Free Access
Issue
Metall. Res. Technol.
Volume 117, Number 3, 2020
Article Number 307
Number of page(s) 8
DOI https://doi.org/10.1051/metal/2020023
Published online 28 May 2020

© EDP Sciences, 2020

1 Introduction

Niobium (Nb) is a strategic resource which is widely used in the fields of medical treatment, superconductivity and steel industry [1]. Though the Bayan Obo mine in China contains amount of resources such as niobium, rare earth and iron, it is difficult to obtain Nb2O5 due to low grade of 0.1–0.14 wt.% and complex embedding [2]. Pyrometallurgy can separate niobium containing slag from iron by direct reduction and slag-metal separation effectively. This method can improve the grade of niobium of slag and is beneficial to further preparation of ferroniobium or other intermediate products. Viscosity is a basic parameter which affects the separation effect and production efficiency. However, the studies on viscosity and structure of Nb-bearing slag are limited. Therefore, it is extremely necessary to study the viscosity of Nb2O5 bearing slag.

As the atomic radius and Pauling electronegativity of Nb5+ and Ti4+ are similar, viscosity studies of slag containing TiO2 should be referenced [3]. Many studies have demonstrated that TiO2 additions less than 10 wt.% could lower the viscosity and activation energy of TiO2 bearing metallurgical slags [48]. However, there are controversial opinion on structure of Ti-bearing slag, Li et al. [5] found that a small amount of TiO2 acted as network formers to increase the polymerization degree but most of Ti4+ mainly existed in the form of [TiO4] tetrahedral coordination in slag. Park et al. [4,68] showed depolymerization behavior of TiO2 additions. Therefore, the effect of TiO2 for silicate network should be further studied. The research on the structure of niobosilicate glass is focused on the structure of quenched glass to interpret melt structure approximately by infrared spectrum and Raman spectrum [3,912]. Nowak et al. [3] considered that Si-O-Si bonds were progressively substituted by Si-O-Nb and Nb-O-Nb bonds by Raman spectrum. Gao et al. [9] confirmed that Nb5+ ions existed in the form of [NbO4] and [NbO6] polyhedra coexisting in the glass network. The [NbO4] could be transformed into [NbO6] forming a network with the [SiO4] tetrahedra as the Nb2O5 content increases. Lopes et al. [10] suggested that Nb5+ ions created cross-links between several oxygen sites, breaking Si-O-Si bonds to form a range of polyhedra [Nb(OM)6 − y(OSi)y], where 1 ≤ y ≤ 5 and M = Ca or others. It is similar with Nowak’s viewpoint.

2 Experimental

2.1 Sample preparation

According to the composition of niobium containing slag obtained in the slag-metal separation process of niobium concentrate, analytical reagents were synthesized. In order to synthesize the sample accurately, CaCO3 powders were calcined in muffle furnace at 1323 K for 10 h to prepare CaO. Other components were dried in a drying oven at 383 K for 10 h to remove moisture. After weighing 50 g of each sample in accordance with Table 1, the mixtures were pre-melted in a molybdenum crucible (height: 80 mm, diameter: 35 mm) under the atmosphere of CO/CO2 = 99 at 1823 K for 10 h. The rate of gas flow is maintained as 0.5 L/min during the experiment. For the same condition, 1 g of pre-melted sample was quenched to obtain the glassy slag. X-ray diffraction (XRD) result, as shown in Figure 1, indicated the quenched slags were amorphous.

Table 1

Chemical compositions of experimental slag samples (wt.%).

thumbnail Fig. 1

XRD patterns of quenched slag samples.

2.2 Viscosity measurement

The viscosity of slag was measured by rotating crucible viscometer (Rheotrnic II viscometer, Theta Ltd, USA). The schematic diagram of the viscometer and the location and material of spindle are shown in Figure 2. During the viscosity experiment, the reducing atmosphere was kept by holding the flow rate of 0.4 L/min in sealed Al2O3 tube.

Before the experiment, the viscometer was calibrated with standard oil (Brookfield standard oil with a viscosity of 1.005 Pa · s) at 298 K. The B-type thermocouple of viscometer furnace was also calibrated with the error of ±3 K. For the experiment, the pre-melted sample was heated to 1823 K and held for 1 h under the same reducing atmosphere to achieve thermal equilibrium, then the molybdenum rotating spindle at 60 rpm was slowly immersed in the molten slag and kept at the distance of 5 mm above the crucible bottom. The crucible and spindle were properly aligned along the axis of viscometer. During the measurement stage, the viscosity was discontinuously recorded at the interval of 20 K, during the measurement of the slag with an equilibration time of 30 min at each temperature. Repetitions of the viscosity measurements showed a good reproducibility.

thumbnail Fig. 2

Schematic diagram of experimental apparatus for viscosity measurement.

