Preparation and characterization of biodiesel waste-derived biomass for the removal of dye from contaminated water

The discharge of dye containing effluents from the textile, leather, paper and plastic industries into the environment poses severe problems to many forms of life. Methylene blue (MB) is the most commonly used chemical substance for dying cotton, wood and silk. Although MB is not strongly hazardous, it can cause some harmful effects. Acute exposure to MB can cause increased heart rate, vomiting, shock, Heinz body formation, cyanosis, jaundice, quadriplegia and tissue necrosis in humans (Theydan & Ahmed 2012). Therefore, the removal of such dyes from processed effluents becomes environmentally important.

The discharge of dyes to the environment is a matter of concern for both toxicological and aesthetic reasons, causing serious water pollution-related problems to aquatic life, e.g., reduced light penetration in water bodies. The presence of even very small amounts of dyes in water, i.e., less than 1 mg L⁻¹ for some dyes, is highly visible and undesirable (Robinson et al. 2001; Rafatullah et al. 2010). There are many techniques for removing dyes from wastewater, such as adsorption, coagulation, filtration, oxidation, sedimentation, precipitation and reverse osmosis (Kumar & Namasivayam 2014).

Adsorption is an efficient and economically feasible process for removing dyes from wastewater using various adsorbents (Kavitha & Namasivayam 2007a, 2007b). Activated carbon is one of the most commonly used adsorbents for dye removal due to its high surface area and large adsorption capacity (Kumar et al. 2017a, 2017b). A number of non-conventional adsorbents have been tested at the lab-scale for the treatment of dye-containing wastewaters. Waste materials from processing industry and agriculture are usually easily available and therefore they are considered as alternative adsorbents (Gupta & Suhas 2009). Such an approach not only converts the waste into a useful material, but it also prevents on-site burning of the waste and saves transportation and disposal costs.

Jatropha curcas is a shrub of significant economic importance due to its several potential industrial and medicinal applications (Herrera et al. 2006). It is a drought-resistant perennial species that grows well in marginal poor soil. Bio-diesel is made from its seeds, generating a large volume of Jatropha husk (JH). Widespread cultivation of J. curcas has been initiated globally. According to the Planning Commission of India, the estimated potential area of J. curcas plantations is 17.4 million hectares and the projected JH production is 350 million tons. As JH is rich in lignin, it has the potential to act as a good precursor for the production of activated carbon (AC), which forms the basis of any adsorbent-related research (Ramakrishnan & Namasivayam 2011).

Waste biomass has appeared in several previous studies for the removal of MB using activated carbons prepared by different activators such as periwinkle shells by KOH (Olugbenga et al. 2008), Arundo donax (Osman 2019), cork and paper waste-based (Novais et al. 2018), thermally activated coir pith carbon (Kavitha & Namasivayam 2007a, 2007b), cotton waste (Tian et al. 2019), neem leaf powder by H₂SO₄ (Patel & Vashi 2013), rice husk by HCl (Muhammad et al. 2012), peanut hull (Renmin et al. 2005), cashew nut shells (Spagnoli et al. 2017), ZnCl₂ activated Buriti shells (Pezoti et al. 2014) and activated JH husk ZnCl₂ (Kumar et al. 2017a, 2017b). Since ZnCl₂ is a dehydrating agent, the formation of tar during the activation of the precursor can be avoided (Caturla et al. 1991).

In the present study, a novel JH carbon-based adsorbent was investigated for its efficiency to remove MB from contaminated water. Batch adsorption experiments were carried out using ZnCl₂ activated Jatropha husk carbon (ZAJHC) and the experimental parameters, namely, pH, contact time, adsorbent dose were optimized. Isotherm, desorption and thermodynamic studies were also conducted in order to ascertain the kinetics and mechanism of the adsorption process and the reusability of the adsorbent.

The washed, dried and crushed JH (450 g) was mixed with 2.0 L of warm water containing 450 g of ZnCl₂ for 1 h. The solution was drained and ZnCl₂ soaked JH was oven dried at 60 °C for 12 h. The material was filled into a steel container and fitted with a tight lid. This container was placed in another concentric steel container with another lid. The inner void space was filled with sand layers up to the brim of the container. This arrangement prevented exposure of the carbonizing material to air, allowing only a limited presence of oxygen trapped in the voids of the material being activated. The activation setup was placed in a muffle furnace for 1 h at 800 °C. After cooling, the carbon was taken out and the excess ZnCl₂ was leached out by immersing it in a hot 1.0 M HCl solution for 24 h and thereafter placed in a hot air oven at 80 °C. It was then repeatedly washed with hot water until the chloride content was completely removed from the wash water (tested by the silver nitrate method). This material was subsequently oven dried at 105 ± 5 °C for 8 h and sieved to 250–500 μm size (60–35 mesh ASTM) and designated as ZnCl₂ activated JH carbon (ZAJHC) and stored in airtight plastic containers. All the chemicals used in this study were of analytical grade (AR) and they were obtained from Loba Chemie, Mumbai.

