Effective Copper and Methylene blue adsorption from industrial effluents onto activated carbons prepared from Rice husk ash and Hazelnut husks modified by Diopside: Equilibrium, Kinetics, and Experimental design

ABSTRACT Activated carbon from rice husk ash (AC-RH) and hazelnut husks (AC-HH) modified by Diopside (AC-RH/Diop, AC-HH/Diop) composites were synthesised and applied for the adsorption of Copper and Methylene blue from industrial effluents. Diopside was provided by the sol-gel method and activated carbon was prepared from rice husk ash and hazelnut husks. The formation of AC/Diop composites was characterised by FT-IR, FESEM, BET, and XRD analyses. Effects of pH, amount of adsorbent, and contact time at 10 mg L−1 of Copper and Methylene blue on removal percentage were studied by Box Behnken design (BBD). Optimal conditions for maximum Copper and MB by AC/Diop adsorbent (95.54% and 99.42%, respectively) were as follows: (pH = 5.45, 7.35) (adsorbent amount: 0.10, 0.45 g), and (contact time: 9.56, 13.49 min) in 20 mL of 10 mg L−1 of pollutants concentration, respectively. In addition, the adsorption kinetics, thermodynamics, and isotherms were examined. Adsorption isotherms (qmax: 184.25 and 289.51 mg g−1 for Copper and MB, respectively) and kinetics were demonstrated that the sorption processes were better described by the Langmuir and pseudo-second-order equation.


Introduction
One of the most important environmental issues in the worldwide is the presence of pollutants such as heavy metals and dyes in industrial effluents and the main characteristics of them are non-biodegradation, toxicity, bioaccumulation, carcinogenesis, biotransportation, and spread to large areas quickly due to their high mobility [1].Manufacturing industries like batteries, textiles, metal plating, dyes, alloys and mining operations are the general heavy metal and dye sources [2][3][4].Copper (Cu), in trace amounts is an essential nutrient, related with liver, brain, respiratory and nervous system diseases especially in children.The maximum industrial effluent release for copper is 2.0 ppm, but for drinking water, the concentration must be less than 1.3 ppm [3].On the other hand, methylene blue (MB), an organic cationic dye, is a heterocyclic compound and the decomposition of MB produces hazardous gases such as carbon dioxide, carbon monoxide, nitrogen oxides, and sulphur oxides that cause anaemia, hypertension, methemoglobinemia, nausea, vomiting, cancer, diarrhoea, precordial pain, gastritis, and allergic dermatitis [5,6].MB is a cationic dye that has been used in industries extensively like acrylic, hemp, dyeing cotton, paper, silk, and making ink due to its colour stability and good water solubility [7].Therefore, the treatment of effluents containing heavy metals and dyes is a major concern due to their different effects on the hydrosphere [8].
Different technologies for the removal of pollutants from water and wastewater like physical adsorption, chemical precipitation, biological decomposition, membrane separation, ion-exchange adsorption have been reported.It is important that the conventional methods have some advantage and disadvantage.For example, the chemical precipitation is cheap, but it produces a large amounts of sludge and causing the secondary pollution problems.Scientists have shown that biological decomposition has high proficiency, on the contrary, it is expensive.Also, membrane separation is indicated very high efficiency, but it is costly, and requires high energy [9].However, adsorption is efficient, faster operation, cheap, more environmentally friendly, and easy-to use.Lots of adsorbents, like zeolite, clay, diatomite, fly ash, red mud, hybrid anion exchanger, and crosslinked fibrous materials have been studied.But, it is necessary to have suitable adsorbent and investigate ideal pollutant removal from contaminated water and wastewater [10].Activated carbon (AC), because of cheap biosources like sunflower piths [11], coconut shell [2], waste rice straw [12], pistachio shell [13], peanut shell [14], coir pith [15], rice husk ash [16], and hazelnut husks [17], besides its huge surface area [11,18] is one of the best adsorbents applied in water and wastewater purification.However, due to the low mechanical strength of AC and its hydrophobic properties, the synthesis of composite with mineral materials to compensate for these defects has been much studied [2,13,19].As well as, the magnesium calcium silicate mineral with the molecular formula of MgCaSi 2 O 6 , diopside, due to slow degradation, its improved mechanical properties, excellent in vivo biocompatibility and bio-functionalities has lots of applications in the dental, bone and biomedical applications [20].The application of diopside in water and wastewater purification is very limited despite the addition of other minerals such as clinoptilolite [21], and hydroxyl apatite [1,18].Easy synthesis method of diopside, in addition to the improved mechanical properties, and its chemical structure, that is rich in superficial hydroxyl groups, are the main target for choosing this material besides AC produced from biosources as an adsorbent for removal of Copper and MB from aqueous solutions.
The design of the experiment (DOE) is used for modelling and optimisation of factors affecting the adsorption process.Response Surface Methodology (RSM) was used to model the adsorption process of the copper and MB by studying the independent factors including pH, adsorbent dosage, and contact time at 10 mg L −1 of Copper and Methylene blue on removal percentage using Box Behnken Design (BBD).Also, DOE presents helpful information about the influence of independent factors that leads to a decrease in experimental error [1].
To the best of acknowledging, no reports on the adsorption of Copper and Methylene blue pollutants by the composite of activated carbon from rice husk ash and hazelnut husks modified by diopside adsorbent can be found.In this project, at the first step, activated carbon was prepared from rice husk ash and hazelnut husks, second, diopside was synthesised andAC/Diopcomposites were prepared by simple mixing of activated carbon and diopside in ethanol.Adsorption results of Copper and MB pollutants from aqueous solutions using AC/Diopnano composites indicated ideal removal percentage of the adsorbent.
The FTIR analysis was done by an FTIR spectrometer (Jasco 680-plus, Tokyo, Japan) in the range of 400-4000 cm −1 .The surface morphology was analyzed using field emission scanning electron microscopy (FESEM) (MiRA3-TESCAN, Czech Republic).The samples were coated with gold film and the acceleration voltage of apparatus was 15.0 kV.Powder XRD analyses were characterised using a diffractometer (Panalytical X'PertPro, Almelo, The Netherlands), with Cu anode.All analyses were run at 40 kV and 30 mA, and the samples were scanned from 10° to 80° at 3°/min.Nitrogen adsorption-desorption isotherms were obtained using a BEL: PHS 1020 sorptometer (China) by using N 2 gas at 77 K after degassing the sample in vacuum at room temperature for 12 h.UV-Vis spectra of MB were obtained by UV-vis 2100 SHIMADZU (Japan) spectrophotometer, and AASpect 203 Flame atomic absorption spectrometer (FAAS) was used for Copper analysis.

