Accepted Manuscript

A novel process on separation of manganese from calcium and

magnesium using synergistic solvent extraction system

Yingsong Wang, Li Zeng, Guiqing Zhang, Wenjuan Guan, Zhen

Sun, Dong Zhang, Jialin Qing

PII: S0304-386X(18)30708-4
DOI: https://doi.org/10.1016/j.hydromet.2019.01.008
Reference: HYDROM 4982
To appear in: Hydrometallurgy
Received date: 20 September 2018
Revised date: 3 January 2019
Accepted date: 17 January 2019

Please cite this article as: Y. Wang, L. Zeng, G. Zhang, et al., A novel process on separation
of manganese from calcium and magnesium using synergistic solvent extraction system,
Hydrometallurgy, https://doi.org/10.1016/j.hydromet.2019.01.008

This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.
 ACCEPTED MANUSCRIPT

1
A novel process on separation of manganese from calcium and
magnesium using synergistic solvent extraction system
Yingsong Wang, Li Zeng*, Guiqing, Zhang, Wenjuan Guan, Zhen Sun, Dong Zhang, Jialin Qing

School of Metallurgy and Environment, Central South University, Changsha 410083, China;

Laboratory of Metallurgical Separation Science and Engineering, Central South University, Changsha 410083, China

PT
RI
Abstract: The synergistic solvent extraction (SSX) system consisting of di-2-ethylhexyl

SC
phosphoric acid (P204) and alkyl-4-pyridinecarboxylate ester (4PC) in sulfonated kerosene was

used to separate manganese from magnesium and calcium and to prepare pure manganese sulfate
NU
solutions with battery grade. In batch tests, over 99.9% of Mn2+ was extracted after a four-stage
MA

counter current extraction using an organic solution consisting of 0.25 mol/L P204 and 0.5 mol/L

4PC in sulfonated kerosene and a feed containing 22 g/L Mn2+, 4 g/L Mg2+ and 0.1g/L Ca2+ with
D

an A/O ratio of 1:6 at 25℃ for 10 min, while extractions of magnesium and calcium were
E

negligible. After scrubbing using 0.025 mol/L H2SO4 at an A/O ratio of 1:8, the loaded organic
PT

solution was stripped using 0.7 mol/L H2SO4 at an A/O ratio of 1:8 to prepare pure manganese
CE

sulfate solutions with battery grade. The results of continuous operation test indicated that the
AC

mass ratios of Ca/Mn and Mg/Mn in strip liquors were below 5.6×10-5 and 2.5×10-4, respectively,

further demonstrating the efficient separation of manganese from magnesium and calcium using

the SSX system.

Keywords: Synergistic solvent extraction; P204; 4PC; Manganese; Magnesium; Calcium

*Corresponding author at: School of Metallurgy and Environment, Central South University, Changsha
410083, China.
Email address: [email protected](L.Zeng).

1
 ACCEPTED MANUSCRIPT

1. Introduction

Manganese-based cathode materials have been recognized as one of the most promising cathode

materials for lithium ion batteries due to their abundant resources, good safety and high energy

density. High quality manganese-based cathode materials can only be obtained when high purity

PT
manganese-based materials guaranteed. No matter manganese sulfate for production of ternary

RI
materials or manganese oxides for production of lithium manganate, the preparation of pure

SC
manganese sulphate is priority since it is the most important and basic manganese source material

for the cathode of manganese-based lithium ion battery (Zhang et al., 2014). The contents of
NU
impurities especially magnesium and calcium in battery-grade manganese sulfate have to be
MA

strictly limited (Tan et al., 2014). Therefore, the complete removal of calcium and magnesium is

the key issue in the process of preparing high purity manganese compounds. During the past
D

decade, much research work was conducted to separate manganese from magnesium and calcium
E

in the process of manganese sulfate purification. A number of processes and technologies for the
PT

separation of manganese from calcium and magnesium have been developed and reported.
CE

Basically, there are four methods including crystallization (Helen et al., 2007; Yuan et al., 2000;
AC

Li et al., 2010), electrolysis (Chen et al., 2008), chemical precipitation (Yan et al., 2014; Bao et al.,

2013; Mei et al., 2014) and solvent extraction (Liu et al., 2008; Yang et al. 2014)to separate

manganese from calcium and magnesium.

Liang et al. (2014) studied two-step leaching of manganese from pyrolusite using sulfuric acid,

resulting in manganese sulfate solution containing low concentrations of magnesium and calcium.

