Effect of the Silicate Structure on Calcium Elution Behaviors of Calcium-silicate Based Mineral Phases in Aqueous Solution

Abstract

In this study, the effect of the silicate structure of calcium-silicate based mineral phases on their Ca elution behaviors into water was investigated. The Ca elution behaviors of Ca-silicate based mineral phases with different skeleton silicate structures in ion-exchanged water were analyzed using the powder leaching test. The elution amount of Ca was in the order alite (Ca₃SiO₅) > belite (γ-Ca₂SiO₄) > rankinite (Ca₃Si₂O₇) > pseudowollastonite (α-CaSiO₃) > wollastonite (β-CaSiO₃) ≈ cuspidine (Ca₄Si₂O₇F₂) > diopside (CaMgSi₂O₆) > hedenbergite (CaFeSi₂O₆) > tremolite (Ca₂Mg₅Si₈O₂₂(OH)₂) > anorthite (CaAl₂Si₂O₈). This suggests that the elution amount of Ca decreased with the skeleton silicate structure became more complicated.

1. Introduction

A large proportion of steelmaking slag has been widely applied in road base course, civil and coastal construction, soil improvement and fertilizers.¹⁾ However, environmental problems have been caused by the application of slag.^2,3) The excess alkaline elution from the steelmaking slag into water has become an important issue to be solved.

To promote the environmentally-friendly application of steelmaking slag, the leaching properties of slag have been studied.^4,5) Tamaki et al.⁶⁾ investigated the alkali elution rate from steelmaking slag and suggested that the alkali elution rate decreased when the weight ratio of dredged soil in the slag increased. In addition, free-CaO accelerated the alkali elution rate when the mass concentration of free-CaO was under 2%. Gao et al.⁷⁾ reported that the solid solution phase 2CaO·SiO₂-3CaO·P₂O₅ also contributed a large amount to the Ca elution among the phases in the steelmaking slag, while little Ca elution was detected from the phases in the CaO–SiO₂–FeO_x system. Tauchi et al.⁸⁾ showed that the complete melting process by slag modification to decrease the basicity suppressed the Ca elution in aqueous solution for dephosphorization slag. Strandkvist et al.⁹⁾ clarified that pure mineral phases like Ca₃MgSi₂O₈, β-Ca₂SiO₄, γ-Ca₂SiO₄, Ca₂MgSi₂O₇ and α-CaSiO₃ were completely dissolved when the pH value of solution was controlled at 7, while the mineral phase CaMgSiO₄ showed better durability than the others. However, research on the Ca elution behavior of a single common mineral phase of steelmaking slag are limited. Because steelmaking slag consists of various silicate minerals and/or glass phases, knowing and controlling the Ca leaching property of each phase may support a perspective to suppress the excess Ca elution of steelmaking slag.

Several factors affect the Ca elution behaviors of mineral phases. Engstrom et al.¹⁰⁾ reported that Ca₃MgSi₂O₈, γ-Ca₂SiO₄, Ca₂MgSi₂O₇, Ca₁₂Al₁₄O₃₃, Ca₂Al₂SiO₇ and Ca₃Al₂O₆ showed a lower dissolution rate at high pH values. As for the temperature dependence of the alkali elution behaviors, Casey et al.¹¹⁾ reported that the dissolution rates of silicate minerals became significantly sensitive to the pH value at elevated temperatures. Moreover, the CaO/SiO₂ mass ratio of minerals also affects the alkali elution behavior of silicate mineral phases. Zhang et al.³⁾ suggested that steelmaking slag with a CaO/SiO₂ mass ratio of 2.0 showed about double the Ca elution amount than that with a CaO/SiO₂ mass ratio of 1.0. However, there remains much to learn about the silicate structural effect on the Ca elution behavior. A few studies described the silicate structural effect on the dissolution behavior of various mineral phases. Casey et al.¹²⁾ reported the difference of dissolution behavior between α-CaSiO₃ and β-CaSiO₃ in acidic solution, where their skeleton silicate structures were a three-membered ring structure and a single chain structure, respectively. Furthermore, α-CaSiO₃ showed a slightly higher dissolution rate than that of β-CaSiO₃. In addition, Westrich et al.¹³⁾ observed that olivine minerals with isolated [SiO₄]⁴⁻ tetrahedral structures showed a more intense elution behavior of metal cations than that of silicates with single chain structures. Thus, an understanding of the silicate structural effect on the alkali elution behavior is essential.