Diopside preparation
The diopside was synthesised by the sol-gel method using Teimouri et al. work [22] and it was utilized for synthesis of the AC/Diop adsorbents.

Preparation of rice husk ash and hazelnut husks-based activated carbon
Rice husk was washed with distilled water and dried in an oven at 110°C for 18 h.The dried sample (100 g) was treated with a mixture of 10% hydrochloric acid and 10% sulphuric acid (1500 mL) for 3 h.Then, to remove the remaining acid content of the rice husk, it was rinsed and stirred for 1 h to neutralise.After that, the sample was dried at 60°C in an oven and ground; it was calcined at 900°C for 3 h, and the calcined husk was sieved and kept for synthesising of composite.Hazelnuthuskwas rinsed several times with distilled water and then placed in the oven at 110°C for 24 h to dry, then ground into fine particles in the range of 150-200 μm.20 g of dried hazelnut husk was added to H 3 PO 4 (30% w/w) and stirred for 2 h at 80°C.The resulting mixture was dried in the oven for 2 h at 110°C and then, it was calcined for 2 h at 900°C.The activated product is cooled to room temperature and washed with distilled water to remove chemical residuals, and placed in an oven at 110°C for 12 h to dry, and finally, the sample is ground and kept for synthesising of composite.

Preparation of AC/Diop composites
AC/Diop (50:50) nanocomposite were prepared simply according to the following method: In a 50 mL beaker with ethanol, an amount of 0.5 g of AC powder was added under stirring.After 20 min, amount of 0.5 g of Diop was added to baker and the mixture was stirred for 24 h at room temperature.Finally, the other ratios of AC/Diop composites with 70:30 and 30:70 were synthesized as before.

