Desalination 436 (2018) 8–14

Contents lists available at ScienceDirect

Desalination
journal homepage: www.elsevier.com/locate/desal

Lithium capturing from geothermal water by hybrid capacitive deionization T

a,⁎ b a
Anna Siekierka , Barbara Tomaszewska , Marek Bryjak
a
Wroclaw University of Science and Technology, Department of Polymer and Carbon Materials, Wyb. St. Wyspianskiego 27, 50-370 Wrocław, Poland
b
AGH University of Science and Technology, Faculty of Geology, Geophysics and Environmental Protection, Mickiewicza 30 Av., 30-059 Kraków, Poland

G RA P H I C A L AB S T R A C T

A R T I C L E I N F O A B S T R A C T

Keywords: We demonstrate a novel mode for hybrid capacitive deionization, HCDI, that allows selective extraction of
Lithium lithium ions from geothermal water. The HCDI cell was comprised with cathode made of lithium selective
Hybrid capacitive deionization material and anode consisting of activated carbon coated with anion-exchange membrane. To confirm ion-
Geothermal water exchange character of the process, based on ions diffusion/migration into electrodes, impedance spectroscopy
Selectivity
and cyclic voltammetry were applied. By optimizing the process mode it was possible to obtain lithium enriched
stream and remove up to 73% of lithium captured by the electrode. The process was characterized by an ex-
tremely high salt adsorption capacity of 800 mg/g and total energy consumption on the level of 0.183 Wh per 1 g
of adsorbed salt.

1. Introduction
Lithium has been of significant interest in many fields of science and
technologies. The global market of lithium salt is growing fast and it is
expected to reach 30 $ billion value by 2020. Increasing rapidly de-
mand of this element is caused by its application for electronics and
batteries mostly [1]. The available lithium resources can be divided into
two main categories: natural like minerals, salt lakes, underground
brines or seawater, and artificial like recycle from electronic waste and
lithium ions batteries [2]. It should be mentioned here that seawater,
the richest resources of lithium, is still out of technological interest due
to low lithium concentration of 0.1–0.2 mg/dm3. From the technolo-
gical point of view, the most promising lithium resource is groundwater
where concentration of Li+ is between 10 and 15 mg/dm3 [3]. That
situation makes geothermal water, and its reverse osmosis (RO) brines,
to be perspective source for lithium extraction.
Geothermal water desalination require an electric or thermal tech-
nology to concentrate or to precipitate particular elements. A variety
technologies have been developed over years. RO and multistage flash
distillation require a large amount of energy that generates almost 50%

⁎
Corresponding author.
E-mail address: anna.siekierka@pwr.edu.pl (A. Siekierka).

https://doi.org/10.1016/j.desal.2018.02.003
Received 20 October 2017; Received in revised form 17 January 2018; Accepted 1 February 2018
0011-9164/ © 2018 Elsevier B.V. All rights reserved.
A. Siekierka et al. Desalination 436 (2018) 8–14

Table 1 concentrate were monitored by the following parameters – tempera-

Comparison of energy consumption for water treatment technologies.
Technology
Electrodeionization (EDI)
Reverse osmosis (RO)
Electrodialysis (ED)
Conventional CDI
Energy consumption
Wh/g kWh/m3
– 7.7
– 1.1
0.99 – [8]
– 0.4
ture, conductivity, and alkalinity – immediately after sampling solution
from the recycling tank. Inorganic components were measured at the
Ref.
accredited laboratory (PCA-AB 1050 and PCA-AB 176) using the in-
ductively coupled plasma mass spectroscopy method ICP-MS (ELAN
6100) and Inductively Coupled Plasma Optical Emission Spectrometer,
[6]
ICP-OES (Optima 7300DV). Details about preparation of samples and
[7]
their measurements are presented elsewhere [16–18].
[9]
2.3. Principles of operation

