See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/374471629 Improving the sorption properties of mesoporous carbons for the removal of cobalt, nickel and manganese from spent lithium-ion batteries effluent Article  in  Separation and Purification Technology · January 2024 DOI: 10.1016/j.seppur.2023.125095 CITATIONS 0 READS 21 2 authors: Naby Conte Complutense University of Madrid 6 PUBLICATIONS   8 CITATIONS    SEE PROFILE J. M. Gómez Complutense University of Madrid 80 PUBLICATIONS   1,071 CITATIONS    SEE PROFILE All content following this page was uploaded by Naby Conte on 05 October 2023. The user has requested enhancement of the downloaded file. https://www.researchgate.net/publication/374471629_Improving_the_sorption_properties_of_mesoporous_carbons_for_the_removal_of_cobalt_nickel_and_manganese_from_spent_lithium-ion_batteries_effluent?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_2&_esc=publicationCoverPdf https://www.researchgate.net/publication/374471629_Improving_the_sorption_properties_of_mesoporous_carbons_for_the_removal_of_cobalt_nickel_and_manganese_from_spent_lithium-ion_batteries_effluent?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_3&_esc=publicationCoverPdf https://www.researchgate.net/?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_1&_esc=publicationCoverPdf https://www.researchgate.net/profile/Naby-Conte?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_4&_esc=publicationCoverPdf https://www.researchgate.net/profile/Naby-Conte?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_5&_esc=publicationCoverPdf https://www.researchgate.net/institution/Complutense_University_of_Madrid?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_6&_esc=publicationCoverPdf https://www.researchgate.net/profile/Naby-Conte?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_7&_esc=publicationCoverPdf https://www.researchgate.net/profile/J-Gomez-4?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_4&_esc=publicationCoverPdf https://www.researchgate.net/profile/J-Gomez-4?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_5&_esc=publicationCoverPdf https://www.researchgate.net/institution/Complutense_University_of_Madrid?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_6&_esc=publicationCoverPdf https://www.researchgate.net/profile/J-Gomez-4?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_7&_esc=publicationCoverPdf https://www.researchgate.net/profile/Naby-Conte?enrichId=rgreq-82d7c7091d094a2b21a7948f1abcd183-XXX&enrichSource=Y292ZXJQYWdlOzM3NDQ3MTYyOTtBUzoxMTQzMTI4MTE5NTk0MTM1OEAxNjk2NTI1NzY4MTUz&el=1_x_10&_esc=publicationCoverPdf Separation and Purification Technology 328 (2024) 125095 Available online 14 September 2023 1383-5866/© 2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). Improving the sorption properties of mesoporous carbons for the removal of cobalt, nickel and manganese from spent lithium-ion batteries effluent N. Conte *, J.M. Gómez * Department of Chemical and Materials Engineering (CyPS Research Group), Universidad Complutense de Madrid, 28040 Madrid, Spain A R T I C L E I N F O Keywords: Activated carbon Conductimetric titration Competitive sorption Cobalt Lithium-ion battery A B S T R A C T The competitive sorption of Co2+, Li+, Ni2+, and Mn2+, strategic metals from spent lithium-ion batteries and their leachates, was studied using activated mesoporous carbons. The mesoporous carbon was synthesized by the replica method using silica gel as a template and exhibited a high surface area with an accessible pore volume due to mesopores (Vmeso > 95%). Fast kinetics and high sorption capacities of these metals were achieved with the chemical activation of mesoporous carbons. The surface modification of the mesoporous carbon was carried out by physical activation with O2 at 450 ◦C and chemical activation under mild conditions (room temperatures) with NaClO2/H2O2 as oxidizing agents. FTIR analysis and conductimetric titration showed that the combination of an initial physical activation step, followed by chemical functionalization, maximized the formation of car boxylic acids (from 0.2 to 0.9 meq/g), due to the complete oxidation of the weakly acidic groups. Those carbons were tested in sorption experiments of the lithium-ion battery metals in monometallic solutions, where physi cally activated and chemically reactivated mesoporous carbon selectively removed over 80% of Co2+, Ni2+, and Mn2+, with sorption capacities over 20 mg/g, while only 20% of Li+ was removed. This carbon stood out amongst the others studied (2 to 11-fold increase in sorption capacities depending on the metal) and was tested in multimetallic solutions, showing fast removal rates, reaching equilibrium within the first 15 min, as well as selectivity towards divalent cations (18 mg/g of Co2+) with insignificant lithium sorption (qLi = 0.33 mg/g). Desorption of metals was carried out using H2SO4, which allowed the recovery and twofold of the initial con centrations of Co, Ni, and Mn. 1. Introduction Water pollution due to heavy metal emissions has been one of the main environmental issues of contemporary global society [1]. Heavy metals presence in water, such as Cd, Ni, Cu, Co, Cr, and Pb are mainly due to industrial discharges [2]. Their danger lies in their high solubility in aqueous media, they can be easily absorbed by living being and affect the trophic chain, causing cancer, growth issues, and irreversible dam age to the respiratory and nervous system [3]. High exposure cases can even cause death [4]. Therefore, it is fundamental for the treatment of aqueous waste flows to reduce the concentration of these metals below the allowed levels [5]. Certain metals, on account of their unique features, low recycling rates, risk of supply, and limited reserves, are commonly known as raw critical materials [6]. Cobalt and lithium fall within this category, as they are crucial in the manufacture of lithium-ion batteries (LiB) [7] for electronic devices and electric vehicles, essential for achieving decarbonization [8]. Cobalt acts as a cathode material, offering great conductivity and high specific energy [9], whereas lithium is the active material used for the electrode reaction, for its high specific heat ca pacity and elevated redox potential value [10]. Nickel is another cath ode active material that generally appears on these batteries in noteworthy concentration, to enhance the storage capacity, at the expense of increasing the costs [11]. Manganese can coexist with the other metals in the structure of batteries, in Li(Mn)O2-based and Li (NiMnCo)O2-based batteries [12], being an important of for increasing the rate capability and life span [13]. The scarcity of these materials make it essential for their recovery from spent lithium-ion batteries, usually undertaken by hydrometallurgy or pyrometallurgy methods [14]. Normally, recovering methods such as leaching and solvent extraction do not achieve a 100% of effectiveness [15], so the outlet aqueous streams need to be extra-purified, and the residual amounts of these metals can be recovered and exploited for further reuse [16]. Many different techniques for the removal of heavy metals from * Corresponding authors. E-mail addresses: nconte@ucm.es (N. Conte), segojmgm@ucm.es (J.M. Gómez). Contents lists available at ScienceDirect Separation and Purification Technology journal homepage: www.elsevier.com/locate/seppur https://doi.org/10.1016/j.seppur.2023.125095 Received 6 July 2023; Received in revised form 31 August 2023; Accepted 11 September 2023 mailto:nconte@ucm.es mailto:segojmgm@ucm.es www.sciencedirect.com/science/journal/13835866 https://www.elsevier.com/locate/seppur https://doi.org/10.1016/j.seppur.2023.125095 https://doi.org/10.1016/j.seppur.2023.125095 https://doi.org/10.1016/j.seppur.2023.125095 http://creativecommons.org/licenses/by-nc-nd/4.0/ http://creativecommons.org/licenses/by-nc-nd/4.0/ Separation and Purification Technology 328 (2024) 125095 2 wastewater can be applied: ionic exchange with resins, adsorption, chemical oxidation, reverse osmosis, flotation, and electrodialysis serve as an example [2]. Adsorption is widely recognized as a reliable, economical, and effective technique for heavy metal removal, achieving a high degree of purified effluent. Moreover, it lacks the disadvantages of other technologies, such as the intensive use of chemical reagents, sludge generation, and high operational costs [5]. Instead, adsorption stands out for its flexibility in the operational design, cost-effectiveness, and the possibility of the regeneration of the adsorbent for further uses [17]. Carbonaceous materials such as activated carbon have been used throughout history as adsorbents for their performance, large surface area (up to thousands of m2/g), diverse porous structure, and multiple applications [18]. They can be produced from different raw materials such as lignite, wood, and peat and by-products and waste materials such as shells, sawdust, straws, ashes, mud, and sludge [19]. Activated carbons are highly efficient in metal cations adsorption because of the presence of oxygenated groups on their surface [20], for instance, carboxyl and hydroxyl groups that enhance the adsorption of cations such as Co2+ [21]. Most commercially activated carbons have a micro porous structure [18] with a pore size <2 nm, which leads to slow ki netics, making it difficult to adsorb large molecules due to diffusional limitations [22]. Working with mesoporous carbons, with a pore size between 2 and 50 nm, allows to overcome those problems, enhancing the adsorption performance by giving access to larger molecules with relative ease. Adsorption kinetics are favored, then operational times are shorter and energy savings can be achieved [23]. Mesoporous activated carbons can possess an ordered structure (OMCs) [22] or disordered (DMCs) [24], depending on the template used to obtain mesoporosity, template that will be impregnated with the selected