2.3 Raman spectroscopy measurement

To determine the various functional group structures of slag, the glassy slag was analyzed by Raman spectroscopy (InVia, Renishaw, UK). The spectra frequency range was 60–1460 cm−1 and the excitation wavelength of confocal laser is 532 nm. The results were analyzed by the assumption of Gaussian distribution, the peaks of different structural units in spectra were fitted by Peakfit software.

3 Result and discussion

As the reducing atmosphere is CO/CO2 = 99, the oxygen partial pressure is calculated as (pθ = 1.01325 × 105 Pa) when the temperature is 1773 K. For the stable existence of Nb2O5 and CeO2 in the melts, researchers studied the thermodynamics of melt by experiments and calculations. Wang et al. [13] studied the stability diagram of the Nb-O system, it can be seen that Nb4+ is the stable valence state for 1773 K in the reducing atmosphere of CO/CO2 = 99. Jeong et al. [14] and Zhao et al. [15] showed that Ce3+ was the main valence state under the experimental conditions.

3.1 Effects of Nb2O5 on the viscosity

The effects of Nb2O5 addition on the viscosity of the CaO-SiO2-Nb2O5-5.0 wt.% CeO2-5.0 wt.% CaF2 slag with fixed basicity (C/S = 1) at the temperature range of 1673–1813 K are shown in Figure 3. The viscosity of all the slags increases when the temperature decreases. There is a break point of temperature in each curve. However, the accurate break temperatures are unclear as the interval of measurement temperature is 20 K. The variations of viscosity are slow when the temperature is higher than break temperature. While the temperature is decreased to be lower than the break temperature, the viscosity increases rapidly and reaches 12 Pa · s in a narrow temperature range. When Nb2O5 contents increase from 0 wt.% to 12 wt.%, the break temperatures are decreased by about 100 K.

thumbnail Fig. 3

Effect of Nb2O5 addition on the viscosity of the CaO-SiO2-Nb2O5-5 wt.% CeO2-5 wt.% CaF2 slag system (C/S = 1).

3.2 Effect of basicity on the viscosity

Figure 4 shows the effect of basicity on the viscosity of niobium-bearing slag system with fixed Nb2O5 addition at the temperature range of 1653–1813 K. The viscosity of slag and break temperatures decrease when the basicity of slags increases from 0.8 to 1.2. When the temperature of slag is lower than the break temperature, the increasing rate of viscosity is less for higher basicity.

thumbnail Fig. 4

Effect of basicity on the viscosity of CaO-SiO2-8 wt.% Nb2O5-5 wt.% CeO2-5 wt.% CaF2 slag system.

3.3 Effect of Nb2O5 and basicity on the activation energy for viscous flow

Activation energy is related to the composition and temperature of slag, which represents the energy barrier that the cohesive flow units must overcome when the melt flows. [1618]. For given melt without crystals, the activation energy keeps constant for given temperature range. In this study, viscosity whose temperature is higher than the break temperature is selected for analysis. Activation energy for viscous flow can be calculated according to the Arrhenius type relationship, as expressed in equation (1): (1) where A, Eη, R, and T denote the pre-exponential factor, the activation energy for viscous flow, the universal gas constant (8.314 J/mol · K), and the absolute temperature, respectively. After taking the logarithm of equation (1), equation (2) shows the relationship with the slope of fitting straight-line (k) and activation energy (Eη). The slope k can be obtained in Figure 5, the activation energy Eη should be calculated by equation (2). (2)

From Figure 5, it can be seen the logarithm of viscosity has a good linear relationship with the reciprocal of temperature, the correlation coefficient is higher than 0.99. It is evident that the slope of all the slags decrease with the Nb2O5 content and basicity increase. Figure 6 shows the activation energy for viscous flow of niobium-bearing slag system with different Nb2O5 contents and basicity. The activation energy is significantly decreased from 0 to 4 wt.% of Nb2O5 addition. Then the decreasing trend tends to be stable with the increase of Nb2O5 addition. For the basicity of slag system, the activation energy decreases when the basicity increases from 0.8 to 1.2. It can be inferred that Nb2O5 addition and basicity are likely to decrease the degree of the polymerization of silicate network structure and provide additional free oxygen ions.

thumbnail Fig. 5

Natural log of viscosity vs. 104/T of CaO-SiO2-Nb2O5-5 wt.% CeO2-5 wt.% CaF2 slag system.

thumbnail Fig. 6

Activation energy for viscous flow of CaO-SiO2-Nb2O5-5 wt.% CeO2-5 wt.% CaF2 slag system.

