The Equilibrium, Kinetics, and Thermodynamics Studies of the Sorption of Methylene Blue from Aqueous Solution Using Pulverized Raw Macadamia Nut Shells

In this study, the potential for pulverized raw macadamia nut shell (MNS) for the sequestration of methylene blue from aqueous media was assessed. The sorbent was characterized using scanning electron microscopy for surface morphology, functional group analysis was performed with a Fourier-transform infrared spectrometer (FT-IR), and Brunauer–Emmett–Teller (BET) isotherm was used for surface area elucidation. The effects of contact time, sorbent dosage, particle size, pH, and change in a solution matrix were studied. Equilibrium data were fitted using Temkin, Langmuir, and Freundlich adsorption isotherm models. The sorption kinetics was studied using the Lagergren pseudo-first-order, pseudo-second-order, Elovich, and intraparticle diffusion models. The feasibility of the study was established from the thermodynamic studies. A surface area of 2.763 m2/g was obtained. The equilibrium and kinetics of sorption was best described by the Langmuir and the pseudo-second-order models, respectively. The sorption process was spontaneous (−ΔG0=28.72 − 31.77 kJ/mol) and endothermic in nature (ΔH0=17.45 kJ/mol). The positive value of ΔS0 (0.15 kJ/molK) implies increased randomness of the sorbate molecules at the surface of the sorbent. This study presents sustainable management of wastewater using MNS as a potential low-cost sorbent for dye decontamination from aqueous solution.


Introduction
Freshwater resources are increasingly under threat from pollution due to anthropogenic activities. Although freshwater is unevenly distributed across the globe, it remains a major resource for livelihood and economic growth. e threat to freshwater quality is majorly caused by human activities and lifestyles. e discharge of untreated and partially treated wastewater remains a major point source of surface water pollution [1]. e impact of polluting freshwater includes loss of source of livelihood, undue stress to aquatic organisms, and negative impacts on the environment [2,3].
One of the recalcitrant pollutants to surface water bodies is dyes. e release of small quantity of dyes in wastewater has a far-reaching effect on the aesthetic value of the receiving watershed. Most dyes are synthetically produced and, therefore, not biodegradable [4,5]. Apart from being persistent in the environment, they are also hazardous, causing physiological changes or death to benthic organisms including fish. ey prevent sunlight from penetrating a water body, thereby affecting photosynthetic plants in the aquatic ecosystem [6,7]. eir presence usually leads to a decrease in dissolved oxygen in the aquatic ecosystem. Dyes production are on the increase due to their application in various industrial processes and products which include but are not limited to textile, cosmetics, tanning, pharmaceutical, pulp, and paper industries. Many textile and allied industries unfortunately do not treat their effluents to regulatory standards before releasing the dye-rich effluents into the receiving watersheds [3,8]. Dyes are recalcitrant in nature and very difficult to remove due to their high solubility and low biodegradability.
Although several conventional methods exist for their removal such as coagulation and floatation, ozonation, membrane separation, and commercially available activated carbon, they often suffer setbacks due to their inability to remove certain dyes from the water stream [9][10][11]. Also, the cost of installation, energy consumption and expertise often limit their use. ere is a continuous search for low-cost, eco-friendly, and sustainable materials for dye pollution control. Attention has shifted to the use of agricultural waste materials for the removal of dyes from environmental media. Several plant-based materials have been used which showed promising results; however, the continuous availability of those materials has limited their potential for use [5,12,13].
South Africa is the third largest producer of macadamia in the world. Most of the fruits and oils are often sold in the international market [14]. Macadamia is widely used in the snacks and food industries for baking, making ice cream, and as confectionery. It is also widely used in the cosmetic industry. One of the waste materials for macadamia processing is the nut shells. e nut shells have been used as a source of energy for local bricks industries in South Africa. It has also been used to augment some building materials and as source of fertilizers.
Few authors have reported on the methods of making activated carbon from macadamia nut shell (MNS) [15][16][17]. Also, few reports have investigated the use of activated carbon from MNS for hexachromium removal [16], tetracycline removal [18] phenol removal [19], and reactive dyes [20]. No study, however, has reported the use of the raw natural product for the decontamination of methylene blue. us, this study investigated the impact of change in water chemistry on the sorption of methylene blue onto raw pulverized MNS.
is study is, therefore, aimed at evaluating the potential of South African macadamia nut shell for decontamination of hazardous dye from aqueous solution; the force needed to break the nut shell is also reported.