Batch adsorption experiments
Adsorption experiments were studied in a batch system.Experiments were done in a baker with 20 mL of 10 mg L −1 pollutant solution at room temperature.The pH of MB solution was adjusted using dilute NaOH and HCl, and pH of Copper solution was adjusted by adding small volumes of dilute HNO 3 andKOH.Then, the adsorbent was added to the baker and stirred.The solution was centrifuged and the absorbance of the dye solution was measured by a spectrophotometer and the Copper solution was analysed by an atomic absorption spectrometer.All experiments were carried out triplicate and the data were averaged.
The adsorption capacity of MB and Copper was calculated according to the following equation: Where C 0 is the initial concentration of MB and Copper, C e is the equilibrium concentration, V is the volume of solution and m is the mass of adsorbent.
And the removal percentage of MB and Copper was determined based on the following equation: Where C 0 is the initial concentration and C e is the concentration of the solution after adsorption process.

Design of experiments
Experimental design was applied to optimise the best conditions for copper and MB removal efficiency.RSM designs let us determine the interaction effects, and give us the idea of the shape of the response surface under research.Box-Behnken design (BBD) has the maximum performance for an RSM problem including three factors (pH (X 1 ), adsorbent dose (X 2 ), and contact time (X 3 )) in three levels (high, middle and low).Hence, the number of experiments is less compared to a CCD.After conducting experiments based on the runs obtained from the experimental design matrix, a regression on the factors and response is done and a model fitted.Quadratic model obtained is given in Eq. (3) 6, [23],: Where Y is the response, β 0 is the constant, β i is the linear coefficient, β ii shows that the quadratic coefficient, β ij is the interaction coefficient, x i is the coded parameter level and i or j is the number of independent parameters.

FTIR analysis
FTIR spectra of Diop, AC-RH, AC-HH., and AC-RH/Diop, AC-HH/Diop composites in the range 400-4000 cm −1 are presented in Figure 1.In the spectrum of Diop, the peaks in the 900-1100 and 650 cm −1 are attributed to the stretching and bending vibrations of the silicate structure, respectively.Diop has three absorption peaks in areas of 955, 1045 and 1086 cm −1 , that are attributed to the vibrations of Si-O bonds (Figure 1(a)) [2,24].As can be seen in Figure 1(d,e), in the spectrum of AC-RH and AC-HH, the peaks at around 3342 and 2954 cm −1 for stretching vibrations of hydroxide ions and aliphatic C-H bonds, respectively.Aromatic stretching vibrations of C = C bonds are found at 1602 cm −1 and the peak at 1211 cm −1 is related to the C-N bond [25].In the spectrum of Diop, the peaks at 650 and 900-1100 cm −1 are related to the bending and stretching vibrations of the silicate structure, respectively.In fact, the diopside has three strong absorption peaks in areas of 948, 1033 and 1073 cm −1 , which are related to the vibrations of Si-O bonds [24].
In the spectra of composites (Figure 1(b,c)), all characteristic peaks of AC-RH, AC-HH, and Diopare are found in the spectrums of AC-RH/Diop and AC-HH/Diopcomposites, clearly.

FESEM Analysis
In Figure 3, the FESEM images of Diop, AC-RH, AC-HH, AC-RH/Diop, and AC-HH/Diop composites show that the average particle size of all the materials is less than 1 μm.Also, uniform and homogeneous dispersion of Diop particles and successful formation of composites were clearly demonstrated.

N 2 adsorption-desorption isotherm
Nitrogen adsorption-desorption isotherms and the pore size distribution of the AC-RH /Diop and AC-HH/Diop composites from the BJH method are analysed and the specific surface area, pore volume, and pore size of the AC-RH/Diop and AC-HH/Diop composites were 120.1, 110.2 m 2 g −1 , 0.22, 0.26 cm 3 g −1 , and 2.16, 4.14 nm, respectively.A comparison between the BET data of Diop, AC-RH, AC-HH, wit hAC-RH/Diop, and AC-HH/Diop composites are presented in Table 1.

Comparison of removal percentage of AC/Diop composites and the components
To demonstrate the synergistic effect of the AC/Diop composites, the removal percentage of AC and Diop were compared with AC/Diop composites in the same conditions (SI Fig. S1).As is obvious from the figure, AC and Diop have removal percentage lower than the AC/Diop composites for removal of Copper and MB dye.