carbon source. Different activation processes can be employed to functionalize carbonaceous materials, to raise the effectiveness of the adsorption operation. Particularly important is the development of oxygen- containing functional groups, such as carboxylic, lactone, carbonyl, phenol, and quinone groups, especially suited for cation adsorption [25]. Gómez et al. synthesized a mesoporous carbon, thermally acti vated in a furnace under an oxidant atmosphere and employed for the adsorption of cobalt, reaching removal rates of Co2+ over 90% in 15 min [26]. Abebaw Mengistie studied the adsorption of Mn2+ onto activated carbon derived from a medicinal plant, achieving a 94% removal of initial concentration in about 2 h, with first-order adsorption kinetics and equilibrium isotherm adjusted to the Freundlich model [27]. Abbas investigated the adsorption of Co2+ and Pb2+ onto Apricot stone acti vated carbon, functionalized with concentrated H3PO4 and heated up to 250 ◦C. The results were promising since adsorption capacities of 81 and 25 mg/g were achieved for Pb2+ and Co2+, respectively [28]. Ahmed Sayed et al. studied the potential of concentrated H2O2 as an oxidant to produce a mesoporous-activated carbon using tea waste as a carbon precursor. The oxidizing agent improved the presence of oxygen surface functional groups on the carbon. That synthesized adsorbent was used to remove Cu2+ and Pb2+ from aqueous solutions, achieving high capac ities, 1.4 mmol/g for Cu and 1.8 mmol/g for Pb in only 30 min [29]. Hu and co-workers employed NaClO2 as a strong oxidizing agent, along with KOH to extract cellulose, lignin, and hemicellulose to produce highly efficient activated carbons. After heating in a furnace under a N2 flow, the carbon developed a great structure with high porosity (0,9 cm3/g) and a large surface area (997 m2/g) [30]. Mohamed et al [31] performed the activation of using H3PO4 followed by pyrolysis at 500 ◦C and then triethoxysilane propylamine and diethylene triamine were employed to dope it with silica and nitrogen, employing that carbon for chromium(VI) adsorption, reaching high adsorption capacities. Abdel naeim et co-workers [32] studied the adsorption of Cu(II) and Cd(II) on an activated carbon prepared from common reed and chemically acti vated using H3PO4, also with treatment at 500 ◦C, reaching adsorption capacities of 47 and 83.4 mg/g for Cu and Cd, respectively. Usually, the activation process has only been studied either physically or chemically, involving some high temperature step. How ever, few references have been found on the combination of both pro cesses to obtain suitable surface groups to improve their adsorptive properties. The present work explores the combination of both activa tion processes. The combined adsorption of the main four metals (Co2+, Ni2+, Li+, Mn2+) present in cathode of spent lithium-ion batteries has not been deeply studied in the literature. In this work, the main aim was to study the improvement of the sorption properties of mesoporous carbons to maximize the sorption of Li+, Co2+, Ni2+ and Mn2+ in multi-metallic solutions, which simulate wastewater from the treatment of spent lithium-ion batteries. The improvement of the sorption properties of mesoporous carbons was necessary to achieve high removals of these metals from wastewater. Suitable surface properties on the mesoporous carbons were developed using different methods: typical physical acti vation with O2 (5% in N2) at 450 ◦C, a new chemical activation under mild conditions with NaClO2/H2O2 and a combination of both methods. An in-depth characterization was carried out to identify the functional groups involved in the metal cation sorption process. Modification of the surface chemistry of mesoporous carbon makes it possible to adjust its properties, allowing it to improve its metal cation sorption performance. It is interesting to note that due to the nature of the mesoporous carbons, the mechanism for removing cations from the aqueous solution could be either by adsorption or ion exchange. Therefore, the term sorption will be used since adsorption and ion exchange are sorption operations. 2. Experimental 2.1. Chemicals Silica gel (SiO2, pore size 150 Å, particle size 75–250 μm) supplied by ACROS Organics, sucrose (C12H22O11, ≥99.5%), hydrofluoric acid (HF, 40%), supplied by Sigma-Aldrich, ethanol (C2H6O, 96%) and sulfuric acid (H2SO4, 98%) from Panreac, sodium chloride (NaCl ≥99.5%) from Fluka, sodium chlorite (NaClO2, 80%) from Thermo Fisher Scientific and hydrogen peroxide (H2O2, 30% w/w) from Scharlab were employed in the mesoporous carbon synthesis and activation. Cobalt (II) nitrate hexahydrate (Co(NO3)2⋅6H2O ≥98%, Sigma- Aldrich), lithium chloride (LiCl, 99%, Alfa Aesar), nickel (II) nitrate hexahydrate (Ni(NO3)2⋅6H2O ≥98%, Panreac) and manganese (II) ni trate tetrahydrate (Mn(NO3)2⋅4H2O ≥97%, Sigma-Aldrich) were used in the preparation of the solutions. Sodium hydroxide (NaOH, 98%) was supplied by Panreac, Nitric acid (HNO3, 69.5%), supplied by Carlo Erba, and hydrochloric acid (HCl ≥37%), by Fluka was employed as well. Deionized water was used throughout the entire process. 2.2. Mesoporous carbons synthesis and activation The synthesis of the disordered mesoporous carbon was performed following the replica method, according to previous works of the research group [33], named MCO1. The silica template was impregnated with an aqueous solution of sucrose, adding sulfuric acid to enhance the dissolution and catalyze the process. The mixture was stirred, then heated up to 100 ◦C for 6 h and up to 150 ◦C for 6 more hours. Afterward the carbon obtained was grounded and sieved <500 μm particle size. Then, it was pyrolyzed in a tubular furnace with a 100 mL/min N2 flow, for 15 min. To remove the silica template, HF (25%) was added to the carbon and stirred for 24 h at room temperature. The obtained sludge was rinsed with water and ethanol and dried in an oven overnight. The activation of the synthesized mesoporous carbon was conducted following various methods (Scheme 1). Physical activation process was carried out in a tubular furnace with temperature control. The non- activated material was placed in a crucible and the heating was star ted, with a temperature ramp of 10 ◦C/min until reaching 450 ◦C, under N. Conte and J.M. Gómez Separation and Purification Technology 328 (2024) 125095 3 an N2 atmosphere. After that, it was switched to an oxidant atmosphere, using a 100 mL/min flow (95% N2 and 5% O2). The activation process was kept at 450 ◦C for 5 h. Chemical activation, also known as wet oxidation [34], was carried out by combining two different oxidizing agents, sodium chlorite and hydrogen peroxide in aqueous solution under mild conditions, a method similar to that used in the oxidation of nanocellulose [35]. Mesoporous carbon (1 g) was suspended at room temperature in a solution with 50 mL of H2O, 2.93 g of NaCl, 1.41 g of NaClO2, and 1.44 g of H2O2. The mixture was stirred for 24 h, keeping the pH above 5 with the addition of 0.5 M NaOH for the first 5 h, to ensure the formation of ClO2 - ions in the media. The obtained slurry was rinsed and washed with deionized water and then dried overnight. The chemical activation process was applied to both non-activated meso porous carbon and previously physically activated carbon. Table 1 presents the nomenclature of the activated and non-activated meso porous carbons employed in the present work. 2.3. Characterization Different techniques for the characterization of mesoporous carbons were selected to study the different properties of the synthesized ma terials. N2 adsorption–desorption isotherms were obtained at 77 K using a Micromeritics ASAP-2020 apparatus. Surface area and pore volume were determined by applying the Brunauer-Emmett-Teller (BET) equa tion and the single-point method, respectively. To obtain the meso porous surface and pore volume, the t-plot method was applied. The surface chemistry of the sorbents was determined by Fourier Transform Infrared Spectroscopy (FTIR) analysis, with a Nicolet iS50 FTIR spec trophotometer in the infrared spectrum (400–4000 cm− 1). Zeta poten tial measurements were carried out in a Zetasizer Nano ZS apparatus, provided by Malvern Instruments. Thermogravimetric analyses were performed using a Labsys EVO STA apparatus, provided by Setaram. Thermogravimetric experiments were conducted under an inert atmo sphere, setting a temperature ramp of 10 ◦C per min, from 35 to 1000 ◦C. Conductometric titration experiments were carried out to determine the presence of acidic functional groups, such as carboxylic acids. Conductivity measurements were undertaken employing a portable 524 Conductivity meter apparatus, provided by Crison Instruments. The procedure was very similar to the used for the determination of the carboxyl content in cellulosic materials [36]. The titration was per formed according to Roman et al. [37] by using 75 mg of mesoporous carbon dispersed in a mixture of 25 mL ultrapure water (conductivity = 55 μS-cm− 1, resistivity = 18.2 MΩ⋅cm) and 2.5 mL of NaCl (0.01 N). The titration started by adding 0.2 mL of NaOH (0.05 N) for the total acidity or NaHCO3 (0.05 N) for partial acidity (only carboxylic acids) to the mixture and measuring the conductivity. The conductivity was cor rected (Eq. (1) since it is dependent on the amount of electrolytic solu tion [38]. ConductivityC = Conductivitym • (V0 + VA) V0 (1) Where conductivityc is the corrected conductivity (mS⋅cm− 1), con ductivitym is the measured conductivity (mS⋅cm− 1), V0 is the initial suspension volume in mL and VA is the added base volume (NaOH or NaHCO3 solution) at each point (mL). The total and partial acidity were calculated by Eq. (2). Acidity ( meq g ) = (V2 − V1) • Nbase m (2) Where V1 and V2 are the volumes of NaOH or NaHCO3 solution (mL) added to reach the equivalent conductivity, which was determined as the points of intersection of the linear regressions of the regions before and after the equivalence point (Fig. 1). NBase is the normal concentra tion (eq/L) of the base used (NaOH or NaHCO3 solution), and m is the mass of the mesoporous carbon (g). 