Journal of Molecular Liquids 215 (2016) 640–646

Contents lists available at ScienceDirect

Journal of Molecular Liquids

journal homepage: www.elsevier.com/locate/molliq

Solvent extraction of lithium ions by tri-n-butyl phosphate using a room

temperature ionic liquid
Chenglong Shi a,b, Yan Jing a, Yongzhong Jia a,⁎
a
Key Laboratory of Comprehensive and Highly Efficient Utilization of Salt Lake Resources, Qinghai Institute of Salt Lakes, Chinese Academy of Sciences, 810008 Xining, China
b
University of Chinese Academy of Sciences, 100049 Beijing, China

a r t i c l e i n f o a b s t r a c t

Article history: Separation among lithium ion and magnesium ion remains challenging. In this work, solvent extraction of lithium
Received 5 October 2015 ions from the Mg(II)-containing salt lake brine into a commonly used ionic liquid 1-butyl-3-methylimidazolium
Received in revised form 4 January 2016 bis(trifluoromethylsulfonyl) imide ([C4mim][NTf2]) using the neutral phosphorus compound tri-n-butyl phosphate
Accepted 7 January 2016
(TBP) as the extractant was investigated. The extraction behavior of lithium ions was investigated as a function of
Available online 27 January 2016
various parameters: acidity of the salt lake brine, dosage of ionic liquid and phase ratio. Under the optimal condi-
Keywords:
tions, the single extraction efficiency of lithium was 92.37%, which indicated that it was an efficient extraction sys-
Ionic liquid tem for the recovery of lithium ions. Based on the study of thermodynamics, thermodynamic parameters of the
Tri-n-butyl phosphate lithium ion extraction reaction were obtained. The extracted species were ascertained from slope analysis method
Lithium which indicated 1:1 (metal:ligand) ratio in the complex. UV–visible and Fourier transform infrared spectroscopy
Extraction (FTIR) studies were also carried out to understand the nature of the extracted species.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction
Lithium, as the lightest metal element, is widely used in various fields
such as lithium batteries, aerospace, ceramics, polymers, lithium-based
lubricating greases and metal additives [1–5]. Its application and con-
sumption is steadily growing in recent years. Lithium resources exist nat-
urally in two forms: as a mineral and as a liquid. And the liquid state
lithium mainly resides in salt lake brine. According to preliminary statis-
tics, the worldwide reserves including lithium content in salt lakes are es-
timated to be 7.3 × 106 tons (as lithium) of which 60% in salt lakes [6].
However, lithium and magnesium are normally associated in the salt
lake brine. And the magnesium presents a major challenge for lithium ex-
traction because these two elements are located in diagonal positions
within the periodic table, resulting in their many chemical similarities.
The partition of Li+ from Mg2+ has drawn considerable attention for a
long time and lithium can be recovered from the salt lakes via a
number of processes, including precipitation, ion exchange, adsorption,
liquid–liquid extraction and supported liquid membrane [7–10]. Among
them, liquid–liquid extraction has been widely studied as a hot technolo-
gy for recovering lithium from the brine sources of a high Mg/Li ratio.
Sometimes, traditional solvent extraction systems which used a
large amount of volatile organic solvent can obtain a satisfied separation
[11–13]. For example, the selective extraction of Li+ with the
phenylphosphonic (H-PHO), phenylphosphinic (H-PHI) and bis(2-
ethylhexyl) phosphoric (H-BIS) acids dissolved in pentanol was
⁎ Corresponding author.
investigated by El-Eswed et al. [14]. And the highest extraction efficien-
cies obtained in the case of H-PHO, H-PHI, and H-BIS were 43.2%,
45.7%, and 90.0%, respectively. The extraction of lithium from salt lake
brine by tributyl phosphate–kerosene–FeCl3 extraction system had
been reported by Sun et al. [15]. The single-stage extraction has been
conducted under the optimum conditions and the extraction efficiency
reached 88%. However, almost all of traditional extraction systems used
large amounts of volatile organic solvents, which may cause some of
the environmental and workers' health problems. Furthermore, most of
the systems were conducted under strong acidic condition and resulted
in severe corrosion to the equipment. For sustainability of the liquid–liq-
uid extraction, it is necessary to find “greener and safer” solvents to re-
place the traditional organic solvents.
Room temperature ionic liquids (RTILs) are a type of organic salts
solely composed of ions and with the melting points near or below
room temperature. They exhibit several properties that make them at-
tractive as a potential solvent for ‘green’ separation processes, including
high thermal stability, negligible volatility, nonflammability, tunable
viscosity and good thermal and radiation stability for solvent extraction
[16–18]. Based on the remarkable properties of ionic liquids, they have
attracted considerable attention for metal extraction from aqueous solu-
tions, since they can replace volatile organic solvents that have been
widely used for this purpose [19–23]. For example, a solvent extraction
system based on combination of ILs and 1-methylimidazole or 2-
methylimidazole was investigated by Shen et al. [24]. Methylimidazole
dissolved in ionic liquid ([Cnmim][NTf2]) is able to extract lanthanides
and yttrium. High selectivity for lanthanides compared to alkali metals
and alkaline earth cations was demonstrated. The solvent extraction of

