Effective Removal of Cd(II) from Aqueous Solutions Using Theobroma cacao Agro-Industrial Waste

Theobroma cacao agro-industrial waste (WTC) has been characterized and tested as an effective biosorbent to remove Cd(II) from aqueous media. At the optimum pH of 5.0, a maximum adsorption capacity of qe,max = 58.5 mg g−1 was determined. The structural and morphological characterization have been conducted by FTIR, SEM/EDX, and TGA measurements. The SEM/EDX results confirmed that the metals are adsorbed on the surface. C-O-C, OH, CH, NH, and C=O functional groups were identified by FTIR. TGA results were consistent with the presence of hemicellulose. Biosorption kinetics were rapid during the first 30 min and then reached equilibrium. The corresponding experimental data were well fitted to pseudo-first and -second order models, the latter being the best. The biosorption isotherm data were also well fitted to Temkin, Langmuir, and Freundlich models, showing that several sorption mechanisms may be involved in the Cd(II) biosorption process, which was characterized as exothermic (ΔH0 < 0), feasible, and spontaneous (ΔG0 < 0). In binary (Cd–Pb and Cd–Cu) and ternary (Cd–Pb–Cu) systems, Cu(II) and particularly Pb(II) co-cations exert strong antagonistic effects. Using HNO3, effective good regeneration of WTC was obtained to efficiently remove Cd(II) up to three times.


Introduction
Nowadays, many drinking water and agricultural irrigation sources are contaminated by heavy metals, representing a threat to human health even when at minimal concentrations within the body [1,2]. Cadmium (Cd) is an extremely dangerous heavy metal for people, animals, and plants [3]. It readily accumulates in the food chain to reach humans, causing severe damage to the renal, reproductive, nervous, hepatic, or pulmonary systems; it also affects growth, bone metabolism, and can even cause cancer [4]. Therefore, the EPA (USA-Environmental Protection) has included cadmium in the list of priority pollutants and the WHO (World Health Organization) has set the maximum permissible concentration at 0.003 mg L −1 in drinking water and industrial wastewater [5].
The environmental presence of Cd comes from various sources, particularly industrial ones related to the production and recycling of batteries [6,7], alloys, coatings (galvanoplasty), solar cells, plastic stabilizers, or pigments. Therefore, Cd removal is a very current topic of interest. There are traditional methods to remove cadmium from contaminated aqueous solutions: filtration, ion exchange, chemical oxidation or reduction, chemical precipitation, evaporative recovery, and electrochemical treatment [3,8]. However, these methods are not very effective for contaminated aqueous solutions with Cd concentrations The thermogravimetric (TGA) analysis ( Figure 2) shows the thermal degradation of WTC with increasing temperature. We can see different stages of mass loss over temperature: (a) Stage I (from room temperature to 200 °C), the mass loss of ~10% could be attributed to the evaporation of moisture present in the sample [21]. (b) Stage II (200 to 357 °C), the loss of mass is abrupt, up to ~44%, which is related to the degradation of the hemicellulose. (c) Stage III (357 to 450 °C), the mass loss is of the order of 10% and is related to the pyrolysis of cellulose. (d) Stage IV (450 to 600 °C) the loss of mass is slower and would be related to the decomposition of lignin. Therefore, TGA analysis shows that hemicellulose, which is the main component of WTC, degrades more easily than cellulose, and this, in turn, degrades more easily than lignin. These results are similar to those obtained by Basu et al. (2017) [22] with cucumber peel biomass.

pH Effects
The initial pH of the solution is one of the important factors for the biosorption process since it affects the surface charge of the biosorbent and the ionization of the metallic sorbate [31]. Figure 4 displays the effect of the initial pH on the Cd(II) adsorption capacity, qe, at different initial concentrations of Cd(II), Co. For all Co concentrations, a rapid increase in qe is observed in the very acidic pH range (pH ≤ 4). In the pH range from 4 to 5, the increase in qe slows down until reaching a plateau at pH 6 for high Co concentrations. Thus, pH 5.0 has been chosen as the most adequate for the optimal Cd(II) biosorption. These results can be interpreted through the electrostatic interaction mechanism between Cd(II) and the charged WTC surface since the point zero-charge pH of WTC is pHPZC = 3.9 [18]. This indicates that for pH < pHPZC = 3.9, the WTC surface is positively charged and the repulsion with Cd(II) cations limits its adsorption capacity. In contrast, for pH ≥ pHPZC, the WTC surface is negatively charged, facilitating Cd(II) adsorption. In addition, under these conditions, the deprotonation of carboxylic (-COOH) and hydroxyl functional groups (identified by FTIR) would occur, for which the pKa is between 4 and 6.

