of activated carbon was predicted through the fit of kinetic and isothermal adsorption models. RSM models was used to optimize the adsorption capacity of AC. Research results show that ACJP has an adsorption mechanism to follow the second- order kinetic model, Elovich and Langmuir. ACJS was showing a high correlation for Elovich, Bangham, Langmuir, and Temkin models. For the RSM optimization model, ACJP and ACJS were Keywords: optimized with influencing parameters such as pH, Activated carbon, Response concentration, and dosage. The optimal adsorption capacity of Surface Methodology, Kinetic, activated carbon is 94 mg.g-1 (ACJS) and 90 mg.g-1 (ACJP) with Isothermal, Jackfruit. a high correlation coefficient R2=1. 1. Introduction Among the most relevant multivariate techniques used in analytical optimization is the response surface method (RSM). The response surface method (RSM) was developed by Box and collaborators in the 1950s (Bezerra et al., 2008; Gilmour, 2006). RSM includes a group of mathematical and statistical techniques based on the fitting of experimental models to experimental data obtained related to the experimental design. RSM defines the effect of the independent variables, alone or in combination in the processes (Montgomery, 2017; Shojaeimehr et al., 2014b). This method has many advantages like being more economical, needing fewer experiments number, studying the interaction between parameters on response, predicting of the response, checking of method adequacy, and requiring a shorter time. The goal is to simultaneously optimize the extent of these variables for the best performance (Bezerra et al., 2008; Hanrahan & Lu, 2006; Teófilo & Ferreira, 2006). RSM model has been applied a lot in scientific research topics in many fields. In which the field of adsorbent materials, RSM model is applied to evaluate the optimal adsorption capacity of materials or used to evaluate the synthesis process in the most optimal way. In the study of Chijioke Elijah Onu and colleagues, the application of RSM to evaluate the optimal adsorption capacity of modified clay materials for black-T color. The results showed that the material adsorbed well at 30 minutes, pH4, 0.4 g.L-1 content, and temperature 35 °C (Onu et al., 2021). In the study of the author group, Surafel Mustefa Beyan applied activated carbon from bagasse to remove BOD and COD in textile dyeing wastewater. The results show that the 22
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 optimal condition from the RSM model is the content of 0.915 g.L-1, pH3.2 in 98.91 minutes, with the ability to remove BOD and COD at 0.0225 mg.g-1 and 0.023 mg.g-1, respectively (Beyan et al., 2021).In addition, the study of Alfarooq O. Basheer and colleagues investigated the Al3+ removal ability of activated carbon with KOH. Research results have shown that the adsorption capacity of 9,958 mg/g is based on the RSM model. The model shows optimal conditions at 655.77 °C for 1.09 hours with immersion ratio of 2.63 to reach an adsorption capacity of 9.95 mg.g-1 (Basheer et al., 2021). Soheila Gholamiyan's study also showed a high degree of agreement (R2 = 0.9697) when optimizing the erythromycin adsorption process of magnetic activated carbon using the RSM model. The results show that the treatment process is optimized at the initial concentration of 65 mg.L-1, the content of 1.55 g.L-1 in the contact time is 76.25 minutes at 35 oC for the treatment efficiency of 95.125% (Gholamiyan et al., 2020). Next, the research of the author group Rahele Bagheri also used the RSM model to optimize the processing of malachite green with activated carbon. The results show a high degree of agreement between the experimental data and the predicted data. The adsorption process was optimized at 75 minutes, pH7, color concentration 19.89 mg.L-1 with the amount of adsorbent 0.027 g (Bagheri et al., 2019). Tahereh Shojaeimehr's research group optimized the Cu adsorption process using the RSM model. The results show that the correlation coefficient R2 = 0.941, with optimal adsorption conditions with pH = 4.6, temperature 50 oC, Cu2+ concentration is 150 mg.L-1 with the adsorbent weight is 50 mg. Optimum adsorption conditions were selected with the highest adsorption capacity reaching 99,289 mg.g-1 (Shojaeimehr et al., 2014a). Rama Rao Karri's research group also used RSM to optimize the Zn adsorption process with a rather high correlation coefficient. The results showed that the time condition of 45 minutes, the concentration