Skip to content
BY-NC-ND 3.0 license Open Access Published by De Gruyter October 5, 2016

Structural Properties of Molten CaO–SiO2–P2O5–FeO System

  • Jiang Diao EMAIL logo , Zhaoqun Ke , Lu Jiang , Zhen Zhang , Tao Zhang and Bing Xie

Abstract

The structural properties of molten CaO–SiO2–P2O5–FeO slag system with varying slag basicity have been investigated by molecular dynamic (MD) simulations using the pairwise potential model. The result shows that more than 95 % Si and 98 % P are four coordinated and form tetrahedral structures. Non-bridging oxygen occupies a predominant position in the system. With basicity increasing from 0.6 to 1.5, the proportion of non-bridging oxygen increases from 66.3 % to 77.3 %, whilst the bridging oxygen decreases from 30.1 % to 10.2 %. Both the result of MD simulations and Raman spectroscopic analysis show the proportion of Q0 increases with increasing slag basicity, whilst Q2 and Q3 decrease. The degree of polymerization of CaO–SiO2–P2O5–FeO system decreases with increasing slag basicity.

Introduction

The physicochemical properties and transport behaviors of molten slag have important influence on most metallurgical smelting processes. Essentially, the physicochemical properties of molten slag are determined by the structure of the slag. Therefore, the structural information of molten slag is of great interest for understanding their physicochemical properties and transport behaviors in metallurgical fields. The structural properties of some simple oxide systems have been investigated through various sorts of experimental techniques, e. g. nuclear magnetic resonance (NMR), X-ray diffraction, X-ray absorption, neutron diffraction, Raman, infrared spectroscopy and so on [1, 2]. However, taking into account most metallurgical slag systems are extremely complex and the difficulties of experiments at high temperatures, molecular dynamic (MD) simulation has been widely employed to calculate the structural and transport properties of molten slag at high temperatures [3, 4, 5].

It is well known that the CaO–SiO2–P2O5–FeO system is one of the most important component parts in molten converter slag. In the early stage of converter steelmaking, the viscosity of molten slag directly affects the dissolution rate of lime, slag–metal interface reaction and mass transfer of phosphorus from hot metal to the slag system [6, 7]. As the viscosity of molten slag is determined by its structure, it is necessary to investigate the structural properties of CaO–SiO2–P2O5–FeO system. So far as is known to the author, most of the previous studies mainly concentrated on the structural information of simple binary or ternary silicate systems and aluminate systems [8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]. However, only a few studies have reported on the steelmaking slag system. Wang et al. [20] investigated the effect of P2O5 and FetO on the viscosity and slag structure in steelmaking slags through viscosity measurement, Raman, magic angle spinning-nuclear magnetic resonance (MAS-NMR) and FTIP spectra measurements. The results show that the viscosity of molten steelmaking slag increases slightly with increasing P2O5 content, while decreases with increasing FetO content. The degree of the polymerization of quenched steelmaking slag is found to increase with increasing P2O5 content and decrease with increasing FetO content. Our previous work [21] studied the structural properties of the ternary molten CaO–SiO2–P2O5 slag system by MD simulations. The results show that the concentration of free oxygen decreases remarkably with the increase of P2O5 content. The P ion has a tendency to promote the polymerization of phosphosilicate melts. Mercier et al. [22] also observed that the glass network structure became more and more polymerized while increasing P2O5 content in SiO2–Na2O–CaO–P2O5 bioglasses using Si and P MAS-NMR.

Because of the importance of the molten CaO–SiO2–P2O5–FeO system, it is necessary to investigate and discuss it thoroughly. Almost all P2O5 concentrates in Ca2SiO4–Ca3P2O8 solid solution in steelmaking slag; therefore, in this paper, we performed a MD simulation and emphatic studied the effect of basicity (mass ratio of CaO/SiO2) on the structural properties of molten CaO–SiO2–P2O5–FeO slag system.

Computational details and experiments

In this work, MD simulations are performed with the parallel package Materials Explorer 5.0 using the born mayer huggins (BMH) function [23, 24]:

(1)Urij=qiqjrij+AijexpBijrijCijrij6

where U(rij) is the interatomic potential; qi and qj are the charges of ion i and ion j; rij represents the distance between ion i and ion j; Aij, Bij and Cij are parameters for BMH function. The first term is Coulombic interaction, the second term is inter-core short-range repulsion and the last term represents van der Waals attraction.

