IOP Conference Series: Earth and Environmental Science

PAPER • OPEN ACCESS

Magnesium-Containing Slurry as Technogenic Alternative Raw Material

for Magnesium Oxychloride Cement
To cite this article: I N Lipunov and I G Pervova 2019 IOP Conf. Ser.: Earth Environ. Sci. 272 032032

View the article online for updates and enhancements.

This content was downloaded from IP address 178.171.6.216 on 21/06/2019 at 18:06

International science and technology conference "Earth science" IOP Publishing
IOP Conf. Series: Earth and Environmental Science 272 (2019) 032032 doi:10.1088/1755-1315/272/3/032032

Magnesium-Containing Slurry as Technogenic Alternative

Raw Material for Magnesium Oxychloride Cement

I N Lipunov1, I G Pervova1
1
Department of physical and chemical technologies in environmental engineering of
Ural State Forest Engineering University, Yekaterinburg, Russia

E-mail: [email protected]

Abstract. It has been shown experimentally that a waste of magnesium metal production is a
realistic alternative to naturally occurring cement raw materials. The waste is the carnallite
slurry forming during the chlorination stage. The slurry contains magnesium oxide (MgO) and
magnesium chloride (MgCl2) in a mass ratio close to their optimal ratio when preparing Sorel
cement. It is found that the magnesium oxide in the slurry is highly reactive. MgO extracted
from the slurry and MgCl2 solution are mixed forming homogeneous magnesium oxychloride
cement (MOC) paste in which bonding crystalline structures typical for Sorel cement are
formed during hardening.

1. Introduction
Magnesium oxychloride cement (MOC) are widely used in the construction industry. Traditional
MOC raw materials are natural minerals such as magnesite, dolomite, and brucite. However, the
shortage of naturally occurring cement raw materials, arising environmental challenges associated
with waste management, as well as the high cost of caustic magnesia binders require to find alternative
raw materials for manufacture of magnesium-based cements. The excellent solution for that is using
magnesium-containing technogenic raw materials [1,2] with a special focus on a MgO- and MgCl2-
containing waste such as the carnallite slurry forming during the chlorination stage of magnesium
metal production.
The aim of this study was to determine chemical, phase, and radionuclide compositions of the
slurry forming during magnesium metal manufacturing, VSMPO-AVISMA Corporation, Russia; to
study the hydration kinetics of MgO containing in the slurry, binding properties of MgO and the
slurry, and to develop technology for recovering and recycling the carnallite slurry to MOC powder.

2. Research methods
The slurry for the study was prepared (regeneration) by mechanical crushing, grinding to a particle
size varying between 100–200 μm and fractionating. A laboratory unit for fractionation consisted of a
jaw crusher and ball mill equipped with a cyclone for dust collection. Screen size gradation was used
to fractionate the powder. The slurry prepared for the study was stored in hermetic containers to avoid
the hydration of MgO containing in the slurry.
Quantitative determination of the chemical composition of the slurry was investigated by chemical
and X-ray diffraction analyses. Chlorides of alkali metals were determined by flame
spectrophotometry; magnesium-containing compounds were determined by complexometric titration.

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
International science and technology conference "Earth science" IOP Publishing
IOP Conf. Series: Earth and Environmental Science 272 (2019) 032032 doi:10.1088/1755-1315/272/3/032032

