Performance of a synthetic resin for lithium adsorption in waste liquid of extracting aluminum from fly-ash

Zhengguo Xu,Xiaochong Wang,Shuying Sun

National Engineering Research Center for Integrated Utilization of Salt Lake Resource,East China University of Science and Technology,Shanghai 200237,China

Keywords:Adsorption Wastewater treatment Lithium Utilization of fly-ash

ABSTRACT In this study,we investigated the performance of a synthetic resin for the adsorption of Li from predesilicated solution which is the waste liquid produced by extracting aluminum from fly ash.The adsorption kinetics and isotherms of the resin were obtained and analyzed.The saturated adsorption sites of the resin were in agreement with the quasi-second-order kinetic model.Then,the pore diffusion model(PDM)was applied to represent the lithium adsorption kinetics which confirming that the external mass is the limiting step.Moreover,we evaluated the adsorption properties of this resin in fixed-bed mode.We established a feasible extraction process for Li from strong alkaline solutions with low Li concentrations.The process parameters,such as the flow rate,initial adsorption solution concentration,water washing process,desorption agent concentration,and flow rate were studied.The desorption rate of the Li+ ions was directly proportional with the concentration of the desorption agent.The time required to accumulate Li decreased as the hydrochloric acid concentration and flow rate increased.Time of the peak appeared increased from 0.5 bed volume (BV) to 2.5 BV as the concentration was increased from 1 to 3 mol?L-1,and the peak increased from 231 to 394 mg?L-1.The resin presented good selectivity for Li+ions and could effectively separate impurity ions from the pre-desilication solution.

1.Introduction

Owing to the rapid development of new energy and new materials industries,Li resources have attracted much attention[1],particularly for the Li battery industry [2].Recently,the use of new energy vehicles has been increasingly promoted,the applications of Li salts have been expanding,and the market demand for Li has been gradually increasing [3].Li resources consist of solid and liquid ores,and the main ones are spodumene,Li mica,and Li in salt lake brines [4].The technologies used to extract Li from spodumene and Li mica include limestone sintering,sodium carbonate digestion,and sulfuric acid,sulfate,or chlorination roasting.Moreover,the technologies used to extract Li from salt lake brine include precipitation,extraction,adsorption,calcination leaching,electrodialysis,and ion sieving [5–7].

Recently,researchers worldwide have started to investigate the distribution of coal-associated Li deposits in different regions on the globe.The distribution of Li in coal is extremely uneven,and the content of Li varies greatly between regions [8].Fly ash,the fine ash collected from the flue gas after the combustion of coal,is the main solid waste discharged from coal-fired power plants[9].Fly ash,an atypical aluminosilicate material that can be considered a potential ‘‘urban mineral”resource,mainly consists of SiO2(35.6%–57.2% (mass)),Al2O3(18.8%–55.0% (mass)),Fe2O3,CaO,MgO,K2O,Na2O,P2O5,TiO2,MnO,and SO3.In addition,fly ash contains trace amounts of valuable elements,such as Ga,Ge,Se,Li,V,Ni,Pt,Cu,and U [10–12].Currently,fly ash is mainly used to produce bricks,concrete blocks,cement additives,and other building materials[13].In terms of high-value utilization,researchers have mainly focused on the extraction and utilization of Al and Si from fly ash;however,the research on the cooperative extraction and utilization of rare and valuable elements from fly ash is still in its infancy.

Although the Li content of fly ash is low,China’s coal-based energy consumption structure is difficult to change.Fly ash presents great development prospects owing to its huge output,and the Li resource is immeasurable.Few studies on the extraction of Li from fly ash have been conducted,mainly because of the low leaching rate of Li during the solid–liquid conversion process,complex composition of the leaching solution,and high concentrations of acids and alkalis required.Many types of industrial technologies are available for the extraction of Li from fly ash,such as sulfuric acid roasting,hydrochloric acid leaching,sodium carbonate sintering,limestone sintering,and sodium carbonate and ammonium sulfate sintering [14–17].

Datang International and Tsinghua Tongfang have jointly developed a process for the extraction of alumina from high-alumina fly ash using pre-desilication soda-lime sintering [18].The world’s first 200,000 t?a-1alumina multi-generation demonstration production line was completed and inaugurated in 2012.The limesoda sintering method increases the Al/Si ratio of fly ashviapredesilication,thereby reducing the energy and material consumption and final slag formation during sintering [19].Li is enriched at different steps of the process,and studies have determined that Li was mostly enriched in the pre-desilication solution,which presented an extraction value of 50 mg?L-1.Extraction [20–22] and adsorption [23,24] are mainly used to extract lithium from low lithium concentration solutions.Compared with extraction,adsorption requires lower equipment investment and energy consumption.However,the structure of some adsorbents can be destroyed in high concentration NaOH solutions.Moreover,problems such as strong corrosiveness and low extraction rate also cause difficulties in industrialization.

