Synthesis Optimization of Activated Carbon Driven from Scrap Tire for Adsorbent Yield and Methylene Blue Removal under Response Surface Methodology

Department of Chemical Engineering, College of Biological and Chemical Engineering, Addis Ababa Science and Technology University, Addis Ababa 16417, Ethiopia Food and Beverage Industry Research and Development Center, Addis Ababa, Ethiopia Department of Environmental Engineering, College of Biological and Chemical Engineering, Addis Ababa Science and Technology University, Addis Ababa 16417, Ethiopia Sustainable Energy Center of Excellence, Addis Ababa Science and Technology University, Addis Ababa 16417, Ethiopia Bioprocess and Biotechnology Center of Excellence, Addis Ababa Science and Technology University, Addis Ababa 16417, Ethiopia Ethiopian Conformity Assessment Enterprise, Addis Ababa, Ethiopia


Introduction
Industries including textile, printing, plastic, cosmetic, paper, and food processing industries use dyes [1,2]. ere are over 100,000 di erent types of commercially available dye throughout the world. According to Fito et al. [3], 700,000 tons of dyes are globally produced per year. During the manufacturing process, approximately, 15% of the dyestu is lost in industrial wastewater [4]. Wastewater that contains dyes is characterized by low biodegradability and high contaminate the soil [31]. Scrap tires are not easily biodegradable in landfills. e volume that they occupy is also a major problem in landfills [32].
Instead of landfilling, recycling scrap tires have its own economic advantage. Scrap tires can be used as part of construction material due to their physical characteristics [33]. Ground scrap tires are used for roofing materials, sports tracks, and noise pollution barriers [32]. According to Dick et al. [34], pyrolysis is the most recommended method of recycling waste tires. e three valuable products of waste tires that are processed through pyrolysis are pyro-oil, pyrogas, and pyro-char [35]. Due to its high carbon content, surface area, and porosity, pyro-char (activated carbon derived from the waste tire) can be used as an adsorbent for removing different pollutants from wastewater [24,[36][37][38].
Increasing the surface area of the activated carbon, which is derived from scrap tires, is the major problem that is faced during the preparation process [24]. Demineralization enhances the adsorption capacity of the adsorbent by removing inorganic minerals from the given adsorbent through a process called leaching [39].
is process increases the availability of the pours by removing the impurities (inorganic minerals) that blocked the pour structure, which in turn increases the surface area, micro, and mesopore volume [39]. Some of the inorganic minerals that are found in waste tires are iron, silicon dioxide, zinc, calcium, potassium, magnesium, chromium, manganese, sodium, nickel, and lead [40]. Moreover, demineralization lowers the ash content and increases the yield of the activated carbon [41]. Response surface methodology is one of the most efficient methods that is used to study and optimize experimental variables [42,43]. It is used to study and optimize experimental variables. is methodology has an advantage in terms of evaluating the controlling effect of each parameter and their interaction by reducing the number of experiments [44].
Previously, research has been studied regarding the use of activated carbon derived from scrap tires for the removal of methylene blue dye from wastewater [2,38,45]. However, it is noted that no research is carried out regarding the synthesis optimization of activated carbon derived from scrap tires under response surface methodology. Teng et al. [46] studied the effect of impregnation ratio, pyrolysis temperature, and holding time on surface area and yield of activated carbon derived from scrap tires. However, it is noted that no research is carried out regarding the demineralization of activated carbon derived from scrap tires using NaOH + H 2 SO 4 (1 : 1) as a demineralizing solvent for surface area enhancement. erefore, this study aimed to optimize the synthesis of activated carbon driven from scrap tires in terms of yield and methylene blue removal.

Pretreatment of Scrap Tire.
e scrap tires were collected from a local market, which is located in Addis Ketema subcity, Addis Ababa, Ethiopia. ese tires were washed several times with detergent and tap water to remove dirt and dried at an ambient temperature 25±1°C. e cleaned tires were then cut into small pieces by sharp knives. e size of the tires was further decreased by a high-speed multifunction miller (Model HC-700) and sieved to a size of 0.2 mm. e scrap tires were then washed several times with distilled water and dried in an oven (Model BOV-T50F) for 5 h at 60°C [47,48].

