Removal of lithium from water by aminomethylphosphonic acid-containing resin

This paper gives an overview of the ability of an aminomethylphosphonic acid-containing chelating resin for the removal of lithium from water. The studies were performed under various conditions, such as resin dose, initial Li+ concentration, solution pH and solution temperature. The results showed that the sorption of Li+ reached equilibrium within 15 min and the experimental data were well-fitted by the pseudo-second-order kinetic model. The Li+ sorption was highly pH dependent, and the optimum pH for Li+ removal was ≥3. Isotherm sorption data displayed good correlation with the Langmuir model, and the maximum monolayer sorption capacity of the resin found to be 13.65 mg g-1. Thermodynamic studies suggested that Li+ sorption onto the chelating resin was an exothermic and spontaneous process in nature. The resin could be regenerated by 0.1 M HCl, NaCl or H2SO4 with > 99 % efficiency. Desorption of Li+ with 0.1 M NaCl resulted in no changes in the uptake capacity through four sequential sorption/desorption cycles.


INTRODUCTION
Lithium is the 25 th most abundant element (at 20 mg kg -1 ) in the earth's crust. 1 The major lithium minerals with commercial value are classified into three major groups, namely, silicates, micas, and phosphates. 2 Lithium finds application in rechargeable lithium ion batteries (LIBs) because of its very high energy density by weight and high electrochemical potential (3.045 V). 1 Besides batteries, lithium compounds are used in ceramics and special glass industries, in primary aluminum production, rocket propellants, nuclear and pharmaceutical industries, in the manufacture of lubricants and greases, synthesis of vitamin A, synthesis of organic compounds, silver solders, underwater buoyancy devices, and batteries.Lithium is alloyed with aluminum and magnesium as light metals to form stronger and lightweight alloys. 3It is recovered from mines and salt 1060 ÇIÇEK, YILMAZ and ARAR lakes, which contain about 17 million t of lithium in total, while seawater is also considered as a vast source of lithium (about 2.5×10 14 kg), although the concentration of lithium is very low, i.e., 0.17 mg L -1 . 4Various methods have been studied for the recovery of lithium from seawater, brine, and geothermal water.These can be classified into three groups: adsorption (ion-exchange), solvent extraction and co-precipitation. 5The efficiency and limits of such methods have been reviewed elsewhere. 3Problems in separation or in the concentration of trace constituents are sometimes encountered.Ion exchange resins with different cation-exchange groups may possess different selectivity coefficients.Among the factors that determine the ionic selectivity of resins, the most important ones are: a) the nature of the acidic functional group and b) the density of the structure, which is largely determined by the degree of cross linkage.Regarding the first point, it is characteristic that the affinity of alkali ions toward strongly acidic groups (sulfonic acid groups, etc.) decreases in the order K >> Na > Li, whereas the order is precisely the opposite on resins with carboxylic, phosphonous, and phosphonic acids (especially the latter) as the functional groups. 6,7he purpose of this study was to evaluate an aminomethylphosphonic acid--containing chelating resin for Li + sorption.The sorption process was optimized by varying different parameters, such as resin dose, initial solution pH, concentration of Li + and temperature.Elution of Li + from the resin was also examined.

Materials
Lewatit TP 260 resin was used in the experiments.It is a weak acidic, macroporous resin contains chelating aminomethylphosphonic acid groups.The properties of resin are listed in Table I. 8 The resin was converted into the sodium form by treatment (100 mL wet resin) with 2 M NaCl (250 mL) solution for 24 h and then washing thoroughly with water.It was dried in an oven and used in the experiments.The pH of the solutions used in the batch test was adjusted to its optimum value by the addition of sufficient amounts of 0.1 M HCl and 0.1 M NaOH solutions.

Lithium analysis
The concentration of Li + was determined using a flame photometer (Jenway PFP7).The concentration ranges of Li + standards for the calibration curve were in the range of 0.1 to 5 mg L -1 .High concentration of Li + (which was used in the isotherm study) was measured after appropriate dilutions.

Batch adsorption tests
Experimental conditions for Li + sorption were summarized in Table II.The removal efficiency (R) and capacity (q) of the resin were calculated according to Eqs. ( 1) and ( 2), respectively: where, c 0 and c e (mg L -1 ) are the initial and equilibrium Li + concentrations, respectively, V is the volume of solution (L) and m is the mass of the resin (g).

Effect of resin dose
The effect of resin dosage on the uptake of Li + was studied to understand the efficacy of the resin for Li + removal.The uptake of Li + plotted as percent removal vs. resin dose is illustrate in Fig. 1, from which it could be seen that on increasing the resin dose from 0.02 to 0.3 g per 25 mL, the removal efficiency improved from 53 to 98 %.This could be attributed to an increase in the availability of more sorption sites as the dose of resin was increased. 9The optimum resin dose was found to be 0.3 g for 25 mL of solution and this amount was used in further experiments.