Experimental design approach for the optimisation of Copper and MB adsorption by AC/Diop adsorbents
To investigate the most important factors affecting the adsorption of Copper and MB by AC/Diopadsorbent, three factors: pH (X 1 ), adsorbent dosage (X 3 ), and contact time (X 3 ) at 10.0 mg L −1 initial concentration of Copper and MB were considered and 15 experiments were designed by applying the Box Behnken design (BBD).Three independent factors with considered range and matrix of the experiments were demonstrated, respectively in Tables 2, 3 and SI Tables 4, 5.The results of the analysis of variance (ANOVA) for the removal percentage of Copper and MB by the AC/Diop adsorbent are shown in Tables 6 and 7.The significance of the factors is determined by the values of P. For the removal percentage of Copper and MB, all three factors in this model are significant because they have P values less than 0.05 [28].The lack of fit (LOF) demonstrates the residual sum of squares fraction occurs becase of the incompleteness of the model.A model is reliable if the P value of LOF is less than 0.05 and it does not have a significant remark [28].The P values of LOF for the removal percentage of Copper and MB are 0.0877 and 0.2199 that demonstrates the quadratic equations are valid.Also, the high correlation coefficient for both dye models (0.9964, 0.9945) confirms that the predicted model is fitted with the data, ideally.The values of R 2 Adj (0.9898, 0.9847) and R 2 Pred (0.9449, 0.9241) show the accuracy of the models for Copper and MB, respectively.Also, the predicted models for Copper and MB have ideal PRESS values (68.00, 97.51) demonstrating the power of the model.The accuracy function for the two models is 4 < 34.631, 29.068 that represents ideal signal-to-noise ratios [28].

Interaction between factors.
As can be seen in Figure 4(a) when the pH and adsorbent dose increase, the percentage of Copper removal is increased.As a result, by increasing the adsorbent mass, active sites of AC/Diop increased and the higher probability of forming electrostatic bond between adsorbent and adsorbate is expected so the removal percentage of Copper increased.The contour plot of the mentioned factors supports the 3D graph, too.The maximum Copper removal percentage done at pH = 5.45 and adsorbent dose of 0.1 g.
Figure 4(b) shows the 3D and contour plots of the simultaneous effect of pH and contact time.When the pH and contact time increased, the percentage of Copper removal is increased.The contour plot of mentioned factors supports the 3D graph and the maximum percentage of Copper removal occurred in 9.56 min.
In Figure 4(c), when the contact time and adsorbent dose increase, the percentage of Copper removal is increased and the maximum Copper removal occurs during the contact time 9.56 min, that it can be derived that the process equilibrium time probably occurs at the end of considered range of time [29].
In Figure 5(a), the 3D and contour plots of the simultaneous influence of the two factors of pH and adsorbent dose on the MB dye removal.As can be seen from MB dye regression equation, the positive effect of adsorbent dose and pH on the MB dye removal can be observed.
In Figure 5(b), the influence of two factors of pH and contact time on the MB dye removal applying the AC/Diop composite is observed.The increase of pH and contact time leads to an increase of MB dye removal percentage, as their signs in the MB dye regression equation also derive the same effect.In the contour and 3D plots of Figure 5(c), it can be seen in the 13.49 min of contact time and 0.43 g of adsorbent mass, the MB dye maximum removal happens.
The pH has a greater effect on the adsorption process of MB dye and based on the coefficients of the factors in the regression equation, can be seen the magnitude of the factors effects.

3.2.2.3.
Optimisation..The optimal conditions for having the maximum removal percentage of Copper and MB were obtained using the BBD.The maximum percentage of pollutants' removal (97.20%, 96.19%) at pH = 5.45, 7.35 using AC/Diopadsorbent amount of 0.10, 0.43 g was achieved with desirability 1.0 at contact time of 9.56, 13.49 min at initial concentration of 10.0 mg L −1 (SI Table 8).
For validation of the model, the pollutants' removal percentage was done four times by AC/Diop adsorbent and experimental values of 96.86% and 97.35% for removal of Copper and MB demonstrates the validity of the model.

Adsorption isotherm
Two important isotherm equations like Langmuir and Freundlich were used to describe the equilibrium characteristics of the adsorption of Copper and MB by the AC/Diop adsorbent.Surface equilibrium adsorption isotherms supply data about the adsorbent capacity.The experimental data of Copper and MB by adsorbent obtained at equilibrium were fitted satisfactorily with Langmuir isotherm and the isotherm parameters are given in SI Table 9.The q m values and the equilibrium adsorption constants by Langmuir model were 184.25, 289.51 mg g −1 and 0.19, 0.21 L mg −1 for Copper and MB, respectively, that showing the adsorption of pollutants onto AC/Diop composite is favourable.The value of n −1 for Freundlich isotherm (0.13, 0.23 for Copper and MB, respectively) presents an ideal adsorption.In order to study the adsorption isotherm, the results (R 2 values) showed that Langmuir adsorption isotherm with R 2 = 0.9754 and 0.9498 is more qualified for both Copper and MB pollutants [30].

