Optimization and Adsorption-Based Recovery of Cobalt Using Activated Disordered Mesoporous Carbons

Kinetic and operating condition studies were studied in the batch mode. MCSG60A has a large and fast adsorption capacity, retaining 90% of the Co2+ in the solution and reaching equilibria in only 15 minutes. Among all the variables studied, the adsorbent dosage and the agitation speed favor the Co2+ adsorption until an agitation speed of 1100 rpm and 15.0 mg/L of adsorbent dosage; higher values do not increase the adsorption capacity. On the other hand, the presence of Na+ ions in the solution and high particle size hinder the adsorption process. The thermodynamic study indicated that this process is exothermic and spontaneous, mainly caused by electrostatic interactions between adsorbent and adsorbate. The adsorption is highly pH dependent: while it is optimum at basic pH, it decreases by a 70% at pH 2. The adsorption process is favored in controlled pH, in spite of the ionic strength that involves the buffer presence in the solution. As to the cobalt recovery, to preconcentrate this metal, it is tested with different acid solutions, proving that lower pH promotes this phenomena, even reaching a recovery around 98%, with HNO3 solution with 0.5 pH. Moreover, if the volume of regenerative solution is decreased to the eight part, the cobalt concentration increases five times, reaching enough concentration to facilitate its recovery in metallic state by other techniques.