http://dx.doi.org/10.1016/j.molliq.2016.01.025
0167-7322/© 2016 Elsevier B.V. All rights reserved.
 C. Shi et al. / Journal of Molecular Liquids 215 (2016) 640–646
U(VI) from aqueous solution through ionic liquids-based systems is
also a hot topic. Yuan et al. [25] reported the extraction of U(VI)
using trioctylphosphine oxide (TOPO) in ionic liquid 1-butyl-3-
methylimidazolium bis(trifluoromethylsulfonyl)imide. The extraction
efficiency of U(VI) in the TBP/IL system was comparable to that in the
TBP/dichloromethane system. Raut et al. [26] demonstrated that 2-
thenoyltrifluoroacetone (HTTA) in ionic liquids enhanced the extraction
efficiency of U(VI) compared with that in a conventional organic sol-
vent. In our previous studies, extraction of lithium ions using three dif-
ferent solvent systems containing TBP in ionic liquids such as
[Cnmim][PF6] (where, n = 4, 6, 8) [27–30]. The trend of lithium ion ex-
traction in the ionic liquids containing PF− 6 counter anion was:
[C4mim][PF6] N [C6mim][PF6] N [C8mim][PF6]. The stripping test was
performed using dilute hydrochloric acid. However, the HPF6 that was
formed during the stripping step can be lost to the aqueous phase due
to the solubility of the acid HPF6 in water. Ionic liquid [C4mim][NTf2]
has high hydrophobicity and the loss of ionic liquid will decrease during
the extraction and stripping process.
In the present study, studies on lithium extraction from the Mg(II)-
containing salt lake brine are carried out using TBP in imidazolium
based ionic liquids [C4mim][NTf2]. The effects of several experimental
parameters such as extractant concentration and phase ratio (O/A) on
the extraction behavior of metal ions have been investigated and the re-
sults illustrated that [C4mim][NTf2] was a promising solvent to replace
traditional organic solvents (VOCs) in liquid/liquid extraction. The
UV–visible and Fourier transform infrared spectroscopies have been
performed to investigate the interactions between ligands and metal
ions. Numerical treatments and graphical methods have been used to
determine the stoichiometry of the complexes extracted, then the stoi-
chiometry has been confirmed by slope analysis method and Fourier
transform infrared spectra. The extraction equilibrium constants have
also been calculated based on the solvent extraction data. The extraction
mechanism was also studied and a cation exchange mechanism be-
tween [C4mim+] and lithium ion was proposed.
2. Experimental
2.1. Materials and apparatus
Tributyl phosphate (98.5% purity) was purchased from Tianjin
Yongda Chemical Reagents Development Center (China) and used with-
out further purification. Room temperature ionic liquids, [C4mim]Cl (1-
butyl-3-methylimidazolium chloride) and [C4mim][NTf2] (1-butyl-3-
methylimidazolium bis(trifluoromethylsulfonyl)imide), were procured
from Lanzhou Institute of Chemical Physics, CAS (Lanzhou, China) and
used as received. Initial concentrations of Li+ and Mg2+ in the salt lake
brine were determined before extraction. And they were maintained at
2.02 g·L−1 and 92.34 g·L−1 for all the studies, respectively. Mg2+ con-
centration before extracted was determined by titration with standard
EDTA using chrome black T as indicator. Li+ concentration was deter-
mined by atomic absorption spectroscopy using a GBC (Melbourne,
Australia) 908 atomic absorption spectrophotometer, and the concentra-
tion in organic phase was got by mass balance. UV–visible spectroscopic
studies were carried out using a Shimadzu (Kyoto, Japan) UV-2600 single
beam spectrophotometer using quartz cells and suitable reference solu-
tions. Fourier transform infrared spectroscopy (FTIR) measurements
were performed on a Thermo Nicolet Corporation 670 spectrometer.
The slope analysis was used to determine the extraction equations. All
other reagents were of analytical grade.
2.2. Liquid–liquid extraction
For the metal extractions, a required amount of [C4mim][NTf2] was
added into TBP to form the organic phase. The organic phase and the
aqueous solution containing Li+ or mixed metal ions (Mg2+, Na+ and
K+) were added to a separatory funnel and were shaken on a
641
spectrophotometer at room temperature (303 ± 1 K) for 20 min to en-
sure equilibrium. Afterwards, the organic phase and the aqueous phase
were separated. And the separated organic phase and aqueous phase
were both clear and transparent. After phase disengagement, the aque-
ous phase was properly diluted and the concentration of lithium ions
was measured using an atomic absorption spectrometer. The concentra-
tion of Mg2+ in the aqueous phase was determined by using chrome
black T as indicator and EDTA as titrant. The metal concentrations in or-
ganic phase were calculated based on mass balances.
The extraction efficiency (E), the distribution ratio (D) and the sepa-
ration factor (β) were calculated according to the following equations:
C0 −Ce
E ð%Þ ¼  100 ð1Þ
C0
C0 −Ce Vaq
D¼  ð2Þ
Ce Vorg
DLi
β¼ ð3Þ
DMg
where C0 and Ce (mg·L−1) are the initial and equilibrated concentra-
tions of metal ion in the aqueous phase, respectively. Vaq and Vorg
(ml) represent the volume of the aqueous phase and organic phase,
respectively.
3. Results and discussion
3.1. Effect of HCl concentration
The extraction behaviors of lithium ions by the TBP/[C4mim][NTf2]
system with various concentrations of hydrochloric acid in the aqueous
phase were studied. The HCl concentration was varied over the range 0–
1.0 mol·L−1. As shown in Fig. 1, the increasing HCl concentration
showed a negative effect on the lithium ions' extraction performance.
At 0.1 mol·L− 1 HCl, the extraction efficiency of Li+ was 76.06%, but
the extraction efficiency gradually decreased to 34.62% at 1.0 mg·L−1
HCl. As a result, the distribution ratio (D) decreased from 3.18 to 0.53.
The acidity dependence can be explained by the competition between
the extraction of protons and the extraction of lithium ions. Extraction
of hydrochloric acid by the TBP can lead to a decrease in the effective li-
gand concentration in the organic phase, thereby decreasing lithium
ions' extraction. Since the pH of the brine is 5.58, the brine can be
used for extracting directly without adding acid to the aqueous phase.
Fig. 1. Effect of HCl concentration on Li+ extraction. Organic phase = 10% [C4mim][NTf2];
O/A = 1/1; T = 303 K; [HCl] = 0–1.0 mol·L−1.
642 C. Shi et al. / Journal of Molecular Liquids 215 (2016) 640–646