The objective of this research was to investigate the effect of different silicate skeleton structures on the Ca elution behaviors of silicate mineral phases into water. Mineral phases with isolated [SiO₄]⁴⁻ tetrahedral structures, dimer [SiO₄]⁴⁻ tetrahedral structures, three-membered ring structures, single and double chain structures, and three-dimensional framework silicate structures were synthesized, and their Ca elution behaviors were analyzed using the leaching test. Moreover, their Ca elution behaviors were also evaluated in terms of the corrected basicity and degree of polymerization of the mineral phases.

2. Experimental

2.1. Target Calcium-silicate Based Mineral Phases

The basic unit of a silicate structure is the [SiO₄]⁴⁻ tetrahedron, and the conventional classification is based on geometrical forms created by the linkage of [SiO₄]⁴⁻ tetrahedrons. These geometrical forms include isolated tetrahedrons, dimer tetrahedrons, silicate rings, silicate chains, silicate sheets and three-dimensional frameworks of tetrahedrons.¹⁴⁾ The Al ion sometimes replaces the Si ion to exist as an [AlO₄]⁵⁻ tetrahedral skeleton unit, while most of the metal ions, Ca²⁺, Mg²⁺ and Fe²⁺, exist in the gaps between the silicate skeleton structures.

Table 1 shows the target mineral phases in this study. A series of calcium silicate based mineral phases, including six kinds of silicate structures, were analyzed. Quasi-binary CaO–SiO₂ phases of alite (Ca₃SiO₅), belite (γ-Ca₂SiO₄), rankinite (Ca₃Si₂O₇), pseudowollastonite (α-CaSiO₃) and wollastonite (β-CaSiO₃), and quasi-ternary phases of the fluorine-bearing phase cuspidine (Ca₄Si₂O₇F₂), magnesium-bearing phases diopside (CaMgSi₂O₆) and tremolite (Ca₂Mg₅Si₈O₂₂(OH)₂), iron-bearing phase hedenbergite (CaFeSi₂O₆) and aluminum-bearing phase anorthite (CaAl₂Si₂O₈) were subjected to the experiments. From the view of silicate skeleton structure, Ca₃SiO₅ and γ-Ca₂SiO₄ demonstrated an isolated [SiO₄]⁴⁻ tetrahedron structure, Ca₃Si₂O₇ and Ca₄Si₂O₇F₂ showed a dimer structure with two [SiO₄]⁴⁻ tetrahedrons linked together. α-CaSiO₃ showed a three-membered ring structure. β-CaSiO₃ was composed of a single chain structure with a three-tetrahedron structural repeating unit, CaMgSi₂O₆ and CaFeSi₂O₆ were composed of a single chain structure with a two-tetrahedron structural repeating unit. In addition, the Ca₂Mg₅Si₈O₂₂(OH)₂ showed a double chain structure. CaAl₂Si₂O₈ showed the most complicated structure, a three-dimensional framework structure with all oxygen atoms shared between two [SiO₄]⁴⁻ tetrahedrons.

Table 1. Target calcium silicate based mineral phases. (Online version in color.)

*1: Since there are quasi-ternary mineral phases, the corrected basicity in this research was defined as ${\displaystyle \frac{M_{(\text{CaO})}+M_{(\text{MgO})}}{M_{({\text{SiO}}_2)}+M_{({\text{Al}}_2{\text{O}}_3)}}}$, where M(i) is the mass percent of i component.

