MATEC Web of Conferences 62, 02004 (2016) DOI: 10.

1051/ matecconf/20166202004
ICCFE 2016

Purification of Food-grade Magnesium Chloride

1 1, a 1 1 1 1
Lianmin Ji , Zhiqi Liu , Lijuan Li , Xuexue Song , Zhongmin Zeng , Feng Nie
1
Key Laboratory of Comprehensive and Highly Efficient Utilization of Salt Lake Resources, Chinese Academy of Sciences
Qinghai Institute of Salt Lakes, Chinese Academy of Sciences, 810008 Xining Qinghai, China

Abstract: The application of the varying weights of bischofite dissolved in the distilled water was investigated. The
effects of the temperature on the rate of evaporation and the thermal precipitation time on the purity of the crystal
products were fully investigated. Two validation tests including magnifying tests and recycling residue were also
studied. Our results demonstrate that the contents of NaClǃKCl and CaSO4 in the filtrate reached a minimum value
after the pretreatment of 350 g bischofite dissolved in 100 ml distilled water. In the crystal products from the second
evaporating stage of the validation tests, the contents of MgCl2·6H2O, SO4 and NaCl+KCl are 99%, ≤0.1±0.01% and
≤0.8±0.04%, respectively. The content of magnesium chloride in the solution was increased to a greater extent, and
the impurities reduced correspondingly through the dissolution pretreatments of bischofite. This could decrease
energy consumption for the impurity removing stage, evaporation and crystallization process, and thus reduce costs
for the industrial production of food-grade magnesium chloride

1 Introduction
The positive correlation between diet and health has led 2 Experimental
health agencies around the world to control the intake of
certain food components that are thought to promote 2.1 Reagents and instrumentation
some disorders [1].According to the national standards
for food safety (GB2760ˉ1996ˉ2005), the additives The raw ore used in this study included bischofite
used in food production must be food grade additives. (Golmud zhenghai magnesium science and technology
Edible magnesium chloride (MgCl2·6H2O) as additive has development Co., Ltd.)
more widely application in food, salt, mineral water, The instruments used for the analyses included stirrer
medicine and other industries [1-4]. (IKA EUROSTAR 60; Shanghai Chuangyi Science and
The Qinghai salt lakes in the qaidam basin are well education equipment Co., Ltd.), oil bath pan (DF-101S;
known for their huge reserves of potassium chloride (KCl) Gongyi Yuhua Instrument Co., Ltd.), an inductively
and magnesium chloride (MgCl2) in China [5]. In recent coupled plasma–atomic emission spectrometer
years, about 4.5 Mt·a-1 potassium fertilizer has been (ICAP6500 DUO; America Thermo Scientific, USA).
produced with 24-30 Mt·a-1 magnesium chloride as
by-product, and a large amount of magnesium chloride 2.2 Experiments
has been left as the by-product or even waste in salt lakes
of Qinghai [6,7]. This has caused not only the waste of
magnesium resources, but also the environmental 2.2.1 Solution analysis
pollution [8]. The production of food grade magnesium
The solution with a certain amount of bischofite dissolved
chloride used bischofite as raw material is another
in the distilled water was mechanically stirred at a
effective utilization way of magnesium resources of salt
frequency of 300 rpm for 30 min at room temperature (20
lake, in which it is playing an active role.
± 2 °C). Subsequently, the residue and filtrate were
This study investigated the performance of the
separated and weighed. The filtrate was analyzed using
varying weights of bischofite dissolved in the distilled
EDTA volumetric analysis to determine the Mg2+, Cl-
water, determined the optimal ratio relationship between
concentrations, and inductively coupled plasma–atomic
bischofite and distilled water under the condition of the
emission spectrometry (ICP) to determine the Na+, K+,
least amount of impurity in the filtrate, and confirmed the
Ca2+, As, Pb, SO42- and B2O3.
reliability and repeatability of the optimal dissolution
ratio.
2.2.2 Evaporative crystallization

a
Corresponding author: [email protected]

© The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons Attribution
License 4.0 (http://creativecommons.org/licenses/by/4.0/).
MATEC Web of Conferences 62, 02004 (2016) DOI: 10.1051/ matecconf/20166202004
ICCFE 2016

(1) First evaporating stage refrigerated to 40 °C for filtration. Finally, the product of
The filtrate was heated up to110±5 °C for some water crystals MgCl2·6H2O was obtained and analyzed using
loss, and then thermally precipitated for 1h, finally, EDTA volumetric analysis to determine the Mg2+
conducted with thermal filtration. concentrations, and ICP to determine the Na+, K+, Ca2+
(2) Second evaporating stage and SO42-concentrations.
In order to obtain crystals MgCl2·6H2O, this filtrate The process flowsheet for purification of food-grade
from the first evaporating stage was heated up to140 °C magnesium chloride was shown in Figure 1.
again for some water loss, and then the thermal