pH Effects
The initial pH of the solution is one of the important factors for the biosorption process since it affects the surface charge of the biosorbent and the ionization of the metallic sorbate [31]. Figure 4 displays the effect of the initial pH on the Cd(II) adsorption capacity, q e , at different initial concentrations of Cd(II), C o . For all C o concentrations, a rapid increase in q e is observed in the very acidic pH range (pH ≤ 4). In the pH range from 4 to 5, the increase in q e slows down until reaching a plateau at pH 6 for high C o concentrations. Thus, pH 5.0 has been chosen as the most adequate for the optimal Cd(II) biosorption. These results can be interpreted through the electrostatic interaction mechanism between Cd(II) and the charged WTC surface since the point zero-charge pH of WTC is pH PZC = 3.9 [18]. This indicates that for pH < pH PZC = 3.9, the WTC surface is positively charged and the repulsion with Cd(II) cations limits its adsorption capacity. In contrast, for pH ≥ pH PZC , the WTC surface is negatively charged, facilitating Cd(II) adsorption. In addition, under these conditions, the deprotonation of carboxylic (-COOH) and hydroxyl functional groups (identified by FTIR) would occur, for which the pKa is between 4 and 6. Figure 5 displays the dosage effect of WTC on both the adsorption capacity (q e ) and removal efficiency (%R) of Cd(II). %R increases with an increasing dose of WTC from 0.5 to 4.0 g L −1 . In contrast, q e decreases in this dose range. At 1.0 g L −1 , %R = 72% and it increased to maximum removal value (90%) at 4.0 g L −1 . This would be due to an increase in the surface area with a large number of biosorption vacant sites. For low values of the dose (range 0.5 to 1.0 g L −1 ), all adsorbent sites would be fully occupied and the surface would be saturated, resulting in high q e (~18 mg g −1 ) values [32]. The significant decrease in q e at doses ≥ 1.0 g L −1 would be due to the splitting effect of the concentration gradient between the biosorbent and the sorbate with increasing WTC dose, which would cause a reduction in the amount of Cd(II) adsorbed per unit mass of WTC [33,34]. A considerable increase in the biosorbent dosage could also cause its agglomeration and therefore a blockage of the available adsorption sites, reducing the interaction of the metal with the surface [15]. Taking into account these results, a WTC dosage of 1 g L −1 was selected as the optimum value that provides an adequate surface area for Cd(II) adsorption.   Figure 5 displays the dosage effect of WTC on both the adsorption capacity (qe) and removal efficiency (%R) of Cd(II). %R increases with an increasing dose of WTC from 0.5 to 4.0 g L −1 . In contrast, qe decreases in this dose range. At 1.0 g L −1 , %R = 72% and it increased to maximum removal value (90%) at 4.0 g L −1 . This would be due to an increase in the surface area with a large number of biosorption vacant sites. For low values of the dose (range 0.5 to 1.0 g L −1 ), all adsorbent sites would be fully occupied and the surface would be saturated, resulting in high qe (~18 mg g −1 ) values [32]. The significant decrease in qe at doses ≥1.0 g L −1 would be due to the splitting effect of the concentration gradient between the biosorbent and the sorbate with increasing WTC dose, which would cause a reduction in the amount of Cd(II) adsorbed per unit mass of WTC [33,34]. A considerable increase in the biosorbent dosage could also cause its agglomeration and therefore a blockage of the available adsorption sites, reducing the interaction of the metal with the surface [15]. Taking into account these results, a WTC dosage of 1 g L −1 was selected as the optimum value that provides an adequate surface area for Cd(II) adsorption.   Figure 5 displays the dosage effect of WTC on both the adsorption capacity (qe) and removal efficiency (%R) of Cd(II). %R increases with an increasing dose of WTC from 0.5 to 4.0 g L −1 . In contrast, qe decreases in this dose range. At 1.0 g L −1 , %R = 72% and it increased to maximum removal value (90%) at 4.0 g L −1 . This would be due to an increase in the surface area with a large number of biosorption vacant sites. For low values of the dose (range 0.5 to 1.0 g L −1 ), all adsorbent sites would be fully occupied and the surface would be saturated, resulting in high qe (~18 mg g −1 ) values [32]. The significant decrease in qe at doses ≥1.0 g L −1 would be due to the splitting effect of the concentration gradient between the biosorbent and the sorbate with increasing WTC dose, which would cause a reduction in the amount of Cd(II) adsorbed per unit mass of WTC [33,34]. A considerable increase in the biosorbent dosage could also cause its agglomeration and therefore a blockage of the available adsorption sites, reducing the interaction of the metal with the surface [15]. Taking into account these results, a WTC dosage of 1 g L −1 was selected as the optimum value that provides an adequate surface area for Cd(II) adsorption.   