After crystallization at high temperature, pure manganese sulfate solution with magnesium and

2
 ACCEPTED MANUSCRIPT

calcium concentrations both less than 100 μg/g was finally obtained. But this method is not

suitable for purification of manganese sulfate solutions containing high concentrations of

magnesium and calcium since multiple crystallizations are needed and therefore, resulting in sharp

decrease of manganese recovery (Chen et al., 2016).

Zhu et al. (2012 and 2013) applied the method of electrolysis to prepare battery grade

PT
manganese sulfate solution from industrial and feed grade manganese sulfate. Both concentrations

RI
of magnesium and calcium in final products were only below 3 μg/g. However, this method has

SC
some drawbacks of long flow, low efficiency and large energy consumption. In addition, high

concentrations of calcium, magnesium and heavy metal impurities in solutions would result in the
NU
decrease of potential of hydrogen evolution and therefore lower the efficiency of electrolysis and
MA

decrease the purity of final product (Xing et al., 2015).

The method of chemical precipitation of magnesium and calcium as fluoride precipitates is the
D

most common one that has been widely used in industry recently due to easy operation. One
E

problem of this method is difficult filtration of colloidal precipitates of MgF2 and CaF2. Another
PT

problem is introduction of new impurity of 𝐹 − . Due to large electronegativity, the ion of 𝐹 − is

CE

easy to be adsorbed in the product or even into lattice of the product, resulting in large excess of
AC

fluorine content in final products. In addition, equipment would be easily corroded if the content

of fluorine reached 0.1% (Zhang et al., 2014).

The method of solvent extraction is regarded as a promising way on recovery of manganese

from acid leach solutions since it exhibits advantages of short flow, high efficiency and low

energy consumption (Shen et al., 2008). But it is far away from industrial application because

there is no suitable extractant with high selectivity of manganese over magnesium and calcium.

3
 ACCEPTED MANUSCRIPT

The separation of manganese from alkali metals mainly involves acidic phosphorus extractants

such as 2-ethylhexyl phosphonic acid mono-2-ethylhexyl ester (P507), P204 and Cyanex272.

However, P204 and P507 cannot be used to selectively extract manganese from magnesium and

calcium (Dreisinger et al., 1984; Cheng et al., 2000; Zhang et al., 2007). Although their

separations can be achieved using Cyanex272, the cost of extractant is too expensive and therefore

PT
uneconomical (Jouni et al., 2011). A possible and promising way to selectively extract target metal

RI
from impurity at low pH is to use synergistic effect of mixed solvent extraction (SSX) system

SC
consisting of two or more commercial extractants. For example, Preston et al. (2010) studied a

SSX system consisting of pyridine carboxylates and carboxylic acid extractants to selectively
NU
extract nickel from acidic solutions, realizing effective separation of nickel from aluminum,
MA

manganese, magnesium and calcium. The SSX systems of Versatic10/LIX63,

Versatic10/LIX63/TBP and Versatic10/4PC proposed by Cheng et al. (2006) exhibited excellent

D

performance on separation of nickel and cobalt from manganese, magnesium and calcium, and
E

have been successfully applied in industrial production. But there is no report on SSX systems for
PT

selective extraction of manganese from magnesium and calcium so far. In this study, a SSX
CE

system consisting of P204 and 4PC was proposed to selectively separate manganese from
AC

magnesium and calcium.

2. Experimental

2.1. Chemicals and reagents

The extractant of 4PC with a purity of 94% was provided by Hunan Hongbang Material

Technology Company. The extractant of P204 with a purity of 95% is purchased from Shanghai

4
 ACCEPTED MANUSCRIPT

LaiYaShi Chemical Company. The commercial sulfonated kerosene was used as diluent. The feed

solutions were synthesized by dissolving analytical grade MnSO4·H2O, MgSO4·7H2O and

CaSO4·2H2O in distilled water.

2.2 Shakeout test procedures

PT
Solvent extraction and stripping experiments were carried out by mechanically shaking the

RI
organic and aqueous phases in separating funnels at room temperature for a certain time. The

SC
acidity of aqueous solution was adjusted using 18 mol/L H2SO4 or 13 mol/L NH3·H2O and

determined by Hanna HI9125 digital pH meter. To determine the metal extraction distribution
NU
isotherms, the organic system containing 0.25 mol/L P204 and 0.5 mol/L 4PC in sulfonated
MA

kerosene was mixed with the aqueous solution using saturated extraction method by repeating

extraction at an A/O ratio of 1/1 and 25 °C for the same organic solution until saturation. To
D

determine the metal stripping distribution isotherms, stripping tests were conducted using the
E

loaded organic solution and a strip solution of 0.7 mol/L H2SO4 at A/O ratios of 1:1, 1:3, 1:5, 1:10,
PT

1:15 and 1:20.