X-ray diffraction analysis was also used to reveal the phase state of individual compounds containing
in the slurry. X-ray patterns of the studied slurry samples were shot on auto-diffractometer STADI-
(STOE, Germany), copper Кα radiation, an interval of angles 2θ from 5° to 80°, step of 0.02°. The
compounds were identified according to “Base powder standards JCPDS-ICDD PDF2” (ICDD, USA,
Release 2003).
Reactivity of MgO extracted from the slurry, was determined by the degree of its hydration at
temperatures of 40, 60, and 80 0C in systems MgO− H2O and MgO− MgCl2−H2O. Magnesium oxide
was extracted from the slurry by hydroleaching the soluble salts (MgCl2, KCl, NaCl); the formed
heterogeneous mixture was separated by filtration, the resulting residue after the filtration was washed
by water, then by acetone, made it air-dried and kept in hermetic containers. According to X-ray
diffraction analysis the MgO content in the residue was 97 % wt. The hydration of MgO was
investigated by thermogravimetric analysis. Derivatograms of hydrated MgO samples were filmed
with derivatograph Q-1500D. The investigated samples were heated in air medium in crucibles to a
temperature of 700 0C at rate of 10 deg∙min‾1. The sample mass was 1000 mg, the sensitivity of DTA–
250, both TGP and TG – 500 μV. The product compositions after the hydration were determined by
X-ray diffraction analysis and infrared spectroscopy.
The experimental values of the degree of MgO hydration were calculated according to
derivatograms. To do this, the established ratio of the mass loss in the samples were related to
stoichiometric amounts of products after thermal decomposition of crystallohydrate structures
Mg(OH)2 and 3Mg(OH)2∙MgCl2∙8H2O. The rate of the MgO hydration in the investigated mixtures
was calculated according to the results of kinetic studies on the equations of Kolmogorov−Erofeev and
Sakovich [3].
The role of MgO in the formation of crystallohydrate structures was revealed experimentally
according to X-ray diffraction analysis. Normal density magnesium mixture, start and final moments
of its setting, formed when mixing MgO with MgCl2 solution, and the slurry – with water, were
determined in accordance with GOST 310.3−76 using Vika tool. Binding properties of MgO and the
slurry were estimated by the compressive strength values of the laboratory samples of magnesium
stone, shaped in the form of cylinders, diameter and height 20mm and seasoned for 3, 7, and 28 days
for strength developing. The samples were subjected to destruction under dynamic load test on a
hydraulic press 2NG−10 with evenly increasing speed of 1 mm/s.
Toxicological tests of the slurry were carried out in the Center of State Sanitary and
Epidemiological Supervision, Sverdlovsk region. The natural radioactivity of the slurry was
determined by the method of γ-spectrometry.

3. Results and discussion

Methods of chemical analysis identified the substance composition of the slurry, which is composed of
98.4 wt. % of NaCl, KCl, MgCl and MgO. Impurity compounds containing in the slurry are SiO, FeO,
AlO and CaCl2 and BaCl2 4.0∙10-1, 2.7∙10-1, 9.0∙10-2, 2.0∙10-3 4.3∙10-4 wt. % respectively.
When studying several sets of the slurry samples taken at the site for dehydration of carnallite
melting in VSMPO-AVISMA Corporation, Russia, it was revealed that the vertical cut of the slurry
ingot is a “layer cake” and each layer is differently coloured: white (upper layer), grey (middle layer)
and yellow (bottom layer). The differently coloured slurry samples were analyzed to detect
quantitative content of the main chemical compounds and their phases by chemical and X-ray
diffraction analyses (table 1).
The white layers were characterized by high content of the water-soluble chemical compounds
MgCl2, KCl and NaCl, which, when cooling the slurry melting, "were getting salted out" to the upper
parts of the ingots. MgO was almost absent in the upper parts.
The grey layers were presented by magnesium components (MgO, MgCl2), their contents were
35.0 и 27.3 wt. % respectively according to chemical analysis.

2
International science and technology conference "Earth science" IOP Publishing
IOP Conf. Series: Earth and Environmental Science 272 (2019) 032032 doi:10.1088/1755-1315/272/3/032032

Table 1. Mass fraction (%) of main chemical compounds in the differently coloured layers and in the
averaged sample.
Chemical analysis X-ray diffraction analysis
Slurry KCl∙MgCl2∙
color MgO MgCl2 NaCl KCl MgO ∙6H2O KCl∙MgCl2 NaCl
a
White 0.7 47.4 2.7 18.1 8.0 80.5 5.9
a
Grey 35.0 27.3 1.7 10.0 31.4 51.5 3.8
a
Yellow 56.6 18.1 1.1 6.3 72.6 46.2 1.3
a
Averaged 42.3 27.9 2.0 16.6 63.3 34.6 2.2
a
the compound is not presented in the studied sludge sample

