Roles of Mineralogical Phases in Aqueous Carbonation of Steelmaking Slag

Mineralogical phases of steelmaking slags have significant influences on the carbonation of the slags. In this paper, the effects of temperature and reaction time on the conversion of calcium-related phases and the carbonation degree of a slag sample were studied. The experimental conditions were a liquid-to-solid ratio of 20 mL/g, a carbon dioxide flow rate of 1 L/min and a slag particle size of 38–75 μm. The results show that the optimum carbonation temperature and reaction time are 60  ̋C and 90 min, respectively, and calcite phase content is about 26.78% while the conversion rates of Ca3Al2O6, CaSiO3, Ca2SiO4 and free CaO are about 40%, 42.46%, 51% and 100%, respectively, and the carbon dioxide sequestration efficiency is about 170 g/kg slag.


Introduction
As known to us, the generation rate of steelmaking slag is about 10%-15% of crude steel production.In China, there are more than 100 million tons of steelmaking slags stored on the ground.Steelmaking slag has been accumulated for a long time without consideration.This not only wastes valuable resources, but also causes some threats.For example, heavy metal in slag can contaminate groundwater.Thus, the development of utilization technologies of steelmaking slag has garnered more attention recently.
The idea of slag carbonation comes from the mineral carbonation firstly proposed by Seifritz [1] and then developed by Lackner [2].CO 2 is hardly released from carbonation products after mineralizing because they are thermodynamically stable [2,3].Furthermore, steelmaking slag is rich in calcium and magnesium, and naturally, steelmaking slag can be a better CO 2 adsorption carrier.The utilization of slag has some advantages including lower cost and minimal environmental impacts through pH-adjustment [4,5].
The existing dominant slag carbonation process has three advantages.Firstly, greenhouse gas carbon dioxide can be sequestrated permanently.Secondly, the reaction process is helpful for improving the stability of slag for construction materials, because active mineralogical phases containing calcium stay stable in the form of calcium carbonate.Lastly, calcium carbonate precipitates in slag particles surface, and the coated surface after carbonation hinders the heavy metal leaching process.
In previous works, the effect of experimental parameters such as the ratio of liquid to solid (L/S ratio), temperature, particle size, stirring intensity, etc., on carbonation efficiency and the slag carbonation mechanism were analyzed when slag was dissolved in water [6,7].In order to improve the carbonation efficiency, conditions such as higher reaction pressure [8][9][10][11][12] and an acid leaching agent [13][14][15] are adopted.Previous researchers focused on the total carbonation degree of steelmaking slag in their research.The effects of mineralogical phases containing calcium on the carbonation degree were less referred to in literature.In order to analyze the CO 2 -binding mineralogical phases' activity with carbon dioxide when slag is applied for carbon dioxide sequestration, in this paper the effects of the main mineralogical phases containing calcium on the carbonation degree have been studied by quantitative X-ray diffraction (XRD) collaborating with X-ray fluorescence spectroscopy (XFR) analysis, and the effects of temperature and time on the conversion rate of mineralogical phases have been analyzed.

Sample Characterization
A steelmaking slag sample was taken from a Chinese steel plant (Baosteel, Shanghai, China).The steelmaking slag sample was ground in a ball mill.Chemical composition of the slag sample was analyzed and is presented in Table 1.It can be seen that calcium oxide content nearly counts for 55.3%, magnesium oxide content is less than 10%, silicon dioxide content of the steelmaking slag is more than 10%, and aluminum oxide is about 30%.Quantitative X-ray diffraction (XRD) test for steelmaking slag was conducted.XRD analysis in a Dron-3M diffractometer (Shimadzu Ltd., Tokyo, Japan) using Ni-filtered Cu-Kα radiation, powdered non-oriented preparations was made.Diffractograms were digitally registered and analyzed by Sirquant code [16] using full-profile Rietveld analysis [17].Specific surface area (SSA) was estimated with BET method at Quanta chrome instruments Autosorb (Automated Gas Sorption1-Win, Version 1.50, CostechMicroanalytical Ltd., Beijing, China), 0.5-1.5 g steelmaking slag with different particle sizes was applied for testing, Nitrogen gas was used for analyzing gas, and analyzing time was set for 200 min, outgas temperature was controlled at 350 ˝C, and liquid density was 8080 kg/m 3 .Scanning electron microscope (SEM) JSM-6301F (JEOL Ltd., Tokyo, Japan) was used for slag surface observations before and after carbonation process.