of total production cost [4,5]. A search for less energy consuming
technique was finalized with launching capacitive deionization, CDI,
based on the use of supercapacitors. The comparison presented in
Table 1 puts CDI as the most economical method.
The development of CDI has initiated a search for other methods
with architecture similar to capacitive deionization but applying dif-
ferent electrode materials. One of such approaches is to use desalination
batteries that consist of cation insertion electrode and chloride-cap-
turing electrode, most frequently made of silver [5].
The goal of this paper was to demonstrate an efficiency of new
hybrid capacitive deionization system, HCDI, for selective recovery of
lithium ions from real geothermal water. The evaluated system was
comprised of lithium-capturing electrode, built with lithium‑tita-
nium‑manganese oxide material, LMTO, and activated carbon electrode
coated with anion-exchange membrane made by poly(vinyl chloride)
modified with ethylene diamine (PVC-EDA). At the first, it was shown
innovative modes for running HCDI process that allowed to selectively
remove lithium ions from multicomponent aqueous solution. To express
the selective properties of HCDI we selected the geothermal water from
Carpathian region in Poland. Secondly, the performances of HCDI were
expressed by the energetic factors and related to selective lithium
desorption in spots of electrochemical measurements.
2. Materials and methods
2.1. HCDI system design
The batch-mode HCDI test cell consisted of two parallel electrodes
separated by a polypropylene spacer of 200 μm thickness to allow liquid
solution flow through a channel. As an cathode material the spinel li-
thium-manganese-titanium oxide, LMTO, characterized in our pre-
viously study [10–13], was used. The LMTO adsorbent is characterized
by a similar crystallinity to a H0.6Li0.08Mn1.72O4 (reference code: 01-
089-0755) [9]. Activated carbon, AC, YP-50F Kuraray Chemical Co.
LTD. Osaka, Japan [14] was applied to form anode. LMTO electrode
was used for selective capturing of Li+ ions and a AC electrode to
capture anions. To prevent the effect of co-ions adsorption in the des-
orption step [4] an anion-exchange membrane was put on the surface of
the AC electrode. As an anion-exchange membrane, AEM, we selected
poly(vinyl chloride) modified by ethylene diamine, PVC-EDA. Details of
membrane modification and preparation can found elsewhere [12,15].
The whole system consisted of HCDI cell, recycling tank, peristaltic
pump (Elan, type 372.C), a digital electrical conductivity meter, pH
meter and DC power supplier (KD3005D Digital-Control DC Power
Supply). All experiments were conducted at constant voltage mode, CV
with value of 2 V, at 25 °C.
2.2. Geothermal water quality
Tests were carried out using geothermal water with high miner-
alisation (TDS 2.3–2.5 g/dm3), > 270 mg CaCO3/dm3 and lithium
concentration > 15 mg Li+/dm3. The physical and chemical composi-
tion of crude water is presented in Table 2. The used geothermal water
could be classified to the hydrogeochemical type 2.6% Cl-Na + I group.
During running the HCDI process, the quality of feed water and
For the evaluation of HCDI system in lithium recovery from geo-
thermal water two sets of process parameters were chosen. The first one
was composed of CV-RCV-ZVC sequence, what meant adsorption at
constant voltage (CV), desorption under reversing voltage (RCV) and
final desorption under zero charge voltage (ZVC). The second set
comprised CV-ZVC-RCV operations, where the desorption steps were
conducted at opposite sequence as for CV-RCV-ZVC mode. The external
voltage applied in the CV and RCV modes was 2 V. The time run of the
HCDI process is presented in Fig. 1. The sequences of used process
operations are juxtaposed in Table 3.
During the first step, the CV mode was applied and ions were re-
moved from feed solution. It took place during first 3 min. The desali-
nated geothermal water was then collected and marked as solution A.
The system was filled with deionized water (DI) and two stages of
desorption process were carried out. Under RCV or under ZVC condi-
tions, both lasting 1 min, adsorbed salts were removed. It was the first
step of desorption. The obtained solutions were collected and marked as
D1 and the system was filled with fresh DI water. In the next step, the
second desorption process, lasting 5 min, was carried out. Charged
electrodes released ions incorporated into volume of electrodes and
produce enriched solution marked D2. Finally, the HCDI system was
refilled with new portion of geothermal water and desalination proce-
dure was repeated. Hence, the innovative system of removal lithium
salt from geothermal water worked with following steps: adsorption A
(performed always in CV mode), first desorption D1 (performed in RCV
or ZVC mode) and second desorption D2 (also performed in relation
ZVC or RCV mode). The indices A, D1 and D2 described the used
electrical mode and were related to different salt concentration; thus A
assigned solution after total salt adsorption while D1 and D2 showed
solution obtained after first and second desorption.
2.4. Process metrics
Salt adsorption capacity (SAC) is a measure of extracted salt nor-
malized to amount of used adsorptive material. When the amount of
adsorbed salts was normalized to the process time and to the area of
electrode (11.25 cm2) we were able to get the average salt adsorption
rate (ASAR), an useful metrics for process description. To characterize
stack working in CV mode, the charge efficiency Λ (arb.uni.) was ap-
plied:
Λ = [(C0 − Cf )·V·F]/ΣM·Q (1)
−1
where F is the Faraday constant, F = 96,485 (C·mol ), ΣM is an
average molar mass of salts in geothermal water, Q is a charge calcu-
lated by integration of current and time of the system charging.
Total energy consumption (Etotal) was the sum of energy for ad-
sorption (Eads) and for desorption (Edes) processes. These metrics are
computed from numerical integration of current versus time relation-
ship [19,20]. Next useful metrics for characterization of HCDI process
was energy normalized adsorption of salt (ENAS), expressed in grams
per Joules of the total energy. That value quantifies the energetic effi-
ciency of the desalination process. Last metrics describing hybrid ca-
pacitive deionization process was the electrical work expressed in Wh
per gram of salt. It showed amount of energy needed by the system to
adsorb or desorb 1 g of salt. To present the selective behaviour of HCDI

9
A. Siekierka et al. Desalination 436 (2018) 8–14

Table 2
Composition of geothermal water.

