Carbon Dioxide Uptake by MOC-Based Materials

Featured Application: The highest potential of magnesium oxychloride cement (MOC) is its capability to be used as a component of low-energy building composite materials while acting as a CO 2 sink. The results of this contribution also show that MOC can be used as a binder in advanced building materials that have particular properties, and therefore speciﬁc application potentials. Formation and hardening of this material are rather fast, so the material can be used in quick repairs as well as a protection layer. This property is also beneﬁcial for use in prefabrication, due to the possibility of unmolding after a shorter time compared to Portland cement (PC) materials, so the whole production process can be considered more e ﬀ ective. Somewhat signiﬁcant importance should be given to its ability to capture CO 2 , which not only makes it more eco-friendly, but also improves its mechanical properties. Abstract: In this work, carbon dioxide uptake by magnesium oxychloride cement (MOC) based materials is described. Both thermodynamically stable magnesium oxychloride phases with stoichiometry 3Mg(OH) 2 · MgCl 2 · 8H 2 O (Phase 3) and 5Mg(OH) 2 · MgCl 2 · 8H 2 O (Phase 5) were prepared. X-ray di ﬀ raction (XRD) measurements were performed to conﬁrm the purity of the studied phases after 7, 50, 100, 150, 200, and 250 days. Due to carbonation, chlorartinite was formed on the surface of the examined samples. The Rietveld analysis was performed to calculate the phase composition and evaluate the kinetics of carbonation. The SEM micrographs of the sample surfaces were compared with those of the bulk to prove XRD results. Both MOC phases exhibited fast mineral carbonation and high maximum theoretical values of CO 2 uptake capacity. The materials based on MOC cement can thus ﬁnd use in applications where a higher concentration of CO 2 in the environment is expected (e.g., in ﬂooring systems and wall panels), where they can partially mitigate the harmful e ﬀ ects of CO 2 on indoor air quality and contribute to the sustainability of the construction industry by means of reducing the carbon footprints of alternative building materials and reducing CO 2 concentrations in the environment overall.


Introduction
Magnesium oxychloride cement (MOC), also known as Sorel cement, was discovered by French engineer Stanislas Sorel in 1867 [1]. Since then, it has often been compared to Portland cement (PC), which was discovered just a few years earlier. It is formed using a mixture of magnesium oxide with an aqueous solution of magnesium chloride in a certain ratio. The ratios in the system MgO-MgCl 2

67
MOC phases and MOC-based composites have been intensively studied in recent years.

68
Synthesis, structure, and the thermal stability of magnesium oxychlorides in particular were studied 69 in detail [6,7]. MOC is a non-hydraulic versatile binder with specific properties, which make it superior 70 to PC. One of the most significant properties is its specific density, which is considerably lower and 71 possesses materially high compressive and flexural strength. Other superior properties are its elevated 72 fire resistance, resistance to abrasion, and low thermal conductivity. All of these properties develop in 73 quite a short curing time, so the material is suitable for quick repairs. Further, MOC is not affected by 74 oils, grease, or paint [8][9][10][11], and it is less alkaline (pH~10) than PC, making it appropriate for use with 75 glass fibers and preventing the aging process as well as with other fillers and aggregates such as tire 76 rubber, synthetic resin, and wood particles [12]. The advantage of MOC also lies in the fact that it has 77 a high bonding capacity with different types of filers including secondary raw materials coming from 78 industrial processes.

79
The versatility and functional performance of MOC-based materials have spurred considerable 80 research into product design, development, and characterization [13][14][15]. Properties of MOC cement-81 based materials make them applicable in many ways-primarily as a flooring material, but also in fire 82 protective systems, grinding wheels, decoration, wall insulation, and others [16,17].