CE
AC

2.3. Data process

The metal concentrations in aqueous phase were determined using Inductively Coupled Plasma

Atomic Emission Spectrometry (ICP-AES). The metal concentrations in loaded organic phase

were calculated by mass balance. Metal extraction (E), distribution ratio (D) and separation

coefficient (β) were calculated according to Eqs. (1)-(3), respectively.

CF VF  CRVR
E  100% (1)
CF VF

5
 ACCEPTED MANUSCRIPT

Corg
D (2)
CR

Mn/ Me  DMn / DMe (3)

where CF, CR and Corg stand for the metal concentration in feed, raffinate and loaded organic

phase, respectively.

PT
3. Results and discussion

RI
3.1. Effect of organic composition

SC
The pH isotherms of metal extraction with 0.25 mol/L P204 alone and a SSX system consisting

of 0.25 mol/L P204 and 0.5 mol/L 4PC are shown in Fig. 1.
NU
It can be seen from Fig. 1 that the metal extraction sequence was found to be Ca＞Mn＞Mg
MA

with the P204 system alone. The pH50 difference between manganese and calcium was only 0.39

( ΔpH50(Mn-Ca) =pH50(Mn) -pH50(Ca) =0.39 ), indicating that the P204 system alone cannot be used to
D

effectively separate manganese from calcium. However, after the addition of 4PC into the P204
E

system, the extraction curve for manganese was shifted left to lower pH values
PT

( ΔpH50(Mn) =0.83 )while that for calcium was shifted right to higher pH values ( ΔpH50(Ca) =-0.3 ). The
CE

curve for magnesium maintained almost unchanged. The metal extraction sequence was changed
AC

to Mn＞Ca＞Mg. The selectivity of manganese and calcium was reversed and the values of

ΔpH50(Ca-Mn) and ΔpH50(Mg-Mn) increases from -0.39 and 0.75 pH units to 0.74 and 1.62 pH units,

correspondingly, indicating that manganese can be selectively extracted over magnesium and

calcium with the synergistic solvent extraction system of P204 and 4PC. The synergistic effect on

manganese could be attributed to the formation of transition metal (Mn) organic complexes with

more stable structure due to exchange of H+ by P204 and coordination by 4PC. And such

6
 ACCEPTED MANUSCRIPT

antagonistic effect on non-transition metals of magnesium and calcium resulted from the

interaction between the acidic (P204) and basic components (4PC) of the extractant mixture which

competes with magnesium and calcium for available P204 required for the formation of the

extractable metal complex (Preston et al., 2000).

100

PT
80

RI
Extraction %

60

SC
1.71 2.15 2.45 3.29 Mn(P204+4PC)
2.54 3.33
Mg(P204+4PC)
40 Ca(P204+4PC)
Mn(P204)
NU
Mg(P204)
20
Ca(P204)
MA

0
1 2 3 4 5
Equilibrium pH
D

Fig. 1 pH isotherms of metal extraction with 0.25 mol/L P204 alone and a SSX system consisting of 0.25 mol/L
E

P204 and 0.5 mol/L 4PC at 25°C (Feed: 0.02 mol/L Mn2+, 0.02 mol/L Mg2+, 0.01mol/L Ca2+, 0.25 mol/L Na2SO4)
PT
CE

3.2. Effect of extractant P204 concentration

AC

The effect of P204 concentration ranged from 0 to 0.4 mol/L at a fixed concentration of 1 mol/L

4PC on metal extraction is shown in Fig. 2.

It can be seen from Fig. 2 that the extraction of manganese increased significantly from 12% to

65% with the increase in P204 concentration from 0.1 mol/L to 0.4 mol/L, while the extractions of

Mg and Ca maintained almost stable at less than 5%. It was of interest that the separation

coefficient of manganese over magnesium and calcium increased with the increase in P204

7
 ACCEPTED MANUSCRIPT

concentration from 0 to 0.25 mol/L and then gradually decreased with further increase in P204

concentration, which could be attributed to the more increases in extraction of calcium and

magnesium than that of manganese with the increase in P204 concentration higher than 0.25

mol/L. Therefore, the separation coefficient of Mn/Mg and Mn/Ca reached maximum values of 50

and 28 at a P204 concentration of 0.25 mol/L, respectively. The P204 concentration of 0.25 mol/L

PT
was thus selected as optimum for subsequent experiment.