of 55 mg.L-1, the adsorbent content of 11 g.L- 1 , the temperature of 50 oC and pH5 gave a high correlation coefficient (R2 = 0.852) with the efficiency. The treatment result reached 90% (Karri & Sahu, 2018). In addition, Nawal Taoufk's research group also applied the surface response model to optimize the alicyclic acid adsorption process by activated carbon material with a correlation coefficient of 0.91, showing a high degree of suitability of the alicyclic acid. experimental data and predictions. From here, the research team has successfully optimized the treatment process with adsorption conditions at 60 minutes, adsorbent content 25 mg, concentration 15 mg.L-1, pH6 (Taoufik et al., 2021). From the model that the studies have given the optimal conditions for the processing. In this study, the optimization process is performed on RSM model. The material used to evaluate the adsorption process is activated carbon. Activated carbon is the most commonly used and versatile adsorbent because of its extremely large surface area and microparticle volume (Do, 1998), large adsorption capacity, fast adsorption kinetics, and relative ease of regeneration (Khosravi et al., 2017). They are manufactured from a variety of carbon-based materials. The choice of precursor is largely dependent on its availability, cost and purity, but the production process and intended application of the product are also important considerations (Liang et al., 2020). Agricultural wastes are considered as a very important source of raw materials because of especially two facts: they are renewable sources and low cost raw materials (Prahas et al., 2008). Many agricultural by-products such as coconut shells (Endut et al., 2017; Khosravi et al., 2017; Liang et al., 2020), olive mills (Marrakchi et al., 2020), coffee bean husks (Rebollo-Hernanz et al., 2021), sawdust rubberwood (Yusoff, 2020), Sugarcane bagasse (Beyan et al., 2021), durian (Latib et al., 2013), jackfruit (Bhushan et al., 2021), was found to be a suitable precursor for activated carbon due to its high carbon content and low ash content. Jackfruit is commonly used in South and Southeast Asian cuisines. Jackfruit originated in India and spread to tropical regions, including Vietnam. Jackfruit can be harvested all year round, corresponding to a huge amount of by-products from jackfruit. Waste from jackfruit peels has no economic value and in fact often causes serious disposal problems for the local environment. Converting jackfruit peels into activated carbon would increase its economic value, help reduce waste disposal costs, and provide a potentially inexpensive raw material for commercial 23
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 activated carbon. In this study, the raw material for activated carbon was collected by collecting jackfruit shell and jackfruit pulp. 2. Experimental 2.1. Chemicals Activated carbon materials were used from the shell and pulp of jackfruit after preliminary processing. Raw materials were taken from Thu Duc Farmers Market, City. Ho Chi Minh. Methylene blue (MB) was acquisition from HIMEDIA (Himedia Laboratories Pvt., Ltd). It was shown molecular formula and the wavelength of maximum absorption (λmax) in Figure 1. MB (C16H18ClN3S) MW: 319.85 g/mol has a maximum wavelength 664nm (Altıntıg et al., 2017; Asfaram et al., 2015; Yagub et al., 2014). Dye wastewater in laboratory is prepared by melting a industrial dye in distilled water. N H3C CH3 + N S N CH3 Cl CH3 Figure 1: The molecular structure of MB 2.2. Preparation of Activated carbon Purchased materials were removed to prevent damage and rinsed with water. The raw materials were cut into small samples of 1 cm x 1 cm size and dried at 100 ℃ for 24 h. The raw material sample after drying was finely ground before conducting activation. NaOH was used to activate the raw materials as the chemical activation method (Foo & Hameed, 2012). 50g of raw materials were impregnated with the ratio of 1:1 (activator mass/raw material mass) for 2 hours. The mixture from raw material and NaOH was dried at 100 °C for 24 h. Then, the mixture was carbonized by microwave assisted method at 600 W for 2 min. After activation, the activated carbon products were collected and cooled to room temperature. The activated carbon products were repeatedly washed with HCl (1M) and distilled water to remove residual NaOH (Foo & Hameed, 2012). Washing was terminated until the constant pH. Finally, the activated carbon from Jackfruit shell (ACJS) and Jackfruit pulp (ACJP) were dried at 110 °C for 24 h and stored in a desiccator for later use in the experiment. 