Elements Na+ K+ Li+ Ca2+ Mg2+ Sr2+ Cl− Br− HCO3− B Si Σ

mg/dm3 10,298 102.1 15.7 63.7 50.3 33.5 11,421 128.9 1462 156 11.4 23,743

Fig. 1. Scheme of desalination geothermal water by Hybrid Capacitive Deionization.

Table 3 system we used efficiency ratio ηion (%) for a particular ion:
General outline parameters for CV-RCV-ZVC and CV-ZVC-RCV sets.
ηion = 1 − [(C0 − Cf )/C0)] (2)
Type of set CV-RCV-ZVC CV-ZVC-RCV

Electrical mode CV RCV ZVC CV ZVC RCV

Type of step A D1 D2 A D1 D2 2.5. Electrochemical evaluations
Time of operation min 3 1 5 3 1 5
Applied Voltage V 2 −2 0 2 0 −2 The electrochemical characterization was carried out in two-elec-
Volume of circulating feed dm3 0.1 trode Swagelok@ system containing: LMTO/graphite current collector
Solution flowrate dm3/h 4
as a working electrode and YP-50F activated carbon as a counter
electrode covered by PVC-EDA membrane. Solution of 2 M LiCl in DI
water was used as electrolyte and disc of fiberglass as a separator. The

Fig. 2. Charge – discharge configurations

CV-ZVC-RCV CV-RCV-ZVC according to high efficiency operating
7,8 ADSORPTION DESORPTION1 DESORPTION 2 A procedure for HCDI. The fluctuation of pH
(A), conductivity (B) and current (C) for
investigated approaches.
7,5
pH

7,2

6,9
45 0 60 120 180 240 300 480 540
B
Cond (mS)

30

0,1
0 60 120 180 240 300 480 540
0,4 C
Current (A)

0,10

0,05
0,00
0 60 120 180 240 300 480 540
Time (sec)