*2: The polymerization degree of skeleton structure was defined as the ratio of the sum of the non-bridging oxygen (NBO) and the free oxygen (FO) to one tetrahedron, (NBO + FO)/T (where T represents tetrahedrally coordinated cations Si or Al, and NBO is an oxygen bonded to only one T atom).¹⁹⁾ The value of (NBO + FO)/T declines with a rise in the polymerization degree.

2.2. Synthesis and Phase Identification

Most of mineral phases were synthesized from high purity reagents by high temperature sintering according to the phase diagrams. However, because CaFeSi₂O₆ and Ca₂Mg₅Si₈O₂₂(OH)₂ were difficult to be synthesized under one atmosphere, the natural minerals were subjected to the experiments.

Table 2 shows the synthetic conditions of the minerals. For starting materials, lime powder was obtained by firing the analytical grade reagent CaCO₃ powder (99.5 mass%, Wako Pure Chemical Industries, Ltd.) at 1000°C for 3 hours. Meanwhile, guaranteed reagents SiO₂ (99.9 mass%, Wako Pure Chemical Industries, Ltd.), MgO (99.0 mass%, Sigma-Aldrich Co. LLC), Al₂O₃ (99.5 mass%, Sigma-Aldrich Co. LLC) and CaF₂ (99.9 mass%, Wako Pure Chemical Industries, Ltd.) were used as starting materials to synthesize calcium silicate based mineral phases by mixing them to the specific composition ratios. In addition, CaFeSi₂O₆ and Ca₂Mg₅Si₈O₂₂(OH)₂ were obtained from Kamioka Mine (Hida, Gifu Prefecture) and Fort Dauphin (Tulear, Madagasucar), respectively. Generally, starting powder reagents were blended homogeneously in an agate mortar. The mixed powder sample was then pressed into tablet form (φ8× h6 mm) and the tablets were sintered in a platinum crucible at a fixed temperature. After the synthesis, the phase identification was conducted by X-ray diffraction (Rigaku RINT2200/PC) with Cu-Kα radiation. Because the amount of impurities in each case was very low, it was suggested that the impurities would have no considerable influence when analyzing the leaching property of the mineral phases.

Table 2. Synthetic conditions of target mineral phases.

Mineral Phase	Temperature (°C)	Crucible	Atmosphere	Time (h)
Ca3SiO5	1600	Pt	Air	18*1
γ-Ca2SiO4	1500	Pt	Air	48
Ca3Si2O7	1400	Pt	Air	60
Ca4Si2O7F2	1350	Pt	Ar*2	2
α-CaSiO3	1400	Pt	Air	72
β-CaSiO3*3	1000	Pt	Air	48
CaMgSi2O6	1325	Pt	Air	10
CaFeSi2O6	Impurity*4: Small quantity of Mg and Mn
Ca2Mg5Si8O22(OH)2	Impurity*4: CaMgSi2O6 < 2.4 mass%, SiO2 < 1.0 mass%
CaAl2Si2O8	1500	Pt	Air	48

*1:  Samples were sintered for 6 hours for each treatment and the process was repeat 3 times.

*2:  In order to prevent the loss of F, the tablet samples were hermetically contained in a platinum capsule and the experiment was conducted under argon gas atmosphere.

*3:  Samples were melted and cooled rapidly to make glass phase, and then the glass phase was put into furnace again and kept 48 h at 1000°C.

*4:  XRD and Electron Probe Micro Analyzer (EPMA) were conducted to confirm the impurities in natural phases of CaFeSi₂O₆ and Ca₂Mg₅Si₈O₂₂(OH)₂, and the percentage area modal analysis was measured to estimate the content of impurities.

2.3. Leaching Test

The elution behavior of each element from the mineral phases were analyzed using the powder sample leaching test,⁷⁾ which was confirmed to be comparable with the standard leaching test defined by Notification No. 46 of the Ministry of Environment of Japan.