2.3. Adsorption experiment (0.5 – 2 g.L-1) material was added to 100 ml of Methylene Blue (MB, concentration 0 – 200 mg.L-1) in erlen. The mixture from the material and the dye solution was shaken on a Thermal Incubation Shaker at 200 rpm for 4 h at 30 °C. The dye solutions were analyzed using an Evolution 60s UV−Vis spectrophotometer (thermal fisher scientific, USA) at 664 nm to record the concentration value. Each experiment was performed three times under identical conditions to determine the mean. The dye adsorption capacity (Qe) was calculated according to the formula (1): (𝐶 𝑜 −𝐶 𝑒 ) 𝑄𝑒 = (1) 𝑊.𝑉 where: Ce is the concentration after adsorption (mg.L-1); Co is the concentration before adsorption (mg.L-1); V (L) is the volume of solution and W (g) is the mass of adsorbent 24
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 2.4. Adsorption kinetics MB adsorption kinetic models were assumed the adsorption process and reaction rates for AC. Experimental data was experimented and described on kinetic models such as Elovich, Bangham, pseudo-first-order and pseudo-second-order in the following non-linear forms: 1 𝑞𝑡 = 𝑙𝑛(1 +  𝛽𝑡) (2) 𝛽 𝑞 𝑡 = 𝑘 𝐵 𝑡 𝐵 (3) 𝑞 𝑡 = 𝑞 𝑒 (1 − 𝑒 −𝑘1 𝑡 ) (4) 𝑒 𝑞 2 𝑘2 𝑡 𝑞 𝑡 = 1+𝑘 (5) 2 𝑡𝑞 𝑒 Where 𝑞 𝑡 (mg.g-1) is the adsorption capacity of adsorbent at time t, 𝛽 is the desorption rate, 𝛼 is the rate of chemical absorption, 𝑘 𝐵 and 𝛼 𝐵 are Bangham constants, 𝑘1 (min-1) and 𝑘2 (g.(mg.min)-1) are the rate constant of the pseudo-first-order and the pseudo-second-order, 𝑞 𝑒 (mg.g-1) is the adsorption capacity at the equilibrium. 2.5. Adsorption equilibrium isotherm The adsorption isotherm is the basis for considering the interaction between adsorbents and adsorbents in gas / liquid / solid interfaces. This may indicate that important parameters related to the adsorption mechanisms and the adsorption processes are provided by the adsorption isotherms. The advantages and behavior of adsorption processes were precisely described by isothermal models such as Langmuir, Freundlich, Temkin and Dubinin- Radushkevich (D-R). Models were calculated by the following formula: 𝑄 𝑚 𝐾𝐿 𝐶𝑒 1 𝑞𝑒 = (𝑅 𝐿 = 1+𝐾 ) (6) 1+𝐾 𝐿 𝐶 𝑒 𝐿 𝐶𝑜 1/𝑛 𝑞𝑒 = 𝐾𝐹 𝐶𝑒 (7) 𝑅𝑇 𝑞 𝑒 = 𝐵 𝑇 ln(𝑘 𝑇 𝐶 𝑒 ) (𝐵 𝑇 = ) (8) 𝑏 2 1 1 𝑞 𝑒 = 𝑄 𝑚 𝑒 −𝐵 (e=√2𝐵; =r.t.ln(1+ 𝐶 )) (9) 𝑒 Where ce (mg.L ) is the amount of dye remaining at equilibrium, q m (mg.g-1) and q e -1 -1 (mg.g ) are the adsorption content at maximum and the time of equilibrium in 1 g adsorbent, k l (L.mg-1) is constant Langmuir, rl is dissociation coefficient and rl > 1 (unfavorable), rl = 1 (linear), 0 < rl < 1 (favorable) and rl = 0 (not reversed). In the Freundlich equation, Freundlich constant and the adsorption intensity are k f [(mg.g-1).(L.mg-1)n] and 1/n with n < 1 (one-layer adsorption) and n > 1 (multi-layer adsorption). In the Temkin equation, the relation to the equilibrium link has the isotherm Temkin constant (k t , L.g-1), the correlation coefficients with adsorption heat and isothermal constants are bt and b (L.mol-1). e (KJ.mol-1) is the adsorption energy, 𝜀 is the Polanyi potentiality described and bt (Mol2 . (Kj2 )−1) is the D-R constant, in D- R equation. 2.6. Response Surface Methodology (RSM) In this study, response surface method (RSM) was used to optimize MB adsorption on activated carbon samples from peel and pulp with influencing factors being solution pH, function amount of adsorbent and initial MB concentration. The evaluation factors of the model are divided into 5 levels: central variable (0), low level (-1), high level (+1) and level ±α (±α). Analysis of variance (ANOVA) was calculated using Design-Expert software (version 11, State Ease, Minneapolis, USA). ANOVA of quadratic linear regression model is used to analyze the influence of input and output variables as well as the correlation of response functions and independent variables (Khuri & Mukhopadhyay, 2010). 25