Adsorption kinetics
The pseudo-first-order and pseudo-second-order kinetic models were tested to investigate the rate of the adsorption of Copper and MB by the AC/Diop composite.The kinetics constants are presented in SI Table 10.From the table, it could be confirmed that the adsorption of Copper and MB using the adsorbent followed the pseudo-second order model and it indicated the adsorption of pollutants on the surface of the AC/Diop composite represented two-phase reactions such as rapid adsorption for shorter duration in the initial stage that is followed by slow adsorption for the longer duration.Several investigators have reported that the fast reaction may be due to chemisorptions involving valence forces through the exchange or sharing of electron between the adsorbent and the adsorbate.The slow response is due to the diffusion of ions into the adsorbent.Hence, the rate limiting step is finalised as chemisorption [44][45].Based on the SITable 10, since the correlation coefficients for the second model for Copper and MB are closer to 1, the kinetics of the adsorption process for both pollutants are a kind of second-order pseudo-model and show that the adsorption rate depends on the concentration of the Copper and MB on the adsorbent surface.

Application of AC/Diopadsorbent on real samples
Table 11 represents the results of Copper and MB using the AC/Diop composite adsorbent on the real sample obtained from textile industry.The sorption experiments were performed with textile waste water spiked with Copper and MB fed with various amounts of the pollutants (30, 60 and 90 mg L −1 ).Afterwards, spiked real samples were treated under the mentioned method.As can be seen from the Table 11, acceptable removal percentage of Copper and MB confirms the ideal performance of AC/Diop adsorbent on real sample.

Comparison with other adsorbents for removal of Copper and MB
Several adsorbents have been reported in the literature for removal of Copper and MB.Table 12 compares the efficiency of different adsorbents for removal of Copper and MB.Based on the results, the performance of AC/Diop is preferable compared to other adsorbents.
Analysis of variance.The dependent factor in the experimental design in our study is the percentage of removal of the Copper and MB by the AC/Diop composite adsorbent.Using the BBD, a second-order coded equation is extracted showing the relation between the response and independent factors affecting the adsorption process.

Figure 4 .
Figure 4.The 3D response surface and contour plots for interactive effects of a) pH and Adsorbent dosage, b) pH and contact time, and c) Adsorbent dosage and contact time for the removal percentage of Copper by the AC/Diopadsorbent.

Figure 5 .
Figure 5.The 3D response surface and contour plots for interactive effects of a) pH and Adsorbent dosage, b) pH and contact time, and c) Adsorbent dosage and contact time for the removal percentage of MB by the AC/Diopadsorbent.

Table 2 .
Experimental factors and levels in the Box Behnken design for the removal percentage of Copper by AC/Diop adsorbent.

Table 3 .
Experimental factors and levels in the Box Behnken design for the removal percentage of MB by AC/Diopadsorbent are presented.

Table 6 .
ANOVA of the second-order polynomial equation for the removal percentage of Copper by the AC/Diop adsorbent.
a Sum of square, b Degree of freedom, c Mean square.

Table 7 .
ANOVA of the second-order polynomial equation for the removal percentage of MB by the AC/Diop adsorbent.Sum of square, b Degree of freedom, c Mean square. a

Table 11 .
Removal percentages of Copper and MB by the AC/Diopcomposite from textile wastewaters (N = 4).

Table 12 .
Performance of some different adsorbents for removal of Copper and MB.FESEM, BET, and XRD techniques, and used for removal of Copper and MB pollutants.The results showed that the Copper and MB adsorption process depends on the adsorbent mass, pH, and contact time.The adsorption kinetics follows the pseudo-second-order model and the isotherm follows the Langmuir model.AC/Diop is superior to AC and Diop, with a higher capacity for the removal of Copper and MB.The use of Diop in the synthesis of AC/Diopcomposites improves the removal percentage of pollutants.It is also more biodegradable and low cost in comparison with the other adsorbents and is not harmful to the environment.As a result, the AC/Diop adsorbents can be used to remove the pollutants effectively.