Aqueous Solution
Preparation. An aqueous solution of 1 g/L was prepared by adding 1g of methylene blue (C 16 H 18 N 3 SCl, molecular weight: 319.85 g/mol) in a 1 L volumetric flask. en, distilled water was filled to the mark of 1 L, and dissolution is performed using a magnetic stirrer. By using the concept of the dilution process, a different working solution was prepared.
e chemical activation process was conducted in the presence of ethanol as a solvent. irty grams of the pretreated scrap tire sample was impregnated with KOH at various impregnation ratios (0-2 g/g) and impregnation time (12-36 hr). e impregnated sample was dried in an oven at 105°C. e sample was then pyrolyzed in a muffle furnace (Model Naberthrem F 330) in an inert environment, which is created by N 2 gas. Pyrolysis was carried out at different carbonization temperatures (700-900°C) for 2 hr [49]. After the activated sample was cooled, it was washed with 0.1 N HCl in 150 ml solution for 30 min.
en, the activated sample was washed several times with distilled water until the pH of the supernatant is 7. e activated sample was then dried in an oven at 105°C for 24 hr. e synthesis process of the activated carbon was optimized by central composite design (CCD) using Minitab software Version 20.1.3 under response surface methodology (RSM). Impregnation ratio, impregnation time, and carbonization temperature were factors. ese factors were fixed based on literature values. SCAC Yield and methylene blue (MB) removal efficiency were response variables.
According to the RSM of CCD, 20 experiments were conducted at different combinations of the three variables. e number of experiments was calculated using.
where N is the number of experimental runs, F represents the factor number, and xo is the number of replicates at the central point. In this study, N is 20, F is 3 and xo is 6. Of the 20 experiments, 5 experiments were replicated at center points to evaluate the error. Based on a set of experiments, the range of the variables, step size, and the central value were chosen as shown in Table 1.
e center points are used to determine the experimental error and the reproducibility of the data. e axial points are located at (±α, 0, 0), (0, ±α, 0), and (0, 0, α). α � 1.0 (facecentered) is the distance of the axial point from the center and makes the design rotatable. e experimental sequence was randomized to minimize the effect of the uncontrolled factor. e experimental results were fitted using a polynomial quadratic equation to correlate the response variables. e general form of the polynomial quadratic equation shown in equation (2) was used to develop a model that predicts (estimates) the STAC yield and MB removal percentage at the designed variable combination. e model is also used to predict the individual and interaction effects of different parameters.
where: W o (g) is dry weight of the pretreated scrap tire and W ac (g) is the dry weight of produced STAC.
To determine MB removal, 200 ml of 20 solutions, which have the same value of pH and initial MB concentration, were prepared in a 250 ml Erlenmeyer flask. e pH, initial concentration, adsorbent dose, and contact time were taken constant for all experiments at 9, 20 mg/L, 0.5 g, and 45 min, respectively [50]. e adsorption experiment was conducted at room temperature and at a mixing speed of 125 rpm [3]. e adsorbent was separated from the solution using Whatman filter paper 42. e final MB concentration was determined using a UV-visible spectrophotometer (JASCO V-770) at a wavelength of 668 nm [6].
MB removal efficiency was calculated by equation (4).
where: Re (%) is removal efficiency, Co (mg L −1 ) is the initial concentration of MB, and Ce (mg L −1 ) is the final MB concentration. e optimized STAC is used in characterization, demineralization, isotherm, and kinetics study.

Characterization of the Activated Carbon.
e moisture content (MC), volatile matter (VM), ash content (AC), and fixed carbon (FC) of the optimized STAC were determined according to ASTM standards (D 2866-2869) [18]. e bulk density of the STAC was determined by the water displacement method.
SEM (Model FEI Inspect F50) was used to determine the surface morphology of the pretreated sample, before activation and after adsorption. Standard procedures were followed in preparing the sample and operating the equipment [51,52].  24 36 Advances in Materials Science and Engineering 3 XRD (Model Olympus BTXH) was used at a diffraction angle of 2ø from 10 to 80°to determine the crystalline and amorphous nature of the sample after activation and after adsorption. e XRD was operated at 15 mA and a scanning rate of 4.2°C/min. e results were analyzed using origin software Version 9.55 [18].
BET surface area analyzer (Horiba, SA-9600) was used to determine the surface area of optimized STAC. e sample was analyzed by taking 0.4 g of STAC in three vacuum tubes for 2 h at 100°C [53].