3.4 Effect of Nb2O5 on the slag structure using Raman spectroscopy

It is known that the structure of cation is mainly determined by ion size and charge. As the radius of Ce3+, Nb5+ and Nb4+ are 103 pm, 70 pm and 74 pm, respectively. All of them are too large to enter the tetrahedral structure of niobosilicate slag. Therefore, it is generally believed that Nb5+/Nb4+ mainly exists in the form of [NbO6] octahedron [19], and some reports show that Nb5+ exists in the structure of [NbO6] octahedron and [NbO4] tetrahedron simultaneously [20]. For the structure of silicate melt containing rare-earth ions, [CeNbO4] structure with niobium ions at high temperature has been observed [21]. Relevant literature shows that the NBO/T (T denotes any network forming, tetrahedrally coordinated cation [22]) increases when the Ce2O3 content increases in the silicate slag [2325]. Ce3+ conducts a role of charge compensator or network modifier rather than network former. CaF2 is generally known as network modifier and breaks silicate network [2628]. Considering the component of niobium concentrate slag, the content of CeO2 and CaF2 are fixed to 5 wt.%, the influence of Ce3+ and F in slag for Raman spectra is ignored.

As the high polarizability and consequent high scattering cross section of the niobite vibrating groups, it is an order of magnitude higher than Si-O bond vibration [11]. Therefore, the bands associated to silicon-oxygen units are usually seen only at low niobium content in niobosilicate slag. In this study, Figure 7 shows that Raman spectra of slag vary with the content of Nb2O5 (Nb2O5 = 0–12 wt.%, C/S = 1). The whole range of Raman spectra can be divided into three parts: low frequency region (< 500 cm−1), intermediate frequency region (500–900 cm−1) and high frequency region (> 900 cm−1). In the intermediate frequency region (500–900 cm−1), a broad peak at about 650 cm−1 can be observed, which is assigned to vibrational mode for Si–O–Si bond coupled vibrations of [NbO6] octahedra with low degree of distortion and without NBO’s [10]. There is an obvious peak appearing at about 800 cm−1, the intensity of peaks increases when the Nb2O5 contents increase. The peak at 790–805 cm−1 is due to highly distorted [NbO6] octahedron linked to chains to form Nb-O-Si bonds. The frequency of [NbO4] vibration is about 890 cm−1, the intensity is covered by vibration of Si-O bonds [10,12]. A small peak appears at the position of about 860 cm−1 and a significant broad peak appears at the position of 965 cm−1 corresponding to Q0 ( in monomer structure unit) and Q1, Q2, Q3 ( in dimer structure unit, in chain structure unit and in sheet structure unit) in silicate network, respectively. Figure 7 shows the increasing trend of vibration peak of Nb-O-Si bond with the content of Nb2O5 increases. The peak of Q0 is evident at 857 cm−1 and no peak is observed at 800 cm−1 without Nb2O5 in slag. When the content of Nb2O5 is 4 wt.%, the relative intensity of Q0 and Nb-O-Si bond are approximately equal. When the content of Nb2O5 is 12 wt.%, Nb-O-Si bond vibration peak is obvious and Q0 peak is nearly covered. The Raman results are consistent with the results of relevant literatures.

The Raman spectra are deconvolved by Gaussian fitting similar to the method used by other researchers. The deconvolution results for the contents of Nb2O5 from 0–12 wt.% and C/S = 1 are shown in Figures 8a8d, for which all Raman spectra are successfully fitted at the frequency range of 600 to 1150 cm−1. The correlation coefficient R2 is above 0.998. The assignments of Raman shift for various structural units are summarized in Table 2. For intermediate frequency area, a shift towards higher frequency (680–720 cm−1) can be seen with the increase of Nb2O5 addition, which corresponds to [NbO6] octahedra with low degree of distortion and without NBO’s. It can be inferred that the distortion degree of [NbO6] octahedron enhances with Nb2O5 content increase. The peak intensity at about 801 cm−1 corresponding [NbO6] octahedra with high degree of distortion is also increased, indicating that the quantity of [NbO6] octahedra with high degree of distortion increases with the Nb2O5 addition. Mysen et al. [5,2931] have assigned the band at 840–860 cm−1, 900–920 cm−1, 960–980 cm−1 and 1040–1060 cm−1 to Q0 – Q3 units, respectively. The frequencies of Q0 – Q3 in this study are similar with the ones reported in these reference and structural unit areas could be measured for a quantitative evaluation of the structure of silicate melt.