3.2. Effect of dosage of ionic liquid

It is well known that the ionic liquids played a key role in the extrac-
tion process. So the volume of ionic liquids used for extraction of metal
ions was investigated. For this purpose, different amounts of
[C4mim][NTf2] were mixed with TBP and taken into the separatory fun-
nel for isolation process of lithium ions. The [C4mim][NTf2] volume per-
centage in organic phase was varied in the range of 0–30% and the
obtained results were presented in Fig. 2. Obviously the extraction effi-
ciency increased rapidly with increasing ionic liquid volume percentage
in the range of 0–10%, then follows a slow decline. The maximum of ex-
traction efficiency was about 81.75% and it decreased from 81.75% to
70.39% by increasing ionic liquid volume percentage to 30%. Meanwhile,
the distribution ratio (DLi) decreased from 4.48 to 2.38. This suggested
that adding too much ionic liquids had diluted the organic phase,
which reduced the probability of reaction between lithium ions and
TBP. It is important to mention here that the extraction of Li+ was
very low with pure TBP though the extraction efficiency of 8.20% was
Fig. 3. Effects of phase ratio on the extraction efficiencies of metal ions. Organic phase =
obtained. This situation could be attributed to the fact that a solvation 10% [C4mim][NTf2]; O/A = 1/2 to 3/1; T = 303 K; pH = 5.58.
type mechanism is dominant with nonionic organic phase. However,
an ion exchange mechanism often occurs during solvent extraction
using imidazolium based ionic liquids as solvents. And the ion exchange
mechanism leads to high extraction efficiency in the extraction of metal evaluate the potential use of ionic liquid to replace traditional volatile
ions in ionic liquid systems. Finally, neither stable emulsion nor third organic solvents in liquid/liquid extraction, the contrast tests were car-
phase formation was observed in this series of experiments. For further ried to study the extraction of lithium ion using tri-n-butyl phosphate
studies of lithium extraction, an organic phase containing 10% of ionic (TBP) in some volatile organic solvents like chloroform and sulfonated
liquid was chosen. (See Fig. 3.) (See Fig. 9.) kerosene under identical experimental conditions. However, the extrac-
tion efficiency was only 7.82 and 7.10%, respectively. This behavior has
been attributed to a change in the extraction mechanism from solvation
3.3. Effect of O/A phase ratio type prevalent in molecular solvents to cation-exchange type in ionic
liquids which helped in the efficient extraction of the metal ion. Consid-
The phase ratio may have considerable influence on the extraction ering the higher extraction efficiency and cost reduction, the O/A phase
process. The effect of phase ratio on the lithium ion extraction was in- ratio of 2 was used for further studies.
vestigated covering different volume ratios of organic phase to aqueous
phase from 0.5 to 3. Obviously the extraction efficiency increased with
the increasing of the O/A phase ratio by ionic liquid [C4mim][NTf2] dilut- 4. Extraction mechanism
ed in TBP. When ratios were higher than 2, the extraction efficiency had
reached 92.37%. Meanwhile we found that the extraction efficiency of Since [C4mim][NTf2] displays a very good extractability for Li+ but
magnesium ion was 2.91%. And the extraction efficiency for Li+ slightly not good for Mg2+, the extraction mechanism of Li+ has been discussed
increased from 92.37% to 94.61% by increasing phase ratio to 3. When here. It is well known that an ion exchange mechanism often occurs dur-
the O/A phase ratio was 2, the DLi and the DMg were 6.05 and 0.015, re- ing solvent extraction using imidazolium based ionic liquids as diluents.
spectively. The separation factor (β = DLi / DMg) reached 403.33, which In order to confirm the existence of the ion-exchange mechanism, the ef-
indicated that this ionic liquid system showed considerable extraction fect of the concentration of relative cation and anion species on the
ability for lithium ion from salt lake brine of a high Mg/Li ratio. To

Fig. 4. Effect of aqueous-phase [C4mim][Cl] concentration on distribution ratios and

Fig. 2. Effect of ionic liquid concentration on lithium ion extraction. Organic phase = 0–30% extraction efficiency of lithium ions using TBP in [C4mim][NTf2]. Organic phase =
[C4mim][NTf2]; O/A = 1/1; T = 303 K; pH = 5.58. 10%[C4mim][NTf2]; O/A = 1/1; pH = 5.58; T = 303 K.
 C. Shi et al. / Journal of Molecular Liquids 215 (2016) 640–646
Fig. 5. Linear relationship between ([log DLi + log[C4mim]+aq −log[C4mim]+org]) and
log[TBP]org in Li+ extraction equilibrium. Organic phase = 10% [C4mim][NTf2]; O/A = 2/
1; pH = 5.58; T = 303 K.
extractability of Li+ was studied. If the Li+ transfer involves C4mim+, the
extractability of Li+ would decrease with increasing the [C4mim+] con-
centration in the aqueous phase based on the equilibrium shift. In
order to check if cation [C4mim+] was released into the aqueous phase
from the organic phase via the involvement of the extraction reaction,
the effect of [C4mim+] on the extraction has been investigated with
the increasing concentration of [C4mim+] by the addition of hydrophilic
ionic liquid [C4mim][Cl]. If the cation exchange mechanism is dominant,
the addition of the cation part of ionic liquid (i.e. [C4mim+]) into the ini-
tial aqueous phase should decrease the extraction efficiency of lithium.
As shown in Fig. 4, the degree of Li+ extraction in the ionic liquid sys-
tem is gradually reduced with increasing [C4mim+] concentration. It
can be concluded that the cation [C4mim+] from ionic liquid phase
should be exchanged into the aqueous phase during the extraction pro-
cess. The increasing concentration of [C4mim+] would prevent the ex-
traction reaction from moving to the right, which leads to the
decreasing extraction. It can demonstrate that cation [C4mim+] was in-
volved into the extraction reaction. Based on the above hypotheses, the
extraction reaction equation can be supposed as follows:
Liþ þ þ þ
aq þ C4 mimorg þ nTBP → ½Li  nTBPorg þ C4 mimaq : ð4Þ
Fig. 6. UV–vis spectra of the aqueous phase before and after extraction. a: before
extraction; b: after extraction.
643
According to Eq. (4), the reaction equilibrium constant (K) can be
expressed as:
 þ
½Li  nTBPþ
org C4 mim aq
K¼ þ  ð5Þ
Liaq C4 mimþ org ½TBPnorg
 