Experimental Setup
The carbonation degree of the steelmaking slag was investigated at room temperature and under normal atmosphere, and the experimental apparatus is displayed in Figure 1.The experimental apparatus includes measurement and control of temperature and pH measurement.The liquid-to-solid ratio was set up as 20 mL/g, and the sample weight was 5 g each time.
The reactor was sealed with preservation film for avoiding solution evaporation; the initial time is obtained when the temperature reached the target temperature.When experiments began, CO 2 gas at a predetermined flow rate was blown into the reactor.The experimental conditions are presented in Table 2.The temperature varied among 25 ˝C, 40 ˝C, 60 ˝C, and 90 ˝C, the reaction time changed among 30 min, 60 min, 90 min, 120 min and 180 min, and CO 2 flow rate was kept at 1 L/min.
The carbonation degree of the steelmaking slag was investigated at room temperature and under normal atmosphere, and the experimental apparatus is displayed in Figure 1.The experimental apparatus includes measurement and control of temperature and pH measurement.The liquid-to-solid ratio was set up as 20 mL/g, and the sample weight was 5 g each time.

Evaluation of Carbonation Degree
In order to evaluate carbonation degree of steelmaking slag and efficiency of absorbing CO 2 by the slag, HCT-1/2 differential thermal analyzer (Henven Scientific Instument Factory, Beijing, China) was adopted.In the test, the crucible weighed m 0 , the weight of the crucible plus the slag sample was m 1 , and then weight of the slag sample was m 1 minus m 0 .
The end temperature for the TGA test was 1000 ˝C, and heating rate was 10 ˝C/min; after the end temperature was achieved, the sample was held at that temperature for 3 min.Because HJ thermal analysis software can record temperature, thermo-gravimetry and differential thermal analysis curve synchronously, it is convenient to determine the thermal characteristics of the slag.
Slag weight loss can be obtained from the thermogravimetric analysis curve, which mainly include three parts, free water evaporation in the temperature range of 25-105 ˝C, organic carbon loss and magnesium carbonate decomposition in the temperature scope of 105-500 ˝C, between 500 ˝C and 1000 ˝C, calcium carbonate decomposition occurs.Carbonation degree can be evaluated with the Equations ( 1) and ( 2) referred in literature [9].
where Ca total represents the content of element calcium in slag, %, m 105 ˝C denotes the weight of slag when heated to 105 ˝C, mg, ∆m (500 ˝C-1000 ˝C) denotes steelmaking slag weight loss when heated from 500 ˝C to 1000 ˝C, mg, η CD denotes carbonation degree of slag, %.
Specific surface areas of the raw slag sample change from 0.51 m 2 /g to 1.57 m 2 /g, depending on the particle sizes.The specific areas of carbonated slag samples vary from 7.56 m 2 /g to 12.78 m 2 /g with the carbonation time increasing at 25 ˝C.The results can be explained by the slag fragmentation model [18]; in other words, during the carbonation reaction process, carbonate calcium produced in slag surface area is peeling from the matrix by stirring, and causing increases of the initial slag surface area.SEM images of the steelmaking slag carbonated at different temperatures and reaction times are shown in Figure 2. The slag sample contains 55.3% CaO, of which 10.6% is free CaO, 29.6% is a Ca-aluminate compound (Ca3Al2O6) and 15.1% is Ca-silicates (5.7% CaSiO3 and 9.4% Ca2SiO4).
Specific surface areas of the raw slag sample change from 0.51 m 2 /g to 1.57 m 2 /g, depending on the particle sizes.The specific areas of carbonated slag samples vary from 7.56 m 2 /g to 12.78 m 2 /g with the carbonation time increasing at 25 °C.The results can be explained by the slag fragmentation model [18]; in other words, during the carbonation reaction process, carbonate calcium produced in slag surface area is peeling from the matrix by stirring, and causing increases of the initial slag surface area.SEM images of the steelmaking slag carbonated at different temperatures and reaction times are shown in Figure 2.

