Abstract

In the quest for a sustainable environment and clean water resources, the efficacy of activated carbons synthesized from Garcinia cola nut shells impregnated with KOH (CBK1/1) and ZnCl2 (CBZ1/1) for the adsorption of indigo carmine (IC) dye was studied using the batch technique. The prepared activated carbons were characterized using iodine number, elemental analysis, scanning electron microscopy (SEM), FTIR spectroscopy, powder X-ray diffraction (XRD), TGA/DTA, Boehm titration, and pH at point of zero charge. The elemental analysis showed a high percentage of carbon in both activated carbons (ACs). FTIR and Boehm titration analysis indicated the presence of several functional groups on the surfaces of both ACs which could influence the adsorption of IC. The primary adsorption mechanisms involved electrostatic interaction, hydrogen bonds formation, and ππ interactions. Maximum adsorption capacity values obtained using the Fritz–Schlunder III three-parameter model were 19.019 mg·g−1 and 18.299 mg·g−1 for CBK1/1 and CBZ1/1, respectively. The Fritz–Schlunder model exponent mFS of value less than 1 showed that the adsorption of IC by the ACs occurred on heterogeneous surfaces. Positive values of ∆Q obtained by the linear and nonlinear forms of the Temkin model indicate the exothermic character of the adsorption process.

1. Introduction

Nowadays, several industries discharge their colored effluents containing synthetic dyes into aquatic environments [1]. These dyes are considered as chief contaminants in water since they pose serious threats to human beings and aquatic organisms [2]. Wastewater is often discharged into rivers and lakes or infiltrates into aquifers, where it can affect the quality of freshwater supplies. Globally, over 80% of the wastewater sent into the environment without treatment causes negative consequences for the marine environment [3]. Water pollution by dyes is becoming a major environmental problem since colouring agents cause significant environmental damage to both human health and aquatic organisms. These dyes are highly toxic, carcinogenic, mutagenic, teratogenic, and stable during aerobic degradation [4]. The discharges of industrial wastewater containing very low concentrations of dyes reduce light penetration through the water surface, precluding photosynthesis of the aqueous flora.

Indigo carmine dye is used in the textile industry for dyeing clothes and in cosmetics, plastics, and paper mills, while in medicine, it is used as an aid to diagnosis [5]. Indigo carmine is considered in the class of the highly toxic indigoid dyes. Its contact with the skin can cause irritation, while with the eye, it can cause both irritation and permanent injury to the cornea and conjunctiva [6].

Consumption of this dye at high concentration has also been shown to be fatal, as it is carcinogenic by nature and can cause reproductive, developmental, neurological, and acute toxicity [6]. It has also been reported to cause mild-to-severe hypertension and cardiovascular and respiratory effects [6]. According to the World Health Organization standards, the limit concentration of indigo carmine is 0.005 mg·L−1 [7]. Thus, indigo carmine is not only a noxious substance to aquatic life but also to us humans. Therefore, the elimination of this type of dye in wastewater or at the exit of industries becomes a major concern.

Several techniques have been discussed in the literature for the removal of dyes from water and wastewater. Among these techniques, adsorption is considered the most economically favorable among those available, such as membrane separation, oxidation, and irradiation, because of its high removal efficiency, low operation cost, and its ability to separate a wide range of contaminants from industrial effluents [8].

Amongst the adsorbents used in the adsorption processes, one of the most widespread for the removal of contaminants from water is activated carbon (AC) [9,10] which, due to its pore structure and various functional groups on its surface, provides exceptional adsorption capacity. Different activating carbons can be manufactured by exploiting a wide variety of carbonaceous precursors such as coal, petroleum, peat, wood, and agricultural wastes [11], resulting in the production of a variety of adsorbents with different characteristics and qualities.

The use of biomass waste for the preparation of porous activated carbon has increased considerably in recent years. Some of the residual biomasses used for the manufacture of activated carbons are coffee waste [9], tucumã seeds [10], Ayous sawdust and Cucurbitaceae peelings [12], rice and coffee husks [13], olives stones and cotton cake [14], Vitellaria paradoxa [15], and rice hull and tamarind (T. indica) seeds [16].