Introduction
Several studies have reported that the increasing industrial activity (mainly mining operations, manufacturing of electronic devices, electroplating, and petrochemistry) has boosted heavy metal concentration in water, what is considered a menace for both environment and society [1], due to their nondegradability and toxicity. Among all these heavy metals, some of them such as gallium, indium, and cobalt are considered strategic metals, due to the increase in their consumption and scarce. As a matter of fact, cobalt is one of the most interesting to be removed and recovered since not only it causes neurotoxicological disorders and cancer [2] but also, due to the increase in its consumption, is mainly used for the manufacture of electronic devices [3].
Cobalt is used in a wide range of applications; however, it is mainly used for manufacturing rechargeable battery electrodes and super alloys, resistant to corrosion and wear, which are used for gas turbine engines. Further applications are catalysts for chemical industries, magnets, steel-belted radial tires, and airbags. Every year, its consumption is increasing and its production is decreasing. Consequently, this increases its economic importance. For instance, in 2015, the cobalt consumption increased 8% more than the year before. However, it is important to bear in mind that recycling is increasing every year as an alternative of the scarcity of this precious metal. In fact, in 2016, the amount of cobalt recycled was 30% compared with the total consumed [4]. Cobalt is obtained as a by-product in mining industry, mainly in Congo, from nickel and copper ores. Moreover, this activity is involved in violence, and labor rights are violated [5].
ere is a wide range of technologies that allow achieving the reduction of water pollutants, such as ion exchange, extraction with solvents, adsorption, and nanofiltration. Adsorption is considered the best option in order to overcome this problem since this method is suitable for the removal of heavy metals from water solutions and it is relatively cheap, depending on the price of the adsorbent. Consequently, in bibliography, a large number of materials have been considered as potential metal adsorbents in aqueous solutions. First of all, clinoptilolite has shown good capacity to adsorb Cu 2+ , Cr 3+ , and Fe 3+ , removing 95%, 96%, and 85%, respectively, at pH 6 [6]. Furthermore, spherical mesoporous silica adsorbed the Cu 2+ and Cd 2+ in water solution, 80% and 20%, respectively [7]. Mohammed and Sahum studied the adsorption of Cr 3+ from aqueous solutions using eucalyptus bark, removing 99% of the Cr 3+ in water solution [8]. As for carbonaceous materials, activated carbon from coconut coir pith has shown a good cadmium adsorption from a solution of 40 mg/L at pH 5 [9]. Among all these materials, carbonaceous materials are good adsorbents, even they are especially useful because of their high surface area and chemical surface. Some activated carbons contain a great deal of oxygenated groups: phenolic, carboxyl, and carbonyl, which promote the adsorption phenomena, due to the fact that they increase the negative charge density of the adsorbent surface.
e main characteristic feature of this sort of material is their porous and disorganized structure, which contributes to their high specific surface [1]. Nevertheless, the dominance of micropores (pore sizes lower than 20Å) in their structure increases material diffusion resistance, which hinders the adsorption process. However, some carbonaceous materials such as carbon nanotubes and ordered mesoporous carbons overcome this obstacle [10]. For instance, ordered mesoporous carbons have both uniform and large pore sizes, which endue them with large pore volumes; consequently, they are used for many applications, including adsorption. In fact, some studies have shown good effectivity for metal adsorption. Barczak et al. studied the adsorption of zinc, lead, copper, and cadmium on ordered mesoporous carbons obtaining fast adsorption rates and high adsorption capacities around 30 and 180 mg/g for zinc and lead, respectively. As for the recovery of the metal ions previously adsorbed, the use of acid solutions, such as HCl or HNO 3 , was effective, recovering 40% of the zinc adsorbed previously and all the adsorbed lead [10]. Also, ordered mesoporous carbons allow their organic functionalization by oxidation, which causes an increase of oxygenated groups on the adsorbent surface, promoting the adsorption capacity of this material [11].
Also, operating conditions such as initial solution pH, adsorbent dosage, foreign ions, (Na + presence), particle size, contact time, and temperature must be considered. In some cases, it could inhibit the adsorption process depending on the mechanism of adsorption that prevails. In the case of adsorption using graphene oxide, the solution pH is recognized as a very influential parameter that dominates the adsorption process, which not only governs the species of heavy metal ions present in the solution but also affects the surface potential of the adsorbent. Besides, presence of foreign ions (Na + cations) also affects the heavy metal ions adsorption process since these additional ions compete for the limited reactive sites [12]. For instance, the Cd 2+ adsorption on coconut coir pith carbon increases 40% from pH 2 to pH 6 [9]. Regarding the temperature influence, it depends on the kind of adsorption and whether the adsorption is exothermic or endothermic. For instance, the adsorption of methylene blue on activated carbon highly increases from 276 to 444 mg·g −1 when the temperature increases from 30 to 60°C since it is an endothermic process, typical of chemisorptions [13].
With regard to obtaining metallic cobalt from water solutions, as well as all strategic metals, it depends on its concentration in the water steam. Some of the implemented methodologies are solvent extraction, cementation, and electrochemical treatment. All of them are quite complex; not only do they affect the environment but also they involve many combined operations. us, a simpler and efficient recovery process linked to a commercial application needs to be studied [14,15].
Adsorption processes are considered the best option to overcome this problem since this method is suitable to preconcentrate metal ions, which were initially in trace amounts in the water steam, in order to facilitate its recovery by other technologies, such as electrochemical treatment and solvent extraction. Moreover, adsorption is a suitable technology to preconcentrate metal ions in water solutions to facilitate its recovery by other techniques that are implemented in industrial scale, which are more complex and expensive [16]. Wu et al. [11] evaluated the regeneration of mesoporous carbons, by using acid solutions such as HCl and HNO 3 .
is methodology allows recovering around 40% of the previously adsorbed Zn 2+ and all the adsorbed Pb 2+ . e adsorption capacity of the material after successive adsorption cycles hardly changes [11]. Suryavanshi and Shukla [17] studied the capacity of gallium adsorption on oxidized coir and its recovery. e regeneration tests were performed with HCl solutions, varying the acid concentration in the regenerative solution between 0.1 and 1.0 M. ey found that the higher the acid concentration, the higher the gallium amount recovered, even with a rate of 78.3 and 98.6% using an acid concentration of 0.1 and 1.0 M, respectively. e novelty of this research is the use of adsorption as a tool to preconcentrate cobalt for its recovery by solvent extraction or electrochemical treatment. Compared with previous studies of our research group, this paper focuses on the influence of adsorption conditions and the recovery of cobalt ions [18]. is could provide a wide range of options to overcome the high complexity of the recovery process of strategic metals from mining waste waters.