10
A. Siekierka et al.
electrochemical impedance spectroscopy (EIS) and cyclic voltammetry
(CV) were conducted on Solarton SI 1260 workstation at 25 °C. To
calculate specific capacitance C (F/g) we divided integral of current and
time by electrical potential and by mass of active material:
C= ∫ Idt/U·mel (3)
2.6. XRD evaluation
To evaluate structural change with relation to the manganese redox
reaction the X-Ray Diffraction was employed. The measurements were
performed in 2θ angle ranges of 5–120°, at room temperature with
0.02° step and at a rate of 3 °C/min using a Philips X'Pert PW 3040/60
diffractometer (Kα = 1.5418 Å) with a Cu lamp (30 mA and 40 kV)
[13].
3. Results and discussion
3.1. Principles of HCDI systems
Phenomena of pH, current and conductivity fluctuations for CV-
RCV-ZVC and CV-ZVC-RCV configurations are presented in Fig. 2. It can
be seen, that adsorption step was not related to the kind of studied
configuration. Differences appeared at the desorption stage. When mass
transport was controlled spontaneously, like in CV-ZVC-RCV config-
uration, D1 solution rapidly and quickly achieved its ionic equilibrium.
In the case of CV-RCV-ZVC configuration, extraction of salts was limited
by Faradic process and charge transfer. Efficiency of salt desorption in
D1 step was similar for both configuration and achieved ~79–81%. In
the D2 step, the salt recovery reached value of ~21–19%. To explain
the different phenomenon of ions adsorption and desorption, the pH
fluctuation was monitored. The adsorption process during the CDI op-
eration could be considered as a combination of transport and ad-
sorption of ions into porous materials, transport of ions from flow
channel to electrode surface and Faradic reaction, according to theory
reported in [21]. If the process is run under capacitive forces (formation
of EDLs in porous materials), the changes of the pH profile will be
negligible [21]. In our case, pH fluctuations were kept at 0.8 unit and
they detected presence of Faradic reaction [21]. That was observed for
D1 case. In CV-RCV-ZVC configuration, reduction of water could gen-
erate OH– ions (2H2O + 2e− ↔ H2 (g) + 2OH−) that should raise the
pH value of the solution. However the reduction of carbon could
compensate that phenomenon (C + H2O ↔ CO2 (g) + 4H+ + 4e−). In
the case of CV-ZVC-RCV configuration the last effect was marginalized.
In the D2 step, the process of ions release from volume/surface of
electrodes was controlled by the Faradic reaction mostly. Hence,
monitoring the process by checking pH vs. time course can be very
useful to control adsorption-desorption behaviour of the HCDI process.
Extremely high salt adsorption capacities are presents in Fig. 3A for
both configuration. The ASAR value in the D1 step was increased by a
2.40 factor. It might be caused by difference of salts concentration in
LMTO material and in DI water. For the D2 step, where salt con-
centration was four times lower, the ASAR decreased 20 times when
compare to the value of the first step, even when it lasted much longer
than D1. Fig. 3B shows the comparison to the literature data [22–32].
Moreover, to elucidate extremely large value of SAC in our in-
vestigations, the maximum adsorption capacity of LMTO was measured.
The LMTO sorbent of 0.11 g was kept in 20 mM LiCl solutions by 24 h.
The maximum salt adsorption capacity was calculated at 114 mg/g.
Hence, the ratio SACHCDI/SACMAX could be high as 7. The explanation
of this phenomenon is given in Sections 3.2 and 3.3.
Energy requirement is one of the most important factor of each
electrosorption process. For CV-ZVC-RCV configuration that value was
two time higher than for CV-RCV-ZVC partner. It was 0.183 Wh and
0.111 Wh per 1 g of salt removal, respectively. In comparison of both
11
Desalination 436 (2018) 8–14
operation modes we can conclude CV-RCV-ZVC system to be more
energy efficient than the CV-ZVC-RCV one (0.061 kWh/m3 against
0.059 kWh/m3). It should be stressed here that the energy consumption
for HCDI is significantly lower that for electrodialysis (ED) – 0.99 kWh/
kg [8], reverse osmosis (RO) – 1,1 kWh/m3 [7] and conventional CDI
with activated carbon electrodes – 0.4 kWh/m3 [9]. These information's
are shown in Table 1.
The concentrations of monovalent ions are given in Table 4. The
ratio of concentration at the beginning was 656:6.5:1 for Na:K:Li, re-
spectively. After one cycle of CV-ZVC-RCV process that value was
changed to 51:0.57:1, it meant it was possible to 12 and 11 times de-
crease concentration of sodium and potassium, respectively. In CV-
RCV-ZVC configuration, the ratio was 273:2.6:1. This comparison al-
lowed us to conclude that the electrical mode operation in desorption
affected crucially the ration of removal monovalent ions from multi-
component solutions.
3.2. Desorption mechanisms
The effect of abnormal ENAS relation to charge efficiency in the CV-
ZVC-RCV configuration and the extremely high SAC values can be ra-
tionalized by formation of thin and highly package layers of ion on the
surface of the electrode. Such phenomenon is usually associated with
formation of solid electrolyte interphases (SEI). The interphase allows
penetration and migration of ions and exhibits high ionic conductivity
[33]. Adsorption step combines either intercalation into volume of
LMTO (~60 m2/g of SSA, crystalline domain of 3.88 Å [6]) and/or
creation of clusters that keep ions on the surface of electrodes (viz. SEI
mechanism). Formation of surface layers is characteristic for the
spontaneous electroconvection process [34]. Generally, mass transport
during adsorption step can be explained by the Faradic process and
accumulation of charge in the electrical double-layer. However, this
mechanism does not provide an inherently high ionic selectivity and is
typical for highly porous materials like activated carbon. In this study,
LMTO was used to build selective electrode that allowed to conduct
redox-reaction and provided ion-exchange processes. Hence, to explain
the following facts: i) extremely high salt adsorption capacity, ii) dif-
ference in adsorption and desorption during the run of HCDI process,
and iii) energy requirements some mechanisms should be merged. We
think that the following phenomena sustained the selectivity of in-
vestigated HCDI system: (1) Faradic processes with formation of electric
double layers (EDLs) [4], (2) spontaneously adsorption related to spe-
cific surface area of adsorbent, (3) redox-reaction of entrapment ions,
(4) formation of multilayer in solid-electrolyte interphases, and (5) ion-
capturing arising with crystalline structure of LMTO.
3.3. Electrochemical performances
The electrochemical performances were tested using evaluated
configurations of HCDI with 2 M LiCl as electrolyte. The CV of LMTO
adsorbent (Fig.4A) showed a decrease of anodic peaks between 1.5 and
2.5 V. The latter peak is assigned to oxidation of Mn+2 to Mn+4 and
also decomposition of LMTO structure [35]. This redox reaction is also
confirmed by XRD evaluation presented in Fig. 5. It can be seen that
structure without intercalated lithium had MnO and Mn2O3 phases and
also small amount of Li2MnO3 [36]. Luther [37] reported that manga-
nese could by oxidised with one-electron transfer that allows formation
of soluble complexes while MnO2 could be reduced with transfer of
two-electrons. However, the conversion reaction at ~2.0 V was attrib-
uted to transformation of MnO/Mn3O4 that confirmed redox reaction
[35,38].This transformation is seen in XRD pattern were the peak of
Mn3O4 was detected in phases with intercalated lithium [39]. It should
be memorized that the active LMTO materials is built of multivalent
manganese oxide that can be oxidised at different voltage range. Ac-
cording to literature the manganese dissolving species are Mn+2,
MnOH+ and anionic form of HMnO2− [41]. In cathodic direction, one
A. Siekierka et al. Desalination 436 (2018) 8–14