This work had a double interest: valorizing the residues coming from a Cameroonian biomass to prepare activated carbons using chemical treatment techniques and testing their performance in the water treatment. In this work, the residual biomass, Garcinia cola nut shells from Cameroon, was proposed as a raw material for the development of activated carbons by chemical activation using KOH and ZnCl2, in order to recycle these wastes that are currently generated in large quantities and are of little use in Cameroon [17].

In this research, the feasibility of producing activated carbons suitable for the efficient removal of indigo carmine was demonstrated.

Specifically, this research was subdivided into three parts:(i)Production of activated carbons from Garcinia cola nut shells by chemical activation with KOH and ZnCl2(ii)Application of the activated carbons as adsorbents for the removal of indigo carmine in aqueous solution and evaluation of their efficacy(iii)Demonstration of the reusability of each adsorbent by studying desorption

2. Materials and Methods

2.1. Preparation of Activated Carbons

ACs were produced by the following procedure: first, 200 g of particles of dried Garcinia cola nut shell (CB) with diameters varying between 250 and 1000 μm was blended in a 1 : 1 weight ratio with the activating agent KOH or ZnCl2. Each resulting paste was dried and placed in a porcelain crucible with lid and introduced into an electric furnace of mark ISUNU and heated from the ambient temperature up to 400°C, using a heating rate of 5°C·min−1 in absence of CO2, H2O, O2, and N2 gases. The temperature of the furnace was kept fixed at the maximum temperature of 400°C for 1 hour. It was then allowed to cool to room temperature, and the activated carbon obtained was washed with distilled water until the excess activating agent was removed. The samples were coded as CBK1/1 and CBZ1/1, respectively, according to KOH and ZnCl2 used as the activating agent.

2.2. Characterization

Activated carbons were characterized to determine the surface functional groups by Fourier-transform infrared (FTIR) spectroscopy using a Nicolet iS5 FTIR Spectrometer. The physical and chemical characteristics of both activated carbons were found out by ASTM methods. Thermogravimetric analysis of both activated carbons was performed on STA 409 CD thermos-balance with a heating rate of 10°C·min−1 in a nitrogen flow and temperature range of 50–1500°C. Elemental analysis of both activated carbons was determined using CHNS elemental analyses from the HEKAtech CHNS Analyzer. Scanning electron images of the two activated carbons were obtained using a scanning electron microscope, furnished by sputter coater analysis. Crystalline structure of the activated carbons was analyzed by X-ray diffraction (XRD) analysis using a STOE Stadi-p X-ray powder diffractometer (STOE & Cie GmbH, Darmstadt, Germany). Figure 1 shows the preparation and characterization steps for the different activated carbons.

2.3. Adsorbate

Stock solution of indigo carmine at 500 mg·L−1 was prepared by dissolving a previously calculated mass of 123.75 mg of the latter in a 250 mL volumetric flask containing 100 mL of distilled water, homogenize until the pollutant is completely dissolved, and then make up to the mark and store in a dark place. The working indigo carmine solution was prepared by diluting the stock solution with distilled water. Figure 2 shows the 3D structure of indigo carmine.

2.4. Adsorption and Desorption Studies

Batch adsorption studies were carried out at room temperature by mechanical agitation at a constant speed of 200 rpm to deduce the optimum pH, contact time, adsorbent mass, initial indigo carmine concentration, and the effect of ionic strength. For each run, 20 mL of indigo carmine dye solution of known initial concentration varying between 30 and 100 mg L−1 was treated with 100 mg each of the activated carbons, CBK1/1 and CBZ1/1, at optimum pH. The pH value of each solution was adjusted by addition of 0.1 M solution of HCl or NaOH. After agitation, the solution was filtered using the Whatman no.1 filter paper, and the filtrate was analyzed to obtain the residual concentration of indigo carmine using a UV/Vis spectrophotometer (Jenway, model 6715) at a wavelength (λ) of 610 nm. Similar measurements were carried out by varying the pH of the solution (2–8), contact time (5–120 min), adsorbent mass (50–400 mg), and ionic strength varying NaCl concentration (0.01–0.06 mol·L−1). The amount of dye adsorbed (Qe) and the percentage removal (%ads) of dye were calculated using the following expressions:where C0 is the initial concentration of the dye, Ce is the concentration of the dye at the equilibrium time, V is the volume of the adsorbate solution, and m is the mass of the adsorbent.