Effect of Carbonation Time on Mineralogical Phases Containing Calcium
The effect of the carbonation time on the content of the main mineralogical phases in the slag sample is shown in Figure 3. Clearly, the peak intensity of the calcite phase increases when the carbonation time increases from 30 min to 60 min, but after that, if extending carbonation time, the peak intensity of the calcite phase decreases.The peak intensity of the calcium aluminate phase decreases with the carbonation time increasing.This is because the calcium aluminate absorbs carbon dioxide dissolved in water, and the Ca-silicate phase, such as wollastonite, also reacts with carbon dioxide.
Quantitative XRD analysis of the slag has a better correlation with chemical test results [16].Figure 4 describes the effect of reaction time on the carbonation degree contribution rate for

Effect of Carbonation Time on Mineralogical Phases Containing Calcium
The effect of the carbonation time on the content of the main mineralogical phases in the slag sample is shown in Figure 3. Clearly, the peak intensity of the calcite phase increases when the carbonation time increases from 30 min to 60 min, but after that, if extending carbonation time, the peak intensity of the calcite phase decreases.The peak intensity of the calcium aluminate phase decreases with the carbonation time increasing.This is because the calcium aluminate absorbs carbon dioxide dissolved in water, and the Ca-silicate phase, such as wollastonite, also reacts with carbon dioxide.
Metals 2016, 6, 117 5 of 8 increases to 26.78% when expanding to an additional 90 min, and the optimum reaction time can take place at 90 min.The carbonation time is less than 90 min, and the carbonation reaction cannot be allowed to complete.If the carbonation time is more than 90 min, the reaction kinetic condition is worse because the carbonate calcium production is coated on the surface of the slag, which hinders the carbonation degree from increasing after 90 min.From Figure 4, one can see that the content of Ca3Al2O6 in the slag decreases from 29.6% to 17.79% when the slag is carbonated for 30 min.When the carbonation time increases from 90 min to 180 min, the content decreases from 12.51% to 11.49%.Clearly, almost 40% of the Ca3Al2O6 can be carbonated in the first 30 min, and 8.15% of the Ca3Al2O6 will be carbonated in the next 60 min.As for the calcium-silicate phase, the CaSiO3 content sharply decreases from 5.7% to 3.28% when the slag is carbonated for 90 min, and further decreases to 3.14% after an additional 90 min; the CaSiO3 conversion rate is 42.46% in the first 90 min, and only 4.3% in the additional 90 min carbonation time.Ca2SiO4 has similar carbonation behavior to CaSiO3.In the first 30 min of carbonation time, the content of Ca2SiO4 in the slag decreases from 9.4% to 4.6%; when the carbonation time increases from 90 min to 180 min, the phase content decreases from 2.45% to 2.02%.In the first 30 min of carbonation, 51.1% of the Ca2SiO4 is carbonated, and only 17.5% of the Ca2SiO4 is carbonated in the Quantitative XRD analysis of the slag has a better correlation with chemical test results [16].Figure 4 describes the effect of reaction time on the carbonation degree contribution rate for calcium-related phases.It is apparent that the calcite content in slag increases from 15.34% to 25.47% when the carbonation time increases from 30 min to 90 min.The calcite content in slag increases to 26.78% when expanding to an additional 90 min, and the optimum reaction time can take place at 90 min.The carbonation time is less than 90 min, and the carbonation reaction cannot be allowed to complete.If the carbonation time is more than 90 min, the reaction kinetic condition is worse because the carbonate calcium production is coated on the surface of the slag, which hinders the carbonation degree from increasing after 90 min.increases to 26.78% when expanding to an additional 90 min, and the optimum reaction time can take place at 90 min.The carbonation time is less than 90 min, and the carbonation reaction cannot be allowed to complete.If the carbonation time is more than 90 min, the reaction kinetic condition is worse because the carbonate calcium production is coated on the surface of the slag, which hinders the carbonation degree from increasing after 90 min.From Figure 4, one can see that the content of Ca3Al2O6 in the slag decreases from 29.6% to 17.79% when the slag is carbonated for 30 min.When the carbonation time increases from 90 min to 180 min, the content decreases from 12.51% to 11.49%.Clearly, almost 40% of the Ca3Al2O6 can be carbonated in the first 30 min, and 8.15% of the Ca3Al2O6 will be carbonated in the next 60 min.As From Figure 4, one can see that the content of Ca 3 Al 2 O 6 in the slag decreases from 29.6% to 17.79% when the slag is carbonated for 30 min.When the carbonation time increases from 90 min to 180 min, the content decreases from 12.51% to 11.49%.Clearly, almost 40% of the Ca 3 Al 2 O 6 can be carbonated in the first 30 min, and 8.15% of the Ca 3 Al 2 O 6 will be carbonated in the next 60 min.As for the calcium-silicate phase, the CaSiO 3 content sharply decreases from 5.7% to 3.28% when the slag is carbonated for 90 min, and further decreases to 3.14% after an additional 90 min; the CaSiO 3 conversion rate is 42.46% in the first 90 min, and only 4.3% in the additional 90 min carbonation time.Ca 2 SiO 4 has similar carbonation behavior to CaSiO 3 .In the first 30 min of carbonation time, the content of Ca 2 SiO 4 in the slag decreases from 9.4% to 4.6%; when the carbonation time increases from 90 min to 180 min, the phase content decreases from 2.45% to 2.02%.In the first 30 min of carbonation, 51.1% of the Ca 2 SiO 4 is carbonated, and only 17.5% of the Ca 2 SiO 4 is carbonated in the next 60 min of carbonation.For free lime in the raw slag sample, after 60 min of carbonation all initial 10.6% of free lime is carbonated.