2.4. Sorption experiments Sorption experiments were carried out in batch mode, using a TR100-G thermoblock with orbital agitation and temperature control, provided by Optic Ivymen System. Eppendorf tubes were employed to perform the experimentation, adding the volume of solution, prepared from the corresponding salt, and the amount of adsorbent desired. Both adsorbate and adsorbent were put in contact and agitation was started. After finishing the sorption experiment, the liquid fraction was sepa rated from the solid by filtration. pH was controlled and measured throughout the experiment since it is a crucial parameter of the process [19]. Metal concentration was measured by atomic absorption spec troscopy, employing a Shimadzu AA-7000 device. The selected parameters to assess the effectiveness of sorption were MCO1NA PHYSICAL ACTIVATION 450 ºC 5 HOURS O2 (5%) + N2 (95%) NaCl NaClO2 H2O2 H2O RT CHEMICAL ACTIVATION MCO1PA MCO1CA MCO1PCA Scheme 1. Activation methods. Table 1 Denomination and description of the sorbents. Carbon naming Description MCO1NA Mesoporous carbon, non-activated MCO1PA Mesoporous carbon, physically activated MCO1CA Mesoporous carbon, chemically activated MCO1PCA Mesoporous carbon, physically activated and chemically reactivated Fig. 1. Example of the conductometric titration curve of MCO1CA. N. Conte and J.M. Gómez Separation and Purification Technology 328 (2024) 125095 4 the sorption capacity (q) and the percentage of metal adsorbed. They were calculated with Eqs. (3) and (4), by mass balance: q = (C0 − C) • V m (3) %M+ adsorbed = (C0 − C) C0 x100 (4) where q (mg/g) is the sorption capacity or the amount of cobalt adsor bed per gram of adsorbent; C0 y C (mg/L) relates to the cobalt concen tration in the liquid phase, at zero time and at t time, respectively; m (g) is the mass of adsorbent added and V (L) is the sample volume. Kinetic sorption curves were determined by measuring the metal concentration in the solution at different times and measured by AAS. Final desorption experiments were carried out in batch mode, using a Hettich Thermomixer MHR thermoblock and increasing the volume of the experiment (10 mL), to recover a higher amount of carbon. After reaching the equilibrium, solid and liquid fractions were separated by centrifugation and filtration. The liquid phase was recovered and measured by AAS, while the solid fraction was put in contact with an acid, H2SO4. Sulfuric acid was selected as an acid eluent in previous works [39]. The acid volume was reduced to a quarter (2.5 mL) to pre- concentrate the recovered solution. To evaluate the recovering yields, Eq. (5) was used, as follows: Metaldesorptionyield = Metalconcentrationinacidrecoveredsolution [mg L ] Initialmetalconcentration [mg L ] (5) Reproducibility studies with these methodologies always showed an error of <5% in adsorption capacity. 3. Results and discussion The characterization of the sorbent makes it possible to identify the surface groups that favor the sorption of metal cations. This allows the synthesis and the modifications of the sorbent to be carried out to obtain these groups and improve its sorption capacity. Furthermore, the char acterization of the sorbent could be used as a predictive tool to know its behavior in the sorption experiments. 3.1. Characterization The N2 adsorption–desorption isotherms and the pore size distributions are displayed in Fig. 2A and 2B, respectively. Table 2 dis plays the textural properties of the sorbents. The isotherm shape of the carbons belongs to type IV [40], typical of mesoporous materials, with a hysteresis loop associated with capillary condensation in the mesopores. That behavior happens when the majority of pores are wider than 40 Å [41], which is the case according to the pore size distribution (Fig. 2B). The main contribution to the specific surface area and pore volume was due to mesopores, being >75% in SBET and 93% in pore volume. Pore size distribution was similar in all the carbons, with a wide distribution within the mesoporous range. Mesoporous carbon without activation (MCO1NA) showed a maximum of around 230 Å which was maintained after the chemical activation. However, after the physical activation, the maximum was decreased to 150 Å. The physical activation process resulted in a large decrease in the mesoporous area (from 470 to 290 m2/g, 40%), while the chemical activation process hardly affected the area (from 470 to 420 m2/g, 8%). The reduction in the thermal activa tion process can be explained by the expansion and the collapse of the small pores into wider ones, due to the oxidation thermolytic process with oxygen at high temperatures. In addition, the partial burning of the carbon during the oxidation also collaborated to this decrease. However, in the chemical activation process, the mass yield was higher with a negligible mass loss (<5%,) maintaining the previous porous structure (NaClO2 and H2O2) [42]. The mesoporosity of our carbons can favor the access of oxidant molecules to the carbon surface, and the formation of different func tional groups during activation. In addition, mesoporosity also allows easy and fast access of adsorbates to the sorption sites. Activation of the carbon surface with oxidants (gas and aqueous phase oxidants) produces oxygenated functional groups such as carboxylic, carbonyl, anhydride, ethers, lactones, phenolic, etc. Such groups are responsible for the sur face acidity necessary for the sorption of metal cations in aqueous 0,0 0,2 0,4 0,6 0,8 1,0 0 200 400 600 800 1000 Q ua nt ity A ds or be d ( c m 3 / g ) Relative Pressure (p/p0) MCO1NA MCO1PA MCO1CA MCO1PCA A 100 1000 0,0 0,5 1,0 1,5 2,0 dV /d lo g( D) ( cm 3/ gA ) Pore Width (A) MCO1NA MCO1PA MCO1CA MCO1PCA B Fig. 2. N2 adsorption–desorption at 77 K (A) and pore size distribution (B) of the carbonaceous materials. Table 2 Textural characterization of carbonaceous materials. MCO1NA MCO1PA MCO1CA MCO1PCA SBET (m2/g) 590 350 540 340 Smeso (m2/g) 470 290 420 260 % Smeso 79.6 82.2 77.8 76.6 Vpore (cm3/g) 1.53 0.92 1.29 0.83 Vmeso (cm3/g) 1.43 0.89 1.24 0.80 % Vmeso 93.5 96.7 96.1 96.4 N. Conte and J.M. Gómez Separation and Purification Technology 328 (2024) 125095 5 solutions. These sorption sites are mainly located at the edges of the basal plane or on the outer surface, with a minor presence on the inner surface of the carbon. Generally, the physical oxidation process leads to an enhancement of the hydroxyl and carbonyl groups while liquid-phase oxidation (chemical process) primarily enhances the presence of car boxylic groups. Fig. 3 displays the FTIR spectra of the mesoporous carbons. The main absorption bands, centered at 3400, 1700, 1590, 1365, 1250, and 1080 cm− 1, were compatible with oxygenated groups formed during the oxidation process [43]. The absorption band at 3400 cm− 1 was assigned to O–H stretching vibration mode of hexagonal groups and adsorbed water. Phenolic groups were assigned to the band at 1080 cm− 1 due to the C–OH stretch. Bands centered at 1250–1365 cm− 1 were due to –C=O stretching vibrational mode being compatible with the presence of lac tones. Finally, the band at 1590 cm− 1 was ascribable to the C=O vi bration of carboxylate, carboxyl, or anhydride groups [44] and the band at 1700 cm− 1 to carboxyl acids (–COOH) moieties [45]. Mesoporous carbon without activation (MCO1NA) showed low in tensity in the absorption bands. The presence of carbonyl groups (1590 cm− 1) and lactones (1200 cm− 1) produced in the pre-carbonization stage (dehydration/oxidation) due to the presence of sulfuric acid stood out. Physical activation (MCO1PA) using oxygen (5% in nitrogen) produced mainly phenolic groups (1080 cm− 1) due to the high tem perature (450 ◦C) of the process, which favors the formation of weak acidic groups such as phenols [25]. Chemical activation using aqueous phase oxidants (NaClO2/H2O2, for the MCO1CA carbon) increased the concentration of the oxygen surface groups absorbing infrared radiation at 1700 cm− 1 (carboxylic acids) and 1590 cm− 1 (carboxylates), as well as the presence of hy droxyls (O–H, 3400 cm− 1). Therefore, these mesoporous carbons showed higher surface acidity, which should favor cation sorption. Finally, the physical–chemical sequential activation (MCO1PCA) further increased the presence of carboxylic and carbonyl groups (1700 and 1590 cm− 1). The previous physical activation generated hydroxyl and carbonyl groups that favored the subsequent formation of carboxylic acids, anhydrides, and more carbonyl groups in the following oxidation in aqueous medium. In addition, the first physical activation produced a greater number of edges in the carbon, areas where the formation of carboxylic groups was favored in the chemical activation. High surface acidity was achieved by using both activation methods sequentially, first physical, and then chemical. The results of the thermogravimetric analysis applied to the syn thesized sorbents are displayed in Fig. 4 as the Derivative Thermog ravimetry (-DTG). Weight loss was continuous with increasing temperature due to the decomposition of the oxygenated groups into CO2 and CO [43]. Non-activated mesoporous carbon (MCO1NA) only showed a 5% of weight loss due to the lesser number of oxygenated groups on its surface. Activated mesoporous carbons, with higher oxygenated group content, exhibited a more pronounced weight loss, decreasing from 14% for the physically activated to 27% for the phys ical–chemical sequential activation (MCO1PCA). The derivative thermogravimetry curve (Fig. 4) showed that the weight loss was more significant at low temperatures, below 150 ◦C, for the chemically activated carbons (MCO1CA and MCO1PCA). This weight loss was associated with the water adsorbed on the surface due to the higher negative charge