83
At the same time, these materials have also some disadvantages, especially when they come into 84 contact with water [18,19]. MOC has a low water resistance due to magnesium chloride leaching, a 85 corrosive compound that causes severe damage when it comes in contact with metals [20][21][22]. The rest 86 of the material is then left in the form of hydrated brucite (Mg(OH)2) as the binding phase. This process 87 can be prevented by improving the stability of MOC in humid conditions, which can be achieved 88 naturally by letting the surface of Sorel cement react with CO2 from the air. The reaction provides 89 magnesium chlorocarbonates to the surface of the original material. Another method of improving 90 MOC is to use a very small amount of additives such as phosphoric acid or soluble phosphates of alkali 91 metals, alkaline earth metals, aluminum, ammonia, or iron. This step causes the formation of insoluble 92 phosphate complexes that enhance the waterproofness of Sorel cement [23,24]. The addition of silica 93 fume or fly ash is also a suitable way of improving MOC-based materials [25].

94
The main environmental advantage of MOC is its ability to store CO2-one of the main 95 greenhouse gasses (GHGs)-in the form of carbonates. MOC thus acts as a CO2 sink [26]. This process MOC phases and MOC-based composites have been intensively studied in recent years. Synthesis, structure, and the thermal stability of magnesium oxychlorides in particular were studied in detail [6,7]. MOC is a non-hydraulic versatile binder with specific properties, which make it superior to PC. One of the most significant properties is its specific density, which is considerably lower and possesses materially high compressive and flexural strength. Other superior properties are its elevated fire resistance, resistance to abrasion, and low thermal conductivity. All of these properties develop in quite a short curing time, so the material is suitable for quick repairs. Further, MOC is not affected by oils, grease, or paint [8][9][10][11], and it is less alkaline (pH~10) than PC, making it appropriate for use with glass fibers and preventing the aging process as well as with other fillers and aggregates such as tire rubber, synthetic resin, and wood particles [12]. The advantage of MOC also lies in the fact that it has a high bonding capacity with different types of filers including secondary raw materials coming from industrial processes.
The versatility and functional performance of MOC-based materials have spurred considerable research into product design, development, and characterization [13][14][15]. Properties of MOC cement-based materials make them applicable in many ways-primarily as a flooring material, but also in fire protective systems, grinding wheels, decoration, wall insulation, and others [16,17].
At the same time, these materials have also some disadvantages, especially when they come into contact with water [18,19]. MOC has a low water resistance due to magnesium chloride leaching, a corrosive compound that causes severe damage when it comes in contact with metals [20][21][22]. The rest of the material is then left in the form of hydrated brucite (Mg(OH) 2 ) as the binding phase. This process can be prevented by improving the stability of MOC in humid conditions, which can be achieved naturally by letting the surface of Sorel cement react with CO 2 from the air. The reaction provides magnesium chlorocarbonates to the surface of the original material. Another method of improving MOC is to use a very small amount of additives such as phosphoric acid or soluble phosphates of alkali metals, alkaline earth metals, aluminum, ammonia, or iron. This step causes the formation of insoluble phosphate complexes that enhance the waterproofness of Sorel cement [23,24]. The addition of silica fume or fly ash is also a suitable way of improving MOC-based materials [25].
The main environmental advantage of MOC is its ability to store CO 2 -one of the main greenhouse gasses (GHGs)-in the form of carbonates. MOC thus acts as a CO 2 sink [26]. This process is called carbon mineralization, and due to the increasing amount of GHG in the atmosphere, it has recently been intensively studied [27][28][29][30][31][32].