Figure 6 displays the Cd(II) biosorption capacity, q t , as a function of the contact time t with the WTC biosorbent (q t vs. t), for different initial concentrations of the Cd(II) cation, C o . For all cases, we can appreciate the rapid increase in q t during the first 30 min. Thereafter, the growth of q t slows to reach equilibrium near t = 120 min. The experimental data obtained were fitted to the pseudo-first and psuedo-second order kinetic models. The corresponding optimized parameters are presented in Table 1. The best fit was obtained with the pseudo-second order model (R 2 ≥ 0.93, χ 2 ≤ 0.17), indicating that the Cd(II) biosorption onto WTC would be mainly a chemisorption process [35]. However, the pseudofirst order model also presents a good fit, particularly at low C o concentrations, (R 2 ≥ 0.89, χ 2 ≤ 1.19), which shows that in the Cd(II) biosorption process, physical adsorption also takes place, with adsorption rate constants k 1 , of almost 0.2 min −1 , very similar to that obtained by Tejada-Tovar et al. (2020) [36], in the Pb(II) biosorption onto sugarcane bagasse. Figure 6 displays the Cd(II) biosorption capacity, qt, as a function of the contact time t with the WTC biosorbent (qt vs. t), for different initial concentrations of the Cd(II) cation, Co. For all cases, we can appreciate the rapid increase in qt during the first 30 min. Thereafter, the growth of qt slows to reach equilibrium near t = 120 min. The experimental data obtained were fitted to the pseudo-first and psuedo-second order kinetic models. The corresponding optimized parameters are presented in Table 1. The best fit was obtained with the pseudo-second order model (R 2 ≥ 0.93, χ 2 ≤ 0.17), indicating that the Cd(II) biosorption onto WTC would be mainly a chemisorption process [35]. However, the pseudo-first order model also presents a good fit, particularly at low Co concentrations, (R 2 ≥ 0.89, χ 2 ≤ 1.19), which shows that in the Cd(II) biosorption process, physical adsorption also takes place, with adsorption rate constants k1, of almost 0.2 min −1 , very similar to that obtained by Tejada-Tovar et al. (2020) [36], in the Pb(II) biosorption onto sugarcane bagasse.     In many cases, there is the possibility that intraparticle diffusion is the rate-limiting step of adsorption, which is usually determined using the equation proposed by Weber and Morris (intraparticle diffusion model) [37] (Table 1). In the q t vs. t 1/2 plots (Figure 7) we can distinguish two well-defined parts. The first part, 0 < t 1/2 < 5.41, shows a rapid growth of q t (t), which would indicate a rapid adsorption of Cd(II) cations onto WTC surface. The lines of fit do not pass through the origin of coordinates, which indicates that intraparticle diffusion would not be the only process that controls the rate of adsorption, but that other processes can also control this rate. In the second part, t 1/2 > 5.41, the slower growth of q t (t) would be related to a gradual adsorption process, where Cd(II) cations would enter and fill the biosorbent pores until the equilibrium is reached [14]. In many cases, there is the possibility that intraparticle diffusion is the rate-limiting step of adsorption, which is usually determined using the equation proposed by Webe and Morris (intraparticle diffusion model) [37] (Table 1). In the qt vs. t 1/2 plots (Figure 7 we can distinguish two well-defined parts. The first part, 0 < t 1/2 < 5.41, shows a rapid growth of qt (t), which would indicate a rapid adsorption of Cd(II) cations onto WTC sur face. The lines of fit do not pass through the origin of coordinates, which indicates tha intraparticle diffusion would not be the only process that controls the rate of adsorption but that other processes can also control this rate. In the second part, t 1/2 > 5.41, the slowe growth of qt (t) would be related to a gradual adsorption process, where Cd(II) cation would enter and fill the biosorbent pores until the equilibrium is reached [14].