The yellow layers were characterized by high MgO content. MgO precipitated to the bottom parts
because of its higher density in comparison to other chemical compounds.
According to X-ray diffraction analysis the chemical composition of white coloured layers is
characterized by the presence of three separate phases: the separate phase NaCl (d = 0.282 and 0.199
nm), the phase of anhydrous carnallite KCl∙MgCl2 (0.285; 0.248; 0.246 and 0.175 nm), and the phase
of hexaquacarnallite KCl∙MgCl2∙6H2O (d = 0.333; 0.298; 0.239; 0.234 nm). The chemical composition
of the grey and yellow coloured slurry layers are also characterized by the presence of three separate
phases: the phase of hexaquacarnallite KCl∙MgCl2∙6H2O, the phase NaCl, and the phase MgO (d =
0.211; 0.149; 0.122 nm).
The uneven distribution of chemical compounds along height and across volume of the slurry ingot
forming in a mold is explained by sedimentation and stratification of the chemical components of the
slurry in mixers and molds during cooling and crystallisation of the slurry melting.
The averaged (regenerated) slurry sample was obtained by mixing representative differently
coloured slurry layers in equal mass ratio, then the mixture was crushed, grinded and thoroughly
mixed towards grey powder with definite particle sizes. According to X-ray diffraction analysis the
chemical components in the averaged sample were presented the same phases as in the grey and
yellow coloured slurry layers, however they were evenly spread throughout the sample. The mass ratio
MgO:MgCl2 for the recovered slurry was 1: 0.66 which is close to the optimal ratio of these
components while forming magnesium cement from caustic magnesite and magnesium chloride
solution (1 : 0.62÷0.65).
Thus, the results of these studies showed the necessity of regeneration of the original slurry to
achieve homogeneity and constancy of the chemical composition, especially magnesium components,
throughout the slurry.
Because of high sensitivity of the magnesium binding materials technology to MgO reactivity, as
well as the lack of understanding the mechanism of underlying processes of magnesium cement
formation, especially when using technogenic materials, we investigated the reactivity of MgO
extracted from the regenerated slurry. The background for such research was the developing
knowledge of the kinetics of the MgO hydration in the system MgO−H2O and MgO−MgCl2−H2O at
different temperatures.
The MgO hydration in the investigated systems was described by the following equations:

MgO + H2O = Mg(OH)2 ,

3MgO + MgCl2 + 11H2O = 3Mg(OH)2∙MgCl2∙8H2O.

The values of the MgO hydration degree were evaluated in accordance with the endothermic
effects in the derivatograms of hydrated samples. It was found the values got higher with temperature
rose and were significantly higher for the system MgO−MgCl2−H2O (table 2). The fact indicates that
technogenic MgO has higher reactivity when it interacts with MgCl 2 solution, which is in good
agreement with the scientific report data concerning the reactivity of caustic magnesia binder [4, 5].

3
International science and technology conference "Earth science" IOP Publishing
IOP Conf. Series: Earth and Environmental Science 272 (2019) 032032 doi:10.1088/1755-1315/272/3/032032

Table 2. Values of the MgO hydration degree (α, %) in the investigated systems at different
temperatures and exposure time 6 h.
Temperature, 0С
40 60 80
MgO−H2O 4.1 17.2 34.4
MgO−MgCl2−H2O 22.5 61.8 75.3

The values of the hydration rate constants calculated by mathematical treatment of kinetic curves of
MgO hydration degree in the investigated systems [4] are given in Table 3.