Fig. 3. Modified Ragone plots for (A) evaluated

10
A B
configurations and (B) comparison results with
12
D1 literature data.

10
1

8
ASAR (mg/g/s)

ASAR (mg/g/s)
0,1
6
A This study
MC-MCDI ref. 22
MOG- AC ref. 23
4 HPAC ref. 24
AC with Ag ref. 25
0,01 HAC ref. 26
CNT -AC ref. 27
AN-CFs ref. 28
2 NC-rGO ref. 29
D2 CV-ZCV-RCV
MoS2 ref. 30
GCM-1.5 ref. 31
NMO ref.32
CV-RCV-ZCV
1E-3
0 200 400 600 800 1000 1 10 100 1000
SAC (mg/g) SAC (mg/g)

Table 4 slightly peak is observed at above 1.5 V, indicating decomposition of

Energy-related factors of processes and concentrations of monovalent ions. electrolyte and formation of a solid electrolyte interface layer [38].
Presence of SEI layer explained the decrease of LMTO capacitance for
Work ENAS Charge CNa+ CK+ CLi+
efficiency about ~30% and breakage of anodic reaction. Fig. 4B and C show
Nyquist plot with insertion of equivalent circuit. According to experi-
Wh/g of mg/J arb.uni. mg/dm3 mental data we investigated an equivalent circuit to which we added
salts formation of SEI layer and charge transfer phenomena with resistance
CV-RCV- A 0.105 2.64 16.2 1089 10.8 4.7 at 18 Ω and 282 Ω, respectively, assuming presence of metal substrate
ZVC D1 0.006 45.9⁎ > 100 (282)⁎ 910 8.8 1.2 in direct contact with electrolyte. The metal could be dissolved into
D2 – – – 179 2 3.5 electrolyte according to: Red ↔ Ox + ne−. In the forward reaction,
CV-ZVC- A 0.105 2.64 16.2 1095 10.9 3.4 electrons entered the reduced and oxidised forms of metal were diffused
RCV D1 – – – 931 9.3 2.8
into the electrolyte. In consequence charge is being transferred. The
D2 0.078 3.53 21.6 164 1.6 0.6
charge transfer reaction had a certain rate that was related to the kind
of reaction, temperature, concentration of the reaction products and

Fig. 4. Electrochemical experiments of LMTO/

60
A -100
B Experimental data PVC-EDA/YP-50F configuration: (A) CV curve at
Fitting results 20 mV/s rate, (B) Nyquist plot with fitting results
and related (C) equivalent circuit, consist of Rcc-
resistance of cables, RSEI – resistance of SEI layer,
40 -80 CPESEI and CPECT – constant phase elements for
SEI layer and for charge transfer, RCT- resistance
of charge transfer, Ws – short Warburg element.

-60
Z'' (Ω)

20
C (F/g)

-40
0
-20

-20 1st cycle

0
2nd cycle 0 40 80 120 160 200
rd
3 cycle C Z' ( )
-40
-2 -1 0 1 2 3
E (V)

12
A. Siekierka et al. Desalination 436 (2018) 8–14

LiMn2O4 Without Li 3.4. Selective properties

Li2MnO3 With Li
Mn2O3
To investigate selective removal of lithium ions from geothermal
Mn3O4 water two configuration of HCDI process were selected. Fig. 6 presents
MnO MnO2
the rate of lithium removal from reservoir during adsorption and its
recovery into two-stage desorption. Both configurations CV-RCV-ZVC
Intensity (arb.uni.)

and CV-ZVC-RCV are characterized by the same rate of salt removal and
similar energetic requirements. The differences are clearly noted in the
A desorption steps. For D1 under RCV mode about 80% of adsorbed ions
(111)
were released without any selectivity. Different patter was observed for
Ref. code 01-089-0755 the second configuration. The ZVC process allowed to wash out from
H0.6Li0.08Mn1.72O4 the system other ions while lithium salt was released in the RCV process
(311)
(400) mostly. In such way, the HCDI system was able to selectively separate
(440) lithium salt from geothermal water. Such phenomenon is better illu-
B (511) (531) (444)
strated in Fig. 7, where efficiency ratio for each ion is related to the
20 30 40 50 60 70 80 90 system configuration. Washing out the adsorbed ions in the ZVC process
2
allowed to remove 70–80% of them. The HCDI system kept lithium salt
Fig. 5. The XRD pattern for adsorbent with and without intercalate lithium (A) and re- mostly. When polarization was reversed, the RVC case, lithium salt was
ference sample (B). recovered mostly. To make that effect more visible the lines were added
to the graph for eye guidance only.
30 Adsorption It may be concluded that SEI layer was formed on both HCDI con-
CV-RCV-ZVC
Desorption figurations during the adsorption step. When CV-ZVC–RVC configura-
Extraction rate (%)