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 Giá trị α được tính theo công thức: 𝛼 = 2(𝑘−𝑝)/4 = 2(4−0)/4 = 2 (10) Tổng số thí nghiệm của Ma trận CCD được tính theo công thức: 𝑁 = 2 𝑘 + 2 × 𝑘 + 𝑐 = 24 + 2 × 4 + 6 = 30 (11) Table 1 Matrix of independent factor and levels. Levels No Independent factor Symbol -α -1 0 +1 +α 1 pH X1 6,32 7 8 9 9,68 2 dosage (g.L-1) X2 0,33 0,4 0,5 0,6 0,67 3 Concentration (mg.L-1) X3 32,95 50 75 100 117,05 3. Results and discussion 3.1. Kinetic and Isothermal Models of Adsorption The best factors used to evaluate the kinetic and isothermal models of adsorption include time (120 min), pH (pH8), temperature (30 oC), content (0.5 g.L-1), and concentration (100 mg.L-1). The adsorption process would be evaluated using pseudo-first-order, pseudo-second- order, Elovich, and Bangham kinetic models. The kinetic models of ACJP and ACJS have been shown in Figure 2. Looking at Figure 2, it could be seen that the distribution of experiments according to the model was very uniform. Besides, the results of the model were shown in Table 2. In the results, ACJP was following the pseudo-second-order model with R2= 0.995, but the Elovich model with a high coefficient (R2 = 0.992). This shows that the adsorption of ACJP was represented by chemisorption with heterogeneous diffusion. For ACJS, it was following the Elovich and Bangham model with the coefficients R2 = 0.998 and 0.999. It could be seen that the adsorption process of ACJS was described by adsorption with intraparticle diffusion with chemical interactions. Chemical interaction was one of the important interactions in the adsorption process. (A) (B) 100 100 Adsorption capacity (mg/g) Adsorption capacity (mg/g) 80 80 60 60 40 40 ACJP ACJS Pseudo first order 20 Pseudo first order 20 Pseudo second order Pseudo second order Elovich Elovich 0 Bangham 0 Bangham 0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350 Time (min) Time (min) Figure 2: Kinetic Models of ACJP (A) and ACJS (B). Table 2 Kinetic parameters for the adsorption of MB on ACJP and ACJS Kinetic Models Parameters ACJP ACJS Pseudo-first-oder qe (mg/g) 94.764 83.107 26
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 k1 (min-1) 0.059 0.070 R2 0.960 0.910 qe (mg/g) 101.334 88.610 k2 (min-1) 9.066 0.001 Pseudo-second-oder R2 0.995 0.969 H=k2qe2 93095 7.852  (g/mg) 0.076 0.091 Elovich  (mg/g.min) 93.476 127.201 R2 0.992 0.998 kB (mL/(g/L)) 41.406 38.090 Bangham B 0.158 0.151 R2 0.984 0.999 The adsorption process would be evaluated by the isotherm model Langmuir, Freundlich, Temkin, DR. The isotherm model of ACJP and ACJS was shown in Figure 3. It can be seen that the distribution of experiments according to the model was very uniform. Besides, the results of the model were shown in Table 3. Looking at the results, ACJP was following the Langmuir model with R2 = 0.988. This shows that the adsorption of ACJP was represented as monolayer adsorption with uniformity of adsorption points on the surface. For ACJS, it was following the Langmuir and Temkin models with the coefficients R2 = 0.988 and 0.983. It can be seen that the adsorption process of ACJS was described by the interaction between the adsorbent and the adsorbent on the monolayer surface and was uniform in terms of adsorption point. Through the process of isothermal evaluation and adsorption kinetics The maximum adsorption capacity was recorded at 243,316 mg.g-1 (ACJP) and 230.331 mg.g-1 (ACJS) according to Langmuir model. The adsorption process was described through how well it fits the models with the proposed adsorption mechanisms. The adsorption process is described in detail through the evaluation of adsorption kinetic and isothermal models. ACJP has an adsorption mechanism to follow the second-order kinetic model, Elovich and Langmuir. ACJS was showing a high correlation for Elovich, Bangham, Langmuir, and Temkin models. 27
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 (A) (B) Model Equation Reduced Chi-Sqr 180 160 Adsorption capacity (mg/g) Adj. R-Square Adsorption capacity (mg/g) Adsorption capacity 160 Adsorption capacity 140 140 120 120 100 100 80 80 60 60 ACJP ACJS 40 40 Langmuir Langmuir Freundlich 20 Freundlich 20 Temkin Temkin 0 DR 0 DR 0 50 100 150 200 0 50 100 150 200 Concentration (mg/L) Concentration (mg/L) Figure 3. Isothermal Models of ACJP (A) and ACJS (B). Table 3 Isothermal parameters for the adsorption of MB on ACJP and ACJS Isothermal Models Parameters ACJP ACJS Langmuir KL(L/mg) 0.012 0.013 Qm (mg/g) 243.316 230.331 R2 0.988 0.988 RL 0.604 0.581 Freundlich KF (mg/g) 9.718 10.100 1/n 1.812 1.855 R2 0.957 0.952 Temkin kT (L/mg) 0.129 0.131 BT 50.946 49.608 R2 0.978 0.983 D-R B (mol2/kJ2) 154.307 139.785 Qm (mg/g) 152.236 147.215 R2 0.915 0.931 E (kJ/mol) 0.057 0.060 3.2. Optimization model From the best adsorption conditions, the RSM model is built based on the central and boundary values. Based on CCD, the experimental results to evaluate the MB adsorption efficiency of the material samples are presented in Table 4. From the RSM matrix, a total of 20 experiments were set up. 28