For the desorption studies, 400 mg of each adsorbent was mixed in a flask with 100 mL of indigo carmine dye solution at an initial concentration of 50 mg·L−1. Mixtures were stirred at a constant speed of 200 rpm for 80 min. After stirring, suspensions were filtered and filtrates were analyzed by measuring absorbance using a UV-Visible spectrophotometer. Indigo carmine loaded with each adsorbent was collected after filtration and was dried in an oven at 105°C, and 50 mg of each of these materials was contacted with 10 mL of desorption solutions of 4 × 10−2 mol·L−1 of NaOH or H2SO4 and 10 mL of H2O. The mixtures were stirred at a constant speed of 200 rpm for 80 min. After stirring, suspensions were filtered and filtrates were analyzed by measuring absorbance using a UV-Visible spectrophotometer. Desorption percentages (%des) were calculated by the following equation [18, 19]:where Cf (mg·L−1) is the initial concentration of dyes loaded with each adsorbent and Cr (mg·L−1) is the final concentration of dyes loaded with each adsorbent.

2.5. Equilibrium and Kinetic Studies

In this work, four two-parameter adsorption isotherm models (Langmuir, Freundlich, Temkin, and Halsey), one three-parameter model (Fritz–Schlunder III), and five kinetic models (pseudo-first order, pseudo-second order, Avrami kinetics, intraparticle diffusion, and Boyd) were tested. In the aim to study the mechanisms of adsorption process and the equilibrium relationship between adsorbents and adsorbate, linear and nonlinear models were modelled using Microsoft Excel 2013 and plotted using OriginPro 9, 64 bit.

2.6. Error Functions

In order to determine the best fitting of nonlinear models, four different error functions were examined by minimizing the respective error functions across the concentration and time range studied. “SOLVER ADD-IN” of Microsoft’s spread sheet was used. Expressions of the error functions are given as follows [20]:where Qe,exp and Qe,cal (mg·g−1) are the equilibrium capacity of adsorption obtained from the experiment and by calculating from the model, respectively, and N is the number of data points.

3. Results and Discussion

3.1. Characterization of Adsorbent
3.1.1. Physical and Chemical Characteristics of ACs and Elemental Analysis

Physical and chemical properties of ACs CBK1/1 and CBZ1/1 activating are reported in Table 1. pHpzc values show that surfaces of the different ACs are dominated by the basic functional group. Bulk densities of the different ACs shown in Table 1 are higher than the limit of 0.25 g·cm−3 according to American Water Works Association. Higher density gives a higher volume activity and normally indicates a better AC quality. Bulk densities of different ACs provide information on the microcrystalline structure and strong interlinking bonds. Moisture content values of different ACs obtained as presented in Table 1 are within the norm, since the traditional value for moisture content ranges from 1 to 5% by mass. Iodine number of the ACs obtained in this work is higher than limit (500 mg·g−1), thus showing the good quality of the ACs obtained, reflecting the presence of micropores. Methylene blue number obtained and presented in Table 1 reflects the presence of mesopores on the surface of the ACs.

In Table 1, the quantity of basic groups is higher than the acidic groups on the surface of CBK1/1 and CBZ1/1 activated carbons, indicating that KOH and ZnCl2 as activating agents for production of activated carbons by Garcinia cola nut shells generate carbons with higher amount of basic functional groups on their surface. The ratios of the sum of the basic groups divided by the sum of the acid groups are 1.269 (for CBK1/1) and 1.253 (for CBZ1/1), indicating that CBK1/1 would be more basic than CBZ1/1, as confirmed by the total basicity of CBK1/1 (3.08 meq·g−1) and CBZ1/1 (3.045 meq·g−1). These results follow the pHpzc values, which imply that, in AC, having a higher ratio of total basic groups divided by total acid group leads to higher pHpzc values. Taking into account the results of the Boehm titration and the pHpzc, it can be stated that these results are in total agreement.