Demineralization.
e optimized STAC was demineralized for enhancing the surface area. e demineralization process was conducted using a solvent that is prepared by 5M NaOH + 98% H 2 SO 4 (1 : 1). 20 g of optimized STAC followed by 1 L distilled water was added to 16 ml of the demineralization solvent in Erlenmeyer flask. e mixture was mixed using a hot plate stirrer (Model P40-HS) at a temperature, mixing speed, and mixing time of 100°C, 400 rpm, and 3 hr, respectively. en the demineralized STAC was separated from the solution using Whatman filter paper 42. e separated demineralized STAC was then washed thoroughly with distilled water and dried in an oven at 110°C for 24 hr [39,54]. e surface area of the demineralized STAC was analyzed by BET method.
Langmuir adsorption isotherm assumes monolayer adsorption of adsorbate on the adsorbent surface and adsorption energy is fixed at all points [55]. e linearized form of Langmuir isotherm model is shown in equation (5).
where q e and q m are the adsorbed amount at equilibrium and Langmuir maximum adsorption, respectively (mg g −1 ), K l is the constants of Langmuir (L mg −1 ), V is the adsorbate volume (L), and m is the mass of adsorbent (g). R L is a dimensionless factor that predicts the appropriateness of the adsorption by the constants obtained from the Langmuir model and calculated using R L .
If R L > 1, the used adsorbent is not appropriate for the adsorption of the adsorbate. If R L � 0, adsorption on the adsorbent will be reversed. If R L � 1, the isotherm is of linear type, and if 0 < R L < 1, the utilized adsorbent is appropriate.
Freundlich isotherm assumes a heterogeneous adsorption process [56]. e linearized form of Freundlich isotherm model is shown in equation (8).
where: q e is the Equilibrium loading (mg g −1 ), C e is the Equilibrium concentration (mg L −1 ), K f is adsorption capacity ((mg g −1 ) (L mg −1 ) 1/n ), and n is the adsorption intensity. e value of the Freundlich constant (n) should lie in the range of 1-10 for favorable adsorption, the higher n value the better the adsorption. Temkin adsorption isotherm assumes that the heat of adsorption for all molecules decreases with the coverage of the adsorbent surface [57,58]. According to Togue Kamga [59], this model is only valid for a range of intermediate ion concentrations. e linearized form of Temkin adsorption isotherm is given by where: b is Temkin constant which is related to the heat of sorption (J mol −1 ) and KT is Temkin isotherm constant (L g −1 ). Dubinin-Radushkevich isotherm is an empirical isotherm model that expresses the distribution of Gaussian energy onto heterogeneous surfaces with an adsorption mechanism [44]. Since this model does not predict Henry's law at low temperatures, it is only suitable for an intermediate range of adsorbate concentration [60]. According to Boubaker et al. [44], the model states that the adsorption potential is variable and the free adsorption enthalpy was related to the degree of pores filling. is model determines whether the pollutant uptake mechanism is physical or chemical [42,61]. e linearized form of Dubinin-Radushkevich isotherm adsorption isotherm is given by equation (10).
where: lnqDRmax is the maximum adsorption capacity of , and T is absolute temperature (K).