Table 3. Values of the MgO hydration rate constants in systems MgO−H2O and MgO−MgCl2−H2O.
Process temperature, 0С
40 60 80
MgO−H2O 3.077∙10-5 5.026∙10-2 1.871∙10-1
MgO−MgCl2−H2O 2.513∙10-2 1.827∙10-1 2.964∙10-1

The values of the hydration rate constants got higher when temperature rose; the values for the
system MgO−MgCl2−H2O appeared to be higher by 1000, 10, and 1.5 times at temperatures 40, 60,
and 80 0С respectively.
Study of hydration kinetics has shown that MgO containing in the slurry is highly reactive and,
being mixed with MgCl2 solution, forms magnesium mixture with following forming crystallohydrate
structures, which are the basis for magnesium cement.
It has been established experimentally that MgO takes part in the forming crystallohydrate
structures of magnesium cement. MgO samples were gauged by MgCl2 solution with different mass
solid: liquid ratios (S:L) to form homogeneous MOC paste. It was found that optimal parameters for
the homogeneous MOC paste formation with following hardening and the crystallohydrate structures
forming: the mass ratio MgO and MgCl2 solution (1.2 g/cm3) - 1.96, temperature - 80 0C, and time of
pre-heating the reaction mixture - 30 min, under these parameters the start and final setting time were
20 and 60 min. respectively, and they correspond to the normative values for caustic magnesite
binders.
Laboratory samples formed from the paste and seasoned for strengthening for 3, 7, and 28 days had
a maximum compressive strength of 18.7; 31.2 and 53.1 MPa respectively.
Kinetic studies of strengthening of the laboratory samples showed that the process of forming
crystallohydrate structures ran intensively during the first 7 days and by the finish of the period the
magnesium cement gained 60 % of the grade strength (Figure 1). After 28 days forming
crystallohydrate structures of magnesium cement was almost completed and the compressive strength
of the cement was maximal.

Figure 1. Relation between ultimate

compressive strength and hardening time
for the laboratory samples formed from the
mixture of MgO and MgCl2 solution at
S:L = 1.96 (1) and slurry with water at
S:L = 0.43 (2).

4
International science and technology conference "Earth science" IOP Publishing
IOP Conf. Series: Earth and Environmental Science 272 (2019) 032032 doi:10.1088/1755-1315/272/3/032032

Forming crystallohydrate structures in the samples aged of 28 days of hardening time was
confirmed by the presence in the X-ray diffraction lines d = 0.478, 0.237, 0.180, 0.157, and 0.148 nm,
corresponding to the hydrated form of the composition Mg(OH) 2 and d = 0.240, 0.226, and 0.198 nm
corresponding to the crystallohydrate composition 3Mg(OH)2 ∙ MgCl2 ∙ 8H2O (Figure 2).

Figure 2. X-ray pattern of the sample, formed from MgO, gaged with MgCl2 solution (ρ = 1.20 g/cm3)
and aged 28 days.

Presence of the hydrated form in the magnesium “stone” was also evidenced by the intensive
absorption band appeared in the IR spectrum of the studied sample in the region of 3.500 to 3.700 cm–
1
, typical for stretching vibrations of associated OH-groups.
Thus, the mixing MgO with MgCl2 solution resulted to the formation of strong magnesium
structures; this fact indicated high reactivity of MgO in the reactions of forming and hardening of
magnesium cements.
The binding properties of the slurry were studied in the system slurry–H2O, since MgCl2 containing
in the slurry interacts with water and forms the internal solution of magnesium chloride, the last and
MgO take part in the formation of magnesium-based cement. The optimal mass ratio of water and
slurry (S: L) was determined experimentally. The laboratory samples of magnesium-based cement,
formed from homogeneous MOC paste with different ratios S:L, were tested for strength
characteristics (Figure 3).

Figure 3. Relation between ultimate

compressive strength and (Rcs) and the ratio
S:L for magnesium stone samples, formed in
the system sludge – H2O at 1) 3-day
hardening time and 2) 28-day hardening
time.