Mesoporous Activated Carbon Synthesis.
e synthesis of the adsorbent studied (MCSG60A) was performed following the replica method, as previous works, using sucrose as the carbon source and silica gel (60Å) as the template. Firstly, silica gel was impregnated with sucrose in an acid aqueous solution which contains sulphuric acid (H 2 SO 4 ). Afterwards, this mixture was precarbonized to 100°C for 6 h and then up to 150°C for another 6 h. Later, the material was further heated at 800°C in N 2 atmosphere during 240 min. Finally, the silica gel template was removed using an aqueous solution of hydrofluoric acid (HF, 25 wt.%). Later, the material was washed with ethanol and deionised water [18].
With regard to its functionalization (activation), the mesoporous carbons were heated to 450°C under inert atmosphere at 10°C/min which consists in a nitrogen stream of 100 mL/min of N 2 . e materials were activated in mild oxidant atmosphere (oxygen diluted at 5% in nitrogen), with a flow of 100 mL/min, for 5 hours at 450°C [19].

Characterization of the Materials.
Firstly, the superficial chemistry of the adsorbent has been tested by its infrared spectrum (400 and 4,000 cm −1 ) using a ermo Nicolet, Avatar 360 FT-IR spectrophotometer. Zeta potential measurements were performed using a MALVERN Zetasizer Nano Series nano Z apparatus. During the essays, 50 mg of the sample was diluted in 20 mL of milli-Q water and the pH was adjusted with solutions of 1.0, 0.1, and 0.01 M of either HCl or NaOH.
In order to characterize the textural properties of the adsorbent, it was evaluated by means of nitrogen adsorption-desorption isotherms at -196°C, using a Micromeritics ASAP-2020 adsorption apparatus. e micropore volume (V micro ) and mesoporous surface (S meso ) were obtained using the t-plot method. e total surface area (S BET ) and the pore volume (V total ) were determined using the BET (Brunauer-Emmett-Teller) equation and the single point method. Pore size distribution (PSD) curves were calculated by the BJH (Barrett-Joyner-Halenda) method with the KJS (Kruk-Jaroniec-Sayari) correction.

Adsorption Experiments.
e adsorption experiments were carried out in the batch mode, in 2 mL Eppendorf tubes, in orbital agitation using a ermomixer Comfort apparatus equipped with temperature control. e experimental procedure is as follows: Tests were performed by adding the appropriate dosage of the carbonaceous adsorbent to 2 mL of a Co(NO 3 ) 2 aqueous solution with the desired amount of Co 2+ . After each experiment, the adsorbent was separated from the aqueous solution by centrifugation (10,900 rpm), using a Spectrafuge 24D apparatus from Labnet International. e cobalt concentration in water was measured by atomic absorption spectroscopy (AAS), by means of an AA-7000 Shimadzu equipment [18]. e kinetic adsorption curves were determined by measuring the Co 2+ concentration in the solution at different times during the adsorption process. Each point of the curve was measured in triplicate. e adsorption capacity of the adsorbent was calculated with equation (1) and the percentage of cobalt removed with equation (2).
where q (mg/g) is the amount of Co 2+ adsorbed per gram of the adsorbent; C 0 and C (mg/L) are the concentrations of Co 2+ in the aqueous phase before and after adsorption, respectively; m (g) is the weight of the adsorbent, and V (L) is the volume of the aqueous phase.

Results and Discussion
e influence of operating conditions has been studied following the previous methodology and changing a definite variable.