The leaching test was conducted in a thermostatic bath (25°C) without controlling the atmosphere. Each silicate mineral sample was finely ground in an agate mortar. For the leaching test, 1 g of mineral powder sample with a particle size less than 53 μm was prepared. Next, 400 mL of deionized water was poured into a high-density polyethylene container. A Teflon stirring rod was inserted in the deionized water to homogenize the solution at 170 rpm. Finally, a pH meter was prepared to detect the pH of the leaching solution. Calibration of the pH meter was conducted with standard buffer solutions of pH = 4.01, 6.86 and 9.18.

To start the leaching test, the polyethylene container was held in a thermostatic bath. The powder sample was then put into deionized water, and the pH value of the leaching solution was recorded continuously. About 5 mL of the solution was collected at certain time intervals. The collected solutions were filtered through a syringe filter (Membrane Solutions SFNY013045 0.45 μm). To ensure that there was no precipitate in the sample solutions, a drop of HCl (6 mol/L) was added to each one. After running the leaching test for 60 minutes, the undissolved powder sample was collected by a suction device. The leaching residue was dried at 60°C for 24 hours. The concentrations of the elements dissolved in the filtered solutions, except for fluorine, were measured by ICP-AES (Optima 3300XL, Perkin Elmer Co., Ltd.). The concentration of fluorine was detected by ion chromatography (IC-2010, Tosoh Co., Ltd.). In addition, ²⁹Si magic-angle-spinning (MAS) NMR spectroscopy analysis was conducted on the Ca₃Si₂O₇ and Ca₄Si₂O₇F₂ to observe their structural change before and after the leaching tests. The samples were packed in zirconia rotors, which were spun at the spinning speed of 7 kHz. ²⁹Si MAS NMR spectra of the samples were recorded using a single pulse acquisition with a JEOL ECA 300 spectrometer (magnetic field: 7 T, flip angle: 30 degree, recycle delay: 40 s). ²⁹Si chemical shift is given in parts per million (ppm) from an external reference material: tetramethylsilane (TMS).

3. Experimental Results

3.1. The pH Value Variation and Elution Behavior of Elements from CaO–SiO₂ Quasi-binary System

A comparison of the time-dependent pH value of the quasi-binary mineral phases is shown in Fig. 1.

Fig. 1.

Time dependence of pH value for the mineral phases in CaO–SiO₂ quasi-binary system.

The pH value of all solutions greatly increased in first 5 min then became stable after about 10 min. Furthermore, Ca₃SiO₅ and γ-Ca₂SiO₄ with an isolated [SiO₄]⁴⁻ tetrahedron silicate structure and Ca₃Si₂O₇ with a dimer structure showed high pH values that reached to 12.2, 11.6 and 11.4 after 1 h, respectively. In contrast, the α-CaSiO₃ and β-CaSiO₃ with a ring structure and a single chain structure showed lower pH values of 10.3 and 9.82, respectively, after 1 h. This suggests that the maximum pH value was affected by the silicate structure.

The elution amounts of each element after 1 h are summarized in Table 3. In the binary system, a mineral phase that showed a high pH value in the leaching test also had a high Ca elution amount. In addition, the Ca elution amount of Ca₃SiO₅ with an isolated [SiO₄]⁴⁻ tetrahedron structure was the highest among all of the binary mineral phases, while the α-CaSiO₃ and β-CaSiO₃ with ring and single chain silicate structures showed much lower values. It was confirmed that mineral phases with more complicated skeleton silicate structures showed lower Ca elution amounts.

Table 3. The pH values of leaching solutions and elution amounts of each element from the all the target mineral phases after 1 h.