A resin was synthesized for the absorption of Li from strong alkaline solutions with low Li concentrations.The extraction rate of lithium in the system with high Na/Li ratio was improved,and the dissolution loss of adsorption material in the system with strong alkali was solved.The effects of the process parameters,such as the inlet flow rate,initial adsorbate concentration,water washing process,desorption agent concentration,and desorption agent flow rate were investigated.The goal of this study was to establish a feasible extraction process for Li from strong alkaline solutions with low Li concentrations.

2.Experimental

The resin was synthetic by our lab.All chemicals used in the study were of high-purity analytical grade (LiCl?H2O,≥97%,Sinopharm Group);moreover,deionized water was used to prepare all solutions.All experiments were performed at (25 ± 2) °C and atmospheric pressure.The ion concentration of the aqueous solution was measured using an inductively coupled plasma optical emission spectrometer (ICP-OES,spectra ARCOS,Germany).

2.1.Experimental equipment and analytical instruments

An ARCOS FHS12 inductively coupled plasma emission spectrometer,a DZF-6090 vacuum drying oven,a PHSJ-3F laboratory pH meter,an AR2140 OHAUS precision analytical balance,a S10-3 constant temperature magnetic stirrer,and a BT1-200E constant current pump were used in this study.

Quantitative determination of elements was done by ARCOS FHS12 inductively coupled plasma emission spectrometer.A series of calibration solutions from 0 to 100 mg?L-1were prepared by 1 mg?ml-1single element standard solution.Test conditions:plasma power=1450 W,coolant flow=12 L?min-1,auxiliary flow=0.8 L?min-1nebulizer flow=0.75 L?min-1.All the element concentration determinations were carried out in triplicates and the average value of the same has been reported in the manuscript.The data having deviation more than ±7% were discarded and repeated.

2.2.Characterization of samples

The sample morphology was analyzed by Scanning Electron Microscope (SEM,Hitachi Corporation of Japan,S-3400 N).Working parameters are as follows:high vacuum resolution of 3.0 nm,low vacuum resolution of 4.0 nm,and amplification range of 5–50,000.

Chemical bonds in adsorbents were analyzed by using the Fourier transform infrared spectrometer (FTIR,Niger,7800–350/cm 0.01/cm/6700),and the chemical bonds in the samples were referring to the standard Infrared spectrograph.The instrument wavelength range is 4000–400 cm-1.

Mercury porosimeter (America,AutoPore 9510) was used to analyze and determine the pore diameter distribution,total pore volume,total pore specific surface area,and sample density of the adsorbent.The maximum pressure of the instrument was 413.7 MPa,and the pore diameter was 3 nm–360 m.

2.3.Experimental methods

2.3.1.Preparation of adsorption solution

The adsorption solution was prepared according to the predesilication method using the fly ash-enhanced desilication-mild alkaline Al-Li-Ga synergetic extraction process proposed by Datang International.A fly ash and 20% NaOH solution mixture with the liquid-to-solid ratio of 4:1 reacted at 90 °C for 2.5 h.The predesilication solution was obtained via filtration,and its composition is summarized in Table 1.

Table 1 Composition of pre-desilication solution

2.3.2.Adsorption kinetics and isotherm

First,15 g of adsorbent was added to a conical flask that contained 150 ml of LiCl solution at different initial concentration(pH=7).Then,the flask was placed in a 25 °C air shaker at 150 r?min-1for 24 h,and the supernatant was collected.The concentrations of Li+and other impurity ions were measured before and after the experiment.After the resin was filtered and dried,its adsorbent capacity and selectivity were calculated as follows:

whereQeis the adsorption capacity of the resin at equilibrium(mg?g-1),C0is the initial concentration of ions (mg?L-1),Ceis the equilibrium concentration of ions(mg?L-1),Vis the volume of liquid(ml),andxis the mass or volume of resin (g or ml).

Then,the adsorption curve was obtained and the kinetic equations were fit.The pseudo-first-order [25] and pseudo-secondorder [26] kinetic models could be described as follows:

and

whereQtis the adsorption capacity of the resin at timet(mg?g-1),k1is the pseudo-first-order kinetic constant (min-1),andk2is the pseudo-second-order kinetic constant (g?(mg?min)-1).