Kinetics.
e optimization of the operation time is very important in order to minimize costs and energy. e adsorption kinetic results are represented in Figure 1. is figure shows the evolution of cobalt removal (%) with time.
e cobalt adsorption process on MCSG60A showed a cobalt removal of 85-90% in only 15 minutes, with an equilibrium adsorption capacity of 1.5 mg/g. Moreover, experiments of 4 days of adsorption time displayed that 86% of the cobalt in the solution has been adsorbed; as a result, the activated carbon does not adsorb more cobalt despite increasing the contact time. e high adsorption rate was mainly due to the porous size distribution of the adsorbent because, as the BET analysis (Table 1) indicates, the carbon possesses a large specific surface area with a noticeable mesoporous surface.
Consider that the size of hydrated ion will determine which ions will be able to pass through the adsorbent pores. In this sense, the average pore size of the adsorbent MCSGG60A is much larger (60Å) than the cobalt solvatation shell size (diameter 8.5Å for [Co(H 2 O) 6 ] 2+ ) [20]. Table 1 shows the textural properties of the adsorbent. Figure 2 shows SEM images of MCSG60A adsorbent.
On the other hand, the large equilibrium adsorption capacity on the mesoporous carbons was due to the high negative charge density of the surface, as a consequence of the presence of a large number of oxygenated groups (COOH, C�O, O-H, etc.); this fact was observed in the FT-IR analysis (Figure 3). e Z potential analysis also shows that the surface of the carbon remains negatively charged at pH values higher than 2.3. Figure 4 shows the in uence of the adsorbent dosage on equilibrium adsorption capacity and on cobalt removal. e initial concentration of Co 2+ has been changed to 50 mg/L instead of 20 mg/L to distinguish better the di erences among each test. e percentage of Co 2+ adsorbed from the solution increases with the dosage, until a certain dosage value (15.0 g/L), from which the amount of Co 2+ adsorbed tends to remain constant (2.0 mg/g). Regarding the equilibrium capacity, it tends to diminish with the adsorbent dosage; this fact may be due to the overlapping or aggregation of the adsorbent particles, which leads to a decrease in the available adsorbent surface area and an increase in the di usional resistances. Moreover, heavy metal ions can rapidly induce aggregation/folding by the strong interaction between them and carboxyl groups [12].

Temperature In uence ( ermodynamic Study).
e in uence of temperature on equilibrium adsorption capacity and on cobalt removal appears in Figure 5.
As Figure 5 shows, the e ect of temperature on the adsorption process is almost negligible. In all the ranges of temperature studied, its equilibrium capacity remains constant (1.4 mg/g) since the capacity decreases very slightly (less than 2%), at high temperatures. So, to make a deep analysis, the thermodynamics parameters were determined. e change in the standard enthalpy (ΔH 0 (kJ/ mol)) and entropy (ΔS 0 , (J/mol·K)) was calculated with the Van't Ho expression (equation (6)), which is developed by equations (4) and (5). Gibbs free energy (ΔG 0 (kJ/mol)) is estimated with equations (3) and (4). In these equations, Kc is the equilibrium constant, C 0 (mg/L) is the initial concentration of Co 2+ , C e (mg/L) is the concentration at    Table 2. In addition, the tting of the equation (6) for experimental data is represented in Figure 6.
ΔH 0 and ΔG 0 are below zero; as a result, the adsorption is an exothermic process. e typical range of enthalpy values for chemisorption is between −80 and −450 kJ/mol. However, the fact that in the present study the enthalpy value is −4.57 kJ/mol indicates that physisorption is the only adsorption mechanism.
is kind of adsorption is mainly attributed to electrostatic interactions. is is corroborated       Advances in Materials Science and Engineering by the ΔG 0 value, which is −5.1 KJ/mol, inside the range between 0 and −20 kJ/mol, proving that cobalt get adsorbed on MCSG60A by physisorption phenomena [21].
Regarding the entropy, its positive ΔS 0 suggests an increase of the randomness, which involves some structural changes in the adsorbate-adsorbent interface. e adsorbed molecules, which were displaced by the surface groups of the adsorbate, gained more translational entropy than the amount, which was lost by the adsorbate molecules [22].