Biosorption Isotherms
The adsorption isotherm provides information about the affinity that a sorbent dis plays for a particular sorbate [22]. Figure 8 shows the Cd(II) biosorption isotherm onto WTC. The experimental data obtained were adjusted (non-linear adjustments) to thre models, Temkin, Langmuir, and Freundlich. The corresponding optimized parameters ar consigned in Table 2. With these three models, reasonably acceptable adjustments (R 2 0.96 and χ 2 ≤ 3.42) have been obtained, although the best adjustment was obtained with the Temkin model.

Biosorption Isotherms
The adsorption isotherm provides information about the affinity that a sorbent displays for a particular sorbate [22]. Figure 8 shows the Cd(II) biosorption isotherm onto WTC. The experimental data obtained were adjusted (non-linear adjustments) to three models, Temkin, Langmuir, and Freundlich. The corresponding optimized parameters are consigned in Table 2. With these three models, reasonably acceptable adjustments (R 2 ≥ 0.96 and χ 2 ≤ 3.42) have been obtained, although the best adjustment was obtained with the Temkin model.
Thus, the Temkin parameters, particularly that related to heat adsorption, b T = 0.22 kJ mol −1 (<8 kJ mol −1 ) indicate that the Cd(II) biosorption onto WTC would be predominantly a physisorption process [38][39][40] where the Cd(II) adsorbates adhere to the WTC adsorbent through weak van der Waals interactions.
The Langmuir model suggests that Cd(II) biosorption onto WTC occurs with the formation of a monolayer, with a maximum adsorption capacity, q max = 58.5 mg g −1 and an affinity constant K L = 0.068 L mg -1 ; which, due to its small value, indicates a typical sorbate-sorbent interaction in physisorption processes [40].
The Freundlich model parameters, K F = 5.63 mg 0.53 g −1 L 0.474 and particularly n = 2.11 (n > 1), indicate the favorable biosorption of Cd(II) onto WTC. The kinetic and isotherm adsorption results show that the Cd(II) biosorption process onto WTC would occur through various adsorption mechanisms, such as physisorption, chemisorption, and intraparticle diffusion.  Thus, the Temkin parameters, particularly that related to heat adsorption, bT = 0.22 kJ mol −1 (<8 kJ mol −1 ) indicate that the Cd(II) biosorption onto WTC would be predomi nantly a physisorption process [38][39][40] where the Cd(II) adsorbates adhere to the WTC adsorbent through weak van der Waals interactions.  The maximum Cd(II) biosorption capacities, q e,max of biomasses similar to that studied in this work are shown in Table 3. We can appreciate that the q e,max of WTC is among the highest values, indicating that this material is viable for the aqueous removal of Cd(II).

Biosorption Thermodynamics
Thermodynamic functions, such as standard Gibbs energy (∆G o ), enthalpy (∆H o ), and entropy changes (∆S o ) for the Cd(II) biosorption process were evaluated using Equation (1) and the Van't Hoff Equation (2). All these values are presented in Table 4.
where K c is the equilibrium constant, K c = C es C e , and C es and C e are the equilibrium Cd(II) concentrations, respectively, in the biosorbent and in the solution, R is the universal gas constant, and T is the temperature of the solution.  Table 4. Thermodynamic parameters of Cd(II) biosorption onto WTC.
Negative values of ∆G o and ∆H o indicate that the biosorption process is spontaneous and exothermic. Since ∆H o values can provide information on characteristic binding energies such as electrostatic (6-80 kJ mol −1 ) or hydrogen bonds (4-13 kJ mol −1 ) [52]; the ∆H o value obtained in this work (−8.92 kJ mol −1 ) indicates that electrostatic interactions and hydrogen bonds would be present in the biosorption of Cd(II) onto WTC, which is in good agreement with the kinetics and adsorption isotherm results.
It can also be observed that ∆G o values increase or are "less spontaneous" with increasing temperature, showing that the biosorption process is less favorable at high temperatures. In contrast, the negative value of ∆S 0 confirms the decrease in randomness at the solid-liquid interface during biosorption [53].