density of these materials by the presence of carbonyl groups (C=O). The physically activated carbon (MCO1PA) showed an insignificant amount of desorbed water since the presence of carbonyl groups was lower. However, this carbon showed several peaks located between 400 and 650 ◦C, which were mainly attributed to the decomposition of phenolic hydroxyl groups [46], more common in the physically activated carbon (as it was seen in the FTIR analysis), along with the decomposition of carboxylic anhydrides and lactones [47]. The MCO1PCA mesoporous carbon showed a peak of around 300 ◦C due to the presence of carboxyl groups [47]. The small peaks located at 850 ◦C in the chemically activated carbons MCO1CA and MCO1PCA was attributed to the decomposition of carbonyl groups [43]. Zeta potential measurement is a predictive tool to understand the electrostatic interactions between the sorbent and the adsorbate. The pH value at which the zeta potential is 0 mV is known as the isoelectric point (IEP). This point is an important physicochemical parameter of many materials, such as zeolites, activated carbons, oxides, etc., and can provide a good estimate of the surface charges of the sorbent in aqueous suspension at different pH values. Sorbent particles become positively charged at pH values below IEP and negatively charged above this point [48]. Fig. 5 shows the pH dependence of the zeta potential. The upper limit to study the sorption was pH 8 since at higher pH cobalt precipitates. Concerning the physically activated mesoporous carbon (MCO1PA), it showed an isoelectric point around pH 1–1.5. The zeta potential decreased rapidly with increasing pH up to 4, from 0 mV to –26 mV, with less variation above this pH, reaching –29 mV at pH 6.5. Chemically activated mesoporous carbon (MCO1CA) showed a similar trend but with lower zeta potential values, from 0 mV to –30 mV in the pH range 1 to 4, reaching –35 mV at pH 7. The IEP was at a pH of 1–1.5. However, the physicochemical activated mesoporous carbon (MCO1PCA) showed a strong decrease of the zeta potential up to higher pH values, from − 9 mV to − 45 mV in the pH range of 1 to 6.5. According to these results, MCO1PCA should have the highest and stronger cation sorption capacity since showed a higher negative surface charge at a pH lower than 8. The number of acidic functional groups of the mesoporous carbons Fig. 3. FT-IR spectra for the synthesized mesoporous carbons. Fig. 4. DTG analysis of the mesoporous carbons. N. Conte and J.M. Gómez Separation and Purification Technology 328 (2024) 125095 6 was calculated by conductometric titration, plotting the conductivity of the suspension as a function of the volume of NaOH and NaHCO3 added (Fig. 1). The acidic surface properties of the mesoporous carbon are due to the presence of carboxyl groups, lactone, phenolic groups (O–H), etc. The total acidity, including phenols, lactonic groups, and carboxylic acids, can be determined with titration of NaOH [49]. However, the carboxylic functional group alone can be determined with titration of NaHCO3 [50,51]. Typical conductometric titration curves (Fig. 1 and Fig. 6) showed discontinuities assigned to the presence of a strong acid (HCl added to adjust pH), a weak acid (i.e., carboxylic acid groups introduced during activation), and the base excess. An example of conductometric titration curves obtained is displayed in Fig. 6. The first part up to volume V1 (first equivalent point) corresponds to the neutralization of the protons due to the HCl addition to adjust the pH. From this point, the acid groups of the mesoporous carbon were neutralized: all with NaOH solution and only the carboxylic acids with NaHCO3 solution. The difference between both were assumed to be the phenolic and lactonic groups. This neutralization continued to reach the V2 volume (second equivalent point), which can be seen by a change in the slope of the curve. The third part, above V2, corresponded to the variation in conductivity due to excess NaOH or NaHCO3. The amount of each acid group was calculated according to equation (1) and it is displayed in Fig. 7. Physical activation with oxygen (MCO1PA) led to a low acid group loading (0.99 meq/g) predominantly the phenolic and lactonic ones (0.779 meq/g). In this mesoporous carbon, according to the FTIR spectra (Fig. 3), phenolic groups were predominant. Chemical activation (MCO1CA) increased twice the presence of carboxylic acids (0.405 meq/ g), this was also observed in the FTIR spectra, with an increase in the intensity of the bands at 1590 and 1700 cm− 1. On the other hand, the amount of phenolic and lactonic groups was slightly lower than for MCO1PA carbon (0.661 meq/g) with lactones predominating as dis cussed above in the FTIR analysis. Finally, when the chemical activation was produced on a previously physically activated carbon, the total acidity increased to 1.64 meq/g (MCO1PCA). The presence of hydroxyl, phenolic and lactonic groups on the carbon surface favored the oxida tion to carboxylic acids, increasing their presence to 0.89 meq/g, four times more than MCO1PA and twice more than MCO1CA. This is in agreement with the FTIR analysis, where an increase in the presence of carboxylic groups was detected (1700 and 1590 cm− 1). On the other hand, the presence of phenolic and lactonic groups was as in the MCO1PA mesoporous carbon (0.74 meq/g), but with a predominance of lactonic groups according to the FTIR spectra. The results of the conductimetric titration were also in agreement with the zeta potential analysis (Fig. 5), with the MCO1PCA mesoporous carbon showing the highest negative charge density due to the higher concentration of carboxylic acids on its surface. Therefore, conductimetric titration has proved to be a simple method to quantify the concentration of carboxylic acids (and other groups) existing on the surface of mesoporous carbons. Since pH is a key parameter in the sorption process, it has been studied how it affects metal cations in solution, as some of them can precipitate at certain pH values. The speciation in aqueous solutions of these strategic metals was evaluated. For cobalt, previous studies demonstrated that at pH values over 8, the precipitation of Co2+ in the form of Co(OH)2 begins, as was reported by Krishman et al [52] and in previous works of our own [53]. Lithium precipitation started at pH 13, observed through own speciation studies conducted [53]. Nickel and manganese precipitation were also studied experimentally. Ni2+ pre cipitation was observed at pH values over 8.2, while Mn2+ precipitation was noticeable at pH above 8.4. The experimental data agreed with the information in the literature for nickel [54] and manganese [55]. In addition, the data gathered through experiments and literature agreed with the simulations executed using the MedusaTM chemical equilibrium simulation software (not shown). For every reason, it was mandatory to keep the system pH below those values to prevent precipitation and to study sorption. 3.2. Monometallic sorption Monometallic sorption experiments were initially performed on the mesoporous carbons. Monometallic sorption was first carried out to study the affinity of those sorbents to the metals studied (cobalt, lithium, nickel, and manganese). One of the main advantages of the mesoporous Fig. 5. Zeta potential of mesoporous carbons. Fig. 6. Conductometric titration curve of the MCO1PCA mesoporous carbon. Fig. 7. Number of acidic groups determined by conductometric titration. N. Conte and J.M. Gómez Separation and Purification Technology 328 (2024) 125095 7 structure was the sorption rate, reaching sorption equilibrium very quickly, in less than an hour, according to previous experiments [53]. Therefore, sorption experiments were carried out for 1 h, enough time to reach the equilibrium point of the mesoporous carbons. The dosage (7.5 g/L) was selected after previous experiment, so that the influence of the main variables could be assessed and determined. Metal initial con centrations were high enough (200 mg/L) to be assimilated with output stream concentrations after recovery treatments of these metals from spent batteries, such as liquid–liquid extraction or precipitation [15]. Results for the monometallic sorption experiments are shown in Fig. 8. The sorption differences achieved by the sorbents studied are note worthy. It can be seen in Fig. 8 that the chemically activated mesoporous carbons (MCO1CA and MCO1PCA) showed higher metal removal for each cation studied (>80% for MCO1PCA and divalent cations), compared to those that were not activated by chemical oxidants (MCO1NA and MCO1PA, <20%). This represented a 2 to 11-fold in crease in the sorption capacity of the chemically activated mesoporous carbon, depending on the sorbent and adsorbate, compared to the physically activated one. For example, the mesoporous carbon MCO1PCA adsorbed 5 times more cobalt, 4 times more nickel and lithium, and 11 times more manganese than the mesoporous carbon MCO1PA. Especially significant were the increases with the phys ical–chemical sequential activation (MCO1PCA), where the sorption capacities were twice as high as with the chemical-only activation (MCO1CA). Sorption capacities above 20 mg/g for Co2+, Ni2+, and Mn2+ were achieved with MCO1PCA, with removals of >80%. However, for Li+ (monovalent cation) the behavior was similar but with much lower sorption capacities. Therefore, the sorption of divalent cations (Co2+, Ni2+ and Mn2+) was substantially higher than that of the monovalent cation (Li+). The removal rates of divalent metals were 3 to 25 times higher than those of lithium. This behavior was explained through the characterization displayed above. As discussed during the characterization of the sorbents, the activation process resulted in the formation of phenolic hydroxyl (physical activation, MCO1PA), lactone, and carboxyl groups (chemical activation, MCO1CA and MCO1PCA). Because of the formation of these