The adsorption equilibrium curve was fitted using the Langmuir[27] and Freundlich [28] adsorption isotherms,as follows:

and

whereq0is the saturated adsorption amount (mg?g-1),bis the Langmuir constant,Kfandnare the Freundlich constant.

2.3.3.Pore diffusion model

The literatures show that the adsorption rate of porous solids includes the following three simultaneous steps:external mass transport,intraparticle diffusion,and adsorption on an active site.As the rapid equilibrium rate was observed,so,it can be assumed that the pore diffusion is the predominant intraparticle mass transfer mechanism [29].In this work,the pore diffusion model (PDM)was applied on the basis of assumptions:the adsorbent diffuses in the uniform adsorbent,the pore diffusion coefficientDpis constant,the process was isotherm[30].So,the overall mass balance can be expressed:

whereVis the volume of solution(L),CandCp,r=rpis the metal concentration at bulk liquid phase and surface of the adsorbent(mg?L-1),Wais the mass of adsorbent (g),rpis the radius of the adsorbent (cm),ρPis the apparent density of adsorbent (g?L-1),kfis the film diffusion coefficient (cm?s-1).

Mass balance in the particle can be expressed:

where εpis the porosity of adsorbent (%),Cis the metal concentration at bulk liquid phase(mg?L-1),qis the lithium concentration at solid phase(mg?g-1),Dpis the pore diffusion coefficient(cm2?s-1),ris the radius of the adsorbent(cm).The initial and boundary conditions are given as follows:

In addition,there is an equilibrium between the lithium ions concentration in particle,Cp,and the concentration in solid phase,q.this equilibrium relationship is represented by the adsorption isotherms based on the fitting of the isotherm models.The mathematical model was solved numerically in gPROMS(Process System Enterprise,UK)using the orthogonal collocation on finite elements method.Fifty intervals with three order polynomials were used in the radial domain of adsorbent.The ‘‘Parameter Estimation”function in gPROMS was employed for the fitting procedure.

2.3.4.Fixed bed experiment

Fixed-bed adsorption experiments were conducted to evaluate the performance of a 0.5 L adsorption column packed(inner diameter was 6 cm,height was 1 m,resin mass was 750 g)with the pretreated resin for the adsorption of Li.The pre-desilication Licontaining solution was fed from the top,and the effluent solution was collected at different times.When the concentration of Li at the outlet was 90%of the initial Li concentration,it was considered that the resin column was saturated.After the adsorption was completed,the solution that remained in the column was drained,and the column was rinsed with deionized water until the concentrations of the other ions in the effluent was lower than 5 mg?L-1.After the adsorption column was washed with water,it was evacuated,and a hydrochloric acid solution was fed from the bottom.The effluent solution was collected at different times.Effects of the flow rate (1.5–6.0 bed volume per hour (BV?h-1)),initial adsorption solution concentration(65–430 mg?L-1),water washing process,and desorption agent concentration (1–3 mol?L-1) was investigated (see Fig.1).

3.Results and Discussion

3.1.Adsorption kinetics

The adsorption kinetic curve obtained from the static shaker experiment is presented in Fig.2.The equilibrium time of the resin was short,which indicated that the rate of the ion exchange reaction was fast.The adsorption capacity of the resin for Li in this system was 0.31,0.54 and 0.82 mg?g-1(with the initial lithium concentration was 50,100 and 200 mg?L-1),respectively.The data were fit using the pseudo-first-and pseudo-second-order kinetic equation models,and the fitting curves for these models are illustrated in Fig.3(a) and (b),respectively.

Li adsorption conformed to the pseudo-second-order kinetic model,which indicated that the resin possessed saturation sites.The pseudo-second-order kinetics model could be used to describe the steps of the adsorption process.In order to explain the mechanism of the adsorption,Fourier transform infrared spectrometer was used to analyzed the chemical bonds.Fig.4 showed the FTIR spectra of the resin before and after adsorption.It was found that the absorption peak of the bond P-OH shifts at 2920.5 cm-1and 2851.3 cm-1were weakened and the absorption peak of the bond N-H shifts at 3420 cm-1moved to 3414 cm-1.It indicated that phosphoamide was the functional group of the material and ion exchange occured during the adsorption process.Na+was exchanged with the H+dissociated from P-OH of the functional group,and the coordination occurred to the N atom of the functional group.Then,Li+exchanged with Na+adsorbed on the functional group.