Elemental analyses of CBK1/1 and CBZ1/1 are presented in Table 1. The prepared ACs have high percentages of carbon, which in itself is a fact that justifies the use of Garcinia cola nut shells as a carbon source. The percentages of nitrogen and oxygen are consistent with the amount of functional groups present on activated carbons CBK1/1 and CBZ1/1.

3.1.2. Fourier-Transform Infrared (FTIR) Spectroscopy

Figure 3 shows that the FTIR spectra of CBK1/1 and CBZ1/1 are very similar to each other and different from that of the raw material, indicating that the activating agent did not influence the surface chemical group of the AC. The FTIR identified chemical groups on the surface of both ACs such as the broad peak around 3349 cm−1 for CBK1/1 and CBZ1/1 which are related to OH group stretching [9]. The band obtained approximately at 1587 cm−1 in both ACs indicates the presence of C = O stretching vibration of lactone and carbonyl groups [17]. This band can also be attributed to the C = C stretching vibration of alkene or aromatic ring [17]. The band around 1374 cm−1 corresponds to the C–H bending vibration or C–N stretching vibration of amines or amides of nitroaromatic compounds (-NO2).

3.1.3. X-Ray Diffraction Analysis (XRD)

Figure 4 shows the XRD of ACs. In this figure, the absence of a sharp peak reveals that both ACs prepared from Garcinia cola nut shells are mainly amorphous structure, which is an advantageous property for well-defined porous adsorbents. However, the small sharp peak presented by the X-ray graph of CBZ1/1 (2θ = 28°) indicates very low crystallinity may be due to the presence of zinc oxide and zinc carbide [17].

3.1.4. Thermogravimetric Analysis (TGA/DTA)

TGA is performed to investigate the thermal stability of AC. The weight loss curves for CBK1/1 and CBZ1/1 as can be seen on the TG curves for both ACs are very similar (Figure 5).

In Figure 5, the TGA/DTA curves of both ACs suggest three typical distinct weight losses. The first weight loss of 2.016% for CBK1/1 and 4.314% for CBZ1/1 at temperature below 200°C corresponds to the elimination of the water content (physisorption phenomenon) in the ACs, which means that there is still physisorbed water remaining in the micropores and mesopores of the various activated carbons. The second weight loss of 20.413% for CBK1/1 and 26.712% for CBZ1/1 within the temperature range of 200°C–900°C may be due to the predominant decomposition of the surface chemical groups of the different ACs and the formation of H2O, CO, and CO2 during the calcination/activation process. In this temperature range, the decomposition of the lactone functions usually takes place between 350 and 600°C.

The third weight loss is secondary degassing which is observed above 900°C, and the resulting intermediate product linked to the carbon skeleton of the ACs involves further decomposition through the formation of residual products such as tar and oxides which are subsequently removed. Of the two ACs, the less thermally stable is CBZ1/1, as its total mass loss is higher (37.04%), and the more thermally stable is CBK1/1 with a lower total mass loss of 30.546%. According to the DTA curve, the thermal decomposition of the two activated carbons is dominated by an endothermic process (0°C–1250°C).

3.1.5. Scanning Electron Microscopic Images

The textural surface properties of CBK1/1 and CBZ1/1 were investigated by SEM (Figure 6). According to this figure, SEM images showing the prepared ACs CBK1/1 and CBZ1/1, respectively, illustrate an irregular and heterogeneous surface morphology with a porous structure developed and fragmented in different sizes [17]. However, it can be seen on the micrographs that the external surface of the ACs shows cracks and crevices [17]. These pores result not only from the evaporation of the chemical reagents (KOH and ZnCl2) during carbonization, leaving empty spaces, but also from the degradation of the lignocellulosic material during carbonization and subsequent removal of volatile materials leaving samples with pores [17].