Particle Size
In uence. For the particle size in uence, the activated carbon has been sieved following the standard ISO-3310.1. e following sizes have been studied: 1.18-1.00, 0.80-0.76, and 0.71-0.56 mm. Figure 7 shows Co 2+ adsorption isotherms obtained with di erent particle size of the adsorbent.
As Figure 7 shows, the higher the particle size, the lower the adsorption capacity. is variable a ects the adsorption process to the point that the adsorption capacity decreases by 18%.
is could be explained by the fact that larger particle sizes o er lower contact surface between adsorbent and adsorbate, which causes a reduction of the super cial groups that could interact with the Co 2+ of the solution [23]. To further analyze, FT-IR analysis was carried out with two di erent particle sizes of MCSG60A (Figure 3). In general, the mesoporous carbons showed carboxyl and phenolic groups; nevertheless, if the particle size is high, the presence of these groups decreases considerably. is is due to the activation process, when the particle is big; the activation is only produced in the external area of the solid, due to internal material di usion resistance. On the other hand, for smaller particles, the most part of its surface is activated. Figure 8 shows Co 2+ adsorption isotherms obtained with di erent stirring rates.

Agitation Speed In uence.
An increment in the stirring speed causes an increase in the adsorption capacity of the adsorbent because when stirring speed is high, the external material di usion is promoted. As can be observed in Figure 8, at 300 rpm, the capacity decreases by 13% compared with the optimum agitation speed (1100 rpm). However, when the stirring speed is increased above 1100 rpm, the capacity remains constant indicating that the external mass transfer resistance is negligible.
us, 1100 rpm is the optimum stirring speed. An increase in the agitation degree involves an increase in the mobility of the system. As a result, the faster the stirrer rotates, the lower the external mass transfer resistance is, which causes an enhancement in the adsorption capacity.

Foreign Ions/Ionic Strength In uence.
In the study of the ionic strength, NaCl has been used to adjust the desired values of this variable since this dissociates to Na + cations and Cl − anions. It is calculated by the following expression: where C i (mg/g) is the concentration of the ion species i, Z i is their charges, and FI (mM) is the ionic strength of the total solution. e results of the in uence of the ionic strength using NaCl, on the cobalt adsorption, are shown in Figure 9.
As Figure 9 shows, the ionic strength a ects negatively the cobalt adsorption on MCSG60A, decreasing the capacity by 60% when the ionic strength is 96 mM, due to the presence of Na + .
is is explained by the fact that Na + ions would compete with metal ions for the adsorption sites, depending on their relative a nities. Considering this, the decrease on its adsorption is a consequence of the competition between Co 2+   Advances in Materials Science and Engineering and Na + . us, the adsorption on MCSG60A is not selective. In fact, this is a characteristic of physisorcion processes. Furthermore, ionic strength can a ect the electrostatic interactions, as well as limit the ions transfer from the solution to the surface of the adsorbent [12]. us, the adsorption process is mainly due to electrostatic interactions, as concluded in the previous thermodynamic study. ese results agree with metal adsorption studies carried out with carbonaceous materials; for instance, Tan and his research group obtained that the adsorption of Cd 2+ , Ni 2+ , and Cu 2+ on graphene oxide was slightly impeded with the increasing presence of Na + , decreasing its adsorption capacity to 8, 11, and 11%, respectively [24]. Regarding the pH evolution of the ionic strength in uence, there is nothing to outline. In all cases, the pH decreases to pH 3.0-3.5; the only di erence is that the initial pH increases with the presence of NaCl, i.e., 6.5 with an initial NaCl concentration of 1 mM and 7.0 at 95 mM.