groups, the negative charge density on the surface of the sorbents was increased, improving the sorption ca pacity of metal cations. These cations can be fixed on the carbon surface by ion exchange with the protons present in the phenolic and acid groups [56], by interactions with the negative charge density of the oxygenated groups, with the delocalized π electrons of the aromatic compounds in the basal planes [57], by the formation of metal com plexes, and/or by chelation with heavy metal ions [25]. The presence of carboxylic groups was necessary to achieve high sorption capacities, as observed for chemically activated mesoporous carbons. The presence of these groups on the MCO1CA and MCO1PCA mesoporous carbons was detected by FTIR analysis (Fig. 3) and conductometric titration (Fig. 7). However, with physical activation (MCO1PA), the phenolic groups were in the majority and the sorption capacity was lower, but higher than in the mesoporous carbon without activation (MCO1NA). One factor to consider was that the charge of the cations (Co, Ni, and Mn) was +2 so that nearby negatively charged groups would be necessary to attach the metal cation to the surface. ClO2 - was able to react with the side groups or with the nonaromatic edges of the basal planes but not with the aromatic ring system [58]. Therefore, the oxygen containing surface groups, such as carboxylic acids, were concentrated on the outer surface or edge of the basal plane, favoring the sorption of these cations with charge +2. However, for the monovalent cation (Li+) the sorption capacity was very low. This order in the sorption capacity for these cations (Ni2+ > Co2+ > Mn2+ ≫>≫ Li+) was explained for the higher charge (+2 versus + 1), greater molecular weights (MWNi = 58.7, MWCo = 58.3, MWMn = 54.9 versus MWLi = 6.9) and larger hydrated ionic radius (rMn = 4.38 Å > rCo = 4.23 Å > rNi = 4.04 Å > rLi = 3.82 Å) [59]. In addition, electronegativity is a key parameter as atoms can Fig. 8. Monometallic sorption experiments for cobalt (A), lithium (B), nickel (C) and manganese (D). Conditions: [M+] = 200 mg/L, carbon dosage = 7.5 g/L, T = 25 ◦C, speed = 1100 rpm, time = 1 h. N. Conte and J.M. Gómez Separation and Purification Technology 328 (2024) 125095 8 attract shared electrons. According to the Pauling scale [60] the order of electronegativity is Ni (1.91) ≥ Co (1.88) > Mn (1.55) > Li (0.98) [61] which is the same as the order of sorption capacity in these mesoporous carbons. Nickel and cobalt will establish stronger bonds than lithium, as lithium will attract less strongly (low electronegativity) the electrons (negative charge density) of the different oxygenated functional groups of the activated mesoporous carbon. The driving force for the sorption was explained by electrostatic interactions between the metal ions and mesoporous carbon sorption sites, hence higher charges and greater molecular weights led to larger sorption capacities towards these diva lent metals. Among the divalent cations, no significant differences be tween the sorption capacities were found. Explained in terms of electrostatic interactions, as the charge, hydrated ionic radius, molec ular weights, and electronegativity were similar, the sorption uptakes were similar in accordance. Mn2+, Co2+, and Ni2+ are three transition metals, almost consecutive in the Periodic Table, then their chemical behavior in adsorption was similar, so the adsorption capacities were considerably alike. Therefore, there were no selective sorption sites for the different metals, and the sorption of metals depended on the func tional groups present on the sorbent surface. A correlation was observed between the number of acid groups (either carboxylic acids or other groups) and the sorption of these metals, as the conductometric titration showed that the chemically activated mesoporous carbons had a large amount of these groups, especially the chemically activated carbon previously physically activated. Since precipitation is an undesired event in terms of studying the sorption process, was evaluated the pH throughout the whole sorption process. Table 3 shows the initial and final pHs of each sorption experiment. As indicated in the previous sections, the worst-case sce nario in terms of precipitation was with cobalt, being mandatory to work below pH 8 to avoid precipitation. Nickel, manganese, and lithium gave a larger scope for action, raising the threshold to 8.2, 8.4, and 13, respectively. The non-chemically activated mesoporous carbons acidi fied the medium mainly due to the presence of phenolic groups (pKa = 9.9) together with a very low concentration of other oxygenated groups. On the other hand, the chemically activated carbons showed a final pH in the sorption experiments between 4.5–6, although the pKa of the carboxylic acids is around 4–5, being stronger acids than phenol. The surface chemistry of these chemically activated mesoporous carbons was complex, showing different functional groups capable of binding and releasing protons, such as carbonyl, anhydrides, and lactones, whose oxygens can attract protons, buffering the pH [62]. This buffering effect is well known, as the carbon takes on an amphoteric character when oxidized, presenting acidic and basic centers [25]. This behavior, keeping the sorption pH around 5–6 increased the efficiency of the sorption process of metal cations due to the increase of the negative charge density on the surface of the chemically activated mesoporous carbons, as was observed in the zeta potential analysis (Fig. 3). Table 4 shows a comparison between the sorption capacities of this work and those found in similar studies in the literature, for the four metals evaluated: cobalt, lithium, nickel and manganese. It can be observed that our physico-chemically activated carbon shows higher adsorption capacities than other carbonaceous sorbents, with very fast kinetics. It is important to note that comparisons are not always easy because studies tend to use low concentrations of metals in aqueous solutions. Real wastewater from leachates from spent lithium- ion battery recovery processes is generally not simulated. In general, other carbonaceous materials showed less effective adsorption results than this work, with some exceptions [69]. In the case of lithium, only lithium-specific ion sieves and zeolites were able to adsorb lithium in significant amounts. 3.3. Multimetallic sorption Once the results for the monometallic sorption experiments were gathered, the next step was to evaluate the combination of those metals in a solution, more orientated to a real wastewater problem where none of those metals are separated in real effluents. Wastewater from a spent lithium-ion battery recovery process was emulated. The concentration of these metals in the outflow after treatment of the leach solution by hydrometallurgical or pyrometallurgical processes [23], liquid–liquid extraction, precipitation, etc., is in the ppm range (20–400 ppm), as although high removal efficiencies are achieved, they are not complete (95% for manganese, 90% for cobalt, 97% for nickel [70]). The con centrations used to prepare the wastewater simulated were 200 mg/L of cobalt, 33 mg/L of lithium, 67 mg/L of nickel, and 33 mg/L of manga nese. These ratios agree with the concentrations of these metals in spent lithium-ion battery samples analyzed in our laboratory. Kinetic experiments were carried out to determine whether the equilibration time was influenced by the presence of various cations in the solution. Fig. 9 displays, as an example, the kinetic curves obtained for Co2+, Li+, Ni2+, and Mn2+ on the MCO1PCA mesoporous carbon, the best sorbent selected. Fig. 9 only shows a fraction of the whole-time interval, since no changes in sorption capacity were found between 100 and 1300 min. It can be seen in Fig. 9 the sorption process was considerably fast, reaching equilibrium within the first 15 min, due to its mesoporous structure. Similar results were obtained with the mono metallic solutions. The average pore size for the MCO1PCA carbon was 244 Å, remarkably bigger than the hydrated ionic radius of those cat ions, approximately 60 times greater. For that reason, steric hindrance and other diffusional limitations are not occurring. Both external and internal diffusional resistances can be easily overcome thanks to a proper agitation system and the control of the porosity of the meso porous carbon, achieved via the template synthesis method. The sorp tion kinetics curves for all the metals were similar to each other, with a rapid increase until the maximum sorption capacity was reached. All of them were removed at the same time, which implies that, despite the different initial concentrations, the activated material had sufficient capacity to adsorb all metals at the same time. The sorption capacity depended on the metal concentration in the initial aqueous solution, being higher for cobalt than for other metals. No selectivity with sorp tion time was detected, apart from the almost non-existent lithium sorption already observed. The removal degree was like those found for monometallic solutions. The metal removal at equilibrium was 69, 66, and 63% for Ni2+, Mn2+, and Co2+, respectively. This removal was considerably higher than that of Li+ (9%). Fig. 10 displays the results of the multimetallic sorption at the equilibrium time for all the sorbents. As in the experiments with monometallic solutions, physically activated carbon (MCO1PA) showed a low removal degree (<10%), increasing with chemical activation (<35%, MCO1CA), and reaching the highest values (>60%) with sequentially activated mesoporous carbon (MCO1PCA). Competitive sorption was observed with significant sorption of Co2+, Ni2+, and Mn2+ cations but negligible sorption of Li+. The selectivity of sorption was due to the charge of the cations, with divalent but not monovalent cations being adsorbed. The sorption capacity obtained was conditioned by the concentration