Effect of solution pH
The solution pH is a one of the important factor in sorption studies, because it can influence the structure of resin (for weak acidic and weak base), the struc-1062 ÇIÇEK, YILMAZ and ARAR ture of target molecule (especially if it has weak acidic or weak basic character) and H + or OH -act as competing ions in the ion exchange process.The effect of initial solution pH values (1-6) on Li + ion sorption onto TP 260 resin was investigated, and the results are shown in Fig. 2. Li + removal was unfavorable in acidic media (pH ≤ 1) but it increased with increasing pH value.There was only 4 % removal at pH 1 that increased to 99 % at pH ≥ 3. The pK a values of aminomethylphosphonic acid (AMPA) are 2.35 and 5.9. 10 At pH 1, the functional group of the resin is in the molecular form, thus removal of Li + was not possible.On the other hand, when the pH of solution was increased, the functional group of the resin ionized and sorption of Li + was enabled.

Isotherm analysis of Li + removal
The sorption isotherm is one of the important parameters for understanding sorption behavior and mechanism.The sorption capacity of the resin for Li + removal was studied at different initial Li + concentrations and the obtained results were applied to the Langmuir and Freundlich models.The Langmuir isotherm model is used to describe a monolayer adsorption process, and the model can be described in linear form by Eq. ( 3): c e /q e = 1/bq 0 + c e /q 0 (3) In this equation, q e / mg g -1 is the amount of Li + sorbed per gram of dry resin at equilibrium, q 0 / mg g -1 , and b / L mg -1 are the Langmuir constants related to the capacity and energy of sorption, respectively. 11,12The Freundlich isotherm model shows that a multilayer of the adsorption process occurs on heterogeneous surfaces, and is expressed by Eq. ( 4): log where K F / dm 3 g -1 is the isotherm constant of the Freundlich model and n is the exponent of the Freundlich model.K F and n are characteristics of the system and are indicators of the sorbent capacity (or affinity for the solute) and sorption intensity, respectively. 13he related parameters of the two models were calculated and are summarized in Table III.From the values of the linear correlation coefficients (R 2 ), the Langmuir model was found more suitable for describing the Li + sorption than the Freundlich model.The results suggest that monolayer sorption of Li + on such resin is the main mechanism.Additionally, the values of q 0 calculated from the Langmuir model was 13.65 mg g -1 .

Dubinin-Radushkevich model
The Dubinin-Radushkevich (D-R) model is another model that is used for clarification of the mechanism of sorption (i.e., physical or chemical).The linear D-R equation is given by Eq. ( 5): l n q e = ln X m -βε 2 (5 where β is a constant related to the sorption energy (mol 2 J -2 ), X m (mol g -1 ) is the D-R monolayer capacity, ε (mol L -1 ) is the Polanyi potential that is calculated as shown in Eq. ( 6): The mean free energy, E (kJ mol -1 ), of sorption can be estimated by using β values as expressed in the following equations: 14 For values of E < 8 kJ mol -1 , physical forces may have an effect on the sorption mechanism while E values between 8 and 16 kJ mol -1 depict sorption controlled by ion exchange and E > 16 kJ mol -1 signifies the process is chemisorption. 15064 ÇIÇEK, YILMAZ and ARAR In the present case, X m was found to be 0.0027 mol g -1 , which is equal to 18.74 mg Li + per g-resin, β was found to be 0.0050 mol 2 J -2 , and mean free energy, E, was 14.1 kJ mol -1 , indicating that removal mechanism was ion exchange.The ion-exchange reaction of the resin with Li + can be expressed as: 16  R CH PO(ONa) 2Li R CH PO(OLi) 2Na

Kinetics of Li + removal
Ion exchange time-dependent experiments were performed to evaluate the sorption kinetics.The kinetic data shown in Fig. 3 indicate that the sorption of Li + increased rapidly within 5 min, followed by a relatively slow process, and then the sorption equilibrium was achieved within 15 min.In addition, no remarkable changes were observed from 15 to 45 min.

Reaction-based model
The kinetic data were analyzed using pseudo-first-order, and pseudo-second--order models.The equations for these models are given as Eqs.( 9) and (10): 17,18 log (q e -q t ) = log q ek 1 t/2.303(9)   t /q t = 1/k 2 q e 2 + t/q e (10) Experimental results were applied to the kinetic models and the results are summarized in Table IV, from which it could be seen that the correlation coefficient (R 2 ) obtained from the pseudo-second-order model was larger than that from the pseudo-first-order model.