Fig.5 showed the SEM of resin before and after adsorption.It can be seen that the resin was a microsphere with a smooth surface,and its structure did not change before and after adsorption.It indicated that the good stability of the resin under strong alkali system.

Dissolution loss of resin was investigated by weight method.Both before and after adsorption–desorption cycles,resin at hygrometric state was weighted after suction filtration.Dissolution loss of resin after different cycles is shown in Table 2.It can be seen that the dissolution loss of resin was always about 0.3%no matter how many successive cycles.The loss can be due to experimental manipulation error,and no dissolution loss has occurred during the adsorption process.

Fig.1.Experimental setup for the study.

Fig.2.Adsorption rate of lithium by adsorbent (initial lithium concentration=50,100,200 mg?L-1,temperature=25 °C).

3.2.Effect of temperature

Experiments using alkaline solutions were performed using different Li concentrations in the range of 25–2000 mg?L-1in Fig.6.The equilibrium concentrations of Li before and after adsorption at 30,40 and 60 °C were measured,and the corresponding linear fitting of adsorption isotherms are depicted in Fig.7(a) and (b),respectively.Moreover,the fitting parameters are summarized in Table 3.

On comparing the correlation coefficients (R2) of the Langmuir and Freundlich adsorption isotherms at 30,40 and 60 °C.It was concluded that the resin adsorption of Li was in agreement with the Langmuir model and the Freundlich constantn>1.It showed that the lithium adsorption capacity decreased with the increasing of the temperature.The adsorption capacity of adsorbent is greatly affected by temperature and initial concentration.The maximum lithium adsorption capacity decreased from 6.95 mg?g-1to 1.51 mg?g-1with the temperature increasing from 30 °C to 60 °C.These results indicated that the adsorption process was an exothermic process and was favorable adsorption.

Fig.4.FTIR spectra of the resin before and after adsorption.

3.3.Pore diffusion model

Pore characteristic parameters were needed to fit the PDM.Fig.8 showed the results of measuring resin pore structure with mercury injection instrument.It can be seen that the pore diameter of the resin was mainly within 200 nm.The analysis results of pore structure are summarized in the Table 4.The porosity of the resin was 0.3418,and the porosity significantly affected the adsorption kinetics.

Table 2 Dissolution loss of resin after different cycles

Table 3 Fitting results of adsorption isotherm equations

Table 4 Summary of pore structure by mercury porosimeter

3.3.1.Calculation of film diffusion coefficient

From the overall mass balance equation[Eq.(6)],the film diffusion coefficient can be estimated through the initial adsorption rate.InitiallyC=C0andCp,r=rp=0 (for finite film mass transfer resistance or finitekf),the initial adsorption rater0is defined:

Eq.(8) provides the relationship for calculation the film diffusion coefficient through the initial adsorption rate and the initial adsorption rate was determined by using the bulk phase concentration change between 0 and 10 min in the present study [31].

3.3.2.Estimation of pore diffusion coefficient

As can be known from the PDM model,Dpis a constant that can be calculated by empirical formula:

Fig.3.(a) Pseudo-first and (b) pseudo-second order kinetic equation models.

Fig.5.SEM of resin before (left × 50) and after (right × 200) adsorption.

whereDMis ion diffusion coefficient of Li+(DM=1.1664 × 10-6cm2?s-1,at the temperature of 303 K),τ is tortuosity of the absorbent.

As the value of void fraction of particle (0.3418) was observed,Dpwas calculated by Eq.(9)about 1.36×10-6cm2?s-1.The experimental kinetics curves of lithium were obtained under different initial concentration (50,100 and 200 mg?L-1).The only unknown parameter,Dp,was estimated by fitting the numerical solution of the PDM to the experimental kinetic data.And Fig.9 depicted the experimental kinetic curves of lithium adsorption and the kinetic curves predicted with the PDM model using the optimal values ofDp,showing the PDM fitted the experimental data very satisfactorily.Different from the common surface diffusion which is the rate control step of the whole lithium adsorption process[32],the resin is mainly affected by pore diffusion.