C4 mimþ aq
K ¼ DLi   ð6Þ
C4 mimþ org ½TBPnorg
   
log DLi þ log C4 mimþ aq − log C4 mimþ org ¼ nlog½TBPorg þ log K ð7Þ
where the subscripts aq and org denote the aqueous phase and organic
phase, respectively. The reaction equilibrium constant K is only related
to the temperature for an extraction reaction. Slope analysis was conduct-
ed as a function of the equilibrium concentrations of TBP in the extracting
phases to determine the stoichiometry of the metal-TBP complex. The
TBP volume percentage in organic phase was varied in the range of 40–
90%. And in order to maintain vol% of the ionic liquid in organic phase,
the dichloromethane was used as another diluent in this part. As shown
in Fig. 5, the plots of ([log DLi + log[C4mim]+aq −log[C4mim]+org]) versus
log[TBP]org showed a straight line with a slope of 0.95, which suggested
that only one molecule of TBP was involved during the extraction process
for lithium ions. In our previous studies, the extraction complex was iden-
tified as [Li·2TBP]+ in TBP/[Cnmim][PF6] (where, n = 4, 6, 8) extraction
systems. However, the extracted Li species in this study is [Li·TBP]+.
This phenomenon would be distributed for at least two main reasons.
On the one hand, both of the [Li·2TBP]+ and [Li·TBP]+ are monovalent
cations. The small cations with higher surface charge density lead to
stronger electrostatic interactions and to the formation of more stable
species with the anion [NTf2]−. On the other hand, as the anionic radius
increased, the steric hindrance of [Li+][NTf− 2 ] increased accordingly,
which leads to only one TBP molecule which was included in the
complex.
In conclusion, the extraction reaction of lithium ions by TBP in
[C4mim][NTf2] can be described as the following equation:
Liþ þ þ þ
aq þ C4 mimorg þ TBP → ½Li  TBPorg þ C4 mimaq : ð8Þ
To further understand the cation exchange mechanism during the ex-
traction of Li+ using the extractant TBP in an ionic liquid [C4mim][NTf2],
UV–visible spectroscopic studies were carried out. According to the cat-
ion exchange mechanism, the concentration of imidazolium cations in
aqueous phase was usually increased as the ionic liquid solubility plus
the exchanged amount of [C4mim+]. So we employed a UV–visible spec-
trometer to record the UV–vis spectra of the aqueous phase before and
after extraction equilibrium of lithium ions into the organic phase
(Fig. 6). As shown in Fig. 6, the aqueous phase showed one ultraviolet ab-
sorption peak (λmax = 211 nm). And the absorption peaks at 211 nm be-
longs to the cation of [C4mim+]. The curve b in Fig. 6 showed that the
enhanced absorbance of the spectra of aqueous phase after extraction
was obtained, though the characteristic spectral features were very sim-
ilar to the aqueous solution spectra before extraction. This result sug-
gested that the transfer of cationic part of the ionic liquid to the
aqueous phase well and the extraction of Li+ by TBP/[C4mim][NTf2] oc-
curred in a cation exchange mode as represented in Eq. (8).
FTIR spectroscopy is a chemical-microenvironment-sensitive and
functional-group-characteristic technique. These measurements were
carried out to clarify the existence of interactions between lithium ion
and TBP. The FTIR spectra of organic phase before and after extraction
have been analyzed and the results were shown in Fig. 7. The bands
from 1210 to 1350 cm− 1 in the IR of the organic phase, which were
assigned to the P_O stretching vibration of TBP [31,32]. A literature sur-
vey showed that the peak around 1282 cm−1 was assigned to the P_O
stretching vibration of TBP and it was found that when a second compo-
nent was involved with TBP phase through the phosphoryl band, P = O,
644 C. Shi et al. / Journal of Molecular Liquids 215 (2016) 640–646

Fig. 7. FT-IR spectra of the organic phase before and after extraction. a. TBP + [C4mim][NTf2] (before extraction); b. TBP + [C4mim][NTf2] (after extraction).