Adsorbate Preparation.
Methylene blue (MB) was purchased from Fisher Scientific from the United States of America. 1000 mg/L stock solution was prepared following standard methods. Working solutions were made by the dilution method from the prepared stock solution. 0.1 M NaOH and HCl were used for pH adjustment.

Sorbent Preparation.
Waste macadamia shells were collected from a processing farm in Vhembe District, Limpopo Province of South Africa. e hard shells were separated from the valuable nuts which are very rich in essential oils. e shells were washed with deionised water and oven dried (Eco erm, Labotec) for 24 hrs at 105°C (Figure 1). e dried shells were pulverized to fine particles using a pulveriser (Retsch RS 200). Mechanical sieves (King-Test VB 200/300) were used to obtain various fractions of the adsorbent.

Instrumentation.
MB in aqueous solution was analysed using a UV spectrophotometer (Orion Aquamate 7000, ermoscientific) at a predetermined wavelength of 664 nm. e mechanical property of the nut was determined using Computerized Ingstrom Machine. Fourier transform infrared (FT-IR) spectra of the sorbent were obtained using a Perkin Elmer 100 FT-IR spectrophotometer (Waltham, MA, USA) with accessories. e spectra were scanned over the wave number range of 4500 to 400 cm −1 . e surface area and pore width were determined by the N 2 gas Brunauer-Emmett-Teller method of analysis using a Micromeritics Chemisorption ASAP 2020 analyzer (Norcross, GA, USA). A temperature-controlled shaking water bath (Eco-Bath, Labotec) was used to agitate the solutions at specific time and temperature. An Orion pH meter ( ermo Orion VersaStar) was used to measure the pH of the solution. e surface morphology of the sorbent was analysed using scanning electron microscopy (SEM). e sample was crushed, mounted on a stud, carbon coated, and irradiated with a beam of electrons at 20 Kv, and the resulting micrograph was recorded.

Batch Sorption Studies.
Several sorption experimental drivers were investigated to determine their effect of the sorption process. e effects of sorption time were investigated by varying the sorption time between 5 and 270 minutes using three different dosages (1.25, 2.5, and 3.75 g/L) of MNS with an initial MB concentration of 30 mg/L. e samples were placed in a shaking water bath set at 303 K and 250 rpm. Samples were withdrawn from the shaking water bath at predetermined intervals, centrifuged at 250 rpm and analysed for residual MB concentration at 664 nm.
Similarly, the effect of sorbent dosage was performed using 0.25-3.75 g/L of MNS with two initial MB concentrations (30 and 70 mg/L). e solutions were equilibrated for 3 hours at 250 rpm. e experiment was performed at 303, 313, and 323 K. All the samples were then centrifuged and subsequently analysed using a UV-Vis spectrophotometer. A control sample without the sorbent was also ran for quality control. e percentage of MB removed was calculated using the relation in equation (1), and the quantity of MB sorbed was calculated using equation (2): where q e is the amount of adsorbate adsorbed at equilibrium (mg/g), C o and C e were the initial and the equilibrium concentrations in mg/L, respectively, V is the volume in liters of the solution used during the experiment, and M is the mass of the adsorbent in Gram.
e effects of solution pH were elucidated by varying the pH of the dye solution (30 mg/L) between 2 and 11 using three different dosages (0.25, 0.5, and 1.25 g/L). e other experimental procedure was the same as stated for the effects of dosage. e sorbent particle size was varied between 10-124 µm (referred to as <125 µm) and 125-250 µm (referred to as >125 µm), to investigate their impact of the sorption process. e final MB concentration was analysed after sorption. e influence of change in water chemistry was investigated by preparing MB solution with water collected from the Mutale River in Vhembe District, South Africa. e water chemistry of the river water was analysed for basic physicochemical parameters using standard methods. e result obtained was compared to that of deionised water.