3.4.Effect of initial Li concentration on the performance of adsorption curves

The adsorption performance of the resin was tested at the initial Li concentrations(C0) of 65,200 and 430 mg?L-1,and the sorption breakthrough curves are illustrated in Fig.10.The decrease inC0caused the decrease in breakthrough time,and that allowed for more solution to be processed.This was attributed to the decreases in the diffusion coefficient and mass transfer coefficient at low Li concentrations.The binding sites of the adsorbent were quickly filled at the higher initial Li concentration,and that resulted in the decrease in the breakthrough time.The treated volume decreased from 18 BV to 12 BV as the initial Li concentration was increased from 200 to 430 mg?L-1.And the adsorption rates at theC0values of 200 and 430 mg?L-1were 25.9% and 31.2%,respectively.The adsorption force of the more concentrated Li solution was stronger than that of the less concentrated Li solution,and that could promote the increase in the adsorbed Li amount.Furthermore,the adsorption capacity of the resin and extraction rate were higher when the initial concentration of Li was higher.These results were consistent with most fixed-bed studies in that the adsorbent gets saturated faster at higher concentrations [33].

3.5.Effect of flow rate on the performance of adsorption curves

Experiments were conducted at the flow rates from 1.5 to 6.0 BV?h-1,and theC/C0versusthroughput volume breakthrough curves are illustrated in Fig.11.As the flow rate decreased from 6.0 to 1.5 BV?h-1,the penetration time was 5,10,15 and 120 min,and the treatment capacity was 0.25,0.375,0.75 and 3.00 BV,respectively.Different from the common breakthrough curve [34],the effect of flow rate on adsorption is different at different period.The amount of alkali leaching solution that could be treated with the adsorbent resin increased with the decrease in flow rate;moreover,the breakthrough point shifted backward,and the time required to reach breakthrough increased as the flow rate decreased.The increase in breakthrough time could be attributed to the increase in contact time between the sorbate and sorbent,which further resulted in the diffusion of a higher amount of solute into the pores of the sorbent.As the flow rate increased,the adsorption rate decreased continuously.When the amount of Li+ions that came in contact with the sorbent decreased,the efficiency of the Li uptake by the resin decreased.As the flow rate through the adsorption column decreased,the depth of the adsorption zone decreased because more time was available for adsorption on each layer.In contrast,a high flow rate would improve the processing efficiency.

3.6.Adsorption performance of impurity ions

Owing to the pre-desilication solution containing large amounts of Na+,Al3+,andions,the changes in the concentrations of these impurity ions during the adsorption and water-washing processes were studied.The adsorption breakthrough curves of the Na+,Al3+,andimpurity ions are presented in Fig.12 and are very similar;moreover,the outlet concentration reached the initial concentration after the adsorption of approximately 0.3 BV of each ion.The shorter breakthrough times of the impurity ions compared with that of Li+ions could be attributed to the better selectivity of the adsorbent for Li+ions and higher Li+ion concentration.The Na+ions did not flow out in the beginning owing to the high concentration of the alkali solution resulting in the Na transformation at the remaining few sites on the resin.

Water-washing was used to wash away the residual metal ions from the packed bed after adsorption and to increase the purity of the desorption liquid.The changes in ion concentrations with water consumption are illustrated in Fig.13.The ion concentrations in the effluent were almost unchanged when the volume ofdeionized water was twice the volume of the adsorbent,which is when the water-washing process was considered complete.

Fig.6.Equilibrium concentrations of Li before and after adsorption at 30,40 and 60 °C.

Fig.7.(a) Langmuir and (b) Freundlich adsorption isotherms.

Fig.8.Cumulative intrusion curves against pore size.

Fig.9.Comparison of experimental data and PDM of lithium kinetics at different initial concentrations.

Fig.10.Effect of initial Li concentration (C0) on the adsorption performance of the resin [flow rate=6 BV?h-1,initial Li concentration (mg?L-1):() 65;() 200;()430].

Fig.11.Effect of flow rate on adsorption curves for adsorption of Li[C0=65 mg?L-1,flow rate (BV?h-1):() 1.5;() 3.0;() 4.5;() 6.0].

Fig.12.Adsorption breakthrough curves of impurity ions [flow rate=6 BV?h-1,initial concentration (mg?L-1):(Na) 101290;(Al) 452.5;(Si) 21540].

Fig.13.Changes in ion concentrations with water consumption [flow rate=6 BV?h-1,initial concentration (mg?L-1):(Li) 65;(Na) 101290;(452.5);(Si)21540].