The charge of Ca, Si, P, Fe and O are 2, 4, 5, 2 and −2, respectively. The Coulombic potential was calculated by using the Ewald summation. All the noncoulombic parts of the pair potentials were subjected to a short-range cutoff of 10 Å. The interatomic potential parameters of the CaO–SiO2–P2O5–FeO slag system in this study were determined by using empirical parameters model. Optimization of the empirical parameters based on Hirao and Kawamura [25] was carried out by Lammps program. The principle of the optimization is energy minimization. The optimized parameters Aij, Bij and Cij are listed in Table 1.

The slag composition, total number of atoms and other detailed information of the CaO–SiO2–P2O5–FeO system are presented in Table 2. To eliminate the edge and surface effects, three-dimensional periodic boundary condition was employed for each simulation process. The Newton motion equations were numerically integrated with a step of 1 fs. The initial temperature was set at 4,000 K for 15,000 steps to eliminate the effect of the initial distribution. Then, the temperature was decreased slowly to 2,000 K, 1,873 K and 1,673 K for 10,000 steps, respectively. The calculation was equilibrated at 1,673 K for another 25,000 steps. All simulations were performed in NVT (constant number of particles, volume and temperature) ensemble. In order to ensure that the CaO–SiO2–P2O5–FeO system is a homogeneous molten slag system, the slag compositions are designed based on the CaO–SiO2–P2O5–FeO phase diagram [26]. It can be seen in Figure 1 that all the slags are located in the liquid phase region at 1,673 K.

To check the validity of the numerical calculations, Raman spectroscopic analysis of the quenched slag samples have been conducted. The slag samples listed in Table 2 were prepared using analytical reagents. The mixture of slag sample was charged into a corundum crucible and melted at 1,673 K. The molten samples were held at this temperature for 1 h. Then, the sample was quenched quickly with liquid nitrogen. Raman spectroscopic analysis was performed on a Horiba LabRAM HR Evolution system.

Table 1:

Potential parameters used in the MD simulation.

ijAij (g Å/fs2)Bij (1/Å)Cij (g Å8/fs2)
OO2.40E − 205.88E + 002.78E − 25
SiSi3.47E − 236.25E + 000
CaCa5.27E − 216.25E + 006.95E − 26
FeFe4.70E − 233.45E + 000
PP4.56E − 227.06E + 000
OSi1.01E − 216.06E + 000
OCa1.15E − 206.06E + 001.39E − 25
OFe6.41E − 225.16E + 000
OP3.04E − 233.45E + 000
SiCa4.28E − 226.25E + 000
SiFe9.22E − 231.29E + 010
SiP1.73E − 231.25E + 014.49E − 25
CaFe3.53E − 236.25E + 000
CaP2.64E − 211.25E + 010
FeP2.05E − 226.25E + 000
  1. MD: Molecular dynamic.

Table 2:

Slag compositions and number of molecules used in the MD simulation.

Mass fraction (mass %)Atomic number
No.CaOSiO2P2O5FeORCaSiPFeOTotalDensity
1362410301.56754201484372,3204,0003.127
232.827.210301.26064711464322,3454,0003.075
328.431.610300.95175371444252,3774,0003.009
422.537.510300.64016241414162,4184,0002.923
Figure 1: Slag compositions in CaO–SiO2–P2O5–FeO system.
Figure 1:

Slag compositions in CaO–SiO2–P2O5–FeO system.

Results and discussion

Bond lengths, angles and coordination numbers

The calculated bond lengths of Si–O, P–O, Fe–O and Ca–O for No. 2 slag are 1.61 Å, 1.53 Å, 2.05 Å and 2.30 Å, respectively. The results are in a good agreement with the previous data in molten CaO–SiO2–P2O5 system [27]. The coordination numbers of Si–O, P–O, Fe–O and Ca–O are equal to 3.86, 3.97, 5.86 and 5.10, respectively. Figure 2 shows the bond angle distributions of O–Si–O and O–P–O of No. 2 slag. The angles of O–Si–O in [SiO4] tetrahedron and O–P–O in [PO4] tetrahedron are 109.5° and 109.0°, respectively. The results are in a good agreement with the theoretical value of 109.5°.

Figure 2: The distributions of O–Si–O and O–P–O bond angles for No. 2 slag.
Figure 2:

The distributions of O–Si–O and O–P–O bond angles for No. 2 slag.