the intensity of the P_O stretching vibration will decrease and the peak
shifted to the lower wavenumber [31,32]. As shown in Fig. 7, it is clear
that some main characteristic peaks' positions of TBP (asymmetric –
CH3 stretching vibration, 2963 cm−1; symmetrical C–H stretching vi-
bration, 2892 cm− 1; asymmetric P–O–C stretching vibration,
1028 cm− 1) were not changed after extraction. However, TBP phase
loading Li+ could lead to shift of the P_O stretching vibration from
1282 to 1258 cm− 1. The shift of the P_O frequency with respect to
the net TBP was about 24 cm−1, which may be attributed to an interac-
tion between lithium ion and TBP. Furthermore, an important vibration-
al characteristic band of anion [NTf− 2 ] is C–F. And this vibrational
characteristic band occurred at 1224 and 1137 cm−1 [33,34]. As seen
in Fig. 7, the presence of lithium ion in the organic phase did not modify
the FTIR vibration bands of the [NTf− 2 ] anion. However, the intensities of
the peaks at 3154 and 3102 cm−1, which were attributed to the C–H
ring stretching vibration of the imidazolium cation [35], decreased
after extraction, indicating a decrease in the [C4mim+] concentration
in the organic phase during lithium ion extraction. Therefore, FTIR