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 Table 4 Table of CCD values of 20 independent experiments. Experimental value Predicted value No pH concentration Dosage ACJP ACJS ACJP ACJS 1 7 50 0,4 81,86 86,00 81,87 86,22 2 9 50 0,4 92,16 86,14 91,68 86,32 3 7 100 0,4 61,94 52,20 62,07 52,40 4 9 100 0,4 55,47 57,00 55,19 57,30 5 7 50 0,6 87,54 92,00 87,86 92,54 6 9 50 0,6 99,94 100,00 99,85 100,64 7 7 100 0,6 76,16 72,00 76,68 72,66 8 9 100 0,6 71,94 84,94 71,97 85,56 9 6,32 75 0,5 80,00 74,00 79,43 73,44 10 9,68 75 0,5 83,23 85,00 83,73 84,37 11 8 32,95 0,5 99,27 98,00 99,43 97,47 12 8 117,05 0,5 59,57 57,00 59,35 56,35 13 8 75 0,33 65,50 69,00 65,88 68,87 14 8 75 0,67 85,48 99,00 85,03 97,95 15 8 75 0,5 85,53 88,50 85,59 88,63 16 8 75 0,5 85,04 88,50 85,59 88,63 17 8 75 0,5 86,36 88,73 85,59 88,63 18 8 75 0,5 85,56 90,56 85,59 88,63 19 8 75 0,5 85,07 87,97 85,59 88,63 20 8 75 0,5 85,96 87,31 85,59 88,63 From the RSM model, the results obtained include: the quadratic equation of the model, residual value plots, line and 3D graphs between factors, and analysis of variance table. The quadratic equation describing the correlation between the response factor (y) and the independent variables is determined as follows: HACJP (%) = 85,59 + 1,28*A – 11,92 *B + 5,69*C – 4,17*AB + 0,54*AC + 2,15*BC – 1,42*A2 – 2,19*B² – 3,58*C² HACJS (%) = 88,63 + 3,25*A – 12,23*B + 8,64*C + 1,2*AB + 2*AC + 3,48*BC – 3,44*A² – 4,14*B² – 1,85*C² The data of analysis of variance and ANOVA for the regression equations are presented in tables 5, and 6. The significance of the regression models is determined by the correlation coefficient (R2), the p-value and F. In general, the smaller the p-value and the larger the F-value, the more statistically significant it is. paradigm. The model is effective when p < 0.05, the model is not effective when p > 0.05. A p value of less than 0.05 shows the statistical significance of the effect factors (95% confidence level). The results of ANOVA analysis show that the model is statistically significant (P 0.75 and the ACJS material R2 = 0.997 > 0.75. Model is compatible with the experiment. 29
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 Table 5 Analysis of variance of mathematical models of ACJP and ACJS. Parameter ACJP ACJS SSres 3,01 10,20 SSlof 1,70 4,26 SSE 1,31 5,94 SStot 2832,19 3756,98 MSE 0,2620 1,19 R² 0,999 0,997 R² correction 0,998 0,995 R² prediction 0,995 0,989 Table 6 ANOVA values of the RSM Sample Parameter Sum of squares Average squared F-value p-value Model 2829,18 314,35 1043,96 < 0,0001 A-pH 22,26 22,26 73,93 < 0,0001 B-concentration 1939,74 1939,74 6441,80 < 0,0001 C-dosage 442,66 442,66 1470,08 < 0,0001 AB 139,36 139,36 462,82 < 0,0001 ACJP AC 2,37 2,37 7,86 0,0187 BC 37,11 37,11 123,24 < 0,0001 A² 28,92 28,92 96,04 < 0,0001 B² 69,23 69,23 229,90 < 0,0001 C² 184,86 184,86 613,91 < 0,0001 model 3746,78 416,31 408,10 < 0,0001 A-pH 144,22 144,22 141,38 < 0,0001 B-concentration 2040,99 2040,99 2000,76 < 0,0001 C-dosage 1020,49 1020,49 1000,38 < 0,0001 AB 11,52 11,52 11,29 0,0072 ACJS AC 32,00 32,00 31,37 0,0002 BC 97,16 97,16 95,25 < 0,0001 A² 170,19 170,19 166,84 < 0,0001 B² 247,44 247,44 242,56 < 0,0001 C² 49,08 49,08 48,12 < 0,0001 From there, the RSM model was used to optimize the experimental conditions such as color concentration (mg.L-1), amount of ACJP or ACJS adsorbent (g.L-1) and pH value. Looking at Figure 4 and Figure 5, the surface response plot can also see the highest point of the model for the optimal value of the material. After the experimental procedure and using the 30
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 RSM model, it can be seen that the ACJP material has a convergence point of pH > 8, the concentration is from 0 to 50 mg.L-1, and the content is in the range < 0.5 g.L-1. For ACJS materials with convergence point with pH value in the range of pH8 to pH10, the concentration is about 20 mg.L-1 to 60 mg.L-1, and the concentration ranges from 0.6 g.L-1 to 1 g.L-1. The surface histograms and the 3D model both show the optimal region of the evaluation model. The interaction between the elements shows convergence in the 3D figure. Design-Expert® Software Factor Coding: Actual H (%) Design Points 55.47 99.94 Design-Expert® Software Factor Coding: Actual X1 = A: pH H (%) Design-Expert® Software X2 = B: concentration (A) (C) Design Points (E) Factor Coding: Actual 55.47 99.94 Actual Factor H (%) C: dose = 0.5 Design Points