Table 3 shows the coordination statistics of Si and P with varying slag basicity. It can be seen that the proportions of 4-coordinated P (PIV) and Si (SiIV) ions are beyond 98 % and 95 %, respectively, which means [SiO4] and [PO4] tetrahedrons are prominent in the molten CaO–SiO2–P2O5–FeO slag system. The variation of slag basicity has little effect on the coordination of P–O and Si–O. It can be concluded that (SiO4) and (PO4) tetrahedrons are extremely stable in this slag system.

Table 3:

Coordination statistics of Si and P, %.

RSiIVSiVSiVIPIVPV
0.696.753.25100
0.996.762.310.9398.111.89
1.295.604.40100
1.596.233.140.63100

Distributions of Onb, Ob and Of

It is well known that the form of oxygen can be divided into bridge oxygen (Ob), non-bridge oxygen (Onb), free oxygen (Of) and tricluster oxygen. However, the proportion of tricluster oxygen is very low (<4 %) according to the calculation of Fan et al. [28]. In the present calculation, the tricluster oxygen was not found in the CaO–SiO2–P2O5–FeO system. The distributions of the aforementioned three type of oxygen are presented in Figure 3. It can be seen that when the basicity increases from 0.6 to 1.5, the proportion of Ob decreases from 30.1 % to 10.2 %, on the contrary, the proportion of Onb increases from 66.3 % to 77.3 %, and the proportion of Of increases from 3.6 % to 12.6 %. This implies that the polymerization degree of slag decreased by increasing CaO content.

Ob can be further subdivided into three groups, i. e. P–O–P, Si–O–P and Si–O–Si. Onb can be further subdivided into two groups, i. e. P–O–Ca, Si–O–Ca. AsFigure 3(b) shows, the proportion of Si–O–Si decreases sharply with increasing slag basicity, which means CaO plays the role of network former in the CaO–SiO2–P2O5–FeO system.

Figure 3: Proportions of different oxygen types with varying slag basicity.
Figure 3:

Proportions of different oxygen types with varying slag basicity.

Distributions of Qn

The polymerization of the slag structure can be estimated by the parameter Qn, where n denotes the number of Ob in a tetrahedron, i. e. Qn(Si) represents the proportion of Ob in [SiO4] tetrahedron, Qn(P) represents the proportion of Ob in [PO4] tetrahedron. The calculated distributions of Qn for P and Si are shown in Figure 4. It can be found that the proportions of Q0(Si) and Q0(P) show a remarkable uptrend with the increase of slag basicity. With the basicity increasing from 0.6 to 1.5, the proportion of Q0(Si) increases from 6.2 % to 44.5 %, and the proportion of Q0(P) increases from 12.3 % to 67.6 %. The proportions of Q2(Si), Q3(Si), Q4(Si), Q1(P), Q2(P) and Q3(P) also decrease obviously with the increase of slag basicity. When basicity increased to 1.5, Q4(Si), Q3(P) and Q4(P) disappeared completely. It can be concluded that CaO can make the complex structure of slag become simple.

Figure 4: Effect of slag basicity on (a) Qn(Si) and (b) Qn(P).
Figure 4:

Effect of slag basicity on (a) Qn(Si) and (b) Qn(P).

Raman spectra

Figure 5 displays the Raman spectra for the CaO–SiO2–P2O5–FeO system. It can be seen that the No. 1 slag has a noticeable 600 cm−1 band. However, it shifts to a higher frequency band with increasing slag basicity. The 600 cm−1 band is assigned to the vibration of a 3-member ring [29]. It is proposed that the strain on the ring would be attacked by the free oxygen added by CaO [30]. The disappearance of this band and subsequent appearance of a higher frequency band with increasing slag basicity have been also observed by Kline et al. [31]. The peak at around 440 cm−1 is assigned to the P–O–P bending [32]; however, it fade away gradually with increasing slag basicity, which means the structure of CaO–SiO2–P2O5–FeO system becomes simple with increasing slag basicity.