Mineral Phase	pH Value	Elution Amount (mg/L)
		Ca	Si	Mg	Fe	Al	F
Ca3SiO5	12.2	185	35.0	−	−	−	−
γ-Ca2SiO4	11.6	128	38.1	−	−	−	−
Ca3Si2O7	11.4	92.7	34.5	−	−	−	−
α-CaSiO3	10.3	19.2	11.1	−	−	−	−
β-CaSiO3	9.82	14.9	10.4	−	−	−	−
Ca4Si2O7F2	10.3	14.1	4.29	−	−	−	3.70
CaMgSi2O6	8.83	9.21	6.84	2.95	−	−	−
CaFeSi2O6	6.61	6.11	1.55	−	0.13	−	−
Ca2Mg5Si8O22(OH)2	7.11	4.61	1.48	1.38	−	−	−
CaAl2Si2O8	6.59	1.20	0.10	−	−	0.03	−

3.2. pH Value Variation and Elution Behavior of Elements from a Quasi-ternary System

The comparison of the time-dependent pH value among the mineral phases in quasi-ternary system is shown in Fig. 2.

Fig. 2.

Time dependence of pH value for the mineral phases in quasi-ternary system.

All maximum pH values were lower than those obtained from the binary minerals except for Ca₄Si₂O₇F₂. Among the quasi-ternary mineral phases, CaAl₂Si₂O₈ with the three-dimensional framework silicate structure showed the lowest pH value. Regarding the silicate structures, Ca₂Mg₅Si₈O₂₂(OH)₂ with a double chain structure and CaAl₂Si₂O₈ with a three-dimensional framework structure showed relatively lower pH values of 7.11 and 6.59 after 1 h, respectively, implying that a complicated silicate structure leads to a lower pH value.

The pH values of leaching solutions and elution amounts of each element from the mineral phases in the quasi-ternary system after 1 h are also summarized in Table 3. All the constituent elements were eluted into the leaching solution. Regarding the silicate structures, CaAl₂Si₂O₈ showed the lowest Ca elution amount, which was similar to the trend with the pH value. It was thus confirmed that the mineral phases in the quasi-ternary system that had more complicated skeleton silicate structures also showed lower Ca elution amounts.

4. Discussion

4.1. Relationship between pH Value and the Ca Elution Amount

For all target mineral phases, compared with other alkali elements, Ca showed the highest elution amount after 1 h. Thus, the elution amount of Ca became an effective indicator for analyzing the alkali dissolution of a mineral phase. The relationship between pH value and the logarithm of the Ca elution amount after 1 h is described in Fig. 3 with the skeleton model of silicate.

Fig. 3.

Relationship between the logarithm of Ca elution amount and the pH value after 1 h in the leaching test. (Online version in color.)

The elution amount of Ca from Ca₃SiO₅ was the highest among all the mineral phases after 1 h. The obvious trend is that mineral phases that exhibited a higher pH value showed a higher Ca elution amount. It was thus suggested that the increase in pH value was mainly caused by the Ca elution, which is the same as the trend reported by Tauchi et al.⁸⁾ The Ca elution behaviors of Ca₃Si₂O₇, α-CaSiO₃, β-CaSiO₃, CaMgSi₂O₆ and CaAl₂Si₂O₈ before the pH reached its maximum value were discussed by Zhu et al.,¹⁵⁾ which showed good agreement with the variation trend in this study. Regarding the silicate structure, the mineral phases with loose skeleton silicate structures, namely, Ca₃SiO₅, γ-Ca₂SiO₄ and Ca₃Si₂O₇ showed higher Ca elution amounts over 90 mg/L. In contrast, the mineral phases with complicated skeleton silicate structures, including those with ring and chain structures showed low values of less than 25 mg/L. CaAl₂Si₂O₈ with a three-dimensional framework structure showed the lowest pH value and lowest Ca elution amount. Some Al ions in CaAl₂Si₂O₈ contribute to the formation of the framework structure with [SiO₄]⁴⁻ tetrahedrons, which was considered a reason for the low Ca elution amount.¹⁶⁾ Therefore, the skeleton silicate structure affected the Ca elution amount, and the Ca elution amount roughly corresponded to the polymerization degree of the silicate structure of the mineral phase.