3.7.Effect of acid concentration

Hydrochloric acid was selected as the desorption agent,and experiments were performed at different hydrochloric acid concentrations,the initial Li concentration of 50 mg?L-1,and under the same adsorption and water-washing conditions.The desorption curves at the hydrochloric acid concentrations of 1,2,and 3 mol?L-1are illustrated in Fig.14,where a proportional relationship between the desorption rate and concentration of hydrochloric acid can be observed.The peak increased from 231 to 394 mg?L-1as the concentration was increased from 1 to 3 mol?L-1.At the highest hydrochloric acid concentration of 3 mol?L-1,the desorption peak appeared much earlier than at the lower concentrations.It was concluded from the peak shape that the time required to accumulate Li decreased from 120 min (4 BV) to 30 min(1 BV)with the increase in the hydrochloric acid concentration.This decrease in desorption time could be attributed to the higher concentration gradient promoting the ion exchange between the H+ions in the solution and binding sites of the resin.

Fig.14.Desorption curves at different hydrochloric acid concentrations [flow rate=2 BV?h-1,hydrochloric acid concentrations (mol?L-1):() 3;() 2;() 1].

3.8.Effect of desorption agent flow rate

The flow rate of a 1 mol?L-1solution of hydrochloric acid was varied in the range of 1–3 BV?h-1for the initial Li concentration of 50 mg?L-1under the same adsorption and water-washing conditions.The desorption curves of hydrochloric acid at different flow rates are presented in Fig.15.The time required for desorption decreased with the increase in flow rate.The complete desorption duration at the three flow rates was different,but the amount of hydrochloric acid used to achieve complete desorption was almost the same regardless of the flow rate.The highest point of the desorption peak decreased from 679 to 336 mg?L-1with the increase in flow rate,which is attributed to the residence time affecting the Li+ion enrichment.Similarly,a previous study [35] also reported that using a high flow rate can shorten the dynamic desorption duration.

At the same time,it was determined that almost no Li desorption occurred during the early stage of desorption.The Na+ion efflux curve during the desorption process is illustrated in Fig.16.During the desorption process,Na+ions were desorbed first followed by Li+ions.This is attributed to the exchange of Na+ions taking precedence over that of Li+ions and the binding sites of the resin presenting stronger adsorption capacity for Li+ions than for Na+ions.In order to verify the proposed mechanism,the ion exchange sequence of the resin was studied.Pure sodium chloride and lithium chloride solutions were used for adsorption.After different experimental sequences,it is found that lithium can replace the sodium on the resin,but not vice versa.indicating that the ion exchange order of the resin is Li+>Na+.

Fig.15.Desorption curves at different hydrochloric acid flow rates [hydrochloric acid concentrations=3 mol?L-1,flow rate (BV?h-1):() 1;() 1.5;() 2].

Fig.16.Efflux curves of()Na+and()Li+ions[flow rate=1 BV?h-1,hydrochloric acid concentrations=3 mol?L-1].

4.Conclusions

The resin in this study could be used in alkaline systems and could effectively separate Li from fly ash pre-desilication solutions.The adsorption curve of resin was related to concentration gradient and residence time,and the adsorption rate increased as both.Moreover,the desorption process is more dependent on the concentration and residence time of desorption agent and the ion exchange sequence of the resin.Owing to the low Li concentration in the pre-desilication solution,the saturated adsorption capacity of the resin was only 0.31 mg?g-1(initial lithium concentration is 50 mg?L-1).The adsorption of Li by the resin was affected by the temperature.The adsorption process was facilitated at (25 ± 2)°C,where the adsorption capacity of the resin for Li was higher.And PDM model illustrated that the pore diffusion is the ratecontrolling step of the whole lithium adsorption process.Moreover,it was determined that the adsorption process was affected by the inlet flow rate and initial concentration of Li.The adsorption rate increased with the increasing inlet flow rate and initial Li concentration.However,to increase the initial Li concentration,the pre-desilication solution must be concentrated,and the energy required for that does not justify the benefit.Therefore,it would be recommended to use the original liquid during the adsorption process in industrial settings.The resin presented good selectivity for Li+ions and could effectively separate the Si and Al impurities present in the pre-desilication solution.In addition,the desorption process was affected by the concentration and flow rate of hydrochloric acid,the preferred desorption agent.The Li desorption rate was proportional to the hydrochloric acid concentration and flow rate,and the Li enrichment time decreased as the hydrochloric acid concentration and flow rate increased.Because the desorption of Li and Na were distinctly segmented,Li and Na could be collected separately in industrial settings to increase the Li concentration in the desorption solution.

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.

Acknowledgements

The research was supported by National Key Research and Development Program of China(2017YFB0603104)and Sponsored by Shanghai Pujiang Program (2019PJD011).The authors also gratefully acknowledge the financial support from the State Key Laboratory of Green Chemistry Synthesis Technology (Zhejiang University of Technology,Hangzhou 310032).