Effect of Carbonation Temperature on Mineralogical Phases Containing Calcium
In order to analyze the effect of carbonation temperature on the carbonation degree and slag phase conversion, TGA-DSC (Netzsch, Gerätebau, Germany) tests were conducted on the slag carbonated, and TGA analysis results at different reaction temperatures are shown in Figure 5.
Metals 2016, 6, 117 6 of 8 next 60 min of carbonation.For free lime in the raw slag sample, after 60 min of carbonation all initial 10.6% of free lime is carbonated.

Effect of Carbonation Temperature on Mineralogical Phases Containing Calcium
In order to analyze the effect of carbonation temperature on the carbonation degree and slag phase conversion, TGA-DSC (Netzsch, Gerätebau, Germany) tests were conducted on the slag carbonated, and TGA analysis results at different reaction temperatures are shown in Figure 5. First, 7.95 g of steelmaking slag after 90 min of carbonation at 25 °C was analyzed.The weight loss of the sample is 10.1% when the heating temperature increases from 500 °C to 1000 °C; the carbonation degree calculated by Equation ( 1) is then 18.5% at 25 °C.The weight loss at 40 °C is 14.5%, and the carbonation degree is 27.9%, and the weight loss at 60 °C is 17.2%, and the carbonation degree is 33.4%.The weight loss at 90 °C is 15.2%, and carbonation degree is 29.5%.It can be concluded that the optimum temperature for maximum carbonation efficiency is about 60 °C, due to the contradicting effects of temperature on the carbonation kinetics and solubility of calcium-related minerals in the water.
Obviously, at ambient reaction temperature and with 90 min of reaction time, the contents of calcite, Ca3Al2O6, CaSiO3 and Ca2SiO4 in the slag are 16.8%, 18.9%, 4.3% and 6.3%, respectively in Figure 6.Calcite in the slag increases from 22.68% to 25.47% when the reaction temperature increases from 40 °C to 60 °C.The calcite content increases to 25.79% when the carbonation temperature rises to 90 °C.The Ca3Al2O6 content in slag decreases from 14.8% to 12.49% when the carbonation temperature increases from 40 °C to 60 °C, but it increases to 13.75% when the reaction temperature First, 7.95 g of steelmaking slag after 90 min of carbonation at 25 ˝C was analyzed.The weight loss of the sample is 10.1% when the heating temperature increases from 500 ˝C to 1000 ˝C; the carbonation degree calculated by Equation ( 1) is then 18.5% at 25 ˝C.The weight loss at 40 ˝C is 14.5%, and the carbonation degree is 27.9%, and the weight loss at 60 ˝C is 17.2%, and the carbonation degree is 33.4%.The weight loss at 90 ˝C is 15.2%, and carbonation degree is 29.5%.It can be concluded that the optimum temperature for maximum carbonation efficiency is about 60 ˝C, due to the contradicting effects of temperature on the carbonation kinetics and solubility of calcium-related minerals in the water.