4.2. Effect of Corrected Basicity and the Polymerization Degree of Silicate Structures on the Ca Elution Amount

There is a close relationship between the polymerization degree of the skeleton structure and the basicity.¹⁷⁾ Among various kinds of the expressions of basicity, the corrected basicity was used in this study to include the effect of MgO, FeO and Al₂O₃ in the quasi-ternary mineral phases. Here, the corrected basicity was proposed by Pretorius et al.¹⁸⁾ to express the behavior of CaO–SiO₂–FeO–MgO–Al₂O₃ slag system and is written as the following equation.   

$$Corrected\mathrm{\hspace{0.17em} } basicity=\frac{M_{(\text{CaO})}+M_{(\text{MgO})}}{M_{({\text{SiO}}_2)}+M_{({\text{Al}}_2{\text{O}}_3)}}.$$(1)

where M_(i) is the mass percent of the i component. The corrected basicity of each target mineral phase is summarized in Table 1. The relationship between the Ca elution amount and the corrected basicity of the mineral phase is shown in Fig. 4.

Fig. 4.

Relationship between the Ca elution amount and the corrected basicity of the target mineral phase after 1 hour in the leaching test. (Online version in color.)

Regarding the silicate skeleton structure, the compact structure generally showed low corrected basicity, and the mineral phase with a lower corrected basicity exhibited lower Ca elution amounts. The same trend was also confirmed when the normal basicity or the optical basicity was used instead of the corrected basicity. Thus, one effective way to reduce the Ca elution amount is to decrease the corrected basicity of the mineral phase.

To evaluate the effect of the silicate structure on the Ca elution behavior more directly and quantitively, the degree of polymerization was introduced, which is characterized by the ratio of the sum of the nonbridging oxygen (NBO) and the free oxygen (FO) to one tetrahedron, (NBO + FO)/T (where T represents tetrahedrally coordinated cations Si or Al, and NBO is an oxygen bonded to only one T atom).¹⁹⁾ The value of (NBO + FO)/T declines with a rise in the polymerization degree.

The polymerization degree of the skeleton structure of each target mineral phase is shown in Table 1. The relationship between the Ca elution amount and the polymerization degree of a skeleton structure is shown in Fig. 5(a). Similar to the variation trend of corrected basicity, the Ca elution amount decreased when the (NBO + FO)/T decreased. In addition, Ca₃SiO₅ and γ-Ca₂SiO₄ with an isolated [SiO₄]⁴⁻ tetrahedron structure, as well as Ca₃Si₂O₇ with a dimer silicate structure, showed much higher Ca elution amounts than that of other mineral phases. Because the Ca amounts of the target mineral phases in 1 g of the sample are different from each other, the relationship between the Ca elution ratio and the polymerization degree of a skeleton structure was also considered. The result is shown in Fig. 5(b). Here, the Ca elution ratio is defined by the Ca amount eluted in the solution divided by that existing in 1 g of the sample. The Ca elution ratios also decreased when the (NBO + FO)/T decreased, where the variation trend is almost the same with that in Fig. 5(a). However, Ca₃Si₂O₇ and Ca₄Si₂O₇F₂, with the same value of the (NBO + FO)/T, showed a significant difference in both Ca elution amount and elution ratio after 1 h. Thus, ²⁹Si NMR spectroscopy analysis was conducted to observe the changes before and after the leaching test.

Fig. 5(a).

Relationship between the Ca elution amount and the polymerization degree of skeleton structure. (Online version in color.)

Fig. 5(b).

Relationship between the Ca elution ratio and the polymerization degree of skeleton structure. (Online version in color.)