Effect of Co-Cations in Binary and Ternary Systems
Wastewater usually contains more than one metal (multimetallic aqueous solutions) and the removal by adsorption of each of them is affected by the presence of the other co-cations, which can produce up to three types of behavior: synergism, antagonism, or non-interaction [54]. Antagonistic effects are usually reported in the adsorption capacities of multimetallic solutions, which would be caused by competition among co-cations for the binding sites of the biomass [55]. Figure 9 displays the influence of the initial concentration of Pb(II) and Cu(II) cocations on the Cd(II) biosorption capacity onto the WTC. As can be seen, the antagonistic effect of co-cations is significant, particularly of Pb(II). Therefore, (a) in the binary systems (Cd-Cu) and (Cd-Pb) (Figure 9a), for low concentrations of co-cations (C o = 25 mg g −1 ), the presence of Cu(II) and Pb(II) reduce q e by up to 33% and 42%, respectively. For high co-cation concentrations (C o = 100 mg g −1 ), the q e reductions can reach up to 71% (Cu) and 82% (Pb). (b) In the ternary system (Figure 9b), the reduction of q e is even more drastic, reaching more than 90%. Similar behavior was reported by Hossain et al. 2014 [44] in a ternary (Cd-Cu-Pb) sorption system of the agro-waste cabbage biosorbent.
hydrogen bonds would be present in the biosorption of Cd(II) onto WTC, which is in good agreement with the kinetics and adsorption isotherm results.
It can also be observed that ΔG o values increase or are "less spontaneous" with increasing temperature, showing that the biosorption process is less favorable at high temperatures. In contrast, the negative value of ΔS 0 confirms the decrease in randomness at the solid-liquid interface during biosorption [53].

Effect of Co-Cations in Binary and Ternary Systems
Wastewater usually contains more than one metal (multimetallic aqueous solutions) and the removal by adsorption of each of them is affected by the presence of the other cocations, which can produce up to three types of behavior: synergism, antagonism, or noninteraction [54]. Antagonistic effects are usually reported in the adsorption capacities of multimetallic solutions, which would be caused by competition among co-cations for the binding sites of the biomass [55]. Figure 9 displays the influence of the initial concentration of Pb(II) and Cu(II) cocations on the Cd(II) biosorption capacity onto the WTC. As can be seen, the antagonistic effect of co-cations is significant, particularly of Pb(II). Therefore, (a) in the binary systems (Cd-Cu) and (Cd-Pb) (Figure 9a), for low concentrations of co-cations (Co = 25 mg g −1 ), the presence of Cu(II) and Pb(II) reduce qe by up to 33% and 42%, respectively. For high cocation concentrations (Co = 100 mg g −1 ), the qe reductions can reach up to 71% (Cu) and 82% (Pb). (b) In the ternary system (Figure 9b), the reduction of qe is even more drastic, reaching more than 90%. Similar behavior was reported by Hossain et al. 2014 [44] in a ternary (Cd-Cu-Pb) sorption system of the agro-waste cabbage biosorbent. Although the mechanisms of multimetallic removal are not sufficiently clarified, the competition between ionic sorbates, such as Pb(II), Cd(II), or Cu(II), for occupying the adsorption sites, is evident. Basu et al. (2017) [22], among others, highlight the importance of metallic sorbate parameters such as electronegativity and hydrated ionic radius. It is known that Pb and Cu have high electronegativity values (1.8 and 1.9, respectively) [56], but a low hydrated ionic radius (410 and 419 pm) [57] in comparison with Cd (1.7 and 426 p.m.). As they are more electronegative and have smaller hydrated ionic radii, both Pb(II) and Cu(II) would be more competitive cations than Cd(II). Therefore, these co-cations would antagonistically affect the Cd(II) biosorption capacity in binary systems and much more in ternary systems. Furthermore, Hossain et al. 2014 [44] have reported that the most Although the mechanisms of multimetallic removal are not sufficiently clarified, the competition between ionic sorbates, such as Pb(II), Cd(II), or Cu(II), for occupying the adsorption sites, is evident. Basu et al. (2017) [22], among others, highlight the importance of metallic sorbate parameters such as electronegativity and hydrated ionic radius. It is known that Pb and Cu have high electronegativity values (1.8 and 1.9, respectively) [56], but a low hydrated ionic radius (410 and 419 pm) [57] in comparison with Cd (1.7 and 426 p.m.). As they are more electronegative and have smaller hydrated ionic radii, both Pb(II) and Cu(II) would be more competitive cations than Cd(II). Therefore, these co-cations would antagonistically affect the Cd(II) biosorption capacity in binary systems and much more in ternary systems. Furthermore, Hossain et al. 2014 [44] have reported that the most binding sites of biosorbent (cabbage) are occupied by Pb(II) ions in competitive binary, ternary and quaternary systems.