Figure 1 shows the molecular structure of MB. It has a molecular formula C₁₆H₁₈CIN₃S with a molecular weight of 319.85. It is water soluble and blue in colour (λ_max, 665 nm). A stock solution of MB was prepared by dissolving an appropriate quantity of MB in distilled water. The stock solution was diluted with distilled water to obtain the desired initial MB concentrations used for the experiments.

Adsorption experiments were carried out using 25 mg of ZAJHC with 50 mL of dye solution, with the desired initial concentration and pH, at 200 rpm and 35 °C in a thermostat controlled orbital shaker (ORBITEK, Chennai, India). The MB concentration was determined spectrophotometrically by measuring the absorbance at 665 nm using an UV-Vis spectrophotometer (Specord 200, Analytic Jena, Germany). The pH was measured using a pH meter (Elico, model LI-127, Hyderabad, India).

The effect of adsorbent dosage was tested by varying the dose from 25 to 200 mg per 50 mL of dye solution, and experiments were carried out until the adsorption reached equilibrium conditions. The effect of pH was envisaged by performing experiments in the pH range of 2.0–11.0. Adsorption experiments were carried out by equilibrating (shaking, stirring) an adsorptive solution of a known composition and volume, with a known amount of adsorbent, at a constant temperature and pressure for a limited period of time. Under such conditions, it was expected that adsorption would reach a steady state or the adsorption rate would no longer change after a period of time. The adsorption of MB reached equilibrium at 120 min for a concentration of 100 mg L⁻¹, 140 min for 120 mg L⁻¹ and 160 min, respectively, for other MB concentrations tested in this study. The MB removal decreased with increasing initial concentrations; however, the actual amount of dye adsorbed per unit mass of carbon increased with an increase in the MB concentration.

The Langmuir, Freundlich and Dubinin–Radushkevich (D-R) isotherms were employed to study the equilibrium adsorption capacity of ZAJHC. Desorption studies were carried out at an initial dye concentration of 100 mg L⁻¹. The dye solution was separated from the adsorbent by centrifugation at 2,500 rpm for 30 min and its absorbance was measured. The dye-loaded adsorbent was filtered using Whatman filter paper and washed gently with water to remove any unabsorbed dye. Thereafter, the spent adsorbent was agitated for 120 min with 50 mL of distilled water and adjusted to different pH values. For estimating the temperature effects, the adsorption of 100 mg L⁻¹ of MB by 25 mg/50 mL of adsorbent was tested at temperatures of 35, 40, 50 and 60 °C in a thermostat controlled orbital shaker (ORBITEK, Chennai, India).

The surface and structural morphologies of the loaded and unloaded ZAJHC were characterized by different analytical techniques. X-ray diffraction (XRD) was performed at room temperature using a PAN analytical X-Pert-Pro diffractometer, Westborough, USA. Infrared spectrum of the samples was obtained by using a Fourier transform infrared (FTIR) spectrometer (Bruker Tensor 27, Germany). Raman scattering was performed on a JY-1058 Raman spectrometer using a 520 nm laser source. The morphology and elemental analysis of the loaded and unloaded ZAJHC were examined using a scanning electron microscope (SEM-JSM, 840A, JEOL, Japan). The carbon samples were filtered using a suction pump on qualitative filter paper and stored in a vacuum desiccator, which was later used for instrumental analysis.

The XRD pattern was used to analyse the crystallinity and phases of the as-prepared ZAJHC from Jatropha husk (JH) activated by ZnCl₂ and the MB adsorbed ZAJHC. Figure 2(a) shows broad peaks at 23.5° and 43°, which correspond to the (002) and (100) planes, respectively. Thus, the results confirm that the ZAJHC was amorphous in structure (Bharath et al. 2014; Rajesh et al. 2014). The XRD pattern of MB adsorbed ZAJHC shows that the intensity of the (002) and (100) planes are slightly suppressed (Figure 2(a)). Borah et al. (2015) reported that the amorphous nature of the entire studied porous carbon (diffused peak was observed at 2θ = 25.4°) corresponded to the (002) diffractions of graphitic carbon. The decreased intensity of the planes may be due to the adsorbed MB on the surface and pores of the ZAJHC.