pH In uence.
e pH of the solution plays an important role in the adsorption process and particularly in the uptake capacity. For this reason, the adsorption of Co 2+ on MCSG60A was studied varying the initial pH from 2 to 8, considering the speciation diagram of cobalt as reference. According to this diagram, at pH values higher than 8, cobalt tends to precipitate in the form of cobalt hydroxide (Co(OH) 2 ) [23].
Two di erent sets of experiments were carried out: a rst group in which the pH was set free and a second one in which the pH was kept constant during the whole process by means of a bu er solution. Figure 10 shows the results obtained with di erent initial pH values, without controlling this variable throughout the experiment. Table 3 consists in the pH evolution in each one of the adsorption experiments.
As Table 3 shows, during the adsorption process, the pH decreases considerably from 5.6 to 3.2. However, this e ect decreases in acid solutions. If the initial pH is higher than 4.0, the nal pH is between 3.2 and 3.5. For instance, at initial pH 2.5, the pH remains steady during the process. is could be due to the presence of some groups, such as phenolic groups in the adsorbent surface that acidify the media without exchange process.
According to Figure 10, the capacity decreases by 70% when the initial pH is varied from 8 to 2. us, the Co 2+ adsorption on MCSG60A is highly pH dependent. is can be explained in terms of isoelectric point (pH IEP ) and how the surface charge of the solid changes with this value, as well as considering that electrostatic forces are the main responsible of the adsorption process. According to the potential Z analysis, if the pH is lower than pH IEP , which is 2.3, the surface of the carbon is positively charged; as a result, it loses its capacity to retain Co 2+ . On the other hand, above pH IEP (2.3), the surface is negatively charged. us, an increase in the initial pH solution will lead to more negatively charged sites, promoting the adsorption of Co 2+ [23]. Despite the fact that cobalt adsorption on MCSG60A is mainly physical, it has to be considered that its super cial chemistry is complex (Figure 3). us, another reason that justi es the observed e ect of the pH on the adsorption could be the suppression of the complexation. In acid solutions, some functional groups of the surface are protonated and in these conditions, it is di cult to dissociate them; as a result, the complexation between cobalt ions and these groups is suppressed [12].   To further analyze this phenomenon, the pH evolution of a solution of Co 2+ and of pure water was measured, using MCSG60A. Both experiments were carried out at pH 6, T 25°C, adsorbent dosage 12.5 g/L, particle size powder, 1100 rpm, and [Co 2+ ] 20 mg/L for the cobalt solution.
e studies showed that the pH changes even without adsorption process (when the solid is dispersed in water), changing its pH from 6.6 to 3.6. is change is similar to the studies in presence of Co 2+ , in which the pH of the media changes from 6.0 to 3.3. As a consequence, the synthesized carbon MCSG60A causes the acidi cation of the media, this could be due to the presence of phenolic and carboxylic groups on the solid surface. is kind of groups release H + in water solution, without adsorption process [25]. e study at controlled pH was carried out using a phosphate bu er solution (K 2 HPO 4 /KH 2 PO 4 ). is solution consists of a mixture of monobasic dihydrogen phosphate and dibasic monohydrogen phosphate. By varying the amount of each salt, a range of bu ers can be prepared. In this case, with a mixture of 8.5 mL of K 2 HPO 4 1 M and 91.5 mL of KH 2 PO 4 1 M, it is possible to bu er well at pH 6. Figure 11 shows the results obtained with di erent initial pH values, controlling this variable throughout the experiment by means of bu er solutions. e initial pH and nal pH are 5.8 and 5.6, respectively; as a result, it could be considered a bu ered assay.
As can be observed in Figure 11, when the pH is controlled, the adsorption capacity does not vary when compared with the free pH test, despite the negative e ect of the ion strength associated to the bu er solution (91 mM). Controlled pH also makes the cobalt adsorption even faster, reaching equilibrium in 3 minutes. e cause could be its super cial charge, which is negative, according to zeta potential results, as previously shown in Figure 3.
is negative charge maintains with the pH controlling, keeping the driving force between the positive ions of cobalt and the negative surface of the activated carbon. Besides, if the pH remains constant, the groups on its surface remain protonated. On the other hand, when the ionic strength rises but without controlled pH, its capacity decreases more than a half compared with the pH controlled adsorption. is proves that the adsorption process is mainly due to electrostatic interactions between adsorbate and adsorbent. e only alternative to promote the adsorption in solutions with high ionic strength is controlling the solution pH.