3.2. Batch Studies
3.2.1. Effect of pH

From Figure 7, the results show that maximum adsorption capacity is obtained for pH 2 and adsorption is favoured for pH < pHpzc. We observe that the adsorption capacity decreases with increasing pH from 8.354 to 3.159 mg·g−1 and from 7.960 to 2.876 mg·g−1 for ACs CBK1/1 and CBZ1/1, respectively. It is generally known that due to the abundant presence of H+ ions, the anionic dyes such as IC are preferentially adsorbed by the adsorbent at lower pH [21]. At that particular pH (=2) (pH < pHpzc), protons are advantageously available for protonation of the AC surface, which increases the electrostatic interaction between the now positively charged adsorbent sites and the negatively charged sites or nucleophilic sites (−SO3-) of indigo carmine. Similar results have also been observed by several authors for the adsorption of IC [22, 23]. At the same time, IC is an anionic dye with two sulphonic acid groups and two aromatic rings; this is probably the other reason for its high removal efficiency at the low pH. A similar trend was reported by Zhao et al. [24] where they use amino acid-doped polyaniline nanotubes as adsorbent for wastewater treatment. However, increased pH (pH > pHpzc) of the adsorbate makes it more soluble and the negative charges on the surface of the ACs increase the electrostatic repulsion force between the adsorbates (sulphonic functional group of the pollutants) and the –OH groups on the surface of the ACs due to the increase in hydroxyd ions. The electrostatic repulsion force between indigo carmine and the adsorbent promotes the decrease in adsorption capacity. A similar result was obtained from the adsorption of thymol blue onto powdered activated carbons from Garcinia cola nut shells impregnated with H3PO4 and KOH [25].

3.2.2. Effect of Contact Time

Figure 8 shows the evolution of adsorption capacity of indigo carmine as a function of the contact time at room temperature. Rapid adsorption of IC takes place in the first 5 minutes for the two ACs; this rapid phase may be attributed to rapid dye attachment on the ACs surface due to either the large amount of surface area available or to the availability of more adsorption vacant sites at the initial stage. Thereafter, the rate of adsorption deceased gradually with the progress of adsorption within 5–80 minutes for the two ACs and reached equilibrium in about 80 minutes. The adsorption capacities at these equilibrium points are 8.60593 mg·g−1 CBK1/1 and 8.03346 mg·g−1 CBZ1/1. This can be explained by the fact that the available sites for adsorption have been greatly reduced by the indigo carmine molecules already attached there and decrease in the total adsorbent surface area and less available binding sites. The quantities adsorbed can be attributed to the ππ interactions, electrostatics attractions, and formation of hydrogen bonding. In the third phase, there is no significant change in the quantity of dye adsorbed after the equilibrium time due to the saturation of the adsorption active sites by the dye molecules [26].

3.2.3. Effect of Adsorbent Mass

Effect of adsorbent mass on the removal of indigo carmine was studied, and the results of this study are shown in Figure 9. In this figure, the percentage removal of indigo carmine by CBK1/1 and CBZ1/1 increased as the adsorbent dosage increased. The removal percentage of indigo carmine increased from 76.8 to 97.7% and 75.1 to 92.7% for CBK1/1 and CBZ1/1, respectively. This is because the increase in the mass of the adsorbent leads to the increase in the adsorption sites and the contact area of each adsorbent becomes free and available for indigo carmine fixation during the adsorption process. This increase in the mass of each adsorbent further promotes the increase in their adsorption percentage. A similar trend was reported in Hameed et al. [27] for the adsorption of chromotrope dye onto activated carbons obtained from the seeds of various plants.