4.3. ²⁹Si NMR Spectroscopy Analysis of Mineral Samples and Leaching Residues

Figure 6 shows the ²⁹Si MAS NMR spectra comparison between the powder sample and the leaching residue for (a) Ca₃Si₂O₇ and (b) Ca₄Si₂O₇F₂. Five structural units related to the [SiO₄]⁴⁻ tetrahedron were denoted as Qⁱ (i = 0, 1, 2, 3, 4), where i represents the number of bridging oxygens around one Si atom.²⁰⁾ Magi et al.²¹⁾ reported that the ²⁹Si NMR isotropic chemical shifts at −74.5 and −76.0 ppm were identified as the Q¹ group in the synthetic Ca₃Si₂O₇. Hansen et al.²²⁾ reported that the isotropic chemical shifts at −79.9 ppm were identified as the Q¹ group in the synthetic Ca₄Si₂O₇F₂. In Fig. 6, the ²⁹Si MAS NMR spectrum of synthetic Ca₃Si₂O₇ and Ca₄Si₂O₇F₂ agreed well with the reference data. It can be observed in Fig. 6(a) that the Q⁰ group with a chemical shift at −71.1 ppm was identified in the leaching residue for Ca₃Si₂O₇ after 1 h. On the contrary, there was no Q⁰ group identified in the leaching residue for Ca₄Si₂O₇F₂ as shown in Fig. 6(b). It can be observed from the comparison of the NMR spectrum that for both of Ca₃Si₂O₇ and Ca₄Si₂O₇F₂, there were only dimer [SiO₄]⁴⁻ tetrahedron structures before the leaching test. However, the isolated [SiO₄]⁴⁻ tetrahedrons were detected in the leaching residue of Ca₃Si₂O₇, and there was no significant change in the leaching residue of Ca₄Si₂O₇F₂. This suggests that, for Ca₃Si₂O₇, some dimer [SiO₄]⁴⁻ tetrahedrons broke into two isolated [SiO₄]⁴⁻ tetrahedrons during the hydration reaction, but Ca₄Si₂O₇F₂ retained its dimer [SiO₄]⁴⁻ tetrahedron structure after 1 h. Because the mineral phase with an isolated [SiO₄]⁴⁻ tetrahedron structure showed a high Ca elution amount in the course of the leaching test, the significant difference in Ca elution amounts between Ca₃Si₂O₇ and Ca₄Si₂O₇F₂ can be explained by the different dissociation mechanisms of the silicate network during the hydration process. Moreover, the difference in the dissociation mechanisms of the silicate network may be caused by the different chemical binding states around the Ca atoms in Ca₃Si₂O₇ and Ca₄Si₂O₇F₂.

Fig. 6.

²⁹Si MAS NMR spectra comparison between the powder samples and the leaching residues for Ca₃Si₂O₇ (a) and Ca₄Si₂O₇F₂ (b).

5. Conclusions

(1) During the leaching tests, the pH value of each solution greatly increased in the first 5 minutes, which was caused by the intense elution behavior of Ca from target mineral phases. The elution amount of Ca was in the order Ca₃SiO₅ > γ-Ca₂SiO₄ > Ca₃Si₂O₇ > α-CaSiO₃ > β-CaSiO₃ ≈ Ca₄Si₂O₇F₂ > CaMgSi₂O₆ > CaFeSi₂O₆ > Ca₂Mg₅Si₈O₂₂(OH)₂ > CaAl₂Si₂O₈.

(2) The Ca elution amount after 1 h showed a decreasing trend when the corrected basicity and the (NBO + FO)/T decreased. The amount of Ca elution during the leaching test is related to the crystal structure of the silicates in the mineral phases.

Acknowledgements

This work was supported by the iron and steel institute of Japan (ISIJ) and the research program of “five-star alliance” in “NJRC Mater. & Dev.” The authors are grateful to all members of the ISIJ Research Committee in 「Control of Solidified Structure of Steelmaking Slag for Suppression of Elution of Alkali」 for fruitful discussion. This work was financially supported in part by a Grant-in-Aid for Scientific Research (B) Grant (No. 19H02487) from the Japan Society for the Promotion of Science (JSPS). The authors would like to thank Ms. Mariko Ando (Tohoku University) for her supports in NMR experiments.