Obviously, at ambient reaction temperature and with 90 min of reaction time, the contents of calcite, Ca 3 Al 2 O 6 , CaSiO 3 and Ca 2 SiO 4 in the slag are 16.8%, 18.9%, 4.3% and 6.3%, respectively in Figure 6.Calcite in the slag increases from 22.68% to 25.47% when the reaction temperature increases from 40 ˝C to 60 ˝C.The calcite content increases to 25.79% when the carbonation temperature rises to 90 ˝C.The Ca 3 Al 2 O 6 content in slag decreases from 14.8% to 12.49% when the carbonation temperature Metals 2016, 6, 117 7 of 8 increases from 40 ˝C to 60 ˝C, but it increases to 13.75% when the reaction temperature rises to 90 ˝C.This is because there is a reversal relationship between CO 2 solubility in water and the reaction temperature.Ca 3 Al 2 O 6 has a lower conversion rate with a lower carbon dioxide solubility at a higher carbonation temperature.The contents of CaSiO 3 and Ca 2 SiO 4 decrease from 2.9% to 2.72% and 5.1% to 3.48%, respectively.When the carbonation temperature increases from 40 ˝C to 60 ˝C, the contents of the two phases increase to 3.11% and 5.18%, respectively.

Conclusions
The carbonation process was experimentally tested on a steelmaking slag sample under the conditions of a ratio of liquid to solid at 20 mL/g, a 5 g sample weight and particle sizes from 38 μm to 75 μm.The optimum carbonation temperature and reaction time are 60 °C and 90 min, respectively.The maximum carbonation degree is about 26.78%.The conversion rates of Ca3Al2O6 phase, CaSiO3, Ca2SiO4 and free CaO are about 40%, 42.46%, 51% and 100%, respectively.

Conclusions
The carbonation process was experimentally tested on a steelmaking slag sample under the of a ratio liquid to solid at 20 mL/g, a 5 g sample weight and particle sizes from 38 µm to 75 µm.The optimum carbonation temperature and reaction time are 60 ˝C and 90 min, respectively.The maximum carbonation degree is about 26.78%.The conversion rates of Ca 3 Al 2 O 6 phase, CaSiO 3 , Ca 2 SiO 4 and free CaO are about 40%, 42.46%, 51% and 100%, respectively.

Figure 3 .
Figure 3.Effect of carbonation time on mineralogy phases containing calcium.

Figure 4 .
Figure 4. Effect of carbonation time on mineralogical phase content of slag carbonated at 60 °C.

Figure 3 .
Figure 3.Effect of carbonation time on mineralogy phases containing calcium.

Figure 3 .
Figure 3.Effect of carbonation time on mineralogy phases containing calcium.

Figure 4 .
Figure 4. Effect of carbonation time on mineralogical phase content of slag carbonated at 60 °C.

Figure 4 .
Figure 4. Effect of carbonation time on mineralogical phase content of slag carbonated at 60 ˝C.

Figure 5 .
Figure 5.Effect of temperature on carbonation degree for steelmaking slag carbonated for 90 min.

Figure 5 .
Figure 5.Effect of temperature on carbonation degree for steelmaking slag carbonated for 90 min.

Figure 6 .
Figure 6.Effect of temperature on content of mineralogical phases of slag carbonated for 90 min.

Figure 6 .
Figure 6.Effect of temperature on content of mineralogical phases of slag carbonated for 90 min.