Desorption
The regeneration of a biosorbent, such as WTC, is crucial for biomass reuse, and metal recovery, and also to reduce the operating costs of any type of water treatment [58]. The literature reports a high desorption yield using HNO 3 as an eluent or desorbing agent [18,59,60]. This acid acts as an effective ion exchange medium, since its released H + replaces the Cd(II) cation on the WTC surface, allowing efficient Cd(II) desorption [35,44]. Figure 10 shows the desorption capacity of WTC using HNO 3 . In the first cycle of Cd(II) removal, a desorption of %D = 58% is reached; in the second cycle, 45%; and in the third cycle, up to 20%. ternary and quaternary systems.

Desorption
The regeneration of a biosorbent, such as WTC, is crucial for biomass reuse, and metal recovery, and also to reduce the operating costs of any type of water treatment [58]. The literature reports a high desorption yield using HNO3 as an eluent or desorbing agent [18,59,60]. This acid acts as an effective ion exchange medium, since its released H + replaces the Cd(II) cation on the WTC surface, allowing efficient Cd(II) desorption [35,44]. Figure 10 shows the desorption capacity of WTC using HNO3. In the first cycle of Cd(II) removal, a desorption of %D = 58% is reached; in the second cycle, 45%; and in the third cycle, up to 20%.

Preparation and Characterizations of WTC Biosorbent
The preparation of the biosorbent is quite simple and friendly to the environment. Agro-industrial waste from Theobroma cacao was collected from the Chanchamayo province located in the Junin-region of Peru. Samples were first washed with water, rinsed with distilled water, and dried at 70 °C for 48 h. Subsequently, the dried adsorbent was ground and sieved using a 70-mesh. All the chemical products used in this work were of analytical grade and from Sigma-Aldrich (Steinheim, Germany) or Merck, Darmstadt, Germany. The Cd(II) stock solution (1000 mg L −1 ) was prepared from Cd(NO3)2.4H2O. To obtain solutions at different concentrations, appropriate amounts of the stock solution were diluted.
Thermogravimetric analysis (TGA) of WTC was performed using Thermo Scientific spectrometer model Evolution 220 (Thermo Scientific Co., Ltd., Waltham, DE, USA) in an N2 atmosphere (100 mL min −1 ). The heating rate of WTC was 10 °C min −1 with temperatures ranging from 20 to 600 °C.