3.2.4. Effect of Initial Indigo Carmine Concentration

The effect of initial concentration on the adsorption of indigo carmine was investigated of 30–100 mg·L−1. Figure 10 shows the quantity of indigo carmine adsorbed at equilibrium increases with the concentration from 5.814 to 11.349 mg·g−1 and 5.756 to 10.823 mg·g−1, respectively, from CBK1/1 and CBZ1/1. This result can be explained by the ππ interactions, electrostatic attraction, and formation of hydrogen bonding between the dyes and the surface functional groups of activated carbons on the one hand and the difference in the pore size of these carbons on the other hand. This can also be explained by the fact that increasing the initial concentration of indigo carmine would increase the driving force of mass transfer and, consequently, the speed at which indigo carmine molecules pass from the solution to the surface of the ACs [28]. In this figure, we observe a saturation of the adsorption sites which is progressively increasing until reaching a plateau caused by the formation of monolayers on the surface of each adsorbent. The isotherm forms of CBK1/1 and CBZ1/1 have been classified as type L which can be better explained by the Langmuir, Freundlich, and Temkin models that show a relatively high affinity between the adsorbate and adsorbent.

3.2.5. Effect of Ionic Strength

The effect of ionic strength on adsorption of indigo carmine onto CBK1/1 and CBZ1/1 was studied in the NaCl solutions with concentrations ranging from 0.01 to 0.06 mol·L−1, and the results are illustrated in Figure 11. When the ionic strength increased, the adsorption capacity of indigo carmine decreased (from 8.102 to 7.860 mg·g−1 and 7.928 to 7.708 mg·g−1) for CBK1/1 and CBZ1/1, respectively. This result can be explained by the fact that the surface of the ACs becomes difficult to access by indigo carmine when the amount of NaCl salt in solution increases. Consequently, the adsorption capacity of indigo carmine decreases. This result is in agreement with the literature, which reports that when the electrostatic attraction forces between the surface of an adsorbent and the ions of an adsorbate are attractive, the increase in ionic strength thus decreases the adsorption capacity [29].

3.3. Adsorption Kinetics

In this work, the data from the kinetic experiments were analyzed using the different linear and nonlinear kinetic models presented in Table 2.

The linear and nonlinear regression graphs of adsorption kinetics for indigo carmine adsorption onto both ACs are shown in Figures 12 and 13, respectively, and the corresponding constants are summarized in Tables 3 and 4.

According to Table 3, the data linear kinetic model indicates that the pseudo-second order model is perfectly appropriate to describe the adsorption kinetics of indigo carmine and on both ACs with a high correlation coefficient (R2 ≥ 0.9993). Moreover, it can be observed that the experimental Qe,exp values obtained during the equilibrium adsorption process are close to the Qe,cal values calculated from the pseudo-second order model, which indicates that the adsorption of indigo carmine on both ACs is a process dominated by the interaction ππ [33]. The values of the initial rate of reaction and the reaction half time determined from pseudo-second order model confirm that the accumulation kinetics of indigo carmine is faster at the surface of AC CBK1/1 than at the surface of CBZ1/1. Figure 12(d) shows that the plot of Qt = f (t1/2) is not a straight line through the origin of the marker. This indicates that the intraparticle diffusion is not the limiting step or the only rate control step in the adsorption process of indigo carmine on ACs. In this case, other kinetic models can also control the adsorption rate to describe other adsorption mechanisms.

As shown in Figure 12(d), the adsorption of indigo carmine on ACs forms two straight line segments over the entire time interval, meaning that this adsorption process can be controlled by at least two mechanisms occurring simultaneously [33]. This can be attributed to the difference in mass transfer rate between the initial stage and the final stage in ACs. It can be said that at least two types of pores, namely, micropores and mesopores, are involved in the adsorption of indigo carmine. The analysis of Table 4 shows that the first step is faster than the second since Kid1 > Kid2. However, the values of boundary layer thickness C for each linear portion not approaching zero (C2 > C1) corresponding to the increase in boundary layer thickness indicate that intraparticle diffusion is not the only step in controlling the adsorption rate in all stages [34].

Figure 12(e) shows that the Boyd model lines do not pass through the origin of the graph, meaning that external mass transfer primarily governs the rate controlling step at the initial stages [18, 35]. In Table 4, the low values of the correlation coefficient obtained do not make it possible in this case to validate Boyd’s model.