Carbonation of MOC Phases is beneficial for the environment as well as for MOC itself, making the material harder and tougher as a result of its denser microstructure [33][34][35][36]. The significance of these issues has been documented by monitoring air pollution [37][38][39]. Using carbonation curing of cement-based materials during the preparation of building materials has attracted wide attention [40][41][42][43]. Similarly, CO 2 uptake within the carbonation of MOC phases can be beneficially used both from an environmental point of view, and in improving the performances of MOC-based materials.
In this study, the uptake of CO 2 by MOC Phases 3 and 5 is examined along with the long-term kinetics of this process. The morphology of both materials was analyzed using scanning electron microscopy, and the chemical composition of the precipitated MOC phases was determined using energy dispersive spectroscopy. The kinetics of the carbonation of formed stable MOC phases has not previously been reported in the literature in regards to the assumed high CO 2 uptake capacity by carbonated minerals and its use in the passive control of indoor air quality. This could partially reduce the concentration of CO 2 in the polluted environments, and thus slow down global warming-related effects.
The ratio of 5 MgO to MgCl 2 ·6H 2 O was used for the synthesis of a stoichiometric phase of MOC with a composition 5Mg(OH) 2 ·MgCl 2 ·8H 2 O (termed as Phase 5, also known as Mg 3 (OH) 5 Cl·4H 2 O). To prepare the first sample, 30.6 g of MgCl 2 ·6H 2 O was dissolved in 19.0 g of deionized water. In the next step, 30.4 g of magnesium oxide was added, and the suspension was intensively stirred for 10 min. Equation (1) summarizes the formation of Phase 5: Similarly, the stoichiometric Phase 3 (3Mg(OH) 2 ·MgCl 2 ·8H 2 O or Mg 2 (OH) 3 Cl·4H 2 O) was prepared by mixing MgO, MgCl 2 ·6H 2 O and H 2 O in the molar ratio 3:1:5. To prepare the second sample, 39.1 g of MgCl 2 ·6H 2 O was dissolved in 17.3 g of deionized water. In the next step, 23.2 g of magnesium oxide was added, and the suspension was intensively stirred for 5 min. The formation of magnesium oxychloride cement MOC 3-1-8 is summarized in the following Equation (2): Suspensions were used to prepare small samples in XRD holders ( Figure 2). Samples were stored indoor at ambient temperature, with an excess of humidity and carbon dioxide. Figure 2 shows the photography of prepared samples in XRD holders.
For the study of the surface morphology, scanning electron microscopy (SEM) was used (Tescan MAIA 3). Elemental composition and mapping were characterized by energy dispersive spectroscopy (EDS) using X-Max150 analyzer equipped with a 20-mm 2 SDD detector (Oxford instruments) and AZtecEnergy software. The sample was put on a carbon conductive tape in order to ensure the conductivity of the experiments. For both SEM and SEM-EDS analyses, the electron beam was set to 10 kV with a 10-mm work distance.
X-ray powder diffraction (XRD) was carried out by a Bruker D2 Phaser, a powder diffractometer with Bragg-Brentano geometry, applying CuK α radiation (λ = 0.15418 nm, U = 30 kV, I = 10 mA) with a rotation of 5 rpm. The step size was set to of 0.02025 • (2θ), and the overall data were acquired in the range of 5-80 • . Measurements were performed on samples matured for 7, 28, 50, 100, 150, 200, and 250 days. After 250 days, the particular samples were manually homogenized in an agate mortar and measured in a powder form to obtain information on the bulk phase composition. Rietveld analysis was performed for the quantification of the identified phases.
Equation (1) summarizes the formation of Phase 5: of MgCl2•6H2O was dissolved in 17.3 g of deionized water. In the next step, 23.2 g of magnesium oxide 124 was added, and the suspension was intensively stirred for 5 min. The formation of magnesium 125 oxychloride cement MOC 3-1-8 is summarized in the following Equation (2):