Desorption 2
20
tion was applied it was possible to obtain brine with domination of
lithium cations. It means, the RVC process allowed lithium to be ex-
tracted from the volume of the LMTO electrode. Additionally, forma-
10
tion of SEI layer enhanced selectivity of the system for monovalent ions.
It was observed in both evaluated configurations that divalent ions
0
Na+ K+ Li+ Ca+2 Mg+2 Sr+2 Cl- Br- HCO3- B Si
were freed during first desorption step.
30
CV-ZVC-RCV
4. Conclusion
Extraction rate (%)

20
In summary it can be said that selection of operation mode for
HCDI, with adsorption and ZVC-RVC desorption processes, allowed to
10 produce enriched stream of lithium ions. Hence, multicomponent geo-
thermal water could be a profitable and attractive sources for lithium
0 salts.
Na+ K+ Li+ Ca+2 Mg+2 Sr+2 Cl- Br- HCO3- B Si

Fig. 6. Effective rate of extraction during A, D1 and D2 for evaluated configurations. Author contributions

Efficiency of D1 Efficiency of D2 The manuscript was written through contributions of all authors. All
100
CV-RCV-ZVC
Vizualization of curves fluctuations authors have given approval to the final version of the manuscript.
80
Funding sources
60
ηion (%)

40
The studies were supported by the Statutory Funds for the Faculty of
20 Chemistry, Wroclaw University of Science and Technology.
0
Na+ K+ Li+ Ca+2 Mg+2 Sr+2 Cl- Br- HCO3- B Si References
100
CV-ZVC-RCV
73%
80 [1] H. Vikström, S. Davidsson, M. Höök, Lithium availability and future production
outlooks, Appl. Energy 110 (2013) 252–266, http://dx.doi.org/10.1016/j.
60
ηion (%)

40
20
0 [3] S. Nishihama, K. Onishi, K. Yoshizuka, Selective recovery process of lithium from
Na+ K+ Li+ Ca+2 Mg+2
Fig. 7. Efficiency of ions elution ratio in both configurations.
potential. Skale et al. [40] reported that in accordance to the Fick first
law diffusion was the phenomenon causing ions transport in field of
chemical potentials. In the simulate experiments it could be deduced
that SEI layer was created on the electrode surface, limited passivation
processes and broke the redox reaction. The peaks at 2.0 V during the
first and the second cycles confirmed exchange phenomenon for LMTO
but did not explain delithiation from the electrode volume.
apenergy.2013.04.005.
[2] P. Meshram, B.D. Pandey, T.R. Mankhand, Extraction of lithium from primary and
secondary sources by pre-treatment, leaching and separation: a comprehensive re-
view, Hydrometallurgy 150 (2014) 192–208, http://dx.doi.org/10.1016/j.
hydromet.2014.10.012.
Sr+2 Cl- Br- HCO3- B Si
seawater using integrated ion exchange methods, Solvent Ext. Ion Exch. 29 (2011)
421–431, http://dx.doi.org/10.1080/07366299.2011.573435.
[4] S. Porada, R. Zhao, A. Van der Wal, V. Presser, P.M. Biesheuvel, Review on the
science and technology of water desalination by capacitive deionization, Prog.
Mater. Sci. 58 (8) (2013) 1388–1442, http://dx.doi.org/10.1016/j.pmatsci.2013.
03.005.
[5] M. Pasta, C.D. Wessells, Y. Cui, F. La Mantia, A desalination battery, Nano Lett. 12
(2012) 839–843, http://dx.doi.org/10.1021/nl203889e.
[6] L. Wang, Y.P. Zhang, J.C. Li, X.Y. Lin, X.G. Luo, Removal of Sr2+ ions from simu-
lated wastewater by electrodeionization, Desal. Wat. Treat. 53 (8) (2015)
2125–2133, http://dx.doi.org/10.1080/19443994.2013.861773.
[7] S. Gahlot, P.P. Sharma, P.K. Jha, V. Kulshrestha, Effective desalination of brackish
water by electrodialysis using SPANI composite cation exchange membranes, Chem.
Select 2 (28) (2017) 8886–8892, http://dx.doi.org/10.1002/slct.201701200.