X1 = A: pH 55.47 99.94 X2 = C: dose Actual Factor X1 = B: concentration B: concentration = 50 X2 = C: dose Actual Factor A: pH = 8 140 H (%) H (%) H (%) 0.95 0.9 60 112 70 0.75 0.7 B: concentration (mg/L) 80 84 C: dose (g/L) C: dose (g/L) 6 0.55 0.5 6 90 100 56 100 100 60 90 90 0.35 0.3 80 28 80 70 60 Design-Expert® Software Factor Coding: Actual H (%) Design points above predicted value Design points below predicted value 55.47 99.94 70 X1 = B: concentration X2 = C: dose 60 Actual Factor A: pH = 8 0 0.15 0.1 Design-Expert® Software Factor Coding: Actual H (%) Design-Expert® Software Design points above predicted value Factor Coding: Actual Design points below predicted value 55.47 99.94 H (%) Design points above predicted value Design points below predicted value X1 = A: pH 55.47 99.94 X2 = C: dose Actual Factor X1 = A: pH B: concentration = 50 X2 = B: concentration Actual Factor 4 6 8 10 12 C: dose = 0.5 4 6 8 10 12 14 16 -120 -68 -16 36 88 140 A: pH A: pH (B) (D) (F) 140 B: concentration (mg/L) 150 120 98 88 100 56 56 50 14 H (%) 24 H (%) H (%) -28 0 -8 -70 -50 -40 0.95 16 0.9 140 12 112 14 0.7 10 0.75 84 12 0.5 8 0.55 10 C: dose (g/L) 140 56 0.3 36 88 8 -16 B: concentration (mg/L) 28 6 A: pH C: dose (g/L) 0.35 A: pH -68 6 0.1 -120 0 4 0.15 4 B: concentration (mg/L) Figure 4. Line diagram and surface plot of ACJP Design-Expert® Software Design-Expert® Software Design-Expert® Software Factor Coding: Actual Factor Coding: Actual Factor Coding: Actual H (%) H (%) H (%) Design Points Design Points Design Points 52.2 100 52.2 100 52.2 100 (A) (C) (E) X1 = A: pH X1 = A: pH X1 = B: concentration X2 = B: concentration X2 = C: dose X2 = C: dose Actual Factor Actual Factor Actual Factor C: dose = 0.5 B: concentration = 75 A: pH = 8 120 H (%) 1.4 H (%) 1.4 H (%) 86 1.1 1.025 B: concentration (mg/L) 6 52 C: dose (g/L) C: dose (g/L) 0.8 0.65 18 95 6 60 95 90 95 90 0.5 0.275 90 6 80 80 -16 80 70 70 70 60 60 Design-Expert® Software Factor Coding: Actual H (%) Design points above predicted value -50 0.2 -0.1 Design points below predicted value 52.2 100 X1 = A: pH X2 = B: concentration Actual Factor C: dose = 0.5 Design-Expert® Software Factor Coding: Actual H (%) Design points above predicted value 4 6 8 10 12 5 7 9 11 13 -55 -11 33 77 121 165 Design points below predicted value 52.2 100 X1 = A: pH X2 = C: dose Actual Factor B: concentration = 75 Design-Expert® Software Factor Coding: Actual H (%) Design points above predicted value Design points below predicted value 52.2 100 X1 = B: concentration X2 = C: dose Actual Factor A: pH = 8 A: pH A: pH B: concentration (mg/L) (B) (D) (F) 100 80 200 150 100 60 100 0 40 50 -100 20 H (%) 0 -200 H (%) 0 H (%) -50 -300 -20 -40 -100 5 1.4 7 1.025 165 4 0.2 9 0.65 121 6 0.5 A: pH 77 -50 -16 8 A: pH 0.8 11 C: dose (g/L) 0.275 33 18 52 10 1.1 -11 B: concentration (mg/L) 86 120 12 C: dose (g/L) 1.4 13 -0.1 -55 B: concentration (mg/L) Figure 5. Line diagram and surface plot of ACJS The optimal values will be calculated and selected based on the optimal area from the influencing factors. From Figure 6, the optimal values of ACJP and ACJS materials are clearly shown with the most optimal values. For ACJP materials, the optimum value was recorded at pH8.97, concentration of 50.5 mg.L-1, content of 0.57 g.L-1. Under optimal conditions, the adsorption capacity predicted from the model is assumed to be 90.82 mg.g-1 with 100% efficiency. For ACJS materials, the optimum value was recorded at pH8.61, concentration 52 mg.L-1, content 0.57 g.L-1. Under optimal conditions, the adsorption capacity predicted from the model is assumed to be 94.04 mg.g-1 with 100% efficiency. From the optimal results, it can be seen that the activated carbon synthesized from the shell or the pulp of the jackfruit gives the same maximum adsorption capacity. 31
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 (A) (B) Figure 6. Optimal values from the model of ACJP (A) and ACJS (B). 