Fig. 8. (a) Effect of temperature on the extraction of Li+; (b) plot of log K versus 1000/T for
Li+ extraction. Organic phase = 10% [C4mim][NTf2]; O/A = 1/1; pH = 5.58; T = 303– Fig. 9. Percentage stripping of lithium ion from the organic phase with different
343 K. concentrations of hydrochloric acid. O/A = 1/2; T = 303 K; [HCl] = 0–3.0 mol·L−1.
 C. Shi et al. / Journal of Molecular Liquids 215 (2016) 640–646
spectroscopy confirmed the presence of a cation exchange equilibrium
between Li+ and [C4mim+] during lithium ion extraction as in Eq. (8).
5. Determination of thermodynamic parameters
To study the effect of temperature on the extraction of Li+ from salt
lake brine by TBP in [C4mim][NTf2], experiments were carried out in a
thermostatic water bath oscillators and the temperature was varied
over the range 303–343 K. Fig. 8a showed that an increase in tempera-
ture from 303 to 343 K leads to a decrease in the percentage of lithium
ion extraction from 92.37% to 86.27%. According to the van't Hoff equa-
tion and other thermodynamic equations, the following equations can
be obtained and the change in enthalpy (ΔH°) during the extraction
can be calculated.
ΔH 1
log K ¼ −  þC ð9Þ
2:303R T
where R is the universal gas constant and C is the integration constant,
which was assumed to be constant at a particular temperature under
the experimental conditions.
Fig. 8b showed that log K versus 1000/T yielded a straight line with a
slope of 0.48 equal to −ΔH°/2.303R. Therefore, the extraction process is
exothermic with ΔH° = −9.19 kJ·mol−1.
The change in Gibbs free energy (ΔG°) was calculated using Eq. (10):
ΔG∘ ¼ −2:303RT log K ð10Þ
where K is the extraction equilibrium constant. According to Eq. (7),
the log K-value is calculated from the intercept of a plot of ([log
D Li + log[C 4 mim]+ aq − log[C 4 mim] + org ]) versus log[TBP] org . The
change in entropy (ΔS°) at a particular temperature can be calculat-
ed using Eq. (11):
ΔH∘−ΔG∘
ΔS∘ ¼ : ð11Þ
T (−ΔH°) with the formation of highly ordered complexes (−ΔS°) in the
The thermodynamic values over the temperature range 303–343 K,
such as enthalpy change (ΔH°), Gibbs free energy change (ΔG°), entro-
py change (ΔS°), were calculated using Eqs. (9), (10) and (11) and listed
in Table 1. The negative values of enthalpy changes (ΔH°) signified the
exothermic nature of the extraction reaction. The negative values of
Gibbs free energy (ΔG°) at all temperatures confirmed the feasibility
of the process and spontaneous nature of the extraction reaction. The
value of ΔS° (− 27.85 J·K− 1·mol− 1) suggested that the degree of
order had increased during the extraction process.
6. Stripping studies
Stripping is an important feature for evaluating an extraction sys-
tem. Since the efficiency of lithium ion extraction by TBP in ionic liquid
reduced with increasing acid concentration, it was likely that all the
loaded metal ions have to be stripped from the organic phase by acidic
solutions. After one stage extraction, the concentration of lithium ion
and magnesium ion in the organic phase was 0.93 g·L and 1.34 g·L, re-
spectively. The stripping test for Li+ extracted by TBP in [C4mim][NTf2]
was performed using different concentrations of dilute hydrochloric
acid. It was observed that the extraction of Li+ was very low with