RI
100 60
Mn
Mg

SC
80 Ca

Separation coefficient β
βMn/Mg
40
Extraction /%

βMn/Ca
NU
60

40
MA

20

20
D

0 0
0.0 0.1 0.2 0.3 0.4
E

P204 concentration/(mol·L-1)
PT

Fig. 2 Effect of P204 concentration on metal extraction from sulfuric acid solution by the mixture of P204/4PC in
CE

sulfonated kerosene at 25 ℃. (Feed: 10.4 g/L Mn2+, 4.62 g/L Mg2+, 0.424 g/L Ca2+; A/O=1:1; t=15 min; Shaking

frequency=200 rpm)
AC

3.3. Effect of 4PC concentration

The effect of 4PC concentration ranged from 0 to 1 mol/L at a fixed concentration of 0.25

mol/L P204 on metal extraction is shown in Fig. 3.

8
 ACCEPTED MANUSCRIPT

50
50

40

Separation coefficient β
40

Extraction /%
30
30
Mn
Mg
20 20
Ca
βMn/Mg

PT
10 βMn/Ca 10

RI
0 0
0.00 0.25 0.50 0.75 1.00

4PC concentration/(mol·L-1)

SC
Fig. 3 Effect of 4PC concentration on metal extraction from sulfuric acid solution by the mixture of P204/4PC in
NU
sulfonated kerosene at 25℃. (Feed:10.9 g/L Mn2+, 4.85 g/L, Mg2+, 0.35 g/L Ca2+; A/O=1:1; t=15 min; Shaking

frequency=200 rpm)
MA

It can be seen from Fig. 3 that the extraction of manganese gradually increased with the
ED

increase in 4PC concentration, while the extraction of magnesium was almost negligible in the
PT

whole tested range of 4PC concentration. However, the extraction of calcium obviously decreased
CE

from 33% to 2.3% with the increase in 4PC concentration from 0 to 0.25 mol/L. Further increase

in 4PC concentration had little effect on metal extractions when the 4PC concentration reached 0.5
AC

mol/L. Therefore, the 4PC concentration of 0.5mol/L was selected as optimum for subsequent

experiment.

3.4. Effect of equilibrium pH

The equilibrium pH was controlled by saponification of extractant using ammonia. The reason

on selection of ammonia as saponifier is to avoid introduction of any other impurities. The effect

9
 ACCEPTED MANUSCRIPT

of equilibrium pH on the extraction of manganese, magnesium and calcium is shown in Fig. 4.

70 50
Mn
Mg
60
Ca
40
βMn/Mg

Separation coefficient β
50
βMn/Ca
Extraction %

PT
30
40

30

RI
20

20

SC
10
10 NU
0 0
0 2 4 6
Equilibrium pH
MA

Fig. 4 Effect of equilibrium pH on metal extraction from sulfuric acid solutions with 0.25 mol/L P204 and 0.5

mol/L 4PC in sulfonated kerosene at 25℃. (Feed: 11.5 g/L Mn2+, 5.02 g/L Mg2+, 0.42 g/L Ca2+; A/O=1:1; t=15
D

min; Shaking frequency=200 rpm)

E
PT

It can be seen from Fig. 4 that the extraction of manganese increased sharply with the increase
CE

in equilibrium pH, while the extractions of magnesium and calcium were negligible at pH<3.
AC

Further increase in equilibrium pH resulted in slight increases of both magnesium and calcium to

less than 6%. It was of interest that the separation coefficient of manganese over calcium reached

maximum value of 32 at pH 3 and then decreased to 18 with the pH increase to 4. Further increase

in pH resulted in steady recovery of separation coefficient of Mn/Ca. This can be attributed to the

more increases in extraction of calcium and magnesium than that of manganese with the pH

increase from 3 to 4. The SSX system exhibits much higher selectivity of manganese over

10
 ACCEPTED MANUSCRIPT

magnesium than that over calcium during the whole tested pH range (∆𝑝𝐻50(𝑀𝑔−𝑀𝑛) = 1.62 ＞

∆𝑝𝐻50(𝐶𝑎−𝑀𝑛) = 0.74). Therefore, the equilibrium pH was controlled around 3 in subsequent

experiments.

3.5. Effect of equilibrium time and temperature

PT
Fig. 5 (a) shows the effect of equilibrium time on metal extractions. It was found that the

RI
extraction of metals reached equilibrium in 5 min, indicating fast kinetics of extraction. It was also

SC
noted that the extraction of manganese was slightly lower at 1 min than that at equilibrium. This

can be explained that part of co-extracted calcium was crowed out by manganese after 1 min. For
NU
subsequent experiment, the extraction time of 10 min was selected to ensure complete equilibrium.
MA

Fig. 5 (b) shows the effect of temperature on extractions of three metals. As seen, the effect of

temperature on metal extractions was negligible. The temperature of 25℃was thus selected as
D

optimum for subsequent experiments.