The FTIR spectra of the MB adsorbed ZAJHC are shown in Figure 2(b). As seen from this figure, the broad band at 3,420 cm⁻¹ can be assigned to the stretching vibrations of hydroxyl groups -OH. The FTIR spectrum of the activated carbon shows weak and broad peaks in the region of 500–4,000 cm⁻¹. The characteristic peaks obtained at 2,926 and 2,850 cm⁻¹ indicate the presence of C-H groups. The MB loaded ZAJHC shows a peak at 1,574 cm⁻¹ which was shifted to 1,570 cm⁻¹ and can be attributed to the N-H vibration coupled with the C-N stretching mode, indicating the saturation of the material with MB (Lua & Yang 2005; Heidarinejad et al. 2018; Jamshaid et al. 2019). The band at 1,065 cm⁻¹ can be assigned to esters (Pezoti et al. 2014). The main change observed after adsorption of the dye molecule was splitting of the strong stretching of C = O into small splits (Figure 2(b)).

Raman spectra (Figure 2(c)) provide structural analysis of ZAJHC and MB adsorbed ZAJHC. ZAJHC shows two prominent peaks at 1,340 and 1,585 cm⁻¹ which correspond to the well documented D and G bands, respectively. The weak D band at 1,340 cm⁻¹ is associated with the vibration of A_{1 g} symmetry of sp³ carbon atoms and associated with defect ordered structural peak. The peak G relates to the first-order scattering of the E_{2 g} phonon of carbon sp² atoms and second-order double resonant process with opposite momentum in the highest optical branch near the K points in the Brillouin zone (Bz) of the carbon atom. The D and G band ratio of ZAJHC (I₁₃₄₀/I₁₅₈₅) was 0.7, which indicates a unique amorphous carbon structure and a high content of lattice edges or plane defects of ZAJHC (Bharath et al. 2014; Rajesh et al. 2014). The intensity ratio (I_D/I_G) of the MB adsorbed ZAJHC was 0.96, which indicates that the defects increased due to the adsorption of MB on the surface of ZAJHC (Figure 2(c)).

The morphology of the ZAJHC and MB loaded ZAJHC was observed through SEM micrographs. Figure 3(a) and 3(b) show tremendous, perfect and constructed pore structures on the surface of ZAJHC. In this study, the carbon was compressed and the appearance of the lanky structure is due to the formation of more interspaces between the monolayers of the carbon, presumably due to activation using ZnCl₂. A higher volume of pores developed from the ZnCl₂ activation acts as a route for the contaminants to enter into the micropores (Ramakrishnan & Namasivayam 2011). Basically, the high SSA and pore structure are the basic parameters for an effective adsorbent. When the porosity of the JHC and ZAJHC increases, the SSA also increases (Mohanty et al. 2005; Borah et al. 2015). After adsorption, the surface turned into a more irregular structure (Kumar et al. 2017a, 2017b). Evidently, the dye molecules were strongly adsorbed onto the surface of ZAJHC (Figure 3(c) and 3(d)).

The effect of pH on the removal of MB is shown in Figure 4(a). The per cent removal was >85% in the pH range of 2.0–11.0. At pH 10.0, negatively charged surface sites on the adsorbent will favour the adsorption of dye cations due to the electrostatic attraction and hence dye removal was high (i.e., >85%). In this study, although the pH was decreased, the removal of MB was still ∼70%. Basically, MB and other cationic dyes produce an intense molecular cation (C⁺) and reduced ions (CH⁺). At high pH, OH^– on the surface of adsorbent will favour the adsorption of cationic dye molecules. Many previous works have also reported that MB adsorption efficiency onto carbon-based adsorbents usually increases as the pH is increased (Kavitha & Namasivayam 2007a, 2007b).

Desorption studies were conducted as a function of pH in order to analyse the possibility of reusing the adsorbent for many adsorption cycles and to make the process more economical. At lower pH, the hydrogen ion (H⁺) concentration increases in solution, which then displaces the adsorbed dye cations into solution. In this study, the desorption efficiency decreased from 35.2% at pH 2.0 to 19.5% at pH 10.0, at an MB concentration of 100 mg L⁻¹ (Figure 4(b)) (Chen et al. 2013). The reversibility of the adsorbed dyes is in agreement with the pH-dependent results obtained. The better desorption efficiency of dye molecules at low pH indicates that adsorption of MB on to ZAJHC may be due to the mechanism of ion-exchange (Ofomaja & Ho 2008).

In order to study the effect of contact time, the initial MB concentrations were varied from 100 to 180 mg L⁻¹ at 35 °C (Figure 5(a)). The adsorption of MB reached equilibrium at 120 min for 100 mg L⁻¹, 140 min for 120 mg L⁻¹ and 160 min for other MB concentrations. The MB removal decreased with increasing initial concentrations, but the actual amount of dye adsorbed per unit mass of carbon increased with an increase in the MB concentration. This implies that the adsorption is highly dependent on the initial concentration of dye. At lower concentration, however, the ratio of the initial number of dye molecules to the available surface area is low. The MB uptake profiles, as a function of time, show a smooth and continuous curve that leads to saturation which suggests the possibility of monolayer coverage of dye on the surface of the adsorbent (Borah et al. 2015). On the other hand, at high MB concentrations, the available sites of adsorption become fewer and hence the removal of MB is dependent upon the initial concentration (Malik 2003). The ZAPHC S_BET (m² g⁻¹) 822 is mentioned in Table 1.