13
A. Siekierka et al.
[8] H. Shemer, R. Semiat, Sustainable RO desalination – energy demand and environ-
mental impact, Desalination 424 (2017) 10–16, http://dx.doi.org/10.1016/j.desal.
2017.09.021.
[9] T.-H. Yu, H.-Y. Shiu, M. Lee, P.-T. Chiueh, C.-H. Hou, Life cycle assessment of en-
vironmental impacts and energy demand for capacitive deionization, Desalination
399 (2016) 53–60, http://dx.doi.org/10.1016/j.desal.2016.08.007.
[10] A. Siekierka, M. Bryjak, J. Wolska, The use of activated carbon modified with
polypyrrole as a supporting electrode for lithium ions adsorption in capacitive
deionization, Desal. Wat. Treat. 64 (2017) 251–254, http://dx.doi.org/10.5004/
dwt.2017.11387.
[11] M. Bryjak, A. Siekierka, J. Kujawski, K. Smolińska-Kempisty, W. Kujawski,
Capacitive deionization for selective extraction of lithium from aqueous solutions,
J. Membr. Sep. Technol. 4 (2015) 110–115, http://dx.doi.org/10.6000/1929-6037.
2015.04.03.2.
[12] A. Siekierka, J. Wolska, M. Bryjak, Kujawski, Anion exchange membranes in lithium
extraction by means of capacitive deionization system, Desal. Wat. Treat. 75 (2017)
331–341, http://dx.doi.org/10.5004/dwt.2017.20431.
[13] A. Siekierka, J. Kujawa, W. Kujawski, M. Bryjak, Lithium dedicated adsorbent for
the preparation of electrodes useful in the ion pumping method, Sep. Purif. Technol.
194 (2018) 231–238, http://dx.doi.org/10.1016/j.seppur.2017.11.045.
[14] Web page http://www.kuraraychemical.com/products/sc/SCcarbon.htm.
[15] A. Siekierka, J. Wolska, W. Kujawski, M. Bryjak, Modification of poly(vinyl
chloride) films by aliphatic amines to prepare anion-exchange membranes for Cr
(VI) removal, Sep. Sci. Technol. (2017) 1–7, http://dx.doi.org/10.1080/01496395.
2017.1358746.
[16] E. Kmiecik, B. Tomaszewska, K. Wątor, M. Bodzek, Selected problems with boron
determination in water treatment processes. Part I: comparison of the reference
methods for ICP-MS and ICP-OES determinations, Environ. Sci. Pollut. Res. Int. 23
(12) (2016) 11658–11667, http://dx.doi.org/10.1007/s11356-016-6328-7.
[17] B. Tomaszewska, M. Rajca, E. Kmiecik, M. Bodzek, W. Bujakowski, K. Wątor,
M. Tyszer, The influence of selected factors on the effectiveness of pre-treatment of
geothermal water during the nanofiltration process, Desalination 406 (2017)
74–82, http://dx.doi.org/10.1016/j.desal.2016.07.007.
[18] K. Wątor, E. Kmiecik, B. Tomaszewska, Assessing medical qualities of groundwater
from the Busko-Zdrój area (Poland) using the probabilistic method, Environ. Earth
Sci. 75 (9) (2016) 1–13, http://dx.doi.org/10.1007/s12665-016-5538-0.
[19] M.E. Suss, S. Porada, X. Sun, P.M. Biesheuvel, J. Yoon, V. Presser, Water desali-
nation via capacitive deionization: what is it and what can we expect from it?
Energy Environ. Sci. 8 (2015) 2296–2319, http://dx.doi.org/10.1039/
C5EE00519A.
[20] A. Hemmatifar, J.W. Palko, M. Stadermann, J.G. Santiago, Energy breakdown in
capacitive deionization, Water Res. 104 (2016) 303–311, http://dx.doi.org/10.
1016/j.watres.2016.08.020.
[21] J.E. Dykstra, K.J. Keesman, P.M. Biesheuvel, A. Van der Wal, Theory of pH changes
in water desalination by capacitive deionization, Water Res. 119 (2017) 178–186,
http://dx.doi.org/10.1016/j.watres.2017.04.039ff.
[22] C. Kim, J. Lee, P. Srimuk, M. Aslan, V. Presser, Concentration-gradient multichannel
flow-stream membrane capacitive deionization cell for high desalination capacity of
carbon electrodes, ChemSusChem 10 (24) (2017) 4914–4920, http://dx.doi.org/10.
1002/cssc.201700967.
[23] Z. Wang, T. Yan, G. Chen, L. Shi, D. Zhang, High salt removal capacity of metal-
organic gel derived porous carbon for capacitive deionization, ACS Sustain. Chem.
Eng. 5 (12) (2017) 11637–11644, http://dx.doi.org/10.1021/acssuschemeng.
7b03015.
[24] K. Tang, J. Chang, H. Cao, C. Su, Y. Li, Z. Zhang, Y. Zhang, Macropore- and mi-
cropore-dominated carbon derived from poly(vinyl alcohol) and poly-
vinylpyrrolidone for supercapacitor and capacitive deionization, ACS Sustain.
Chem. Eng. 5 (12) (2017) 11324–11333, http://dx.doi.org/10.1021/
acssuschemeng.7b02307.
[25] H. Yoon, J. Lee, S. Kim, J. Yoon, Hybrid capacitive deionization with Ag coated