of the divalent cation in the aqueous solution ([Co2+]≫ [Ni2+] > [Mn2+]). Co2+ showed the highest sorption capacity (18 mg/ g), as it was the one with the highest concentration, followed by Ni2+ (7 mg/g) and then Mn2+ (3.0 mg/g) The ratio between the sorption Table 3 Variation of the pH in the monometallic sorption experiments. Initial pH Final pH MCO1NA MCO1PA MCO1CA MCO1PCA Cobalt (Co2+) 5.7 3,6 3,1 5,2 4,9 Lithium (Li+) 5.1 3,8 3,3 5,9 5,6 Nickel (Ni2+) 5.6 3,6 3,1 5,0 4,8 Manganese (Mn2+) 5.8 3,6 3,1 5,0 4,8 N. Conte and J.M. Gómez Separation and Purification Technology 328 (2024) 125095 9 capacities of the different metals (Co2+/Ni2+:2.6, Co2+/Mn2+:6.0 and Ni2+/Mn2+:2.3 for the MCO1PCA) were very similar to those of the initial solution concentrations (Co2+/Ni2+: 3.0, Co2+/Mn2+: 6.1 and Ni2+/Mn2+: 2.0), indicating that the sorption capacity was conditioned by the concentrations of each divalent metal in the initial solution. Thus, not only the competitive effect towards the available sorption sites was present in the system, but also repulsive electrostatic forces between the cations adsorbed can be found [68]. Furthermore, the sum of the sorp tion capacities for the chemically activated mesoporous carbons was the same as the maximum achieved in the monometallic sorption. For the mesoporous carbon MCO1PCA the maximum capacity was 24.2 mg/g (Ni2+), and the sum was 25.1 mg/g (Co2++Ni2++Mn2+) and for MCO1CA it was 13.1 mg/g (Ni2+), the sum was 12.2 mg/g (Co2++Ni2++Mn2+). These maximum sorption capacities were limited by the concentration of carboxylic acids on the surface of the meso porous carbon, as their increase significantly enhanced the removal of bivalent cations. Therefore, saturation was reached for the sorbent doses used and will need to be increased to achieve complete removal of co balt, nickel, and manganese from the wastewater (important to remember that the dose was selected in order to be able to see the in fluence of the variables studied). In the first section of the work, we stated that research and industrial work should focus on two objectives: purifying the wastewater polluted by the incorrect disposal of these metals and fully recovering the metals for reuse. A final experiment was conducted, aiming for both goals. The carbon dose of MCO1PCA was notably increased (20 g/L) to thoroughly remove the metal cations from the aqueous solution. After sorption, the desorption step was performed with H2SO4, using a quarter of the initial volume. In this way, the metal recovered was preconcentrated for the subsequent recovery treatments. Fig. 11 shows the results of the sorption and subsequent desorption experiments. The number of surface sorption sites increased as the dose Table 4 Comparison of metal removal adsorption capacities of different adsorbents. Adsorbent Metal Dosage (g/L) Initial concentration (mg/L) qmax (mg/g) Equilibrium time (min) Reference Ordered mesoporous carbon (PCK) Co2+ 0.3 0.7 1.6 6 [63] Dolomite Co2+ 0.1 10 2.8 120 [64] Microporous F-400 activated carbon Co2+ 12.5 20 3 5760 [26] Lithium ion-sieve adsorbent Co2+ 2 2000 0.08 20 [65] Physicochemically activated mesoporous carbon Co2+ 7.5 200 22.8 15 This work K+ modified zeolite Li+ 1.5 10 17 15 [66] Lithium ion-sieve adsorbent Li+ 2 2500 31.2 60 [65] Physicochemically activated mesoporous carbon Li+ 7.5 200 4.6 15 This work Activated carbon from almond husk Ni2+ 5 25 4.9 50 [67] Acid-Treated Activated Carbon Ni2+ 50 100 5.8 N/A [68] Lithium ion-sieve adsorbent Ni2+ 2 2000 0.3 20 [65] Physicochemically activated mesoporous carbon Ni2+ 7.5 200 24.2 15 This work Activated carbon from Birbira leaves Mn2+ 4 20 3.4 120 [27] Iron oxide modified activated carbon Mn2+ 0.1 350 30 30 [69] Acid-Treated Activated Carbon Mn2+ 50 100 1.5 N/A [68] Physicochemically activated mesoporous carbon Mn2+ 7.5 200 21 15 This work Fig. 9. Kinetic curves for sorption of battery metals. Conditions: [Co2+] = 200 mg/L, [Li+] = 33 mg/L, [Ni2+] = 67 mg/L, [Mn2+] = 33 mg/L, carbon dosage = 7.5 g/L, T = 25 ◦C, speed = 1100 rpm. Fig. 10. Multimetallic sorption experiments on mesoporous carbons. Sorption capacities (A) and metal removal (B). Conditions: [Co2+] = 200 mg/L, [Li+] = 33 mg/ L, [Ni2+] = 67 mg/L, [Mn2+] = 33 mg/L, carbon dosage = 7.5 g/L, T = 25 ◦C, speed = 1100 rpm, time = 1 h. N. Conte and J.M. Gómez Separation and Purification Technology 328 (2024) 125095 10 did so, achieving almost complete removal of the metals. As can be seen in Fig. 11A, 98% of Co2+, Ni2+, and Mn2+ were successfully removed with increasing dosage, while lithium remained in the solution, with a low sorption capacity (qLi = 0.33 mg/g). The subsequent desorption showed promising results for metal recovery and pre-concentration. By reducing the desorption volume (2.5 mL of H2SO4 0.3 M) to the quarter, more concentrated metal aqueous streams can be achieved. The desorb aqueous solution reached a higher concentration of cobalt, nickel, and manganese (Fig. 11B), compared to the initial concentrations of the simulated solution. Metal concentrations twice as high as the initial ones were found ([Co2+] = 455 mg/L, [Ni2+] = 140 mg/L, and [Mn2+] = 70 mg/L). These higher concentrations are more suitable for further re covery processes. The degree of lithium recovery was lower but reached a significant level, around 80% from the initial concentration, however not comparable to the other metals, which were over 200% from the initial concentration. 4. Conclusions Fast kinetics and high sorption capacities of metals present in spent lithium-ion batteries, such as cobalt, nickel, and manganese, have been achieved with the chemical activation of mesoporous carbons. Chemical activation carried out under mild conditions (room temperature), with H2O2 and NaClO2, increases the negative charge density on the surface of the mesoporous carbons due to the significant presence of carboxylic acids and carbonyl groups. The most effective sorption sites for metal cations are the carboxylic acids formed during the chemical activation on the edges of the basal plane or over the outer surface. The concen tration of carboxylic acids (and other groups) can be quantified by a simple method such as conductimetric titration and is a good predictive tool for the adsorption process of these metals, complementing zeta potential measurements and FTIR analysis. Sequential activation, first physically and then chemically, leads to greatly improved Co2+, Ni2+, and Mn2+ sorption capacity by promoting the development of carboxylic acid groups on the surface of the mesoporous carbon. This mesoporous carbon reached the highest sorption capacities for Co2+ (22.3 mg/g), Ni2+ (24.2 mg/g), and Mn2+ (21.0 mg/g). In multimetallic solutions it removed over 60% of these metals at a dosage of 7.5 g/L, increasing to 100% for a dosage of 20 g/L in only 60 min. These sorption sites are selective to divalent cations (Co2+, Ni2+, and Mn2+) and allow the total removal of these metals from the wastewater by selecting the appro priate adsorbent dosage. Sorption capacity is conditioned by the con centrations of each metal in the initial solution due to competitive sorption. The subsequent desorption has shown promising results for the pre-concentration and recovery of these metals by adsorp tion–desorption cycles. CRediT authorship contribution statement N. Conte: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. J.M. Gómez: Conceptualization, Methodology, Formal analysis, Writing – original draft, Writing – review & editing, Supervision, Project administration, Funding acquisition. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Data availability Data will be made available on request. Acknowledgements The authors thank the Correlation Spectroscopy Research Centre of the Complutense University of Madrid for helping with the character ization. We also thank the company Minería Urbana S.L. for supplying the black mass samples. Funding This work was supported by the contracts of research assistant from the Community of Madrid through the program “Garantía Juvenil” [PEJ- 2020-AI/IND-17675]. References [1] M. & Eddy, Ingeniería de aguas residuales, McGraw-Hill, 2001. [2] S. Gunatilake, Methods of removing heavy metals from industrial wastewater, J. Multidicipl. Eng. Sci. Stud. 1 (2015). [3] M. Rafique, S. Hajra, M.B. Tahir, S.S.A. Gillani, M. Irshad, A review on sources of heavy metals, their toxicity and removal technique using physico-chemical processes from wastewater, Environ. Sci. Pollut. Res. 29 (2022) 16772–16781, https://doi.org/10.1007/s11356-022-18638-9. [4] K.H.H. Aziz, F.S. Mustafa, K.M. Omer, S. Hama, R. Fayaq Hamarawf, K. Othman Rahman, Heavy metal pollution in the aquatic environment: efficient and low-cost Fig. 11. Metal sorption (A) and desorption (B) results for cobalt, lithium, nickel, and manganese. Sorption conditions: [Co2+] = 200 mg/L, [Li+] = 33 mg/L, [Ni2+] = 67 mg/L, [Mn2+] = 33 mg/L, carbon dosage = 20 g/L, T = 25 ◦C, speed = 650 rpm, time = 1 h, desorption conditions: time = 1 h, [H2SO4] = 0.3 M, acid volume = 2.5 mL. N. Conte and J.M. Gómez http://refhub.elsevier.com/S1383-5866(23)02003-8/h0005 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0010 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0010 https://doi.org/10.1007/s11356-022-18638-9 Separation and Purification Technology 328 (2024) 125095 11 removal approaches to eliminate their toxicity: a review, RSC Adv. 13 (2023) 17595–17610, https://doi.org/10.1039/D3RA00723E. [5] F. Fu, Q. Wang, Removal of heavy metal ions from wastewaters: a review, J. Environ. Manage. 92 (2011) 407–418, https://doi.org/10.1016/j. jenvman.2010.11.011. [6] G. Blengini, C. Latunussa, U. Eynard, C. Matos, K. Georgitzikis, C. Pavel, S. Carrara, L. Mancini, M. Unguru, D. Blagoeva, F. Mathieux, D. Pennington, Study on the EU’s list of Critical Raw Materials (2020) Final Report, 2020. 