In the region of 800–1,200 cm−1, a peak at around 950 cm−1 is most evident, which can be attributed to the Q2 unit. With increasing slag basicity, the peak fades away gradually. The peak for Q2 and Q3 turn to lower frequency band, which also implies that the structure of the slag became simple. To quantitatively investigate thestructure of CaO–SiO2–P2O5–FeO system, deconvolution of the Raman curves was performed in the range of 800–1,150 cm−1. As the [SiO4] and [PO4] tetrahedrons may be copolymerized, and the phosphorus-related peaks are very weak, it is not easy to obtain the phosphorus-related bands by deconvoluted results [20]. Therefore, the Qn(P) are ignored in the present experimental study. Figure 6 displays the deconvoluted results. Based on the curves, the mole fractions of different structural units can be calculated. It can be seen that with the slag basicity increasing from 0.6 to 1.5, the proportion of Q0(Si) increases from 6.2 % to 43.9 %, the proportion of Q2(Si) and Q3(Si) decreases from 64.3 % to 39.9 % and 12.2 % to 5.4%, respectively. Both the calculated Q1(Si) and experimental Q1(Si) show a trend of first increasing and then decreasing. Therefore, the proportions of structural units showed in Table 4 are consistent with the numerical modeling results showed in Figure 4. It also can be concluded that the degree of polymerization of CaO–SiO2–P2O5–FeO system decreases with increasing slag basicity.

Figure 5: Raman spectra of the slag samples.
Figure 5:

Raman spectra of the slag samples.

Figure 6: Deconvoluted results of the Raman spectral curves at different slag basicities.
Figure 6:

Deconvoluted results of the Raman spectral curves at different slag basicities.

Table 4:

The proportions of structural units.

Q0Q1Q2Q3
No. 16.217.164.312.2
No. 29.543.640.36.5
No. 310.850.132.96.2
No. 443.910.639.95.4

Conclusion

The molten CaO–SiO2–P2O5–FeO system plays an important role in metallurgical process. As this slag system has not been studied adequately, a MD simulation was carried out to calculate the structural information of this quaternary slag system. The simulated results demonstrate that the [SiO4] and [PO4] tetrahedrons are prominent in this molten system. With slag basicity increasing from 0.6 to 1.5, the proportion of Ob decreases, while the proportion of Onb increases. The polymerization degree of slag decreases by increasing slag basicity. The proportions of structural units observed by Raman spectroscopic analysis are in agreement with the data calculated by MD simulation. The results would help us to better understand the fundamental structural properties and physicochemical properties of molten metallurgical slags. Future work will characterize the structural information through experiments.

Funding statement: This work was partially supported by the Fundamental Research Funds for the Central Universities (Project CDJZR 14130001) and National Basic Research Program of China (No. 2013CB632604).

References

[1] K. Zheng, Z. Zhang, F. Yang and S. Sridhar, ISIJ Int., 52 (2012) 342–349.10.2355/isijinternational.52.342Search in Google Scholar

[2] M. Morishita, K. Koyama, A. Hatamoto, M. Morinaga and H. Adachi, ISIJ Int., 36 (1996) 1259–1263.10.2355/isijinternational.36.1259Search in Google Scholar

[3] D. Belashchenko, Russ. Chem. Rev., 66 (1997) 733–762.10.1070/RC1997v066n09ABEH000236Search in Google Scholar

[4] T. Matsumiya, A. Nogami and Y. Fukuda, ISIJ Int., 33 (1993) 210–217.10.2355/isijinternational.33.210Search in Google Scholar

[5] K. Xu, G. Jiang, S. Huang and J. You, Sci. China Ser. E, 42 (1999) 77–82.10.1007/BF02917061Search in Google Scholar

[6] K. Shiamauchi, S. Kitamura and H. Shibata, ISIJ Int., 49 (2009) 505–511.10.2355/isijinternational.49.505Search in Google Scholar

[7] J. Diao, B. Xie, Y. Wang and X. Guo, ISIJ Int., 50 (2010) 768–770.10.2355/isijinternational.50.768Search in Google Scholar

[8] V. Hoang, D. Belashchenko and V. Thuan, Phys. B, 348 (2004) 249–255.10.1016/j.physb.2003.12.001Search in Google Scholar

[9] D. Belashchenko and O. Ostrovski, Thermochim. Acta, 372 (2001) 143–152.10.1016/S0040-6031(01)00452-XSearch in Google Scholar

[10] R. Mead and G. Mountjoy, J. Phys. Chem. B, 110 (2006) 14273–14278.10.1021/jp0628939Search in Google Scholar PubMed

[11] Y. Sasaki, M. Iguchi and M. Hino, ISIJ Int., 47 (2007) 638–642.10.2355/isijinternational.47.638Search in Google Scholar