Regeneration Studies
In order to study the reuse of MCSG60A and preconcentrate cobalt in the water steam, after the adsorption assays, the adsorbent was immersed in acid media. e acid solutions tested where HF (pH 1.5) recirculated from the synthesis step and HNO 3 aqueous solutions in a range of pH between 0.5 and 1.5. Figure 12 displays the regeneration results, expressed as Co 2+ recovered from the saturated adsorbent. e percentage of Co 2+ recovered was estimated by mass balance, as shown by the following equations: %Co 2+ recovery C adsorbed − C recovered C adsorbed · 100.
As it can be observed in Figure 12, lower pH promotes the regeneration of the adsorbent. For instance, when the initial pH of regenerative solution decreases from 1.5 to 0.5, the amount of Co 2+ recovered from the adsorbent increase around 23%. is is caused by the negatively charged adsorbent surface; this agrees with the potential Z analysis results: if the pH decreases, the adsorbent surface tends to be positively charged; therefore, the Co 2+ ions adsorbed by   After the regeneration of the adsorbent is achieved, the following step is to concentrate Co 2+ so that the water stream is conditioned for its recovery by solvent extraction or electrochemical treatment. To address this, further regeneration experiments were performed decreasing the volume of regenerative dilution (HNO 3 , 0.5 pH). Figure 13 shows the experiment results.
As displayed in Figure 13, it is possible to concentrate the Co 2+ recovered ve times, reaching a nal concentration of 92 mg/L, if the volume of regenerative solution decreases to the eighth part. Despite cutting the regenerative solution rate, the mass of Co 2+ recovered in these cases is between 0.035-0.023 mg. For instance, when the volume of regenerative solution is greater than the half part (1.0 mL), the recovered cobalt mass is 0.035 mg, whereas, for the quarter and the eight parts (0.5 and 0.25 mL), it recovers 0.030 and 0.023 mg of cobalt, respectively. All taken into account, it could be concluded that intraparticle forces promotes the particle agglomeration, preventing cobalt recovery when the amount of adsorbent particles per volume is higher. us, it reduces even 35% the mass of cobalt recovered when the volume of regenerative solution is eight times lower.

Conclusions
In this research, the in uence of kinetic and operating conditions of cobalt adsorption has been studied using a disordered mesoporous activated carbon (MCSG60A). is material showed good properties for the Co 2+ removal, adsorbing 90% of the Co 2+ in optimum conditions. ese conditions are shown in Table 4: Moreover, the adsorption of this metal using MCSG60A is fast reaching equilibria in 15 minutes. is adsorbent was employed to assess the e ect of adsorption conditions on Co 2+ adsorption. Adsorption of the metal ions was found to be highly pH dependent, being promoted at basic pH. Besides, the adsorbent dosage was found to be directly related to the adsorption capacity, the optimum dosage being 15.0 g/L. On the other hand, when the particles of activated carbon are larger, the adsorption is impeded and the temperature does not hinder the Co 2+ adsorption on MCSG60A.
e thermodynamic studies showed that the adsorption is mainly physical since not only ΔH 0 proves this but also ΔG 0 has the characteristic value of an adsorption process which is mainly due to electrostatic interactions. Furthermore, the presence of foreign cations (Na + in the experiments performed) in the solution decreases the cobalt adsorption capacity; this fact is characteristic of physical adsorption since this a ects the electrostatic interactions between adsorbent and cobalt.
Finally, the pH-controlled test showed that despite the ionic force that involves the bu er solution, the adsorption is not inhibited due to the negatively charged surface that remains steady, keeping the driving force.
Finally, the regeneration results showed that the solution of HNO 3 (0.5 pH) allows to recover nearly all the cobalt previously adsorbed (98%). Also, it is possible to concentrate this metal ve times by decreasing the volume of regenerative solution eight times. us, adsorption is a suitable option to preconcentrate this metal and facilitate its recovery by solvent extraction or electrochemical treatment.
Data Availability e data used to support the ndings of this study are included within the article.

Conflicts of Interest
e authors declare that they have no con icts of interest.
Acknowledgments is work has been nanced by the Ministry of Economy and Competitiveness CTQ2014-59011-R (REMEWATER) and CTM2014-53485-REDC (TRAGUANET).  Advances in Materials Science and Engineering 9