Based on the nonlinear regression data of the different models shown in Table 3 and on the low values obtained from SCE and other error functions, it can be concluded that the retention kinetics of indigo carmine are best described by the nonlinear Avrami model. The values of the Avrami exponent nAv < 1 show that the adsorption mechanism follows a multiple kinetic order during the adsorption process of indigo carmine on the two ACs. Thus, in this work, the pseudo-first order shows a multilayer adsorption on the surface of the adsorbent based on a weak interaction between the adsorbate and the adsorbent like the Van der Waals forces and the pseudo-second order model allows us to say that the interaction ππ [33] also intervenes in this adsorption process of indigo carmine on the two ACs, thus showing an adsorption of chemical nature.

3.4. Adsorption Isotherms

Adsorption isotherms can contribute to elucidating adsorption mechanisms. In this work, data from the isotherm experiments were analyzed using the different linear and nonlinear kinetic models presented in Table 5.

The linear and nonlinear regression graphs of adsorption isotherms for indigo carmine adsorption onto both ACs are shown in Figures 14 and 15, respectively, and the corresponding constants are summarized in Tables 6 and 7.

The linear regression and separation factor graphs of adsorption isotherms obtained for indigo carmine adsorption onto both ACs are shown in Figure 14. The corresponding constants obtained from the linear regressions are presented in Table 6. According to this figure, the values of correlation coefficient (R2) in the Langmuir model were closer to unity showing fitness of model on the adsorption experiment. This result indicates that adsorption of indigo carmine could take place at homogeneous adsorption sites and that monolayer adsorption occurs on the surface of the adsorbent [41]. The values of the separation factor falling between 0 and 1 (Figure 14(e)) show a favorable adsorption of IC by the two ACs.

Parameters calculated for Freundlich and Halsey isotherms by employing their linear and nonlinear forms are given in Tables 6 and 7. The results show that the Freundlich and Halsey isotherms have a very good coefficient of determination (r2 > 0.96) and very low values of the error functions (χ2, RMSE, and SCE), thus better describing the phenomenon of indigo carmine adsorption. This assumes that the adsorbent surface is heterogeneous with a non-uniform distribution of the heat of adsorption over this surface showing multilayer adsorption [42]. The values of 1/n evaluated from the Freundlich model are less than unity, meaning that adsorption is favorable for dye [43] and physical process is favorable [44]. This confirms the heterogeneity of the adsorbents as found by the SEM.

Maximum adsorption capacities were obtained by the nonlinear Fritz–Schlunder III model and are 19.019 mg·g−1 and 18.299 mg·g−1 for CBK1/1 and CBZ1/1, respectively. The Fritz–Schlunder model exponent mFS value obtained by the nonlinear Fritz–Schlunder III isotherm of less than 1 means that the adsorption of indigo carmine by the two ACs cannot be reduced to the Langmuir isotherm showing that adsorption takes place on heterogeneous surfaces.

3.5. Possible Adsorption Mechanisms

Surface characteristics of adsorbents play an important role in adsorption processes [49]. Thus, the interaction of any molecule in the adsorption process is strongly influenced by the presence of functional groups on the surface of the adsorbent and also by the pH of the adsorbate solution [50]. The probable mechanism of adsorption of the dye on carbon is presented in Figure 16.

The equilibrium study showed the possibility of multilayer adsorption, in which the adsorbate-adsorbent interactions are carried out up to the saturation of the monolayer. Furthermore, based on the chemical structure of the adsorbent surface, the dyes, and considering the condition of the pH of the medium below the pH of zero charge point, which leads to a positively charged adsorbent surface, we can say that, in the adsorption process, there occurs the electrostatic interaction (attraction) between the protonated groups under acidic conditions (e.g., −OH2+ and −COOH2+) and anionic sulphonate groups (−SO3) of the dye on the one hand. If electrostatic interaction dominates the adsorption process of IC in aqueous solution, the ACs will exhibit lower adsorption capacity due to electrostatic repulsion between the adsorbent and the adsorbate at pH 2 because the functional groups (−OH2+ and −COOH2+) of the ACs (pHpzc = 8.324 for CBK1/1; 7.589 for CBZ1/1) and indigo carmine (pKa = 12.2) are positively charged in such an acidic environment. This does not correspond to the experimental results obtained at pH = 2; we have a high adsorption capacity. On the other hand, intermolecular hydrogen bonding of dye molecules generally leads to multilayer adsorption [24]. Adsorption of indigo carmine follows the multilayer adsorption pattern heterogeneous surfaces until saturation of the monolayer. This indicates that hydrogen bonding could be the dominant force for adsorption accompanied by the ππ interaction between the aromatic rings of the dye and the ACs, which is also considered responsible for the adsorption process [51]. Electrostatic attraction also plays an important role in indigo carmine adsorption.