Adsorption Kinetics and Intraparticle
Diffusion. e adsorption kinetics and intraparticle diffusion were studied in 200 ml solution at a fixed value of pH 9, adsorbent dose of 0.5 g, and mixing speed of 125 rpm. e experiments were carried out at room temperature and at different contact times (25,50,100,200,250, and 300 min) in a 250 mL Erlenmeyer flask [62]. In the solid-solution interface, the adsorbate removal rate is expressed by adsorption kinetics.
e experimental data were fitted on pseudo-first-order and pseudo-second-order adsorption kinetics models [63]. e integral form of pseudo-first-order and pseudo-second-order equations are expressed in equations (12) and (13), respectively.
where qe and qt (both in mg g −1 ) are the amounts of MB adsorbed at equilibrium and at time t, respectively; K1 (min −1 ) is the rate constant of pseudo-first-order; K2 (g mg −1 min −1 ) is the rate constant of pseudo-second-order, V (L) is the volume of the solution, m (g) is the mass of adsorbent, and t (min) is contact time.
Weber-Morris intraparticle diffusion model was used to study the diffusion mechanism of MB [63]. e model equation is expressed in equation (16).
where k id (mg. g −1 min 0.5 ) is the intraparticle diffusion rate constant and c is a constant number.

Synthesis Optimization of Activated
Carbon. e synthesis of STAC was optimized by using the impregnation ratio (IR), impregnation time (IT), and carbonization temperatures (T) as a factor. STAC yield and MB removal efficiency were response variables as shown in Table 2.
e maximum STAC yield (68.35%) was obtained at an impregnation ratio of 2 g/g, impregnation time of 12 hr, and carbonization temperature of 700°C. e MB removal efficiency at these factors was 89.96%. e maximum MB removal efficiency (90.48%) was obtained at impregnation ratio, impregnation time, and carbonization temperature of 0 g/g, 12 hr, and 900°C, respectively. e STAC yield at these factors was 45.33%. e minimum STAC yield (34.03%) was obtained at impregnation ratio, impregnation time, and carbonization temperature of 0 g/g, 36 hr, and 900°C, respectively. e minimum MB removal efficiency (89.18%) was obtained at impregnation ratio, impregnation time, and carbonization temperature of 2 g/g, 36 hr, and 800°C, respectively.
e factor values that give relatively both maximum STAC yield and MB removal are considered optimum factor values that are used to synthesize STAC. erefore, an impregnation ratio of 2 g/g, impregnation time of 12 hr, and carbonization temperature of 700 C was taken as optimum factor values for synthesizing STAC. According to Teng et al. [46], a higher value of yield is obtained at carbonization temperature, impregnation ratio, and carbonization time of 500 C, 4, and 1 hr, respectively. Moreover, the maximum value of BET surface area was obtained at a carbonization temperature of 700 C, impregnation ratio of 4, and carbonization time of 1 hr. e maximum MB removal efficiency was compared with previous studies as shown in Table 3. e effect of carbonization temperature, impregnation time, and impregnation ratio on STAC yield and MB removal is presented in Figure 1.
As shown in Figures 1(a)-1(c), the STAC yield decreases with the increment of carbonization temperature. is might be due to a higher degradation of rubber and other tire components at an enhanced temperature, which results in loss of weight [72]. STAC yield decreased with the increment of Impregnation time. is might be due to the collapse of pores and a decrease in surface area [73]. Increasing the impregnation ratio has facilitated the increment of the yield due to the addition of extra mass to the precursor [74].
As shown in Figures 1(d)-1(f), MB removal decreased as carbonization temperature increased. is might be due to the collapse of pores and carbon matrix at high temperatures [75]. MB removal increased with the increment of impregnation ratio. is might be due to the increment of surface area by chemical activation [46]. As the impregnation time increased, the MB removal efficiency decreased continuously. is might be due to the shrinkage of char structure, widening and combining of some micropores into mesopores, which decreased the adsorption capability of the adsorbent [73].
Due to the sorption capacity and pore structure of the activated carbon, the MB molecules can be adsorbed on the surface of the adsorbent [76]. According to Gao et al. [77], the presence of the micropores present in the activated carbon and the attraction of the Vander Waals forces are the main adsorption mechanisms of pollutants.

