Adsorption of the hazardous dye malachite green (MG) by Acacia nilotica (AN) waste was investigated. Batch process variables for the adsorption of MG by AN were optimized. The mechanisms involved in the adsorption of MG by AN were explored using isotherms and kinetic models. The thermodynamic parameters were calculated to determine the spontaneity and thermal nature of the MG adsorption reaction. The maximum equilibrium adsorption capacity of AN was found to be 113.26 mg/g at 30 °C. The MG adsorption data revealed that AN adsorbs MG by multilayer adsorption, as shown by the better fit of the data to the Freundlich and Halsey models (R2 = 0.99) rather than to the Langmuir model. Multilayer adsorption involves physisorption, which was confirmed by the E value (mean free energy of adsorption) of the Dubinin–Radushkevich model (6.52 kJ/mol). Surface diffusion was found to be the main driving force for MG adsorption by AN. The MG adsorption reaction was endothermic, based on the enthalpy, and was controlled by the entropy of the system in the T1 temperature range (30 to 40 °C), while the opposite trend was observed in the T2 range (40 to 50 °C). Moreover, MG adsorption by AN was found to be nonspontaneous at all temperatures.
Malachite green (MG) is a cationic dye, being positively charged in aqueous solution, and belongs to the triarylmethane dye class. This dye has been used to control diseases and parasites in fish farms (Zhang et al. 2008) because of its properties as a disinfectant and ectoparasiticide. In addition, MG is used to color silk, cotton, and leather (Gupta et al. 2004; Kumar et al. 2005). However, this dye can damage aquatic life via its carcinogenic and mutagenic properties, which have been confirmed in animal model studies (Culp et al. 1999; Srivastava et al. 2004; Yonar & Yonar 2010). Moreover, the intense color of this dye imparts an unpleasant color to water bodies, and it is recalcitrant to degradation by light and other oxidizing agents, increasing environmental concerns and making its removal problematic.
Conventional methods of chemical precipitation (Gao et al. 2007), advanced oxidation (Azbar et al. 2004; Yetilmezsoy & Sakar 2008), and biological oxidation (Kapdan et al. 2000; Ghoreishi & Haghighi 2003) have been used to remove contaminating dyes from water, but, for dilute concentrations of dyes, their use is not practical. Moreover, foam filtration (Lu et al. 2010) and flocculation (Wei et al. 2009; Riera-Torres et al. 2010) have also been used to remove dyes from textile industrial waste, but these methods are also ineffective. Therefore, over the last two decades, focus has shifted towards using adsorption techniques for removing hazardous dyes because of their effectiveness and low cost (Kant 2012; Amin et al. 2016). Adsorption techniques are more economical than their alternatives if the adsorbent is indigenous to the treatment area and requires no treatment prior to its application to wastewater treatment (Amin et al. 2015). Although many bioadsorbents have been used for the removal of MG dyes from aqueous streams, for example, sugarcane waste (Kattri & Singh 1999), neem sawdust (Khattri & Singh 2000), maize cob (Sugumar & Gopalan 2000), Parthenium weed (Rajeshwarisivaraj & Subburam 2002), jackfruit peel (Inbaraj & Sulochana 2002), fresh water algae (Kumar et al. 2005), rice husks (Guo et al. 2003), fly ash (Janoš et al. 2003), and spent tea leaves (Akar et al. 2013), the search for low-cost waste materials to remove hazardous dyes from waste streams prior to disposal continues. Furthermore, the use of these waste materials is important from an economic point of view (Hameed et al. 2008).
This study focused on the use of Acacia nilotica (AN) sawdust for the removal of MG in batch experiments. AN is a waste adsorbent which is generated from the wood industry, and AN wood sawdust is abundantly available in the wood processing industry as a solid waste material (Jain et al. 2011). In this study, we optimized various process variables, and kinetics and isotherm studies were performed to elucidate the mechanism of MG dye adsorption. In addition, thermodynamic parameters were calculated to determine the stability of the material and the spontaneity and thermal nature of the MG dye adsorption reactions.
MATERIALS AND METHODS
The chloride salt of MG was used as the model pollutant in this study. A stock solution of 1,000 mg/L was prepared, and, subsequently, the working solutions were prepared by the dilution of the stock solution with deionized water.
Collection and preparation of the waste adsorbent
AN waste biomass was collected from local wood companies in the Kingdom of Saudi Arabia. The adsorbent was washed with deionized water, dried, ground, and sieved, and the required mesh sizes were used in the batch experiments.
Characterization of the AN adsorbent
Scanning electron microscopy (SEM, Hitachi S-3000N, Japan) analysis was conducted to analyze the texture of the AN adsorbent material before MG dye loading. Furthermore, the porosity of the AN surface was analyzed from the SEM micrographs taken before MG dye loading and at various magnifications (250×, 500×, and 1,000×). These micrographs were taken at a 5-kV acceleration voltage, and a 25-mm working distance was maintained. Moreover, energy dispersive X-ray (EDX) analysis was performed to determine the elemental composition of the AN adsorbent. Fourier-transform infrared (FTIR) measurements were made (model 460 plus) in the region of 4,000–400 cm−1, and the types of functional groups on the surface of the AN adsorbent were determined before MG dye loading.