3.6. Desorption

In this study, the desorbing agents such as H2O, NaOH, and H2SO4 4 × 10−2 mol·L−1 were used to regenerate both ACs. The results presented in Figure 17 show that the higher desorption percentage, 61.259% for IC CBK1/1 and 55.08% for IC CBZ1/1, is obtained in NaOH solution. This result can be explained by the phenomenon of anionic exchange between OH of NaOH solution and the activated carbons charged by indigo carmine. However, the low percentage of desorption obtained with CBZ1/1 compared to CBK1/1 is due to the strong bond formed between the IC and the surface of activated carbon CBZ1/1 [18, 52]. A similar result was obtained by Ngaha et al. [19] in the indigo carmine and 2,6-dichlorophenolindophenol removal using cetyltrimethylammonium bromide-modified palm oil fiber.

4. Conclusion

In this study, activated carbons were prepared from Garcinia cola nut shell by chemical activation using KOH and ZnCl2. They were used for removal of indigo carmine from aqueous solution by the batch process. SEM and N2 adsorption-desorption analysis showed that the surfaces of the ACs were heterogeneous and contained with micropores and mesopores. The adsorption process of IC onto the adsorbents was strongly dependent on the pH of the dye solution, contact time, adsorbent dosage, initial concentration of IC, and the presence of ionic strength. Linear and nonlinear kinetic and equilibrium isotherm models were applied to the adsorption of IC onto activated carbons obtained. Nonlinear kinetic models better describe the mechanism of adsorption process than linear kinetic models which show that the adsorption mechanism follows a multiple kinetic order during the adsorption process of indigo carmine on the ACs according to Avrami exponent nAv lower than 1. Nature and the mechanism of the adsorption processes were studied by employing the intraparticle diffusion model and Boyd plot. External mass transfer primarily governs the rate controlling step at the initial stages for the adsorption of the IC. Linear and nonlinear equilibrium isotherm models show that IC adsorption is done on heterogeneous surface and forms nonideal monolayer. This heterogeneity was also confirmed by the constants of Fritz–Schlunder III. Comparison between linear and nonlinear isotherm models with two parameters shows that transformations of nonlinear isotherm equations to linear forms implicitly alter their error structure. Nonlinear forms better describe adsorption process than linear forms. Desorption using NaOH solution as the desorbing agent recovers a maximum quantity of IC and considers the reuse of activated carbons. In future work, we plan to study the life cycle of the biomass used, to use these activated carbons for biogas purification, and finally to study the influence of the NaOH concentration on the desorption percentage and the desorption kinetics of the two activated carbons.

Data Availability

The data used to support the findings of this study are available from the corresponding author upon request and have been deposited in the Mendeley repository (https://www.mendeley.com/reference-manager/library/all-references/).

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was funded by the Government of Cameroon by Research Modernization grants to researchers of higher education. The authors are grateful for technical assistance rendered by the researchers of Materials and Process Engineering Team/RU-NOGEE of the Department of Chemistry, Faculty of Science, University of Dschang, Cameroon. They are also grateful to Kamdem Tamo Arnaud of the Freiburg Materials Research Center (FMF), University of Freiburg, Stefan Meier Strasse 21, 79104, Freiburg, Germany, and Doungmo Giscard of the Institute of Inorganic Chemistry, Christian-Albrechts-Universität zu Kiel, Max-Eyth-Straβe 2, 24118 Kiel, Germany.