E

40
PT

35

30
CE

25
Exration %

Mn
Mg
AC

20
Ca
15

10

0
0 5 10 15 20
Time(min)

(a)

11
 ACCEPTED MANUSCRIPT

40

35

30

25

Exration %
Mn
20 Mg
Ca
15

PT
10

RI
0
15 20 25 30 35 40 45
Temperature ℃

SC
(b)
NU
Fig. 5 Effect of time and temperature on metal extraction from sulfuric acid solutions with 0.25 mol/L P204 and

0.5 mol/L 4PC in sulfonated kerosene. (Feed: 11.5 g/L Mn2+, 5.02 g/L Mg2+, 0.42 g/L Ca2+; A/O=1:1; Shaking
MA

frequency=200 rpm)
E D

3.6. Manganese extraction distribution isotherm

PT

The manganese extraction distribution isotherm and its McCabe-Thiele diagram with the SSX
CE

system consisting of 0.25mol/L P204 and 0.5mol/L 4PC in sulfonated kerosene and the synthetic

solution containing 22 g/L Mn2+, 4 g/L Mg2+ and 0.1 g/L Ca2+ are shown in Fig.6.
AC

12
 ACCEPTED MANUSCRIPT

Mn concentration in organic(g/L)
4

3
1

2 A/O=1/6

PT
1
2

RI
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Mn concentration in raffinate (g/L)

SC
Fig. 6 Manganese extraction distribution isotherm and its McCabe-Thiele diagram with 0.25 mol/L P204 and 0.5
NU
mol/L 4PC in sulfonated kerosene at 25℃. (Feed: 22g/L Mn2+, 4g/L Mg2+, 0.1g/L Ca2+; A/O=1:1; Equilibrium

pH=3; t=10 min; Shaking frequency=200 rpm)

MA

As can be seen, the saturated extraction capacity of manganese reached 4 g/L with the SSX
E D

system at equilibrium pH of 3. It was also found that three theoretical extraction stages are needed
PT

to extract almost all the manganese at an A/O ratio of 1/6. To verify this, a four-stage counter
CE

current extraction test was conducted at an A/O phase ratio of 1/6 and ambient temperature. The

phase separation was good and finished in 10 mins. The results indicated that more than 99.9% of
AC

manganese was extracted, while the extractions of magnesium and calcium were only 0.3% and

10.8%, respectively, further demonstrating that the SSX system can be used to selectively extract

manganese from magnesium and calcium.

3.7. Scrubbing

The dilute sulfuric acid was used to scrub loaded organic phase to minimize the co-extraction of

13
 ACCEPTED MANUSCRIPT

calcium and magnesium. The effect of H2SO4 concentration in the range of 0-0.025 mol/L on

scrubbing of metals with a loaded organic solution containing 4.13 g/L Mn2+, 0.0018 g/L Mg2+ and

0.0018 g/L Ca2+ at an A/O ratio of 1/8 is shown in Fig. 7.

It can be seen from Fig.7 that the scrubbing of magnesium reached 70% when de-ionized water

was used as scrubbing agent, while the scrubbing of calcium was only less than 7%, indicating

PT
that most of magnesium was entrained into the organic phase during extraction. The scrubbing of

RI
calcium increased from 6.8% to 48% with the increase in H2SO4 concentration from 0 to 0.025

SC
mol/L. The scrubbing of magnesium maintained around 75% in the whole tested range of H2SO4

concentration. But the scrubbing of manganese was only 4.4% even when sulfuric acid
NU
concentration increased to 0.025 mol/L. Further increase in sulfuric acid concentration would
MA

result in increase of scrubbing of manganese. Therefore, the sulfuric acid concentration of 0.025

mol/L was selected as optimum.

D

80
E

70
Mn
PT

Mg
60
Scrubbing rate %

Ca
50
CE

40

30
AC

20

10

0
0.000 0.005 0.010 0.015 0.020 0.025
H2SO4 concentration (mol/L)

Fig. 7 Effect of H2SO4 concentration on metal scrubbing at 25℃ (A/O=1:8; t=10 min; Shaking frequency=200

rpm).