High values of the ultimate compressive strength were achieved for samples of 3- and 28-day
hardening time when the ratio of the MOC paste was S:L = 0.37 (72.3 and 89.1 MPa, respectively).
However, the setting time for such MOC paste was less than 5 min. With the ratio, the paste cannot be
used in the technological process, for example, to prepare wood-mineral composition because of very
short lifetime of the MOC paste.
The ratio of W:T = 0.43 is likely to be considered the optimal ratio when the start and finish of
hardening time for the magnesium mixture is 20 and 40 min, respectively, this is typical for the
formation of the magnesium mixture on the basis of caustic magnesite and magnesium chloride

5
International science and technology conference "Earth science" IOP Publishing
IOP Conf. Series: Earth and Environmental Science 272 (2019) 032032 doi:10.1088/1755-1315/272/3/032032

solution. The ultimate compressive strength values for samples of 3- and 28-day hardening time were
32.4 and 50.0 MPa respectively.
Hardening the magnesium-based cement formed in the system at the optimum ratio of S: L ran
intensively during the first 7 days (Figure 1, curve 2). Similar shapes of the kinetic curves describing
the hardening the magnesium paste in the open air indicated that adding water into the slurry resulted
to forming the internal MgCl2 solution, which took part in gauging MgO containing in the slurry with
forming hydrated structures.
The formation of hydrated compounds containing in magnesium-based cement, produced on the
basis of the slurry and water at the optimum mass ratio, was confirmed by X-ray diffraction analysis.
In the X-ray diffraction pattern of the 28-day hardening sample there was a number of both diffraction
lines typical for the hydrated compounds 3Mg(OH)2 ∙ MgCl2 ∙ 8H2O and Mg(OH)2 and diffraction
lines, d = 0.238, 0.192 and 0.183 nm, typical for the crystallohydrate 5Mg(OH)2 · MgCl2 · 2H2O.
However, the concentration of magnesium pentoxihydrochloride was significantly lower in compared
with trioxihydrochloride indicating involving MgO into the hydration as an intermediate metastable
phase.
Mineral binders used in the construction material industry should comply with the radiation safety
criteria in according to radiation safety standards of Russian Federation NRB-99/2009. Specific
effective radioactivity of the slurry was 430 Bq∙kg-1 so the waste can be referred by radiation safety to
the materials of class II. Composite mixture of the slurry and sawdust with weight ratio 1:1 showed
290 Bq∙kg-1.Such construction mixes refer to the class I of danger on radiation safety so they are
suitable for the manufacturing construction materials for residential and public buildings.
The toxicological tests of the slurry and its mixture with sawdust, weight ratio 1:1, has shown that
they are non-toxic. MgO and KCl ∙ MgCl2 ∙ 6H2O are potentially hazard in the slurry and its dust.
While using, the cement doesn’t emit any chemical components to the air exceeding the threshold
limit values (TLV). Sanitary-epidemiological conclusion certificate No.66.01.10.530 T. 000055.06.01
(issued 02.07.2001) confirmed the compliance of the material to the state sanitary-epidemiological
regulations.
The experimental results of the balance experiments to determine output of effective fractions
when regenerating slurry, magnesium components and their mass ratio have provided the basis for
developing novel technology and the unit to recycle the slurry to the MOC powder.

4. Conclusion
The comprehensive study has proved that the carnallite slurry forming during the chlorination stage
can be used to produce magnesia binder materials. It is found that the magnesium oxide in the slurry is
highly reactive and, being mixed with magnesium chloride solution, it forms a homogeneous MOC
paste with the subsequent formation of hydrate crystal structures which are the MOC basis. The
characteristics of the slurry comply with radiation and sanitary-and-epidemiological safety
requirements of Russian Federation because there is no any emission of the chemical components
exceeding the maximum permissible concentrations in the atmospheric air. The studied technogenic
raw material is alternative substitution of acoustic magnesites in manufacturing wood mineral
composites for construction industry.

5. References
[1] ZyryanovaV N, Berdov G I, VereshchaginV I 2009 Izvestija vuzov. Stroitel'stvo - The Bulletin of
Russian Universities. Construction 21 (In Russ.)
[2] ZyryanovaV N, Berdov G I 2005 Izvestija vuzov. Stroitel'stvo - The Bulletin of Russian
Universities. Construction 46 (In Russ.)
[3] Boldyrev V V 1959 Methods to study the kinetics of thermal decomposition of solids (Moscow,
Metallurgizdat Publ) p 62 (In Russ.)
[4] Chau C K, Li Z J 2008 Advances in Cement Research 20 85
[5] Li G, Yu Y, Li J 2008 Construction and Building Materials 22 521