10.2873/11619. [7] J.M. da Cunha, L. Klein, M.M. Bassaco, E.H. Tanabe, D.A. Bertuol, G.L. Dotto, Cobalt recovery from leached solutions of lithium-ion batteries using waste materials as adsorbents, Can. J. Chem. Eng. 93 (2015) 2198–2204, https://doi.org/ 10.1002/cjce.22331. [8] D.L. Thompson, J.M. Hartley, S.M. Lambert, M. Shiref, G.D.J. Harper, E. Kendrick, P. Anderson, K.S. Ryder, L. Gaines, A.P. Abbott, The importance of design in lithium ion battery recycling – a critical review, Green Chem. 22 (2020) 7585–7603, https://doi.org/10.1039/D0GC02745F. [9] K.C. Michaels, The Role of Critical Minerals in Clean Energy Transitions, in: 2021, p. 7. [10] B. Swain, Recovery and recycling of lithium: a review, Sep. Purif. Technol. 172 (2017) 388–403, https://doi.org/10.1016/j.seppur.2016.08.031. [11] P.R. Gismonti, J.F. Paulino, J. Afonso, Recovery of metals from electroactive components of spent Ni-MH batteries after leaching with formic acid, Detritus (2021) 68, 10.31025/2611-4135/2021.14063. [12] J. Sterba, A. Krzemień, P. Riesgo Fernández, C. Escanciano García-Miranda, G. Fidalgo Valverde, Lithium mining: accelerating the transition to sustainable energy, Resour. Policy 62 (2019) 416–426, https://doi.org/10.1016/j. resourpol.2019.05.002. [13] J.-G. Wang, D. Jin, H. Liu, C. Zhang, R. Zhou, C. Shen, K. Xie, B. Wei, All- manganese-based Li-ion batteries with high rate capability and ultralong cycle life, Nano Energy 22 (2016) 524–532, https://doi.org/10.1016/j.nanoen.2016.02.051. [14] J. Piątek, S. Afyon, T.M. Budnyak, S. Budnyk, M.H. Sipponen, A. Slabon, Sustainable Li-ion batteries: chemistry and recycling, Adv. Energy Mater. 11 (2021) 2003456, https://doi.org/10.1002/aenm.202003456. [15] E. Asadi Dalini, G.h. Karimi, S. Zandevakili, Treatment of valuable metals from leaching solution of spent lithium-ion batteries, Miner. Eng. 173 (2021), 107226, https://doi.org/10.1016/j.mineng.2021.107226. [16] X. Chen, H. Ma, C. Luo, T. Zhou, Recovery of valuable metals from waste cathode materials of spent lithium-ion batteries using mild phosphoric acid, J. Hazard. Mater. 326 (2017) 77–86, https://doi.org/10.1016/j.jhazmat.2016.12.021. [17] M. Renu, K. Agarwal Singh, Heavy metal removal from wastewater using various adsorbents: a review, J. Water Reuse Desalin. 7 (2016) 387–419, https://doi.org/ 10.2166/wrd.2016.104. [18] G. Zhang, B. Lei, S. Chen, H. Xie, G. Zhou, Activated carbon adsorbents with micro- mesoporous structure derived from waste biomass by stepwise activation for toluene removal from air, J. Environ. Chem. Eng. 9 (2021), https://doi.org/ 10.1016/j.jece.2021.105387. [19] E. Worch, Adsorption Technology in Water Treatment: Fundamentals, Processes, and Modeling, Walter de Gruyter, 2012. [20] J. Goscianska, A. Olejnik, I. Nowak, M. Marciniak, R. Pietrzak, Stability analysis of functionalized mesoporous carbon materials in aqueous solution, Chem. Eng. J. 290 (2016) 209–219, https://doi.org/10.1016/j.cej.2016.01.060. [21] M. Abbas, S. Kaddour, M. Trari, Kinetic and equilibrium studies of cobalt adsorption on apricot stone activated carbon, J. Ind. Eng. Chem. 20 (2014) 745–751, https://doi.org/10.1016/j.jiec.2013.06.030. [22] S. Mehdipour-Ataei, E. Aram, Mesoporous carbon-based materials: a review of synthesis, modification, and applications, Catalysts 13 (2023) 2, https://doi.org/ 10.3390/catal13010002. [23] M. Barczak, K. Michalak-Zwierz, K. Gdula, K. Tyszczuk-Rotko, R. Dobrowolski, A. Dąbrowski, Ordered mesoporous carbons as effective sorbents for removal of heavy metal ions, Microporous Mesoporous Mater. 211 (2015) 162–173, https:// doi.org/10.1016/j.micromeso.2015.03.010. [24] A.B. Fuertes, D.M. Nevskaia, Control of mesoporous structure of carbons synthesised using a mesostructured silica as template, Microporous Mesoporous Mater. 62 (2003) 177–190, https://doi.org/10.1016/S1387-1811(03)00403-7. [25] A. Rehman, M. Park, S.-J. Park, Current progress on the surface chemical modification of carbonaceous materials, Coatings 9 (2019) 103, https://doi.org/ 10.3390/coatings9020103. [26] J.M. Gómez, E. Díez, I. Bernabé, P. Sáez, A. Rodríguez, Effective adsorptive removal of cobalt using mesoporous carbons synthesized by silica gel replica method, Environ. Process. 5 (2018) 225–242, https://doi.org/10.1007/s40710- 018-0304-9. [27] D.A.A. Mengistie, Adsorption of Mn(II) ions from wastewater using activated carbon obtained from Birbira (Militia ferruginea) leaves, Global J. Sci. Front. Res. 12 (2012) 5–12. [28] M. Abbas, Modeling of adsorption isotherms of heavy metals onto Apricot stone activated carbon: two-parameter models and equations allowing determination of thermodynamic parameters, Mater. Today:. Proc. 43 (2020) 3359–3364, https:// doi.org/10.1016/j.matpr.2020.05.320. [29] S.A.S. Ahmed, L.B. Khalil, T.h. El-Nabarawy, Modified activated carbons from tea waste for the removal of Cu(II) and Pb(II) Ions, Egypt. J. Chem. 58 (2015) 183–202. [30] S. Hu, Y.-L. Hsieh, Preparation of activated carbon and silica particles from rice straw, ACS Sustain. Chem. Eng. 2 (2014) 726–734, https://doi.org/10.1021/ sc5000539. [31] G.M. Mohamed, S.A. Sayed Ahmed, N.A. Fathy, Activated Carbon doped with silica and nitrogen as novel adsorbent for enhancing adsorption capacity of Cr(VI), Chem. Africa (2023), https://doi.org/10.1007/s42250-023-00709-0. [32] M.Y. Abdelnaeim, I.Y. El Sherif, A.A. Attia, N.A. Fathy, M.F. El-Shahat, Impact of chemical activation on the adsorption performance of common reed towards Cu(II) and Cd(II), Int. J. Miner. Process. 157 (2016) 80–88, https://doi.org/10.1016/j. minpro.2016.09.013. [33] J. Galán, A. Rodríguez, J.M. Gómez, S.J. Allen, G.M. Walker, Reactive dye adsorption onto a novel mesoporous carbon, Chem. Eng. J. 219 (2013) 62–68, https://doi.org/10.1016/j.cej.2012.12.073. [34] Z. Jiang, Y. Liu, X. Sun, F. Tian, F. Sun, C. Liang, W. You, C. Han, C. Li, Activated carbons chemically modified by concentrated H2SO4 for the adsorption of the pollutants from wastewater and the dibenzothiophene from fuel oils, Langmuir 19 (2003) 731–736, https://doi.org/10.1021/la020670d. [35] P. Wamea, M.L. Pitcher, J. Muthami, A. Sheikhi, Nanoengineering cellulose for the selective removal of neodymium: towards sustainable rare earth element recovery, Chem. Eng. J. 428 (2022), 131086, https://doi.org/10.1016/j.cej.2021.131086. [36] C. Fraschini, G. Chauve, J. Bouchard, TEMPO-mediated surface oxidation of cellulose nanocrystals (CNCs), Cellulose 24 (2017), https://doi.org/10.1007/ s10570-017-1319-5. [37] M.-M. Roman, I. Ciurdea, C. Lazarev, L. Bulgariu, Determination of mineral acids concentration from mixtures by condutometric titration, Bull. Polytechnic Institute of Iasi, Sec. Chem. Chem. Eng. 62 (2016) 21–31. [38] E.J. Foster, R. Moon, U. Agarwal, M. Bortner, J. Bras, S. Camarero Espinosa, K. Chan, M. Clift, E. Cranston, S. Eichhorn, D. Fox, W. Hamad, L. Heux, B. Jean, M. Korey, W. Nieh, K. Ong, M. Reid, S. Renneckar, J. Youngblood, Current characterization methods for cellulose nanomaterials (2018). [39] N. Conte, E. Díez, B. Almendras, J.M. Gómez, A. Rodríguez, Sustainable recovery of cobalt from aqueous solutions using an optimized mesoporous carbon, J. Sustain. Metall. (2023), https://doi.org/10.1007/s40831-022-00644-3. [40] M. Thommes, K. Kaneko, A.V. Neimark, J.P. Olivier, F. Rodriguez-Reinoso, J. Rouquerol, K.S.W. Sing, Physisorption of gases, with special reference to the evaluation of surface area and pore size distribution (IUPAC Technical Report), Pure Appl. Chem. 87 (2015) 1051–1069, https://doi.org/10.1515/pac-2014-1117. [41] M. Eddaoudi, Characterization of Porous Solids and Powders: Surface Area, Pore Size and Density By S. Lowell (Quantachrome Instruments, Boynton Beach), J.E. Shields (C.W. Post Campus of Long Island University), M.A. Thomas, M. Thommes (Quantachrome In-struments). Kluwer Academic Publishers: Dordrecht, The Netherlands. 2004. xiv + 348 pp. $159.00. ISBN 1-4020-2302-2., J. Am. Chem. Soc. 127 (2005) 14117–14117. 10.1021/ja041016i. [42] M. Sánchez, A. Macías-García, D. Angeles, E.M. Cuerda-Correa, J. Gañán-Gómez, A. Nadal-Gisbert, Preparation of activated carbons previously treated with hydrogen peroxide: study of their porous texture, Appl. Surf. Sci. 252 (2006) 5984–5987, https://doi.org/10.1016/j.apsusc.2005.11.022. [43] P.E. Fanning, M.A. Vannice, A DRIFTS study of the formation of surface groups on carbon by oxidation, Carbon 31 (1993) 721–730, https://doi.org/10.1016/0008- 6223(93)90009-Y. [44] S. Biniak, A. Światkowski, M. Pakula, Electrochemical studies of phenomena at active carbon-electrolyte solution interfaces, Chem. Phys. Carbon 27 (2000) 125–225. [45] K. Le Van, T.T. Luong Thi, Activated carbon derived from rice husk by NaOH activation and its application in supercapacitor, Prog. Nat. Sci.: Mater. Int. 24 (2014) 191–198, https://doi.org/10.1016/j.pnsc.2014.05.012. [46] S. Kundu, Y. Wang, W. Xia, M. Muhler, Thermal stability and reducibility of oxygen-containing functional groups on multiwalled carbon nanotube surfaces: a quantitative high-resolution XPS and TPD/TPR study, J. Phys. Chem. C 112 (2008) 16869–16878, https://doi.org/10.1021/jp804413a. [47] L.A. Alves, A.H. de Castro, F.G. de Mendonça, J.P. de Mesquita, Characterization of acid functional groups of carbon dots by nonlinear regression data fitting of potentiometric titration curves, Appl. Surf. Sci. 370 (2016) 486–495, https://doi. org/10.1016/j.apsusc.2016.02.128. [48] B. Zhu, P. Xia, W. Ho, J. Yu, Isoelectric point and adsorption activity of porous g- C3N4, Appl. Surf. Sci. 344 (2015) 188–195, https://doi.org/10.1016/j. apsusc.2015.03.086. [49] L. Mahardiani, S. Saputro, F. Baskoro, N.M. Zinki, M. Taufiq, Facile synthesis of carboxylated activated carbon using green approach for water treatment, IOP Conf. Ser.: Mater. Sci. Eng. 578 (2019), 012003, https://doi.org/10.1088/1757-899X/ 578/1/012003. [50] Y.S. Kim, C.R. Park, One-pot titration methodology for the characterization of surface acidic groups on functionalized carbon nanotubes, Carbon 96 (2016) 729–741, https://doi.org/10.1016/j.carbon.2015.08.078. [51] H.P. Boehm, Some aspects of the surface chemistry of carbon blacks and other carbons, Carbon 32 (1994) 759–769, https://doi.org/10.1016/0008-6223(94) 90031-0. [52] A. Krishnan, T. Anirudhan, Kinetic and equilibrium modelling of cobalt(II) adsorption onto bagasse pith based sulphurised activated carbon, Chem. Eng. J. 137 (2008) 257–264, https://doi.org/10.1016/j.cej.2007.04.029. [53] N. Conte, J.M. Gómez, E. Díez, P. Sáez, J.I. Monago, A. Espinosa, A. Rodríguez, Sequential separation of cobalt and lithium by sorption: sorbent set selection, Sep. Purif. Technol. 