[12] D. Belashchenko, O. Ostrovski and S. Saposznikova, Metall. Mater. Trans. B, 29B (1998) 105–110.10.1007/s11663-998-0012-3Search in Google Scholar

[13] Y. Sasaki and M. Iguchi, ISIJ Int., 49 (2009) 602–604.10.2355/isijinternational.49.602Search in Google Scholar

[14] D. Belashchenko, O. Ostrovski and L. Skvortscov, Thermochim. Acta, 372 (2001) 153–163.10.1016/S0040-6031(01)00451-8Search in Google Scholar

[15] W. Seo and F. Tsukihashi, ISIJ Int., 44 (2004) 1817–1825.10.2355/isijinternational.44.1817Search in Google Scholar

[16] K. Shimoda and K. Saito, ISIJ Int., 47 (2007) 1275–1279.10.2355/isijinternational.47.1275Search in Google Scholar

[17] Y. Sasaki and K. Ishii, ISIJ Int., 44 (2004) 43–49.10.2355/isijinternational.44.43Search in Google Scholar

[18] Y. Sasaki, M. Iguchi and M. Hino, ISIJ Int., 47 (2007) 643–647.10.2355/isijinternational.47.643Search in Google Scholar

[19] D. Belashchenko and L. Skvortscov, Inorg. Mater., 37 (2001) 571–576.Search in Google Scholar

[20] Z.J. Wang, Q.F. Shu, S. Sridhar, M. Zhang, M. Guo and Z.T. Zhang, Metall. Mater. Trans. B, 46B (2015) 758–765.10.1007/s11663-014-0270-1Search in Google Scholar

[21] J. Diao, G.Z. Fan, X. Liu and B. Xie, Metall. Mater. Trans. B, 45B (2014) 1942–1947.10.1007/s11663-014-0092-1Search in Google Scholar

[22] C. Mercier, C. Follet-Houttemane, A. Pardini and B. Revel, J. Non-Cryst. Solids, 357 (2011) 3901–3909.10.1016/j.jnoncrysol.2011.07.042Search in Google Scholar

[23] L. Woodcock, C. Angell and P. Cheeseman, J. Chem. Phys., 65 (1976) 1565–1577.10.1063/1.433213Search in Google Scholar

[24] H. Luo, S. Xiao, S. Wang, P. Huai, H. Deng and W. Hu, Comp. Mater. Sci., 111 (2016) 203–208.10.1016/j.commatsci.2015.09.052Search in Google Scholar

[25] H. Hirao and K. Kawamura, Material Design Using Personal Computer, Shokabo, Tokyo (1994), p. 52.Search in Google Scholar

[26] Verein Deutscher Eisenhuttenleute, Slag Atlas, 2nd edition, Verlag Stahleisen mBH, Dusseldorf (1995), p. 174.Search in Google Scholar

[27] G. Fan, J. Diao, L. Jiang, Z. Zhang and B. Xie, Mater. Trans., 56 (2015) 655–660.10.2320/matertrans.M2014363Search in Google Scholar

[28] G. Fan, S. He, T. Wu and Q. Wang, Metall. Mater. Trans. B, 46B (2015) 2005–2013.10.1007/s11663-015-0354-6Search in Google Scholar

[29] F.L. Galeener, Solid State Commun., 44 (1982) 1037–1040.10.1016/0038-1098(82)90329-5Search in Google Scholar

[30] P. Boudjouk, C. Kapfer and R. Cunico, Organometallics, 2 (1983) 336–343.10.1021/om00074a023Search in Google Scholar

[31] J. Kline, M. Tangstad and G. Tranell, Metall. Mater. Trans. B, 46B (2015) 62–73.10.1007/s11663-014-0194-9Search in Google Scholar

[32] Y. Sun, Z. Zhang, L. Liu and X. Wang, J. Non-Cryst. Solids, 420 (2015) 26–23.10.1016/j.jnoncrysol.2015.04.017Search in Google Scholar

Received: 2016-3-14
Accepted: 2016-6-25
Published Online: 2016-10-5
Published in Print: 2017-9-26

© 2017 Walter de Gruyter GmbH, Berlin/Boston

This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Downloaded on 2.6.2024 from https://www.degruyter.com/document/doi/10.1515/htmp-2016-0054/html
Scroll to top button