Bischofite
Residue

Filtrate The first Thermal precipitation,

Water Dissolve Filter Residue
evaporating stage filter

Cooling crystallization The second

Crystal product Filter
evaporating stage

Figure 1. Process flowsheet for purification of food-grade magnesium chloride

The experiments were conducted to investigate the effect
of the solubility of bischofite on the contents of soluble
3 Results and Discussion impurities in the filtrate. The bischofites with weights
varying from 100 g to 500 g were dissolved in the
3.1 Materials analysis distilled water of 100 ml. After filtering, the density of
filtrate, the concentrations of MgCl2 and the components
In order to obtain the composition of the bischofite, the of soluble impurities were analyzed respectively.
solution with 50.0390 g bischofite dissolved in the As is shown in Figure 2, the contents of residue
distilled water was transferred into a 500 ml volumetric gradually increased with increasing the weights of
flask, and then analyzed using EDTA volumetric analysis bischofite. The contents increased rapidly with bischofite
and ICP. The result of such analysis is presented in table weights higher than 350 g. Notably, a large amounts of
1. The original magnesium chloride content (wt. %) is mineral MgCl2˜6H2O dissolved and remained in the
44.77 %. The contents of impurities As, B2O3 and Pb in residue. As observed in the Figures 2 and 3, the weight
bischofite are unavailable due to extremely low values. contents of MgCl2 steadily increased with increasing
The total contents of insoluble matter and soluble bischofite weight. However, the contents of NaCl and
impurities (NaCl+KCl+CaSO4) (wt. %) are ˘ 0.10 % KCl initially increased and then rapidly decreased with
and 4.6 %, respectively. increasing bischofite weights. While the CaSO4 contents
declined straightly. When the bischofite weight was
Table 1. Result of materials analysis higher than 350 g, all the NaCl, KCl and CaSO4 contents
Constitutes Content (wt. %) reached minimum levels. Therefore, the relationship of
As 0.00006595 optimal ratio between bischofite and distilled water was
B2O3 0.0003732 determined as 350 g/100ml.
Ca2+ 0.2314
K+ 0.8693
Na+ 0.8469
Pb 0.00002598
SO42- 0.5372
MgCl2 44.77
NaCl 2.15
KCl 1.66
CaSO4 0.79
Insoluble matter 0.095

3.2 Varying solubility experiment

Figure 2. Relationship between MgCl2 and residue with
bischofite dissolved in the distilled water of 100 ml

2
MATEC Web of Conferences 62, 02004 (2016) DOI: 10.1051/ matecconf/20166202004
ICCFE 2016

3.4 Effect of the thermal precipitation time

According to the optimal ratio of 350 g bischofite to 100
ml distilled water, the filtrate was obtained and heated up
to 110±5 °C for some water loss, and then thermally
precipitated for 1 a 5 h. Finally, it was conducted with
filtration. The result for the effect of the thermal
precipitation time on the purity of crystal product is listed
in the Table 2. Table 2 shows that the purities of crystal
product reached maximum levels when the thermal
precipitation time was 1-3h. However, the purity value
reduced as the thermal precipitation time was 5 h. Based
on the energy consumption consideration, the optimum
Figure 3. Relationship between soluble impurities with time is 1 h.
bischofite dissolved in the distilled water of 100 ml
Table 2. Effect of the thermal precipitation time
3.3 Effect of temperature Precipitation time (h) MgCl2 (wt. %)
1 46.30
To confirm the temperature for the second evaporating
2 46.18
stage, the studies were carried out through varying
temperature from 140 to 170 °C for 240 min. The results 3 46.39
are shown in Figure 4 a and b. The evaporative loss of 5 45.37
water initially decreased sharply and then gently declined Precipitation time (h) MgCl2 6H2O (wt. %)
with lengthening time. 1 98.86
The average speeds of evaporation during the time of 2 98.61
0 a 75 min were 0.46, 0.49, 0.53 and 0.55 g/min, 3 99.06
respectively. While the average speeds during the time of 5 96.88
75 a 240 min were 0.051, 0.054, 0.063 and 0.053 g/min
respectively. As observed, all the evaporation rates 3.5 Validation
became slow after 75 min, and had nothing to do with the
temperature. Therefore, the optimal evaporation
temperature was 140 °C for the purpose of reducing 3.5.1 Magnifying tests
energy consumption. The solutions were obtained with 1750 g bischofite
dissolved in 500 ml distilled water, 5 times the original
weights based on the optimal ratio in the section 3.2, and
conducted to the first and second evaporating stages after
filtration. The analyzed results of the products are shown
in the tables 3 and 4. In those two products, the varying
evaporation losses (17.32 g and 16.25 g) for the first stage
gave rise to the different weight levels of crystals (12.68 g
and 3.63 g, respectively) (Table 3). However, the total
contents of MgCl2·6H2O, SO4 and NaCl+KCl ˄wt. %˅
were ˚ 99%, ˘ 0.1% and ˘ 0.6%, respectively
(Table 4). This reveals that the varying evaporation losses
for the first stage have no influence on the contents of the
crystal products for the second stage.
Table 3. Weights of evaporation loss and crystal