Preparation and Characterizations of WTC Biosorbent
The preparation of the biosorbent is quite simple and friendly to the environment. Agro-industrial waste from Theobroma cacao was collected from the Chanchamayo province located in the Junin-region of Peru. Samples were first washed with water, rinsed with distilled water, and dried at 70 • C for 48 h. Subsequently, the dried adsorbent was ground and sieved using a 70-mesh. All the chemical products used in this work were of analytical grade and from Sigma-Aldrich (Steinheim, Germany) or Merck, Darmstadt, Germany. The Cd(II) stock solution (1000 mg L −1 ) was prepared from Cd(NO 3 ) 2 .4H 2 O. To obtain solutions at different concentrations, appropriate amounts of the stock solution were diluted.
Thermogravimetric analysis (TGA) of WTC was performed using Thermo Scientific spectrometer model Evolution 220 (Thermo Scientific Co., Ltd., Waltham, DE, USA) in an N 2 atmosphere (100 mL min −1 ). The heating rate of WTC was 10 • C min −1 with temperatures ranging from 20 to 600 • C.
The functional groups on the WTC surface were identified using Perkin Elmer Frontier Model Spectrometer (PerkinElmer Inc., Wellesley, MA, USA) equipped with a Platinum ATR accessory. Spectral data were collected over the wavenumber range from 400 to 4000 cm −1 . The morphological analysis and elemental composition before and after Cd(II) biosorption were obtained using SEM/EDX (Thermo Scientific Co., Eindhoven, The Netherlands).

Biosorption Assays
pH, dose of biomass, and initial Cd(II) concentration effects on the adsorption capacity of this metal were evaluated. At room temperature (20 • C), amounts of WTC biomass between 0.0125 and 0.1 g were mixed with 25 mL of Cd(II) solution with a concentration that varied from 10 to 200 mg L −1 (Figure 11). The pH of these solutions was adjusted in the range from 3 to 6 by adding appropriate amounts of 0.1 M HNO 3 or 0.1 M NaOH. The obtained suspension was stirred at 300 rpm for 120 min at room temperature. The initial concentration before and after Cd(II) adsorption in each experiment was determined by atomic absorption spectroscopy (Shimadzu-AAS 6800, Kyoto, Japan). The adsorption capacity, q e , and removal efficiency, %R, were determined by Equations (3) and (4), respectively: where C o and C e (in mg·L −1 ) are the initial and equilibrium final Cd(II) concentrations, respectively; V (in L) is the volume of solution; and m (in g) is the biosorbent mass.

Biosorption Assays
pH, dose of biomass, and initial Cd(II) concentration effects on the adsorption capacity of this metal were evaluated. At room temperature (20 °C), amounts of WTC biomass between 0.0125 and 0.1 g were mixed with 25 mL of Cd(II) solution with a concentration that varied from 10 to 200 mg L −1 (Figure 11). The pH of these solutions was adjusted in the range from 3 to 6 by adding appropriate amounts of 0.1 M HNO3 or 0.1 M NaOH. The obtained suspension was stirred at 300 rpm for 120 min at room temperature. The initial concentration before and after Cd(II) adsorption in each experiment was determined by atomic absorption spectroscopy (Shimadzu-AAS 6800, Kyoto, Japan). The adsorption capacity, qe, and removal efficiency, %R, were determined by Equations (3) and (4), respectively: where Co and Ce (in mg·L −1 ) are the initial and equilibrium final Cd(II) concentrations, respectively; V (in L) is the volume of solution; and m (in g) is the biosorbent mass. All experiments were performed in triplicate. The mean and corresponding standard deviation are reported. Both kinetics and biosorption-isotherm experimental data were adjusted according to the different models considered and are described in Table 5. The quality of the adjustments was evaluated with the chi-square χ 2 and correlation coefficient R 2 parameters. The adjustment criteria that indicate the suitability of the model were low values for χ 2 and close to unity for R 2 [61]. Table 5. Kinetic and isotherm models to evaluate Cd(II) biosorption onto WTC.
All experiments were performed in triplicate. The mean and corresponding standard deviation are reported. Both kinetics and biosorption-isotherm experimental data were adjusted according to the different models considered and are described in Table 5. The quality of the adjustments was evaluated with the chi-square χ 2 and correlation coefficient R 2 parameters. The adjustment criteria that indicate the suitability of the model were low values for χ 2 and close to unity for R 2 [61]. Table 5. Kinetic and isotherm models to evaluate Cd(II) biosorption onto WTC.