References

• 1)   A. I.  Al-Negheimish,  F. H.  Al-Sugair and  R. Z.  Al-Zaid: J. King Saud Univ. Eng. Sci., 9 (1997), 39.
• 2)   K.  Horii,  N.  Tsutsumi,  T.  Kato,  Y.  Kitano and  K.  Sugahara: Nippon Steel Sumitomo Met. Tech. Rep., 109 (2015), 5.
• 3)   X.  Zhang,  H.  Matsuura and  F.  Tsukihashi: ISIJ Int., 52 (2012), 928.
• 4)   X.  Gao,  T.  Ito,  H.  Nasugawa and  S.  Kitamura: ISIJ Int., 56 (2016), 1103.
• 5)   T.  Futatsuka,  K.  Shitogiden,  T.  Miki,  T.  Nagasaka and  M.  Hino: Tetsu-to-Hagané, 89 (2003), 382 (in Japanese).
• 6)   H.  Tamaki,  U. M.  Azhar,  Y.  Kato and  K.  Takahashi: Tetsu-to-Hagané, 99 (2013), 676 (in Japanese).
• 7)   X.  Gao,  N.  Maruoka,  S. J.  Kim,  S.  Ueda and  S.  Kitamura: J. Sustain. Metall., 1 (2015), 304.
• 8)   S.  Tauchi and  Y.  Kashiwaya: Tetsu-to-Hagané, 104 (2018), 43 (in Japanese).
• 9)   I.  Strandkvist,  B.  Bjorkman and  F.  Engstrom: Can. Metall. Q., 54 (2015), 446.
• 10)   F.  Engstrom,  D.  Adolfsson,  C.  Samuelsson,  A.  Sandstrom and  B.  Bjorkman: Miner. Eng., 41 (2013), 46.
• 11)   W. H.  Casey and  G.  Sposito: Geochim. Cosmochim. Acta, 56 (1992), 3825.
• 12)   W. H.  Casey: Nat. Mater., 7 (2008), 930.
• 13)   H. R.  Westrich,  R. T.  Cygan,  W. H.  Casey,  C.  Zemitis and  G. W.  Arnold: Am. J. Sci., 293 (1993), 869.
• 14)   T.  Zoltai: Am. Mineral., 45 (1960), 960.
• 15)   Z.  Zhu,  X.  Gao,  S.  Ueda and  S.  Kitamura: REWAS 2019: Manufacturing the Circular Materials Economy, Proc. TMS2019 148th Annual Meeting and Exhibition, Springer International Publishing, New York, (2019), 215.
• 16)   M.  Okuno,  F.  Marumo,  J.  Ossaka,  N.  Ishizawa and  I.  Minato: J. Jpn. Assoc. Mineral. Petrol. Econ. Geol., 77 (1982), Issue Special 3, 19 (in Japanese).
• 17)   Y. S.  Lee,  D. J.  Min,  S. M.  Jung and  S. H.  Yi: ISIJ Int., 44 (2004), 1283.
• 18)   E. B.  Pretorius and  R. C.  Carlisle: 56th Electric Furnace Conf., Iron and Steel Society, Warrendale, PA, (1998), 275.
• 19)   B. O.  Mysen,  D.  Virgo and  F. A.  Seifert: Rev. Geophys., 20 (1982), 353.
• 20)   Y.  Sun,  Z.  Zhang,  L.  Liu and  X.  Wang: J. Non-Cryst. Solids, 420 (2015), 26.
• 21)   M.  Magi,  E.  Lippmaa,  A.  Samoson,  G.  Engelhardt and  A. R.  Grimmer: J. Phys. Chem., 88 (1984), 1518.
• 22)   M. R.  Hansen,  H. J.  Jakobsen and  J.  Skibsted: Inorg. Chem., 42 (2003), 2368.