pure water though the extraction efficiency of 6.33% was obtained.
And 91.02% of Li+ was stripped back from [C4mim][NTf2] phase with a
Table 1
Thermodynamic parameters (ΔG°, ΔH° and ΔS°) for Li+ extractions from salt lake brine
using TBP as an extractant in [C4mim][NTf2] at 303 K.
Metal ion log K ΔH° (kJ·mol−1) ΔG° (kJ·mol−1) ΔS° (J·K−1·mol−1)
Li+ 0.13 −9.19 −0.75 −27.85
645
0.5 M HCl solution. In particular, the lithium stripping percentage rose
significantly to 97.58% using 2 M hydrochloric acid solution, which indi-
cates that Li+ is easily stripped from the loaded ionic liquid phase. At the
same time, the stripping rate of magnesium ion had been determined
and it was close to 100%. The Mg/Li ratio was 1.48 in the aqueous
phase after stripping which has dropped 96.78% when it compared
with the initial value. The reason for efficient transfer of Li+ to the
strip liquor under acidic conditions was exchange of Li+ ions by H+
ions in the organic phase. By working at high acid concentrations, the
extraction of acids is favored over the extraction of metal ions and the
extracted metal ions are released from the organic phase to the aqueous
phase. When the Mg/Li ratio was less than 6, the precipitation technique
can effectively separate Li and Mg. However, once the Mg/Li ratio
exceeded 6, the precipitation technique appeared so limited [36]. In
this system, the Mg/Li ratio was less than 6 in the aqueous phase after
stripping. After concentrating the aqueous phase and precipitating the
magnesium ion with alkali, the magnesium hydroxide could be separat-
ed after filtration. Lithium carbonate will be obtained by adding sodium
carbonate to the aqueous phase.
7. Conclusions
This systematic experimental study of the extraction of lithium ion
by the neutral extractant TBP in ionic liquid with the bis(trifluorometh-
ylsulfonyl)imide anion showed that it was an efficient extraction sys-
tem for the recovery of lithium ion from the solution containing
magnesium ion due to the better extraction selectivity. Cation exchange
mechanism between Li+ and [C4mim+] had been found to be responsi-
ble for the extraction of lithium ion into the organic phase, which was
further supported by the UV–visible spectroscopic study. Based on the
study of thermodynamics, thermodynamic parameters of the lithium
ion extraction reaction were obtained. The calculated thermodynamic
parameters (ΔH°, ΔS° and ΔG°) suggest that the extraction of lithium
ion was spontaneous (−ΔG°) in nature and the process was exothermic
organic phase. The stripping of the lithium ion from the loaded ionic liq-
uid phase has been investigated. The mechanism of the stripping reac-
tion is exchange of lithium ion by protons in the organic phase. The
present findings suggest that the application of ionic liquid as
alternative media for extracting phases can broaden the potential utility
of liquid–liquid extraction processes and the recovery of valuable
metals from salt lake brine.
Acknowledgments
This research was supported by the National Natural Science Foun-
dation of China (U1407717 and U1407205).
References
[1] T. Sinha, M. Ahmaruzzaman, A. Bhattacharjee, M. Asif, V.K. Gupta, J. Mol. Liq. 201
(2015) 113–123.