Batch adsorption studies
Adsorption isotherms, kinetics, and thermodynamic studies
The experimental data for the adsorption of MG onto AN at initial dye concentrations ranging from 25 to 200 mg/L (Table 1) were fitted with different adsorption kinetic and isotherm models. In contrast, the batch parameters, i.e., pH (7), temperature (30 °C), adsorbent dose (0.05 g/L), shaking speed (220 rpm), and contact time (40 min), were kept constant. We used the Langmuir, Freundlich, Halsey, Dubinin–Radushkevich (D-R), and Temkin isotherm models. Different kinetic theories (the pseudo-second-order kinetic and intra-particle diffusion models) were also applied to explain the driving force behind the mass transfer of the pollutants from the liquid to the solid phases. The corresponding correlation coefficients and constants of the model plots were calculated. Moreover, the Gibbs free energy, enthalpy, and entropy of the MG dye adsorption process were investigated at the solid–solution interface. To analyze the energy of the system and the optimum temperature for MG adsorption, the Gibbs free energy, enthalpy, and entropy of the reaction system were studied. For this purpose, equilibrium adsorption studies of MG dye onto AN were carried out at variable temperatures (30, 35, 40, 45, and 50 °C), while keeping the dye concentration fixed at 50 mg/L, the adsorbent dose at 0.5 g/L, the pH at 7, and the agitation at 220 rpm. The enthalpy and entropy of the system were calculated from the slope × R and intercept × R of the (lnKd vs. 1/T) plot, where R is the universal gas constant. The Gibbs free energy was calculated using Equation (5).
|ΔH/(kJ/mol) .||ΔS/(kJ/mol) .||ΔG (J/mol)|
|303 K .||308 K .||313 K .||318 K .||323 K .|
|ΔH/(kJ/mol) .||ΔS/(kJ/mol) .||ΔG (J/mol)|
|303 K .||308 K .||313 K .||318 K .||323 K .|
Multiple replicates (minimum of three) for each batch experiment were conducted, and the data are presented as the mean ± standard error. The one-way analysis of variance was performed to analyze the effects of different variables on the MG adsorption capacity of the AN adsorbent. Different letters on the mean error bar indicated the least significant difference of means at p = 0.01.
RESULTS AND DISCUSSION
SEM and EDX analyses of the AN adsorbent
EDX analysis revealed the elemental composition by weight of the AN adsorbent: C (58.22%), O (41.46%), and Al (0.12%) (see Figure 1(c)). After MG loading onto the AN surface (Figure 1(d)), the increases in C (59.12%), N (4.73%), and O (43.87%) contents confirmed that the dye has been successfully adsorbed on the adsorbent surface.
The peak at 3,616 cm−1 is pertinent to O-H stretching vibrations (Wu et al. 2016). The band at 2,216 cm−1 might be due to –C ≡ N– stretching vibrations. The reduction in peak intensity at 3,616 and 2,216 cm−1 indicates the formation of interactions with the hydroxyl and nitrile groups, respectively, during the adsorption of MG onto AN (Shakoor & Nasar 2016).
Optimization of batch process variables
Effects of the contact time on the adsorption of MG by AN
Effects of pH on the adsorption of MG by AN
MG is a cationic dye and is positively charged in the aqueous phase. Thus, MG is an ionic species, and its adsorption onto the AN surface is primarily influenced by the surface charge of the adsorbent, which in turn is influenced by the solution pH. In this study, the effect of pH on MG adsorption was studied, while the agitation speed, solution temperature, initial dye concentration, amount of AN adsorbent, contact time, and particle size were kept constant. i.e., 220 rpm, 30 °C, 50 mg/L, 0.05 g, 40 min, and 40 mesh, respectively. The effects of pH were investigated over the pH range of 4 to 10 pH (Figure 3(b)). MG adsorption was found to be significantly (p = 0.01) reduced (23.80 mg/g, 47.60%) at low pH (4) compared to that (27.40 mg/g, 54.81%) at high pH (11). This is because the excess H+ ions at low pH can compete with the positive MG ions for active adsorption sites. In contrast, at high pH, the positive charges at the solid–liquid interface decrease and the adsorbent surface becomes negatively charged; this results in high MG dye removal (Pan & Zhang 2009; Baek et al. 2010). Similar pH effects have also been reported for the adsorption of cationic dyes by rattan sawdust (Hameed & El-Khaiary 2008).