127
Suspensions were used to prepare small samples in XRD holders ( Figure 2). Samples were stored 128 indoor at ambient temperature, with an excess of humidity and carbon dioxide. Figure 2 shows the 129 photography of prepared samples in XRD holders.

Results and Discussion
Two MOC phases were prepared and characterized. The phases under study with the compositions 3Mg(OH) 2 ·MgCl 2 ·8H 2 O and 5Mg(OH) 2 ·MgCl 2 ·8H 2 O are referred to as Phases 3 and 5, respectively. The phase purities of both phases were determined using XRD analysis after 7 days (see Figure 3). A fast formation of Phase 3 was confirmed, and a single-phase composition was obtained (ICDD 00-007-0412). Similar results were also obtained for Phase 5. The formation of Phase 5 (ICDD 00-007-0420) was confirmed with no significant impurities. Figure 3 shows the XRD patterns measured after 7 days for MOC Phases 3 and 5.      Another XRD measurement was performed after 50 days of samples maturing. In both samples, chlorartinite was formed on the surface due to carbonation (see Figure 4). Phase 3 reacted with carbon dioxide forming Mg 2 (CO 3 )(OH)Cl·2H 2 O (ICDD 00-061-0391), while Phase 5 formed Mg 2 (CO 3 )(OH)Cl·3H 2 O (ICDD 00-00-0278). A tubular crystal structure of the dihydrate formed by MgO 6 octahedra interconnected by carbonate anions with chloride anions and water molecules located inside the channels is depicted in Figure 5. It is apparent from the comparison with the initial structures of Phases 3 and 5 (Figure 1) that the CO 2 incorporation into the system required a complete reconstruction of the MgO 6 network and, in the case of Phase 5, the formation of another phase (see Equation (4) below). Hence, a relatively slow kinetics of the carbonation process can be anticipated. Figure 4 shows the XRD pattern of MOC Phases 3 and 5 after 50 days of maturing. Figure 5 shows the crystal structure of the chlorartinite phase.   was even lower. Figure 6 shows the surface phase composition of Phases 3 and 5 after 7, 50, 100, 150,   Figure 7 shows the phase composition of the surface and bulk XRD measurements with subsequent Rietveld analysis were performed again after 100, 150, 200, and 250 days of maturing. In both samples, the content of chlorartinite increased with the aging time of the samples. However, as can be seen from Figure 6, the rate of chlorartinite formation was different for Phases 3 and 5. In the case of Phase 3, MOC almost disappeared after 250 days, while chlorartinite represented the dominant phase. The carbonation of Phase 5 was slower than Phase 3, likely due to the phase separation. After 150 days, the surface of the sample contained similar amounts of Phase 5 and chlorartinite. After 250 days, this sample contained 46% of Phase 5 and 54% of chlorartinite. The XRD patterns acquired on 250-day samples are shown in Figure 7 for both surfaces and bulk. It is obvious from Figure 7 that the carbonation took place mainly on the surface. Although the Phase 3 sample contained over 95 wt% chlorartinite, in the whole volume the content of chlorartinite did not exceed 15 wt%. Similar results were obtained for Phase 5, where the content of chlorartinite in the bulk was even lower. Figure 6 shows the surface phase composition of Phases 3 and 5 after 7, 50, 100, 150, 200, and 250 days of sample ageing. Figure 7 shows the phase composition of the surface and bulk (powder) of Phases 3 and 5 assessed for 250-day samples. SEM micrographs were acquired to study the microstructure of the 250-day MOC samples. The typical needle-like crystalline structure of Phases 3 and 5 were obtained in the bulk at high magnification (see Figure 8). The needles were 1-5 µm long and approximately 0.5 µm wide. On the other hand, the morphology of the surface was different. Due to carbonation (formation of chlorartinite), the surface contained a lower amount of needles in both samples. More needles (MOC phase) were detected in Phase 5 than Phase 3, which is in good agreement with the XRD results. The EDS chemical composition was also measured at three different locations in the sample (surface, center, and bottom). High purity of both samples was confirmed when only magnesium, oxygen, chlorine, and carbon were detected. The EDS data confirmed the highest carbon content on the surface of the tested samples. The carbon content in the center and on the bottom of the analyzed samples was significantly lower. Figure 8 shows the SEM micrographs of the surface and bulk (fracture surface) of the Phases 3 and 5 samples scanned after 250 days of materials ageing. The EDS data measured from the surface, center, and bottom of the samples are also provided.  Both MOC phases analyzed in this study exhibited fast formation of chlorartinite as a result of their reaction with carbon dioxide. The carbon footprint of MOC cement-based materials will be thus significantly and quickly reduced. Based on the great sorption activity of Phase 3 towards CO 2 , it is estimated that 1 kg of Phase 3 could theoretically incorporate 211 g of CO 2 , as formulated in Equation (3): Similarly, assuming carbonation of Phase 5 in accordance with Equation (4), where maximum calculated theoretical CO 2 uptake capacity of Phase 5 was 331 g/kg. However, these are theoretical assumptions only, requiring a high concentration of CO 2 in the surrounding air, and full-volume carbonation of MOC phases. On the other hand, increased global CO 2 emissions pose a threat to the Earth's atmospheric environments. It was reported that the annual CO 2 concentration at the Earth's surface in 2017 reached 405 ppm [44], which is approximately quadruple that of the early 1960s [45]. Therefore, any attempt at a reduction of the carbon footprint of construction materials is of particular importance from a sustainability point of view. The production of Portland cement, as a material that is competitive to MOC cement, represents 5-8% of global CO 2 emissions [46,47]. In this sense, alternative approaches to reducing CO 2 emissions associated with the manufacture of the binder phase in concrete and other construction materials are in demand [48]. Seo et al. [44] reported on the potential of CO 2 reduction materialized by carbonation curing. He calculated the CO 2 uptake capacity of Portland cement during an accelerated carbonation test as 12.3%. The computing was done based on the thermogravimetry (TG) and calculation of the difference between the weight loss at 500-720 • C of carbonated and noncarbonated cement paste samples. This was a similar level of CO 2 uptake as observed in a study published by Jang and Lee [41]. Nevertheless, this CO 2 uptake capacity corresponds with the 5% CO 2 concentration within the accelerated carbonation test. The normal level of CO 2 concentration in the air is 250-400 ppm [49]. In occupied spaces with good air exchange, the concentration of CO 2 varies in the 400-1000 ppm range. This means that the applied CO 2 concentration in accelerated carbonation tests was typically even 200 times higher than the normal CO 2 concentration in air. The significant reduction of CO 2 uptake by Portland cement paste can be therefore be anticipated under standard atmospheric conditions. Moreover, the sequestration of CO 2 as a result of the carbonation of Portland cement paste in normal environments is extremely slow [50], contrary to the fast formation of chlorartinite from MOC phases.
Based on the data provided and analysis conducted in this study, it can be concluded that MOC cement represents an alternative construction binder for the production of low carbon materials that can partially mitigate the atmospheric concentration of CO 2 , which is the main constituent of greenhouse gases related to global warming issues [51].