4. Conclusions & recommendations The factors affecting the adsorption process were evaluated such as time, pH, temperature, dosage, and concentration. Maximum adsorption capacity was recorded at 243.316 mg.g-1 (ACJP) and 230.331 mg.g-1 (ACJS) according to Langmuir model with best conditions such as time (120 min), pH (pH8), temperature (30 oC), content (0.5 g.L-1), and concentration (100 mg.L-1). The adsorption process is described in detail through the evaluation of adsorption kinetic and isothermal models. ACJP has an adsorption mechanism to follow the second-order kinetic model, Elovich and Langmuir. ACJS was showing a high correlation for Elovich, Bangham, Langmuir, and Temkin models. Besides, RSM model is used to optimize the adsorption process. Optimum parameters of ACJS were recorded at pH8.61, concentration 52 mg.L-1, content 0.57 g.L-1 with correlation system R2 = 1(H = 100%, Q = 94 mg.g-1). Optimum parameters of ACJP were recorded at pH8.97, concentration 50.5 mg.L-1, content 0.57 g.L-1 with correlation system R2 = 1 (H = 100%, Q = 90 mg.g-1). ACKNOWLEDGEMENTS This research has been funded by a grant from the Science and Technology Development Foundation of Nguyen Tat Thanh University (2021.01.170/HĐ-KHCN). The study was supported by The Youth Incubator for Science and Technology Programme, managed by Youth Development Science and Technology Center - Ho Chi Minh Communist Youth Union and Department of Science and Technology of Ho Chi Minh City, the contract number is " 40/2021/ HĐ-KHCNT-VƯ " signed on 8th, December, 2021. References Altıntıg, E., Altundag, H., Tuzen, M., & Sarı, A. (2017). Effective removal of methylene blue from aqueous solutions using magnetic loaded activated carbon as novel adsorbent. Chemical Engineering Research and Design, 122, 151–163. https://doi.org/10.1016/j.cherd.2017.03.035 Asfaram, A., Ghaedi, M., Hajati, S., Rezaeinejad, M., Goudarzi, A., & Purkait, M. K. (2015). Rapid removal of Auramine-O and Methylene blue by ZnS:Cu nanoparticles loaded on activated carbon: A response surface methodology approach. In Journal of the Taiwan Institute of Chemical Engineers (Vol. 53, pp. 80–91). https://doi.org/http://dx.doi.org/10.1016/j.jtice.2015.02.026 Bagheri, R., Ghaedi, M., Asfaram, A., Alipanahpour Dil, E., & Javadian, H. (2019). RSM-CCD design of malachite green adsorption onto activated carbon with multimodal pore size distribution prepared from Amygdalus scoparia: Kinetic and isotherm studies. Polyhedron, 171, 464–472. https://doi.org/10.1016/j.poly.2019.07.037 Basheer, A. O., Hanafiah, M. M., Alsaadi, M. A., Al-Douri, Y., & Al-Raad, A. A. (2021). Synthesis and optimization of high surface area mesoporous date palm fiber-based nanostructured powder activated carbon for aluminum removal. Chinese Journal of 32
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 Chemical Engineering, 32, 472–484. https://doi.org/10.1016/j.cjche.2020.09.071 Beyan, S. M., Prabhu, S. V., Sissay, T. T., & Getahun, A. A. (2021). Sugarcane bagasse based activated carbon preparation and its adsorption efficacy on removal of BOD and COD from textile effluents: RSM based modeling, optimization and kinetic aspects. Bioresource Technology Reports, 14(February), 100664. https://doi.org/10.1016/j.biteb.2021.100664 Bezerra, M. A., Santelli, R. E., Oliveira, E. P., Villar, L. S., & Escaleira, L. A. (2008). Response surface methodology (RSM) as a tool for optimization in analytical chemistry. Talanta, 76(5), 965–977. https://doi.org/10.1016/j.talanta.2008.05.019 Bhushan, B., Nayak, A., & Kotnala, S. (2021). Green synthesis of highly porous activated carbon from jackfruit peel: Effect of operating factors on its physico-chemical characteristics. Materials Today: Proceedings, 44, 187–191. https://doi.org/10.1016/j.matpr.2020.08.554 Do, D. D. (1998). Adsorption analysis: Equilibria and kinetics (with cd containing computer MATLAB programs) (Vol. 2). World Scientific. Endut, A., Abdullah, S. H. Y. S., Hanapi, N. H. M., Hamid, S. H. A., Lananan, F., Kamarudin, M. K. A., Umar, R., Juahir, H., & Khatoon, H. (2017). Optimization of biodiesel production by solid acid catalyst derived from coconut shell via response surface methodology. International Biodeterioration & Biodegradation, 124, 250–257. https://doi.org/10.1016/j.ibiod.2017.06.008 Foo, K. Y., & Hameed, B. H. (2012). Potential of jackfruit peel as precursor for activated carbon prepared by microwave induced NaOH activation. Bioresource Technology, 112, 143– 150. https://doi.org/10.1016/j.biortech.2012.01.178 Gholamiyan, S., Hamzehloo, M., & Farrokhnia, A. (2020). RSM optimized adsorptive removal of erythromycin using magnetic activated carbon: Adsorption isotherm, kinetic modeling and thermodynamic studies. Sustainable Chemistry and Pharmacy, 17(May), 100309. https://doi.org/10.1016/j.scp.2020.100309 Gilmour, S. G. (2006). Response Surface Designs for Experiments in Bioprocessing. Biometrics, 62(2), 323–331. https://doi.org/10.1111/j.1541-0420.2005.00444.x Hanrahan, G., & Lu, K. (2006). Application of Factorial and Response Surface Methodology in Modern