Models Equation Parameters
Pseudo-first order q t = q e 1 − e −k 1 t q e (mg g −1 ): adsorption capacity q t (mg g −1 ): the amount of Cd (II) retained per unit mass of biosorbent in time t. k 1 (min −1 ): the first-order kinetic rate constant k 2 (g mg −1 min −1 ): rate constant adsorption h: initial sorption rate (mg g −1 min -1 ) Pseudo-second order q t = q e 2 k 2 ·t 1+q e ·k 2 ·t h = k 2 q 2 e Intraparticle diffusion q t = k id t 1/2 + C k id (mg g −1 min −1/2 ): intraparticle diffusion rate constant C (mg g −1 ): constant related to the thickness of the adsorbent boundary layer

Isotherm Models Equation Parameters
Langmuir q e = q max K L * C e 1+k L C e C e (mg L −1 ): adsorbate concentration in equilibrium q max (mg g −1 ): Langmuir constant related to the maximum biosorption capacity K L : Langmuir constant related to the affinity between sorbent and sorbate Freundlich q e = K F C e 1 /n K F (L 1/n mg (n−1)/n g −1 ): constant equilibriumn n: constant related to the affinity between sorbent and sorbate.
Temkin q e = Bl n(K T C e ) B = RT/b T , R is the gas constant (8.3145 J mol −1 K −1 ), T absolute temperature b T : Temkin constant related to the heat of adsorption (J mol −1 ) K T : Temkin isotherm equilibrium binding constant (L g −1 )

Competitive Effect of Co-Cations
The experiments were carried out in a mixed system, varying the Cd(II) initial concentrations from 10 to 150 mg L −1 , keeping constant the initial concentration of Pb(II) and/or Cu(II) co-cations (25 and 100 mg L −1 ). A WTC dose of 0.025 g and 25 mL of mixed solution at pH 5.0 were used.

Desorption Experiments
First, 50 mg of WTC, previously loaded with Cd(II) (C o = 100 mg L −1 ), was subjected to the desorption process by adding 50 mL of 0.1 M HNO 3 eluent and then stirring at 300 rpm for 120 min. Subsequently, the biosorbent was washed with distilled water and reused again. The adsorption/desorption operation was repeated up to three times. The concentration of Cd(II) adsorbed and desorbed was analyzed by an atomic absorption spectrophotometer (described above). The Cd(II) desorption efficiency (%D) of WTC was calculated using the following expression [35]:

Conclusions
The feasibility and efficiency of Theobroma cacao agro-industrial waste (WTC) for Cd(II) biosorption in aqueous solutions has been studied. This eco-friendly and low-cost biomass was characterized by SEM/EDX, TGA, and FT-IR techniques. TGA analysis indicates the presence of hemicellulose in the WTC structure. SEM results show that the WTC morphology changes Cd(II) biosorption. FT-IR spectra show the bands associated with the OH, NH, CH, and C-O-C functional groups typical of polysaccharides; and to C=O carboxylic groups. After Cd(II) adsorption, several of these bands were red-shifted and changed in intensity, revealing the interaction between Cd(II) and the WTC surface. It has been determined optimal biosorption conditions: pH 5.0, WTC dosage = 1.0 g L −1 , adsorption equilibrium time, 120 min. Kinetic experimental data fit quite well both to the pseudo-second and pseudo-first order models, which implies that physisorption and chemisorption processes would be involved in the Cd(II) biosorption. The biosorption isotherm data fit well with the Temkin, Langmuir, and Freundlich models, which shows that several sorption mechanisms would be involved in the Cd(II) biosorption process onto WTC. From the adjustment to the Langmuir model, the maximum biosorption capacity was derived, q e,max = 58.5 mg g −1 ; this is one of the highest values reported in the literature for similar biomasses coming from agro-industrial waste. The thermodynamic study indicates that the Cd(II) biosorption process is exothermic (∆H o = −8.9 kJmol −1 ), spontaneous (∆G o < 0), and with decreasing randomness (∆S o < 0) at the solid-liquid interface. The presence of Pb(II) and Cu(II) co-cations in binary and particularly in ternary systems significantly reduces the Cd(II) biosorption capacity onto WTC; with Pb(II) being the cocation that exerts the strongest antagonistic effect. The desorption experiments, using HNO 3 as the eluent, showed that the recovery of WTC is feasible for reused up to three times.