Diffusion-based model
The intraparticle diffusion model proposed by Weber and Morris in 1963 was also taken into account in the experiments.They concluded that the sorption is proportional to the square root of the contact time: 19 q t =k id t 1/2 (11)   where k id is the intraparticle diffusion rate constant (mg g -1 min -0.5 ).When the intraparticle diffusion model controls the sorption, the graph of q t against t 0.5 should be a straight line passing through the origin.The rate constant could be calculated from the slope of the line, The liquid film diffusion equation is given as: where F is the fractional attainment of equilibrium F = (q t /q e ), k fd (1/min) is the rate constant.A linear plot of −ln (1 − F) vs. t with zero intercept would suggest that the adsorption process was controlled by liquid film diffusion. 19he experimental data were fitted in Eqs. ( 9)-( 12) and the obtained results are summarized in Table IV.The R 2 revealed that the retention process is the film diffusion controlled mechanism.

Thermodynamic parameters
The effect of the temperature on the removal of Li + from water was examined by changing the solution temperature from 30 to 60 °C under the optimized conditions of resin dose and solution pH.Changes in the free energy (ΔG), entropy (ΔS) and enthalpy (ΔH) were estimated by the usual procedure. 20The calculated values are summarized in Table V.The negative values of ΔG shows that ion exchange reaction is spontaneous.The negative value of ΔH suggests the exothermic nature of the sorption.The positive value of ΔS suggests increased randomness at the solid/solution interface during the sorption of Li + onto the resin.

Regeneration of the resin
Regeneration experiments were performed as explained in the literature. 21ecovery of Li + from the resin was checked with HCl, H 2 SO 4 and NaCl solutions at various concentrations.The regeneration efficiency (RE / %) was calculated using Eq. ( 13): Desorbed amount of Li from the resin (%) 100 Sorbed amount of Li onto the resin RE The obtained results are summarized in Table VI, from which it could be seen that 0.1 mol dm -3 HCl, H 2 SO 4 or NaCl was enough for complete regeneration of the resin.

Multiple sorption/regeneration tests
Multiple sorption/regeneration tests were performed to determine the reusability of the ion exchange resin and the recovery of sorbed Li + .The sorption test was realized by contacting 0.3 g resin with 25 mL of Li + solution (5 mg Li L -1 , pH 6) for 2 h.After decantation of the solution, the resin was washed with pure water until the conductivity of the eluting water reached that of pure water conductivity value.The elution step was performed by contacting 0.1 mol dm -3 25 mL of regenerant solution with the washed resin for 2 h.At the end of this time, the solution was decanted and washed with pure water as explained above.This sorption/regeneration cycle was repeated 4 times and the obtained results are summarized in Table VII As can be seen from Table VII, the type of the regenerant influences the sorption capacity.When NaCl was used for regeneration, the sorption capacity of the resin did not change but when HCl was used, sorption capacity decreased.After regeneration with HCl, the resin was converted to the H-form and used in the next sorption cycle.In the second sorption process, Li + replaces H + and thus, the H + concentration in the solution increased and the pH of the solution decreased.As explained in effect of solution pH section, when the pH of solution decreased and the sorption capacity of resin decreased.Previously in the literature, various research studies were conducted for lithium removal/recovery.The results obtained in such research studies are summarized in Table VIII.The capacities of the sorbents vary from 4.07 to 62.5 mg g -1 sorbent.The removal performance of such materials strongly depended on the experimental conditions, particle size, and loaded metal.

CONCLUSIONS
In this study, the removal of lithium ions from aqueous solutions onto an ion-exchange resin was investigated under various experimental conditions, such as resin dose, initial solution pH, and temperature.The results clearly demonstrated that the phosphonic acid groups contributed to the sorption mechanism through electrostatic interactions between the phosphonic acid of the resin and the Li + .The ion exchange process was quite fast and equilibrium was established within 15 min.The percentage removal of Li + was pH dependent, decreasing with decreasing initial pH of the solution.The equilibrium sorption behavior Li + onto Lewatit TP 260 resin followed the Langmuir adsorption isotherm with a maximum theoretical sorption capacity of 13.65 mg g -1 resin.The sorption of Li + onto the resin was found to be mainly based on ion-exchange interactions, and these were confirmed by the Dubinin-Radushkevich model.
Regeneration and reuse of resin work well for NaCl regenerant.Regeneration of resin with acid solution decreased the removal efficiency of resin.
Temperature variations were used to evaluate enthalpy, entropy and free energy changes.The negative value of free energy change showed the spontaneous nature of the adsorption.In the temperature range 303-333 K, enthalpy change was negative, and the ion exchange reaction was exothermic.

TABLE I .
Physicochemical properties of Lewatit TP 260

TABLE II .
Experimental parameters for Li + removal

TABLE III .
Calculated isotherm constants of Langmuir and Freundlich models

TABLE IV .
The calculated parameters of pseudo first and pseudo second order kinetic model

TABLE V .
Thermodynamic parameters for Li + sorption

TABLE VI .
Effect of acid concentration on the desorption of Li + from the resin

TABLE VIII .
Capacities of sorbents used for Li + removal/recovery