14
Desalination 436 (2018) 8–14
composite electrode, Desalination 422 (2017) 42–48, http://dx.doi.org/10.1016/j.
desal.2017.08.010.
[26] W. Hu, F. Xie, Y. Li, Z. Wu, K. Tian, M. Wang, L. Pan, L. Li, Hierarchically porous
carbon derived from PolyHIPE for supercapacitor and deionization applications,
Langmuir 33 (46) (2017) 13364–13375, http://dx.doi.org/10.1021/acs.langmuir.
7b03175.
[27] M. Moronshing, C. Subramaniam, Scalable approach to highly efficient and rapid
capacitive deionization with CNT-thread as electrodes, ACS Appl. Mater. Interfaces
9 (46) (2017) 39907–39915, http://dx.doi.org/10.1021/acsami.7b11866.
[28] L. Zhang, Y. Liu, T. Lu, L. Pan, Cocoon derived nitrogen enriched activated carbon
fiber networks for capacitive deionization, J. Electroanal. Chem. 804 (2017)
179–184, http://dx.doi.org/10.1016/j.jelechem.2017.09.062.
[29] M. Wang, X. Xu, J. Tang, S. Hou, Md. Shahriar A. Hossaun, L. Pan, Y. Yamauchi,
High performance capacitive deionization electrodes based on ultrathin nitrogen-
doped carbon/graphene nano-sandwiches, Chem. Commun. 53 (2017)
10784–10787, http://dx.doi.org/10.1039/c7cc05673g.
[30] F. Xing, T. Li, H. Zhu, N. Wang, X. Cao, Chemically exfoliated MoS2 for capacitive
deionization of saline water, Nano Energy 31 (2017) 590–595, http://dx.doi.org/
10.1016/j.nanoen.2016.12.012.
[31] L. Li, L. Liao, Q. Meng, B. Cao, High performance graphene composite microsphere
electrodes for capacitive deionisation, Carbon 90 (2015) 75–84, http://dx.doi.org/
10.1016/j.carbon.2015.04.009.
[32] J. Lee, S. Kim, C. Kim, J. Yoon, Hybrid capacitive deionization to enhance the de-
salination performance of capacitive techniques, Energy Environ. Sci. 7 (2014)
3683–3689, http://dx.doi.org/10.1039/c4ee02378a.
[33] C. Wan, S. Xu, M. Hu, R. Cao, J. Qian, Z. Qin, J. Liu, K.T. Mueller, J.-G. Zhang,
J. Hu, Multinuclear NMR study of the solid electrolyte interface formed in lithium
metal batteries, ACS Appl. Mater. Interfaces 9 (17) (2017) 14741–14748, http://dx.
doi.org/10.1021/acsami.6b15383.
[34] V.V. Nikonenko, V.I. Vasil'eva, E.M. Akberova, A.M. Uzdenova, M.K. Urtenov,
A.V. Kovaleno, N.P. Pismenskaya, S.A. Mareev, G. Pourcelly, Competition between
diffusion and electroconvection at an ion-selective surface in intensive current re-
gimes, Adv. Colloid Interf. Sci. 235 (2016) 233–246, http://dx.doi.org/10.1016/j.
cis.2016.06.014.
[35] J. Chen, X. Wu, Y. Gong, P. Wang, W. Li, Q. Tan, Y. Chen, Synthesis of Mn3O4/N-
doped graphene hybrid and its improved electrochemical performances for lithium-
ion batteries, Ceram. Int. 43 (5) (2017) 4655–4662, http://dx.doi.org/10.1016/j.
ceramint.2016.12.138.
[36] G.C. Silva, F.S. Almeida, A.M. Ferreira, V.S.T. Ciminelli, Preparation and applica-
tion of a magnetic composite (Mn3O4/Fe3O4) for removal of As (III) from aqueous
solutions, Mater. Res. 15 (3) (2012) 403–408, http://dx.doi.org/10.1590/S1516-
14392012005000041.
[37] G.W. Luther III, Manganese (II) oxidation and Mn (IV) reduction in the environ-
mental – to one electron transfer steps versus a single two-electron step,
Geomicrobiol J. 22 (3–4) (2005) 195–203, http://dx.doi.org/10.1080/
01490450590946022.
[38] S.-W. Lee, C.W. Lee, S.-B. Yoon, M.-S. Kim, J.H. Jeong, K.-W. Nam, K.C. Roh, K.-
B. Kim, Superior electrochemical properties of manganese dioxide/reduce graphene
oxide nanocomposites as anode materials for high-performance lithium ion-bat-
teries, J. Power Sources 312 (2016) 207–215, http://dx.doi.org/10.1016/j.
jpowsour.2016.02.049.
[39] G. Jian, Y. Xu, L.-C. Lai, C. Wang, M.R. Zachariah, Mn3O4 hollow spheres for li-
thium-ion batteries with high rate and capacity, J. Mater. Chem. A 2 (2014)
4627–4632, http://dx.doi.org/10.1039/C4TA00207E.
[40] S. Skale, V. Dolecek, M. Slemnik, Substitution of the constant phase element by
Warburg impedance for protective coatings, Corros. Sci. 49 (3) (2007) 1045–1055,
http://dx.doi.org/10.1016/j.corsci.2006.06.027.
[41] J.D. Hem, Chemical equilibria and rates of manganese oxidation, Chemistry of
Manganese in Natural Water, Geological Survey Water-Supply Paper 1667-A,
United States Government printing office, Washington, 1963.