14
 ACCEPTED MANUSCRIPT

3.8. Manganese stripping distribution isotherm

It was observed that phase separation of stripping was difficult when the sulfuric acid

concentration was higher than 1 mol/L, indicating that phase disengagement of stripping was

acidity dependent. In addition, the effect of H2SO4 concentration on manganese stripping indicated

that the stripping of manganese maintained almost stable with further increase in H2SO4

PT
concentration from 0.7 to 1 mol/L. Thus, the stripping tests were conducted using the scrubbed

RI
organic solution and 0.7 mol/L H2SO4 at A/O ratios of 1:1, 1:3, 1:5, 1:10, 1:15 and 1:20,

SC
respectively. The stripping distribution isotherm of manganese and its McCabe–Thiele diagram

are shown in Fig. 8. It was found that four theoretical stripping stages are needed to strip almost
NU
all the manganese at an A/O ratio of 1/8.
MA

35
Mn concentration in strip liquor (g/L)

30
D

25 1
E

20

A/O=1/8
PT

15

10 2
CE

5
3
0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
AC

Mn concentration in organic(g/L)

Fig. 8 Manganese stripping isotherm using 0.7 mol/L H2SO4 at 25℃ and its McCabe-Thiele diagram (t=10 min;

Shaking frequency=200 rpm)

To verify this, a four-stage counter current stripping test was conducted with a loaded organic

solution containing 3.92 g/L Mn2+, 0.00045g/L Mg2+ and 0.00093g/L Ca2+ at an A/O ratio of 1/8

and 25 ℃. The results are shown in Table 1. As can be seen, the stripping of manganese reached

15
 ACCEPTED MANUSCRIPT

as high as 99.4%. The concentration of manganese in strip liquors reached 38 g/L, which was 6

g/L higher than that of theoretical value of 32 g/L. This can be attributed to 16% volume loss of

strip liquors.

Table 1 Results of four stages counter current stripping with 0.7 mol/L H2SO4 at 25℃
. (A/O=1/8; t=10 min and

shaking frequency=200 rpm)

PT
Metal concentrations in strip liquor (g/L) Metal stripping /%
No.

RI
Mn Mg Ca Mn Mg Ca

SC
R8 37.5 0.0013 0.0090 99.1 97.3 97.2

R9 39.8 0.00099 0.0084 100 98.0 96.9

NU
R10 37.3 0.00097 0.0092 99.6 97.8 96.8
MA

R11 38.4 0.0012 0.0097 99.4 97.5 97.1

D

4. Continuous operation test

E
PT

Based on optimum conditions of batch tests, a principle process flowsheet to selectively extract

manganese over calcium and magnesium from sulfuric acid leach solutions using SSX technology
CE

is proposed in Fig. 9 and a continuous operation test conducted to recover manganese from an
AC

authentic acid leach solution containing 42.2 g/L Mn2+, 7.1 g/L Mg2+ and 0.08 g/L Ca2+ with the

SSX system. The SSX process consists of five stages extraction, two stages scrubbing and five

stages of stripping. Since the concentration of manganese in feeds was two times as that in

previous counter current extraction experiment, the SSX system consisting of 0.5 mol/L P204 and

1 mol/L 4PC was used in the test for 72 h. The process was very stable and easy to control. The

typical results are shown in Table 2.

16
 ACCEPTED MANUSCRIPT

Acid leach solution containing Mn,Ca,Mg

Synergistic solvent extraction

Raffinate
(P204+4PC, pH 3)

Saponification Scrubbing
[25(v/v) ammonia] (0.025 M H2SO4)

Stripping
(1~2 M H2SO4)

PT
Pure MnSO 4 solution

RI
Fig. 9 Principle process flowsheet for the selective extraction of Mn over Mg and Ca using SSX

SC
Note: the flow of aqueous solution
the flow of organic solution
NU
system
MA
E D

Table 2 Typical result of continuous running test on extraction of Mn with the SSX system
PT

Item Mn Mg Ca
CE

Feed (g/L) 42.20 7.10 0.08

Raffinate (g/L) 0.72 6.71 0.069

AC

Extraction (%) 98.30 1.33 9.38

Scrub liquor (g/L) 2.29 0.18 0.010

Strip liquor A (g/L) 158.62 0.034 0.009

Strip liquor B (g/L) 88.68 0.022 0.005

Removal (%) / 99.9 96.7

Note: A/O ratio: 1/5(extraction), 1/10(scrubbing) and 1/10(stripping), H2SO4 concentration for scrubbing:

17
 ACCEPTED MANUSCRIPT

0.1mol/L

It can be seen from Table 2 that the extraction of manganese reached 98% with the mass ratios

of Ca/Mn and Mg/Mn in strip liquors below 5.6×10-5 and 2.5×10-4, respectively. It was noted that

the metal concentration in strip liquor A was 2 times higher than those of theoretical value, which