Effect of agitation speed on the adsorption of MG by AN
Agitation speed plays a key role in the mass transfer processes during batch adsorption. It had a significant (p = 0.01) impact on the adsorption capacity of the AN adsorbent (Figure 4(a)). MG adsorption was increased significantly from 2.28 mg/g (5.70%) at 100 rpm to 12.60 mg/g (31.50%) at 220 rpm. This can be ascribed to the increased turbulence and degree of mixing that reduce the thickness of the boundary layer surrounding the adsorbent particles. Similar observations have been reported in the literature (Fil et al. 2014). Therefore, an agitation speed of 220 rpm was selected for the batch adsorption experiments.
Effect of the initial MG dye concentration on its adsorption by AN
Temperature and thermodynamics
In the T1 temperature range, the value of ΔH was 21.79 kJ/mol, demonstrating the endothermic nature of reaction between the MG dye molecules and the adsorbent surface. The positive value of ΔS (0.06 kJ/mol) confirms the increased disorder at the solid–liquid interface during MG adsorption. This is because the mobility of the adsorbate ions/molecules in the solution increases with increasing temperature. However, the MG adsorption reaction is unfavorable with respect to the enthalpy but favorable with respect to the entropy, indicating that the MG adsorption reaction is an entropy-driven process.
The values of ΔG in the T1 temperature range indicate the nonspontaneous nature of MG adsorption onto the surface of the AN adsorbent. However, the nonspontaneity of the absorption of MG by AN is reflected by the decreasing values of the Gibb's free energies with increasing temperature (Table 1).
In the T2 temperature range, the values of ΔH and ΔS were found to be −30.95 and −0.01 kJ/mol, respectively (Table 1). These values indicate that the adsorption of MG dye is favorable with respect to the enthalpy of the system but unfavorable with respect to the entropy. The values of ΔG in the T2 temperature range indicates the nonspontaneous nature of MG adsorption on the surface of the AN adsorbent at all temperatures.
Equilibrium adsorption isotherms
These isotherms are mathematical models for the optimization of a system with extensive applications. These models also indicate the relationship between the contaminant ions that are adsorbed on the adsorbent surface and those present in the solution at equilibrium (Shanmugaprakash et al. 2014). The Langmuir (Langmuir 1918), Freundlich (Freundlich 1906), Halsey (Halsey 1948), D-R (Dubinin & Radushkevich 1947), and Temkin (Temkin & Pyzhev 1940) isotherms were employed to elucidate the mechanism of MG adsorption.
The equilibrium data were well-fitted (R2 = 0.9943) to the Langmuir model. A dimensionless constant called the equilibrium parameter, RL (1/(1 + KadsC0)), can be used to prove the suitability of AN adsorbent for the removal of MG. The adsorption of any pollutant is regarded favorable if 0 < RL < 1, unfavorable if RL > 1, linear if RL = 1, and irreversible if RL = 0. In the present the study, the RL value was found to be 0.49, indicating that AN is a useful material for the adsorption of MG dye.
According to this model, qt varies linearly with the square root of time. Although this model showed a poor fit (R2 = 0.9627) to the data compared to the pseudo-second-order model (Figure 8(b)), the larger values of Ci (12.23 mg/g) reflect the contribution of surface diffusion processes in the rate controlling step. Moreover, the intra-particle diffusion model is valid only if qt vs. t0.5 passes through the origin; otherwise, other adsorption mechanisms are involved, alongside the intra-particle diffusion model.
In this work, the batch adsorption conditions of MG by AN were optimized. An increased removal of MG dye (about 10%) was observed with increasing contact time, from 20 to 240 min, and equilibrium adsorption of MG dye was attained after 40 min. The adsorption of MG was significantly reduced (about 47%) at low pH (4) compared to that at high pH (11). In addition, MG adsorption increased significantly (about 25%) on increasing the speed of agitation from 100 to 220 rpm. Among the other parameters affecting adsorption, a high MG adsorption capacity (about 56%) was observed at a maximum initial MG dye concentration of 200 mg/L compared to that at lower initial MG dye concentration (25 mg/L), indicating the dependency of AN adsorption on the initial adsorbate concentration in the solution. The maximum equilibrium adsorption capacity of AN was 113.26 mg/g at 30 °C. The MG adsorption data indicate multilayer adsorption because the data were fitted better by the Freundlich and Halsey models (R2 = 0.99) than by the Langmuir model. Using the D-R model, we confirmed the involvement of physisorption during multilayer adsorption, which was determined from the E value of the D-R model (6.52 kJ/mol). Surface diffusion was found to be a possible mechanism for the adsorption of MG by AN. The MG adsorption reaction was endothermic, based on the enthalpy, and was controlled by the entropy of the system in the T1 temperature range (30 to 40 °C), while the opposite trend was observed in the T2 range (40 to 50 °C). Moreover, MG adsorption was found to be nonspontaneous at all temperatures.
The project was financially supported by King Saud University, Vice Deanship of Research Chairs.
CONFLICT OF INTEREST
The authors declare no conflict of interest.