303 (2022), 122199, https://doi.org/10.1016/j. seppur.2022.122199. [54] A. Habib, S. Serniabad, M.S. Khan, R. Islam, M. Chakraborty, A. Nargis, M. E. Quayum, M.A. Alam, V. Rapozzi, M. Tabata, Kinetics and mechanism of formation of nickel(II)porphyrin and its interaction with DNA in aqueous medium, J. Chem. Sci. 133 (2021) 83, https://doi.org/10.1007/s12039-021-01945-y. N. Conte and J.M. Gómez https://doi.org/10.1039/D3RA00723E https://doi.org/10.1016/j.jenvman.2010.11.011 https://doi.org/10.1016/j.jenvman.2010.11.011 https://doi.org/10.1002/cjce.22331 https://doi.org/10.1002/cjce.22331 https://doi.org/10.1039/D0GC02745F https://doi.org/10.1016/j.seppur.2016.08.031 https://doi.org/10.1016/j.resourpol.2019.05.002 https://doi.org/10.1016/j.resourpol.2019.05.002 https://doi.org/10.1016/j.nanoen.2016.02.051 https://doi.org/10.1002/aenm.202003456 https://doi.org/10.1016/j.mineng.2021.107226 https://doi.org/10.1016/j.jhazmat.2016.12.021 https://doi.org/10.2166/wrd.2016.104 https://doi.org/10.2166/wrd.2016.104 https://doi.org/10.1016/j.jece.2021.105387 https://doi.org/10.1016/j.jece.2021.105387 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0095 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0095 https://doi.org/10.1016/j.cej.2016.01.060 https://doi.org/10.1016/j.jiec.2013.06.030 https://doi.org/10.3390/catal13010002 https://doi.org/10.3390/catal13010002 https://doi.org/10.1016/j.micromeso.2015.03.010 https://doi.org/10.1016/j.micromeso.2015.03.010 https://doi.org/10.1016/S1387-1811(03)00403-7 https://doi.org/10.3390/coatings9020103 https://doi.org/10.3390/coatings9020103 https://doi.org/10.1007/s40710-018-0304-9 https://doi.org/10.1007/s40710-018-0304-9 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0135 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0135 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0135 https://doi.org/10.1016/j.matpr.2020.05.320 https://doi.org/10.1016/j.matpr.2020.05.320 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0145 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0145 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0145 https://doi.org/10.1021/sc5000539 https://doi.org/10.1021/sc5000539 https://doi.org/10.1007/s42250-023-00709-0 https://doi.org/10.1016/j.minpro.2016.09.013 https://doi.org/10.1016/j.minpro.2016.09.013 https://doi.org/10.1016/j.cej.2012.12.073 https://doi.org/10.1021/la020670d https://doi.org/10.1016/j.cej.2021.131086 https://doi.org/10.1007/s10570-017-1319-5 https://doi.org/10.1007/s10570-017-1319-5 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0185 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0185 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0185 https://doi.org/10.1007/s40831-022-00644-3 https://doi.org/10.1515/pac-2014-1117 https://doi.org/10.1016/j.apsusc.2005.11.022 https://doi.org/10.1016/0008-6223(93)90009-Y https://doi.org/10.1016/0008-6223(93)90009-Y http://refhub.elsevier.com/S1383-5866(23)02003-8/h0220 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0220 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0220 https://doi.org/10.1016/j.pnsc.2014.05.012 https://doi.org/10.1021/jp804413a https://doi.org/10.1016/j.apsusc.2016.02.128 https://doi.org/10.1016/j.apsusc.2016.02.128 https://doi.org/10.1016/j.apsusc.2015.03.086 https://doi.org/10.1016/j.apsusc.2015.03.086 https://doi.org/10.1088/1757-899X/578/1/012003 https://doi.org/10.1088/1757-899X/578/1/012003 https://doi.org/10.1016/j.carbon.2015.08.078 https://doi.org/10.1016/0008-6223(94)90031-0 https://doi.org/10.1016/0008-6223(94)90031-0 https://doi.org/10.1016/j.cej.2007.04.029 https://doi.org/10.1016/j.seppur.2022.122199 https://doi.org/10.1016/j.seppur.2022.122199 https://doi.org/10.1007/s12039-021-01945-y Separation and Purification Technology 328 (2024) 125095 12 [55] M. Pipí̌ska, Z. Trajteľová, M. Horník, V. Frišták, Evaluation of Mn bioaccumulation and biosorption by bacteria isolated from spent nuclear fuel pools using 54Mn as a radioindicator, Radiochim. Acta 106 (2017), https://doi.org/10.1515/ract-2017- 2836. [56] J. Jaramillo, V. Gómez-Serrano, P.M. Álvarez, Enhanced adsorption of metal ions onto functionalized granular activated carbons prepared from cherry stones, J. Hazard. Mater. 161 (2009) 670–676, https://doi.org/10.1016/j. jhazmat.2008.04.009. [57] S. Sato, K. Yoshihara, K. Moriyama, M. Machida, H. Tatsumoto, Influence of activated carbon surface acidity on adsorption of heavy metal ions and aromatics from aqueous solution, Appl. Surf. Sci. 253 (2007) 8554–8559, https://doi.org/ 10.1016/j.apsusc.2007.04.025. [58] P. Vinke, M. Eijk, M. Verbree, A.F. Voskamp, H.V. Bekkum, Modification of the surfaces of a gasactivated carbon and a chemically activated carbon with nitric acid, hypochlorite, and ammonia (1994), https://doi.org/10.1016/0008-6223(94) 90089-2. [59] E.R. Nightingale, Phenomenological theory of ion solvation. Effective radii of hydrated ions, J. Phys. Chem. 63 (1959) 1381–1387, https://doi.org/10.1021/ j150579a011. [60] L. Pauling, The Nature of the Chemical Bond and the Structure of Molecules and Crystals: An Introduction to Modern Structural Chemistry, Cornell University Press, 1960. [61] C. Tantardini, A.R. Oganov, Thermochemical electronegativities of the elements, Nat. Commun. 12 (2021) 2087, https://doi.org/10.1038/s41467-021-22429-0. [62] B.E. Reed, J.N. Jensen, M.R. Matsumoto, Acid-base characteristics of powdered- activated-carbon surfaces, J. Environ. Eng. 119 (1993) 585–590, https://doi.org/ 10.1061/(ASCE)0733-9372(1993)119:3(585). [63] M.N. Siddiqui, B. Chanbasha, A.A. Al-Arfaj, T. Kon’kova, I. Ali, Super-fast removal of cobalt metal ions in water using inexpensive mesoporous carbon obtained from industrial waste material, Environ. Technol. Innov. 21 (2021), https://doi.org/ 10.1016/j.eti.2020.101257. [64] A. Ghaemi, M. Torab-Mostaedi, S. Shahhosseini, M. Asadollahzadeh, Characterization of Ag(I), Co(II) and Cu(II) removal process from aqueous solutions using dolomite powder, Korean J. Chem. Eng. 30 (2013) 172–180, https://doi.org/10.1007/s11814-012-0113-1. [65] H. Wang, K. Huang, Y. Zhang, X. Chen, W. Jin, S. Zheng, Y. Zhang, P. Li, Recovery of lithium, nickel, and cobalt from spent lithium-ion battery powders by selective ammonia leaching and an adsorption separation system, ACS Sustain. Chem. Eng. 5 (2017) 11489–11495, https://doi.org/10.1021/acssuschemeng.7b02700. [66] J.-M. Park, S.-K. Kam, M.-G. Lee, Adsorption characteristics of lithium ion by zeolite modified in K+, Na+, Mg2+, Ca2+, and Al3+ forms, J. Environ. Sci. Int. 22 (2013) 1651–1660, https://doi.org/10.5322/JESI.2013.22.12.1651. [67] H. Hasar, Adsorption of nickel(II) from aqueous solution onto activated carbon prepared from almond husk, J. Hazard. Mater. 97 (2003) 49–57, https://doi.org/ 10.1016/S0304-3894(02)00237-6. [68] M.A. Abu-Daabes, E. Abu Zeitoun, W. Mazi, Competitive adsorption of quaternary metal ions, Ni2+, Mn2+, Cr6+, and Cd2+, on acid-treated activated carbon, Water 15 (2023) 1070, https://doi.org/10.3390/w15061070. [69] N. Rachel, N. Nsami, B. Placide, K. Daouda, A. Abega, T. Benadette, K. Mbadcam, Adsorption of manganese(II) ions from aqueous solutions onto granular activated carbon (GAC) and modified activated carbon (MAC), Int. J. Innov. Sci. Eng. Technol. 2 (2015). [70] Y.-C. Tang, J.-Z. Wang, Y.-H. Shen, Separation of valuable metals in the recycling of lithium batteries via solvent extraction, Minerals 13 (2023) 285, https://doi. org/10.3390/min13020285. N. Conte and J.M. Gómez View publication stats https://doi.org/10.1515/ract-2017-2836 https://doi.org/10.1515/ract-2017-2836 https://doi.org/10.1016/j.jhazmat.2008.04.009 https://doi.org/10.1016/j.jhazmat.2008.04.009 https://doi.org/10.1016/j.apsusc.2007.04.025 https://doi.org/10.1016/j.apsusc.2007.04.025 https://doi.org/10.1016/0008-6223(94)90089-2 https://doi.org/10.1016/0008-6223(94)90089-2 https://doi.org/10.1021/j150579a011 https://doi.org/10.1021/j150579a011 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0300 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0300 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0300 https://doi.org/10.1038/s41467-021-22429-0 https://doi.org/10.1061/(ASCE)0733-9372(1993)119:3(585) https://doi.org/10.1061/(ASCE)0733-9372(1993)119:3(585) https://doi.org/10.1016/j.eti.2020.101257 https://doi.org/10.1016/j.eti.2020.101257 https://doi.org/10.1007/s11814-012-0113-1 https://doi.org/10.1021/acssuschemeng.7b02700 https://doi.org/10.5322/JESI.2013.22.12.1651 https://doi.org/10.1016/S0304-3894(02)00237-6 https://doi.org/10.1016/S0304-3894(02)00237-6 https://doi.org/10.3390/w15061070 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0345 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0345 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0345 http://refhub.elsevier.com/S1383-5866(23)02003-8/h0345 https://doi.org/10.3390/min13020285 https://doi.org/10.3390/min13020285 https://www.researchgate.net/publication/374471629 Improving the sorption properties of mesoporous carbons for the removal of cobalt, nickel and manganese from spent lithium- ... 1 Introduction 2 Experimental 2.1 Chemicals 2.2 Mesoporous carbons synthesis and activation 2.3 Characterization 2.4 Sorption experiments 3 Results and discussion 3.1 Characterization 3.2 Monometallic sorption 3.3 Multimetallic sorption 4 Conclusions CRediT authorship contribution statement Declaration of Competing Interest Data availability Acknowledgements Funding References