Conclusions
The issue of global warming has resulted in a climate crisis faced by people all over the world. Annually, billions of metric tons of Portland cement are produced around the world, which represents a serious environmental burden due to the release of huge amounts of CO 2 within the decomposition of limestone and burning of coal in cement production plants. As Portland cement and cement-based materials are the most widespread materials in the construction industry, there is a need to develop and discover alternative cement materials with similar or better functional properties than Portland cement, with a lower negative environmental impact.
In this work, MOC was studied as an alternative to Portland cement binder. Based on the conducted tests and analyses, the fast formation and hardening of both stable MOC phases was documented, which is an advantageous property and contrary to the significantly slower hardening of Portland cement-based materials. This could be beneficially exploited in repair applications, prefabrication, and the speeding up of construction processes. Specific attention was paid to the ability of MOC to capture CO 2 from the environment, making it eco-friendly, but also resulting in a more dense and mechanically resistant structure than was anticipated. Both MOC phases proved their fast mineral carbonation, whereas the maximum theoretical CO 2 uptake capacity was high and significantly greater than that of Portland cement paste. Nevertheless, one must consider the surface and bulk carbonation effects, relative humidity values, and CO 2 concentration available for carbonation reaction in this respect. MOC-based composites could find use in applications where a higher concentration of CO 2 in the environment is expected (e.g., in areas and construction sites exposed to higher traffic density, industrial air pollution, etc.). Here, they can be applied in industrial flooring and wall panels. In this case, the concentration of the prevailing component of greenhouse gases will be safely reduced by CO 2 uptake in newly formed carbonated MOC phases which will be more durable and resistant against mechanical loads. Based on the free carbonation capacity of MOC-based products, they could also serve as the passive moderators of indoor air climates with respect to the mitigation of excessive CO 2 concentrations (which can cause serious health problems).