ANOVA and Development of Regression Model Equation.
e regression coefficient for STAC yield (R 2 � 0.957) and MB removal (R 2 � 0.992) indicates that the quadratic regression model best fits the experimental data for both e Predicted R 2 value of the STAC yield is 0.872, which is in reasonable agreement with the adjusted R 2 value of 0.919. For MB removal, the predicted R 2 value is 0.972 and the adjusted R 2 value is 0.984. is indicates that the predicted and adjusted value is in satisfactory agreement. e implication and suitability of the models for both STAC yield and MB removal were also tested by analysis of variance (ANOVA) as shown in Table 4. e mean squares were calculated by dividing the sum of the squares of each of the various sources and the mode and the error variance were calculated by the respective degrees of freedom. e F-value is the ratio of the  mean square contributing to regression to the mean square contributing to error. e greater the F-value, the more is the significance of the corresponding variable to cause an effect. As shown in Table 4, the model F-value of 25.18 for STAC yield and 137.97 for MB removal implies that this model was significant. e lack of fit F-value of 0.62 for STAC yield and 0.47 for MB removal implies that the lack of fit is not significant relative to the pure error. e significance of the model is also tested by P value. If the P values of the model terms are less than 0.05, they are statistically significant. e ANOVA of the quadratic model for STAC yield indicates that T, IT, IR, T 2 , IT 2 , and T * IR are statistically significant (P < 0.05) but IR 2 , T * IT, and IT * IR are not statistically significant (P > 0.05). For MB removal, the model indicates that T, IT, IR, T 2 , IT 2 and T * IR, IR 2 , and IT * IR are statistically significant (P < 0.05) but T * IT is not statistically significant (P < 0.05). e regression model equation was developed by removing the insignificant model terms. e model equation that is developed by three factors or twenty experiments for STAC yield and MB removal is described in equations (16) and (17), respectively.
where: T is Carbonization temperature, IR is Impregnation ratio, IT is Impregnation time, T * IR is the interaction effect of time and impregnation ratio, IT * IR is the interaction effect of impregnation time and impregnation ratio, MBR is Methylene blue removal (%) and Y is STAC yield (%).

e Interaction Effect of Factors on STAC Yield and MB
Removal. Out of the three interaction effects, only one interaction (T * IR) is statistically significant (P < 0.05) for STAC yield and two interactions (T * IR and IT * IR) are statistically significant (P < 0.05) for MB removal as shown in Table 4. e three-dimensional response surface plot of the statistically significant (P < 0.05) interaction effects on STAC yield and MB removal is presented in Figure 2.

Proximate Analysis.
e MC, VM, AC, FC, and bulk density of the STAC were 6.13%, 9.42%, 5.34%, 79.11%, and 0.89 mg/L, respectively. e low value of MC, VM, AC, and the higher value of FC indicates that the STAC has a good adsorbent characteristic [78].

Scanning Electron Microscope (SEM).
e surface morphology of scrap tires before activation (raw), after activation, and after adsorption are presented in Figure 3. In all three SEM micrograph images, a magnification of 10 µm was used. Compared to the raw scrap tire, the activated scrap tire shows a highly porous morphological structure with heterogeneous, irregular, and small pores of various shapes and sizes. e internal structure of the STAC was irregular with many gullies and openings. is might be due to the chemical and thermal activation of the scrap tire [53]. After adsorption, the number of pours decreased. is might be due to the adsorption of MB on the internal and external pore structure of the STAC [79].

X-Ray Diffraction (XRD).
e X-ray diffraction of scrap tires after activation and after adsorption is presented in Figure 4. e large hill of A, from 2θ � 19.56°to 36.56°c orresponding to side spacing 4.5347Å to 2.4558Å indicated the existence of amorphous carbon together with other crystalline compounds. ese crystalline compounds were found to be ZnO (zincite) and β-ZnS (wurtzite). Similar results were reported by Ilnicka et al. [80]; López et al. [81]; Undri et al. [82]. ZnO (zincite) is the major component of the scrap tire. However, according to Amirza et al. [83],   respectively. e presence of crystalline structure is indicated by these high-intensity peaks [85]. On the other hand, similar peaks were observed by MB adsorbed STAC samples, with the exception of increased intensity and full width half maximum (FWHM) values. As 2θ values increased in the spectra pattern, the peak intensity decreased, which indicates the presence of an amorphous carbon arrangement [18].