Isotherm Study of the Sorption Processes.
e data obtained from the batch sorption studies were fitted into three equilibrium isotherms. e Langmuir, Freundlich, and Temkin adsorption isotherms were used. e Langmuir isotherm often estimates the maximum adsorption capacity corresponding to complete monolayer coverage on the adsorbent surface. e linearized equation is represented as follows [21]: where C e is the equilibrium concentration of MB (mg/L), q e is the quantity of MB dye adsorbed at equilibrium (mg/g), q max is the maximum amount of MB sorbed (mg/g), and b is the adsorption constant (L/mg). A plot of 1/C e versus 1/q e will give a straight line, and q max and b can be obtained from the intercept and slope of the plot, respectively. e Freundlich model assumes a heterogeneous sorption surface with unequal available sites with different energies of sorption. Its linearized equation is given as follows [22]: e plot of log q e versus log C e provides the intercept K f and the slope 1/n f , where q e is the amount of adsorbate adsorbed at equilibrium (mg/g), C e is the equilibrium concentration of adsorbate (mg/L), K f is the Freundlich constant, and n f is the Freundlich exponent. e Temkin isotherm model assumes that the heat of adsorption of all the molecules on the adsorbent surface layer would decrease linearly with coverage due to adsorbate-adsorbate interactions, and this fall in the heat of adsorption is not logarithmic as implied in the Freundlich equation [23]. e linearized form of the Temkin isotherm is represented as where q e is the amount of adsorbate adsorbed at equilibrium (mg/g), B is the constant related to the heat capacity (L/mg), R is the universal gas constant (8.314 J/mol·K), T is the absolute temperature (K), K T is the equilibrium binding constant (L/mg), and C e is the equilibrium concentration of adsorbate (mg/L). A plot of q e against log C e is linear, and the constant B (L/g) and K T will be determined from the slope and intercept of the plots, respectively.

ermodynamic Study of the Sorption Processes.
In order to determine the thermodynamic feasibility of the sorption process, the standard Gibbs free energy change (ΔG 0 ), the standard entropy change (ΔS 0 ), and the standard enthalpy change (ΔH 0 ) were calculated. ΔG 0 was determined from the relation below: where K 0 is the equilibrium constant (L/mol) determined from the Langmuir constant b. ΔS 0 and ΔH 0 were estimated using the van't Hoff equation [24]: where T is the absolute temperature (K) and R is the gas constant (8.314 J·mol −1 ·K −1 ). e plot of InK 0 as a function of 1/T should give a linear relationship with the slope of ΔH 0 /R and an intercept of ΔS 0 /R.

Kinetic Study of the Adsorption Processes.
e kinetics of sorption describes the rate of MB uptake on MNS, and this rate controls the equilibrium time. e pseudo-first-order, pseudo-second-order, Elovich, and intraparticle diffusion models were used in this study. e linearized forms of these models are given in equations (8)-(11) [25][26][27][28]. e Lagergren pseudo-first-order kinetic model equation is given by Log q e − q t � Log q e − k 1 2.303 t.
Pseudo-second-order kinetic equation is given by e Elovich model is given by e Webber Morris intraparticle model is given by where q e is quantity of MB sorbed at equilibrium, qt is the quantity of MB sorbed at time t in minutes, K 1 is pseudofirst-order rate constant, K 2 is pseudo-second-order rate constant, a is the initial adsorption rate (mg/gmin), β is the desorption constant (g/mg), and K p (mg g −1 min −1/2 ) is the intraparticle diffusion rate constant.

Results and Discussion
e FT-IR peaks recorded are presented in Figure 2. e hydroxyl, amine, and carbonyl groups were predominant in the sorbent (Table 1). Figure 3 shows the SEM micrograph of the sorbent; surface particle agglomeration was observed with some slight voids capable of adsorbing the dye molecules.
e Brunauer-Emmett-Teller (BET) surface area of 2.763 m 2 /g ( Table 2) was determined for the sorbent material.
is is higher than the surface areas reported for Dicerocaryum eriocarpum leaves (1.85 m 2 /g), activated carbon made from macadamia nut shell (1.083 m 2 /g), and activated carbon from maize tassels (2.52 m 2 /g) [19,29,30], although other sorbent materials with higher surface area have been reported in the literature [18,20]. e surface area value will favour strong affinity between a sorbate and sorbent. e sorbent is a macroporous material based on average pore diameter ( Table 2). Due to this pore dimension, low energy is required for sorption of the sorbate by the sorbent [31,32]. e load at break was 80.4 N while compressive stress at break was 0.15 MPa (Figure 4). Based on the stress-strain analysis of a nut under diametrical compression, material behaves in a brittle manner.