[2] A.V. Vertkov, I.E. Lyublinski, M.Y. Zharkov, V.V. Semenov, E.A. Azizov, V.B. Lazarev,
S.V. Mirnov, Fusion Eng. Des. 89 (2014) 996–1002.
[3] J.S. Chae, M.R. Jo, Y.I. Kim, D.W. Han, S.M. Park, Y.M. Kang, K.C. Roh, J. Ind. Eng. Chem.
21 (2015) 731–735.
[4] L. Li, S.T. Song, X.X. Zhang, R.J. Chen, J. Lu, F. Wu, K. Amine, J. Power, Sources 272
(2014) 922–928.
[5] P. Fiflis, A. Press, W. Xu, D. Andruczyk, D. Curreli, D.N. Ruzic, Fusion Eng. Des. 89
(2014) 2827–2832.
[6] A.H. Hamzaoui, B. Jamoussi, A. M'nif, Hydrometallurgy 90 (2008) 1–7.
[7] L.I. Barbosa, J.A. Gonzalez, M.D. Ruiz, Thermochim. Acta 605 (2015) 63–67.
[8] S. Zandevakili, M. Ranjbar, M. Ehteshamzadeh, Hydrometallurgy 149 (2014)
148–152.
[9] G. He, L.Y. Zhang, D.L. Zhou, Y.W. Zou, F.H. Wang, Ionics 21 (2015) 2219–2226.
[10] N. Um, T. Hirato, Hydrometallurgy 146 (2014) 142–148.
[11] V.I. Kuz'min, N.V. Gudkova, Solvent Extr. Ion Exch. 33 (2015) 183–195.
[12] J.L. Li, H.F. Zhu, M. Wang, L.J. Shi, Y.J. Zhao, H.T. Zhang, F. Ge, W.Q. Kang, J. Gao, Chin. J.
Inorg. Chem. 30 (2014) 2389–2393.
[13] Z.Y. Zhou, W. Qin, W.Y. Fei, J. Chem. Eng. Data 56 (2011) 3518–3522.
646 C. Shi et al. / Journal of Molecular Liquids 215 (2016) 640–646
[14] B. El-Eswed, M. Sunjuk, Y.S. Al-Degs, A. Shtaiwi, Sep. Sci. Technol. 49 (2014)
1342–1348.
[15] S.Y. Sun, F. Ye, X.F. Song, Y.Z. Li, J. Wang, J.G. Yu, Chin. J. Inorg. Chem. 27 (2011)
439–444.
[16] C.V. Manohar, T. Banerjee, K. Mohanty, J. Mol. Liq. 180 (2013) 145–153.
[17] J. Han, Y. Wang, C. Chen, W.B. Kang, Y. Liu, K.K. Xu, L. Ni, J. Mol. Liq. 193 (2014)
23–28.
[18] M.T. Coll, A. Fortuny, C.S. Kedari, A.M. Sastre, Hydrometallurgy 125 (2012) 24–28.
[19] M. Matsumiya, Y. Kikuchi, T. Yamada, S. Kawakami, Sep. Purif. Technol. 130 (2014)
91–101.
[20] K. Larsson, K. Binnemans, Green Chem. 16 (2014) 4595–4603.
[21] J. Castillo, M.T. Coll, A. Fortuny, P.N. Donoso, R. Sepulveda, A.M. Sastre, Hydrometal-
lurgy 141 (2014) 89–96.
[22] K. Sasaki, T. Suzuki, T. Mori, T. Arai, K. Takao, Y. Ikeda, Chem. Lett. 43 (2014)
775–777.
[23] J. Fu, Q.D. Chen, T.X. Sun, X.H. Shen, Sep. Purif. Technol. 119 (2013) 66–71.
[24] Y.L. Shen, W.K. Li, J.R. Wu, S. Li, H.M. Luo, S. Dai, W.S. Wu, Dalton Trans. 43 (2014)
10023–10032.
[25] L.Y. Yuan, M. Sun, X.H. Liao, Y.L. Zhao, Z.F. Chai, W.Q. Shi, Sci. China: Chem. 57 (2014)
1432–1438.
[26] D.R. Raut, P.K. Mohapatra, Sep. Sci. Technol. 50 (2015) 380–386.
[27] C.L. Shi, D.P. Duan, Y.Z. Jia, Y. Jing, J. Mol. Liq. 200 (2014) 191–195.
[28] C.L. Shi, Y. Jing, J. Xiao, F.L. Qiu, Y.Z. Jia, CIESC J. 66 (2015) 265–271.
[29] C.L. Shi, Y.Z. Jia, C. Zhang, H. Liu, Y. Jing, Fusion Eng. Des. 90 (2015) 1–6.
[30] C.L. Shi, Y.Z. Jia, Y. Jing, CIESC J. 66 (2015) 253–259.
[31] S.F. Shen, Z.D. Chang, J. Liu, X.H. Sun, X. Hu, H.Z. Liu, Sep. Purif. Technol. 53 (2007)
216–223.
[32] D.H. Fatmehsari, D. Darvishi, S. Etemadi, A.R.E. Hollagh, E.K. Alamdari, A.A. Salardini,
Hydrometallurgy 98 (2009) 143–147.
[33] T.N. Viet, J.C. Lee, J. Jeong, B.S. Kim, G. Cote, A. Chagnes, Ind. Eng. Chem. Res. 54
(2015) 1350–1358.
[34] Y.Y. Ao, J. Peng, L.Y. Yuan, Z.P. Cui, C. Li, J.Q. Li, M.L. Zhai, Dalton Trans. 42 (2013)
4299–4305.
[35] Y.J. Zheng, W.J. Eli, G. Li, Colloid Polym. Sci. 287 (2009) 871–876.
[36] Z.W. Zhao, X.F. Si, X.H. Liu, L.H. He, X.X. Liang, Hydrometallurgy 133 (2013) 75–83.