Surface Area Study of Optimized and Demineralized
Activated Carbon. A demineralization experiment was conducted to enhance the surface area of the optimized STAC. In this study, inorganic minerals such as ZnO were removed from the pores and polymeric resins of rubber, which could lead to enhancement of the surface area. e surface area of optimized STAC was 54.93 m 2 /g. A surface area of 88 m 2 /g was reported by Jankovská et al. [86] for activated carbon driven from scrap tires. e surface area of optimized STAC that is demineralized by NaOH + H 2 SO 4 was 260.26 m 2 /g. is indicates that the surface area of optimized STAC increased due to the demineralization process [39]. A higher surface area value of 493 m 2 /g was reported by Ali et al. [24] for activated carbon derived from scrap tires.

Adsorption Mechanism
3.6.1. Adsorption Isotherm. Langmuir, Freundlich, Temkin, and Dubinin-Radushkevich adsorption isotherm models were used to study the adsorption mechanism. Regression coefficient (R 2 ) was used to indicate the isotherm model that best fits with experimental data. For both optimized and demineralized STAC, Temkin adsorption isotherm model with R 2 values of 0.974 and 0.982 best describes the adsorption mechanism as shown in Table 5. is indicates that the heat of adsorption of all molecules in the layer reduces linearly [87].
is results from the increase of surface  Table 6: Adsorption kinetics and intraparticle diffusion.

Kinetics model Optimized Demineralized
Pseudo-first-order Pseudo-second-order coverage [57]. Temkin adsorption isotherm also indicates that up to maximum binding energy, the adsorption is characterized by a uniform distribution of binding energies [61,88].

Adsorption Kinetics and Intraparticle Diffusion.
e adsorption kinetics of the adsorption process was examined by pseudo-first-order and pseudo-second-order models. For both optimized and demineralized STAC, pseudo-second-order adsorption kinetics model with R 2 values of 0.996, and 0.999 best fits with the experimental data as shown in Table 6.
is indicates that the adsorption process is controlled by chemical reactions [89]. e diffusion mechanism was studied by the intraparticle diffusion model. From the linear regression equation, it can be noted that values of C of both the optimized and demineralized STAC are not equal to zero. is indicates that intraparticle diffusion was not the only rate-controlling step [90].

Conclusion
e STAC synthesis was optimized using central composite design under response surface methodology. Impregnation ratio of 2 g/g, impregnation time of 12 hr, and carbonization temperature of 700°C was taken as optimum factor values for synthesizing STAC. e regression coefficient for STAC yield and MB removal with R 2 values of 0.957 and 0.992 indicates that the quadratic regression model best fits the experimental data for both response variables. From ANOVA, it can be noted that the interaction effect of carbonization temperature and impregnation ratio on STAC yield is the only significant (P < 0.05) interaction. Moreover, for MB removal, the interaction effect of carbonization temperature and impregnation ratio and the interaction effect of impregnation time and impregnation ratio is the only significant (P < 0.05) interactions. e proximate analysis indicates that the STAC has a good adsorbent characteristic. e surface area of optimized STAC was enhanced from 54.93 m 2 /g to 260.26 m 2 /g by the demineralization process using NaOH + H 2 SO 4 at (1 : 1) as a solvent. Temkin adsorption isotherm with R 2 values of 0.974 and 0.982 best fits with the experimental data for both optimized and demineralized STAC respectively. is indicates that the heat of adsorption of all molecules in the layer reduces linearly with coverage. For both optimized and demineralized STAC, pseudo-second-order adsorption kinetics with R 2 value of 0.996 and 0.999 best fits the experimental data, respectively.
is indicates that the adsorption process is controlled by chemical reactions. From the intraparticle diffusion model, it can be noted that intraparticle diffusion was not the only ratecontrolling step for both optimized and demineralized STAC. Generally, it can be concluded that the optimized STAC that is demineralized by NaOH + H 2 SO 4 solvent has a promising potential to be used as a low-cost adsorbent in developing countries including Ethiopia. However, further investigation needs to be conducted before scaling up at the industrial level.

Data Availability
All the data used to support the findings of this study are included in the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.