Test 1 stages g
Evaporation loss for the first stage 17.32
Crystal for the first stage 12.68
Evaporation loss for the second stage 45.65
Crystal for the second stage 194.79
Test 2 stages g
Evaporation loss for the first stage 16.25
Crystal for the first stage 3.63
Evaporation loss for the second stage 45.34
Crystal for the second stage 203.84

Figure 4. Effect of the temperature on the evaporation rate (a

and b)

3
MATEC Web of Conferences 62, 02004 (2016) DOI: 10.1051/ matecconf/20166202004
ICCFE 2016

Table 4. Contents of crystal products for the second stage Table 6. Contents of crystal products for the second stage

Product 1 wt. % Product 1 wt. %

MgCl2 46.41 MgCl2 46.20
MgCl2·6H2O 99.10 MgCl2·6H2O 98.65
NaCl 0.47 NaCl 0.61
KCl 0.12 KCl 0.17
CaSO4 0.022 CaSO4 0.14
SO4 0.015 SO4 0.099
Product 2 wt. % Product 2 wt. %
MgCl2 46.47 MgCl2 46.36
MgCl2·6H2O 99.23 MgCl2·6H2O 98.99
NaCl 0.46 NaCl 0.64
KCl 0.12 KCl 0.20
CaSO4 0.031 CaSO4 0.16
SO4 0.022 SO4 0.11

3.5.2 Recycling residue 4 Conclusions

(1) The solution with 350 g bischofite dissolved in the (1) The optimal ratio relationship between bischofite and
distilled water was mechanically stirred for 30 min, and distilled water was 350 g : 100 ml.
then filtered. (2) As shown in the validation results for magnifying
(2) The residue from the section (1) was weighted and tests and recycling residue, the products obtained
mixed with bischofite into mixture of total 350 g. Then, according to the optimal ratio meet the product
repeat the operation (1). qualification. In the products, the total contents of
(3) Repeat the above (1) and (2) for 2 times. Finally, MgCl2·6H2O, SO4 and NaCl+KCl (wt. %) are 99%
the obtained filtrate was conducted to the first and second approximately, ≤ 0.1±0.01% and ≤ 0.8±0.04%,
evaporating stages after filtration. The analyzed results of respectively.
the products are shown in the tables 5 and 6. In the two
products, the varying evaporation losses (22.87 g and
29.62 g) for the first stage resulted in the crystals (27.78 g Acknowledgments
and 41.92 g, respectively) (Table 5). However, the total
contents of MgCl2·6H2O, SO4 and NaCl+KCl (wt. %) are The authors appreciate the financial support of Key
99% approximately, ≤ 0.1±0.01% and ≤ 0.8±0.04%, Project supported by QingHai Science Technology
respectively (Table 6). This also indicates that the varying Department (2014-GX-C10) and (2015-HZ-812).
evaporation losses for the first stage have no influence on
the contents of the crystal products for the second stage. References
Table 5. Weights of evaporation loss and crystal 1. C. Horita, M. Morgano, R. Celeghini, M. Pollonio,
Test 1 stages g Meat Sci. 89, 426 (2011).
Evaporation loss for the first stage 22.87 2. W. Q. Yang, Chem. Technol. Market 32, 33 (2009).
Crystal for the first stage 27.78 3. S. Barbut, C. Findlay, J. Food Sci. 56, 180 (1991).
Evaporation loss for the second stage 33.16 4. K. Toda, K. Yagasaki, K. Takahashi, Biosci.,
Crystal for the second stage 261.34 Biotechnol., Biochem. 72, 2824 (2008).
Test 2 stages g 5. P. Zhang, B. Zhang, Y. Tang, C. Yang, S. Huang, J.
Evaporation loss for the first stage 29.62 Wu, Sci. Press 99, 107 (1999).
Crystal for the first stage 41.92 6. Y. Zhou, L. J. Li, Z. J. Wu, X. Li, Chem. Res. 25,
Evaporation loss for the second stage 30.15
1613 (2013).
Crystal for the second stage 240.57
7. W. T. Cheng, Z. B. Li, G.P. Demopoulos, Chin. J.
As shown in the validation results for magnifying tests Chem. Eng. 17, 661 (2009).
and recycling residue, the products obtained meet the 8. P. Ma, Adv. Earth. Sci. 15, 365 (2000).
product qualification (GB 25584-2010), according to the
optimal ratio relationship between bischofite and distilled
water (350 g: 100 ml).