Table 2 shows the values of k₁, which were calculated from the slopes of the linear plots derived from the computed results obtained from the first- and second-order kinetic models along with the experimental determined q_e values which are also shown in Table 2. k₁ and k₂ are the rate constants for pseudo first-order and second-order equations, respectively. Table 2 shows the kinetic data parameters which revealed that MB for pseudo first-order was higher than that of second-order. This implies that the adsorption of MB by ZAJHC adsorbent could be represented by the pseudo first-order kinetic model. In addition, the theoretical q_e cal values for pseudo first-order model were found to be closer to the experimental (q_e exp) values. This result is also in agreement with previously reported literature (Martins et al. 2006).

According to the results shown in Figure 5(b), the initial curved portion can be attributed to the bulk diffusion effect, the linear portion to the intraparticle diffusion effect and the plateau to the equilibrium condition. The plots of q_t vs k_id have the same general features as shown in Figure 5(b). The linear portions of the plots do not pass through the origin indicating that intraparticle diffusion is not the only rate controlling step for the adsorption process. The slopes of the linear portions of the plots of q_t vs t^1/2 give the values of k_id (Table 2). The values of the intercept provide information about the thickness of the boundary layer, i.e., the larger the intercept, the greater the boundary layer effect (Kannan & Sundaram 2001). The linear portions of this plot can be attributed to the instantaneous utilization of the most readily available adsorbing sites on the adsorbent surface.

The Langmuir constants Q₀ and b were found to be 500 mg g⁻¹ and 0.056 (L mg⁻¹), respectively. The essential characteristics of the Langmuir isotherm can be expressed by a dimensionless constant called the equilibrium parameter R_L. The R_L values were found to vary between 0 and 1 which indicates a favourable adsorption process (Table 3). The Freundlich constants, k_f and n were calculated from the linear plot of log q_e vs log C_e (Table 3). Table 4 compares the Langmuir, Freundlich and D-R constants for adsorption of MB onto various adsorbents as reported in the literature. The Freundlich model assumes heterogeneous adsorption due to the diversity of sorption sites or the diverse nature of the adsorbate adsorbed, free or hydrolyzed species. On the other hand, the adsorption capacity based on the D-R isotherm is independent of the operating temperature, but it varies depending on the nature of adsorbent and the adsorbate.

The mean free energy of adsorption gives information about the chemical ion-exchange mechanism. The E value varied between 8 and 16 kJ mol⁻¹ and thus the adsorption process follows the ion-exchange mechanism. On the other hand, if E < 8 kJ mol⁻¹, the adsorption process is physical in nature. The mean energy of the MB adsorption was calculated as 11.18 kJ mol⁻¹, which indicates that the adsorption of MB on the ZAJHC occurred by ion-exchange mechanism. The D-R constants of various adsorbents for MB reported in the literature are presented in Table 4. Figure 5(c) presents the different adsorption isotherms fitted to the experimental data of this study. Based on the results shown in Table 3, the Freundlich isotherm has the lowest Δq and, therefore, it is most suitable to represent the adsorption of MB onto ZAJHC than the other isotherms.

The removal of MB by ZAJHC at different adsorbent doses (25–200 mg/50 mL) and different initial concentrations of MB from 100 to 180 mg L⁻¹ was studied. It was observed that an increase in the adsorbent dose increased the removal of MB and this can be attributed to the greater surface area and the availability of more adsorption sites on ZAJHC (Arivoli et al. 2008). An increase of temperature increased the MB removal (Figure 5(d)). The positive value of ΔH° (3.86 J mol⁻¹K⁻¹) shows the endothermic nature of the adsorption process. The negative values of ΔG° indicate the spontaneous nature of adsorption for MB. Table 5 shows positive values of ΔS° (65.86 J mol⁻¹K⁻¹) which suggest an increased randomness at the solid/solution interface during the adsorption of dye by ZAJHC (Namasivayam & Sumithra 2005).

The pure ZAJHC and MB-adsorbed ZAJHC were characterized using XRD, FTIR and Raman spectroscopy. The prepared ZAJHC shows an excellent dye removal capacity at initial concentrations <180 mg L⁻¹. The pH of the dye solution strongly affected the chemistry of both the dye molecules and the adsorbent in aqueous solutions. The adsorption process was governed by the mechanism of ion-exchange and it followed Lagergren's first-order kinetics.

This research was supported by the National Natural Science foundation of China (Grant No. 51650410657). ERR thanks IHE-Delft (The Netherlands) for providing staff time support to collaborate with researchers from China.