PT
can be attributed to 50% volume loss of strip liquors when 2 mol/L H2SO4 was used as stripping

RI
agent during the first 14 h. To solve this problem of incomplete phase disengagement, the way of

SC
stripping agent addition was changed to 1 mol/L H2SO4 in the first stage of stripping and 0.2V mol

concentrated H2SO4 in every other stage after 14 h operation (V stands for the volume of stripping
NU
agent), resulting in complete and fast phase separation in the next 58 h continuous operation.
MA

5. Conclusion
D

(1) The solvent extraction system of P204 and 4PC shows a synergistic effect on manganese
E

and an antagonistic effect on calcium. The values of ΔpH50(Ca-Mn) and ΔpH50(Ca-Mn) increased from
PT

-0.39 and 0.75 with P204 alone to 0.74 and 1.62 with the SSX system, correspondingly, which
CE

could realize the selective extraction of manganese over magnesium and calcium.
AC

(2) More than 99.9% of Mn2+ was extracted with an A/O ratio of 1:6 at equilibrium pH of 3 in a

four-stage counter current extraction using a SSX system consisting of 0.25mol/L P204 and

0.5mol/L 4PC in kerosene and a synthetic feed containing 22 g/L Mn2+, 4 g/L Mg2+ and 0.01 g/L

Ca2+. After scrubbing using 0.025 mol/L H2SO4, the loaded organic phase was stripped using 0.7

mol/L H2SO4 at an A/O ratio of 1:8 in a four-stage counter current stripping. The concentrations of

manganese, magnesium and calcium in strip liquors were 38 g/L, <0.0015 g/L and <0.01g/L,

18
 ACCEPTED MANUSCRIPT

respectively, resulting in mass ratios of Ca/Mn and Mg/Mn below 5.6×10-5 and 2.5×10-4,

respectively, which could meet the requirement of battery grade manganese sulfate.

(3) The results of continuous operation test on recovery of manganese from acid leach solutions

containing 42.2 g/L Mn2+, 7.1 g/L Mg2+ and 0.08 g/L Ca2+ using the SSX system of 0.5 mol/L

P204 and 1 mol/L 4PC in kerosene demonstrated that the SSX system can be used to completely

PT
separate manganese from magnesium and calcium.

RI
Acknowledgment
SC
NU
The authors are thankful to the National Natural Science Foundation of China (NSFC51404297)
MA

for financial support.

E D
PT
CE
AC

19
 ACCEPTED MANUSCRIPT

REFERENCES

Bao, X.J., Wang, Z.J., Liu, J.B., Su, Z.F., Tian, S.Y., Zhou, D.B., Wang, H.E., 2013. Exploratory

Research on Deep Removal of Ca (Ⅱ), Mg (Ⅱ) and Fe (Ⅱ) from Industrial Manganese

Sulfate (in Chinese). Mining and Metallurgical Engineering, 33(03): 90-93+97.

Cheng, C.Y., 2000 Purification of synthetic laterite leach solution by solvent extraction using

PT
D2EHPA. Hydrometallurgy, 56(3): 369-386.

RI
Cheng, C.Y., 2006. Solvent extraction of nickel and cobalt with synergistic systems consisting of

SC
carboxylic acid and aliphatic hydroxyoxime. Hydrometallurgy, 84(1–2):109-117. 237-248.

Chen, N.X., Liao, Z.W., 2008. The Influence of the Electrolytic Composition on EMM (in
NU
Chinese). China's Manganese Industry, 26(4):5-8.
MA

Chenli, Z., Lian, F., Ma, L., 2016. A novel method to remove Ca2+ and Mg2+ impurities from

manganese sulfate solution. International Conference on Measurement, Instrumentation and

D

Automation.
E

Dreisinger, D,B., Cooper, W.C., 1984. The solvent extraction separation of cobalt and nickel using
PT

2-ethylhexylphosphonic acid mono-2-ethylhexyl ester. Hydrometallurgy, 12(1):1-20.

CE

Helen, E.F., Geoffrey A.L., Erica J.W., 2007. Solubility of calcium sulfate salts in acidic
AC

manganese sulfate solutions from 30 to 105 °C. (86): 13-21.

Jouni, P., Erkki, P., 2011. Effect of temperature on Mn–Ca selectivity with organophosphorus acid

extractants. Hydrometallurgy, 106(3):159-164.

Li, X.H., Ji, D.D., Su H.F., Pan, L.P., Li, J., Li, J., Wen, Y.X., Study on production of manganese

sulphate by high-temperature crystallizing process (in Chinese). Inorganic Chemicals

Industry, 42(12): 16-19.