Effects of Time.
Sorption time usually plays a great role in the uptake of the sorbate by the sorbent. In this study, there was initially an increase in MB uptake as time progresses which can be linked to the available external surface on the sorbent. At 60 minutes of sorption time, >65% removal of initial MB concentration was recorded for a     Figure 5). is increase continued until equilibrium was attained at 150 minutes (>95% uptake). After this point, no significant uptake of MB was recorded. is is due to the saturation of the sorbent surface. Similar results have been recorded in several other plantbased sorbents for MB sequestration [33,34].

Effects of Solution pH.
Solution pH plays a great role in sorption reactions because it often determines the ionisation states of the functional groups present in a sorbent. In this study, higher sorption was recorded as the pH of the solution changes from acidic to more alkaline. e pH of highest sorption was 11 ( Figure 6). Our previous study using Marula seed husks for uptake of MB showed high uptake at the pH of 6; however, this is slightly different with the findings of this study [6]. Similarly, Singh et al. [4] also showed that alkaline pH favored the sorption of MB onto Ginkgo biloba leaves, but 98% MB removal was recorded at the pH of 4 which negligibly increased up to the pH of 11.

Effects of Sorbent Dosage.
A general increase in MB uptake was recorded as the sorbent dosage increased from 0.5 to 4.0 g/L (Figure 7). is is due to the increase in the corresponding sorption sites present on the sorbent. Similar trends have been reported in the literature [17,29,35].

Effects of Particle
Size. e smaller particle size (<125 um) of the sorbent recorded higher sorption than the bigger size of >125 um (Figure 8).
is has been supported by several scholars that the smaller the particle size of the sorbent, the more the surface area. At higher sorbent dosage, the impact of particle size became negligible [6].

Influence of Change in Water Chemistry.
e physicochemical water quality parameters of the Mutale River used in this study recorded the pH of 6.9, conductivity of 24.0 (μS/cm), and turbidity of 14 NTU. e levels of major cations are sodium (4.66 mg/L), potassium (0.35 mg/L),  e sorption was favored in natural water than in deionised water (Figure 9). is shows that the sorbent can be used on real wastewater or polluted water.   e equilibrium data fitted into the three models are already described in equations (3)-(5). e linearized coefficient (R 2 ) was used to determine the isotherm that best described the sorption process. Linear plots were obtained as described earlier, and R 2 was determined from the plots (Figures 10-12). Maximum sorption capacity as well as the isotherm constant was derived from the slopes and intercepts of the respective plots (Table 3). In this study, the Langmuir equilibrium model best described the sorption process with an R 2 factor of 0.93 at 303 K. e R 2 factor for other models ranges from 0.73 to 0.88.

ermodynamics of the Sorption Process.
is was estimated using the relation in equations (6) and (7), and negative values for ΔG were computed which imply that the sorption is spontaneous and thermodynamically feasible (Table 4). e values of ΔS and ΔH were obtained from the van't Hoff plot (Figure 13). e change in enthalpy was positive, which indicates an endothermic sorption process, while the change in entropy was also positive, implying an increased randomness of the sorbate molecules on the surface of the sorbent. Similar results have been reported by several scholars on MB sorption onto plant-based materials.

Kinetic Models.
e kinetics of the sorption was studied using four models. e linearized forms of the models were plotted, and the results are presented in Figures 14-17. e sorption can be described by most of the models used because their linear regression line was greater than 0.7; the pseudo-second-order kinetic model best described the kinetics of the sorption. Results obtained from the pseudo-second-order plot suggest that sorption proceeds by chemisorption which is the formation of bonds between the sorbate and the sorbent. is was further supported from the equilibrium model as the Langmuir model assumes a chemisorption process rather than a physisorption process. e kinetic constants obtained from the plots are presented in Table 5. e intraparticle model by Weber and Morris is more of a mechanistic-based model which is used to ascertain the mechanism of the rate-determining step. If a linear plot of qt versus t0.5 passes through the origin, it implies that the intraparticle diffusion is the sole rate-determining step and if not as obtained in this study, it suggests that the mechanism of the sorption is multilinear and the rate-determining step is controlled by both film diffusion and intraparticle diffusion [29].

Data Availability
Majority of the data used in this study are included in the article. Other data can be made available upon request from the corresponding author.

Conflicts of Interest
e authors declare no conflicts of interest regarding the publication of this paper.