Experimental Design and Optimization. Critical Reviews in Analytical Chemistry, 36(3–4), 141–151. https://doi.org/10.1080/10408340600969478 Karri, R. R., & Sahu, J. N. (2018). Process optimization and adsorption modeling using activated carbon derived from palm oil kernel shell for Zn (II) disposal from the aqueous environment using differential evolution embedded neural network. Journal of Molecular Liquids, 265, 592–602. https://doi.org/10.1016/j.molliq.2018.06.040 Khosravi, R., Azizi, A., Ghaedrahmati, R., Gupta, V. K., & Agarwal, S. (2017). Adsorption of gold from cyanide leaching solution onto activated carbon originating from coconut shell—Optimization, kinetics and equilibrium studies. Journal of Industrial and Engineering Chemistry, 54, 464–471. https://doi.org/10.1016/j.jiec.2017.06.036 Khuri, A. I., & Mukhopadhyay, S. (2010). Response surface methodology. Wiley Interdisciplinary Reviews: Computational Statistics, 2(2), 128–149. https://doi.org/10.1002/wics.73 Latib, E. H. A., Mustfha, M. S., Sufian, S., & Ku Shaari, K. Z. (2013). Methylene Blue Dye Adsorption to Durian Shell Activated Carbon. Key Engineering Materials, 594–595, 350– 355. https://doi.org/10.4028/www.scientific.net/KEM.594-595.350 Liang, Q., Liu, Y., Chen, M., Ma, L., Yang, B., Li, L., & Liu, Q. (2020). Optimized preparation of activated carbon from coconut shell and municipal sludge. Materials Chemistry and Physics, 241, 122327. https://doi.org/10.1016/j.matchemphys.2019.122327 Marrakchi, F., Hameed, B. H., & Bouaziz, M. (2020). Mesoporous and high-surface-area activated carbon from defatted olive cake by-products of olive mills for the adsorption kinetics and isotherm of methylene blue and acid blue 29. Journal of Environmental Chemical Engineering, 8(5), 104199. https://doi.org/10.1016/j.jece.2020.104199 33
Chuyên san Phát triển Khoa học và Công nghệ số 8 (4), 2022 Montgomery, D. C. (2017). Design and analysis of experiments. John wiley \& sons. Onu, C. E., Nwabanne, J. T., Ohale, P. E., & Asadu, C. O. (2021). Comparative analysis of RSM, ANN and ANFIS and the mechanistic modeling in eriochrome black-T dye adsorption using modified clay. South African Journal of Chemical Engineering, 36, 24– 42. https://doi.org/10.1016/j.sajce.2020.12.003 Prahas, D., Kartika, Y., Indraswati, N., & Ismadji, S. (2008). Activated carbon from jackfruit peel waste by H3PO4 chemical activation: Pore structure and surface chemistry characterization. Chemical Engineering Journal, 140(1–3), 32–42. https://doi.org/10.1016/j.cej.2007.08.032 Rebollo-Hernanz, M., Cañas, S., Taladrid, D., Benítez, V., Bartolomé, B., Aguilera, Y., & Martín-Cabrejas, M. A. (2021). Revalorization of Coffee Husk: Modeling and Optimizing the Green Sustainable Extraction of Phenolic Compounds. Foods, 10(3), 653. https://doi.org/10.3390/foods10030653 Shojaeimehr, T., Rahimpour, F., Khadivi, M. A., & Sadeghi, M. (2014a). A modeling study by response surface methodology (RSM) and artificial neural network (ANN) on Cu2+ adsorption optimization using light expended clay aggregate (LECA). Journal of Industrial and Engineering Chemistry, 20(3), 870–880. https://doi.org/10.1016/j.jiec.2013.06.017 Shojaeimehr, T., Rahimpour, F., Khadivi, M. A., & Sadeghi, M. (2014b). A modeling study by response surface methodology (RSM) and artificial neural network (ANN) on Cu2+ adsorption optimization using light expended clay aggregate (LECA). Journal of Industrial and Engineering Chemistry, 20(3), 870–880. https://doi.org/10.1016/j.jiec.2013.06.017 Taoufik, N., Elmchaouri, A., El Mahmoudi, S., Korili, S. A., & Gil, A. (2021). Comparative analysis study by response surface methodology and artificial neural network on salicylic acid adsorption optimization using activated carbon. Environmental Nanotechnology, Monitoring & Management, 15(February), 100448. https://doi.org/10.1016/j.enmm.2021.100448 Teófilo, R. F., & Ferreira, M. M. C. (2006). Quimiometria II: planilhas eletrônicas para cálculos de planejamentos experimentais, um tutorial. Química Nova, 29(2), 338–350. https://doi.org/10.1590/S0100-40422006000200026 Yagub, M. T., Sen, T. K., Afroze, S., & Ang, H. M. (2014). Dye and its removal from aqueous solution by adsorption: A review. In Advances in Colloid and Interface Science (Vol. 209, pp. 172–184). https://doi.org/10.1016/j.cis.2014.04.002 Yusoff, N. R. N. (2020). Removal of Al (III) and Cu (II) from Textile Wastewater using Activated Carbon Prepared from Rubberwood Sawdust and Kenaf Core Fiber. Malaysian Journal of Microscopy, 16(2). 34