20
 ACCEPTED MANUSCRIPT

Liang, M., Luo, C.L., Liu, C.M., Liu, D.Z., Li, G.L., Deng, Y.G., Chen, F.M., Long, H.P., 2014.

The preparation of high purity manganese sulfate from pyrolusite by reduction method,

(CN103011296B) (in Chinese)

Liu, H.G., Zhu, G.C., 2008. Study on Removal of Calcium and Magnesium from Leaching

Solution of Manganese Ore by Solvent Extraction (in Chinese). China's Manganese Industry,

PT
(01): 34-37.

RI
Mei, J., Ye, H., Li, F.Z., 2014. A Study of the Preparation of Battery-level Manganese Sulfate (in

SC
Chinese). China's Manganese Industry, 32(02): 20-24.

Preston, J.S., Preez, A.C.D., 2010. Solvent extraction of nickel from acidic solutions using
NU
synergistic mixtures containing pyridinecarboxylate esters. Part 1. Systems based on
MA

organophosphorus acids. Journal of Chemical Technology & Biotechnology Biotechnology,

66(1): 86-94.
D

Preston, J.S., Preez, A.C.D., 2000. Separation of nickel and calcium by solvent extraction using
E

mixtures of carboxylic acids and alkylpyridines. Hydrometallurgy, 58(3):239-250.

PT

Shen, Y.F., Xue, W.Y., Niu, W.Y., 2008. Recovery of Co(II) and Ni(II) from hydrochloric acid
CE

solution of alloy scrap. Transactions of Nonferrous Metals Society of China,

AC

18(5):1262-1268.

Tan, L.M., Hu, H.P., Liao, J.M., Wang, C.X., Li, C., 2014. Preparation of High-purity Manganese

Sulfate for Battery by Solvent Extraction (in Chinese). Nonferrous Metals (Extractive

Metallurgy), (09): 62-65.

Xing, X.Y., Ning, S.M., 2015. The Research Status and Analysis on Manganese Sulphate Solution

Purification Technology (in Chinese). Yunnan Metallurgy, 44(6): 22-26.

21
 ACCEPTED MANUSCRIPT

Yan, S., Qiu, Y.R., 2014. Preparation of electronic grade manganese sulfate from leaching solution

of ferromanganese slag. Transactions of Nonferrous Metals Society of China,

24(11):3716-3721.

Yang, W.J., Zhu, L.J., Qu, X.J., 2014. Study on removal of calcium and magnesium ions from

leaching solution of poor pyrolusite by extraction. Journal of Hunan University (Natural

PT
Science), 41(12):83-88.

RI
Yuan, M.L., Qiu, G.Z., 2000. Study on the removing of MgSO4 from MnSO4 solution based on the

SC
aqueous system phase diagram (in Chinese). Journal of Central South University of

Technology, (03):212-214.
NU
Zhang, H., Zhao, K., Chen, F.Y., Jiang. Q., 2014. Application Prospect and Development of High
MA

Pure Manganese Sulfate Monohydrate in Battery Level (in Chinese). China's Manganese

Industry, 32(02): 6-8.

D

Zhang, W.S., Cheng, C.Y., 2007. Manganese metallurgy review. Part II: Manganese separation and
E

recovery from solution. Hydrometallurgy, 89(3-4): 160-177.

PT

Zhu, Y.W., Zhang, T.Q., Huang, C.Y., Mu, J.J., Sang, K.X., Ma, L.B., Chen, Y.Z., 2013. A novel
CE

method to prepare manganese sulfate solution with ultra low concentration of calcium and
AC

magnesium. (CN102849802A) (in Chinese).

Zhu, Y.W., Zhang, T.Q., Qu, Z.S., Zheng, X.Y., Huang, C.Y., Mu, J.J., 2012. A novel method to

prepare manganese sulfate solution with ultra low concentration of calcium and magnesium.

(CN102674464A) (in Chinese).

22
 ACCEPTED MANUSCRIPT

Highlights

 The SSX system consisting of P204 and 4PC exhibited a synergistic effect on

manganese and an antagonistic effect on calcium and magnesium.

 More than 99.9% of Mn2+ was selectively extracted over Ca2+ and Mg2+ at

equilibrium pH of 3 in a four-stage counter current extraction using the SSX

PT
system.

RI
 The mass ratios of Ca/Mn and Mg/Mn in strip liquors were 5.6×10-5 and 2.5×10-4,

SC
respectively.
NU
MA
E D
PT
CE
AC

23
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9