The polymerization degree can be represented by the change in the relative contents and conversing of various structural units. Integration of the deconvoluted spectra provides a semiquantitative value of the various Qn species. The mole fractions of Qn species in the silicate network are calculated according to the equation (3) [32]. (3) where, Xn denotes the mole fraction of Qn species (n ranges from 0 to 3), θn and An denote Raman scattering coefficient and structural unit area of Qn, respectively. Considering the θn to be constant, Xn should be studied by comparing the ratios of structural unit area of An. According to Park et al. [4], as is shown in Figure 9, the sum of the Q0 and Q2 increases and the sum of the Q1 and Q3 decreases with Nb2O5 additions. The decrease in the sum of Q1 and Q3 is an indication of depolymerization occurring within the silicate structure. Mysen et al. [29] proposed the depolymerization mechanism can be expressed as equation (4). (4)

The depolymerization mechanism explains the polymerization relation of silicate slag. The whole slag structure is simpler with the Nb2O5 addition. More free oxygen ions break the complex silicate network into simpler structures, which reduce the degree of polymerization and thus the viscosity. Meanwhile, as [NbO6] octahedra with low degree of distortion and without NBO’s exists in slag structure, simpler structural units decrease the viscous resistance. The depolymerization and simpler structural units could be the reason for viscosity decreases.

thumbnail Fig. 7

Raman spectra of as-quenched samples from 1823 K with varying Nb2O5 content (C/S = 1).

thumbnail Fig. 8

Deconvoluted Raman spectra of as-quenched samples from 1823 K with varying Nb2O5 content (C/S = 1).

Table 2

Assignments of Raman bands in the spectra of CaO-SiO2-Nb2O5-CeO2-CaF2 slag system.

thumbnail Fig. 9

Relative fraction of Qn (0, 1, 2, 3) as a function of Nb2O5 content (C/S = 1).

3.5 Effect of basicity on the slag structure using Raman spectroscopy

The same method was conducted to deconvolve the Raman spectra, the results of different basicity of as-quenched sample were shown in Figures 10a10c. The correlation coefficient R2 is 0.999. It can be seen that the peak position of the highest intensity is still around 800 cm−1. The peak position and relative area of Si-O-Si deformation vibration and [NbO6] octahedra with different degree of distortion change slightly, indicating that basicity does not affect the degree of [NbO6] octahedral distortion. With the increase of basicity, the frequency of the structural unit of Qn changes little, while the intensity of the structural unit of Qn changes obviously. Q2 units are still the main structure units in silicate melt. The quantity of Q0 is much more when the basicity increases from 0.8 to 1.2. Figure 11 shows the relative fraction of Qn (0, 1, 2, 3) as a function of basicity, the result is similar with the relative fraction of Qn by varying Nb2O5 contents. The sum of Q1 and Q3 decreases and the sum of Q0 and Q2 increases when the basicity from 0.8 to 1.2, indicating that more simple structure units exist in the molten slag. This is consistent with the tendency of the measured viscosity of slags.

thumbnail Fig. 10

Deconvoluted Raman spectra of as-quenched samples from 1823 K with varying basicity (Nb2O5 = 8 wt.%).

thumbnail Fig. 11

Relative fraction of Qn (0, 1, 2, 3) as a function of basicity (Nb2O5 = 8 wt.%).

4 Conclusions

The viscosity, break temperature and activation energy for viscous flow of CaO-SiO2-Nb2O5-5 wt.% CeO2-5 wt.% CaF2 slag system decrease with the Nb2O5 content and basicity increase in this study.

Raman spectra show that Nb4+ ions exist in the form of [NbO6] octahedron with different distortion and little [NbO4]. The distortion of [NbO6] octahedron increase with the content of Nb2O5. The basicity doesn’t affect the distortion of octahedron.

Both the Nb2O5 additions and basicity could reduce the viscosity by affecting the silicate structure, where the complex silicate network depolymerized into simpler silicate structures.

Acknowledgements

The authors gratefully acknowledge financial support by Major projects of Inner Mongolia Natural Science Foundation (2018ZD07), the open project for key basic research of the Inner Mongolia Autonomous Region (20140201), Inner Mongolia Autonomous Region University Scientific Research Project (No. NJZZ19124), Baotou Science and Technology Project (No. 2017Z1009-2) and Inner Mongolia University of Science and Technology Innovation Fund (No. 2016QDL-B26).

References

  1. D.A.R. Mackay, G.J. Simandl, Geology, market and supply chain of niobium and tantalum – A review, Miner. Deposita 49, 1025–1047 (2014) [CrossRef] [Google Scholar]
  2. B. Zhang, X.X. Xue, X.W. Huang, H. Yang, G.J. Chen, Study on recycling and leaching valuable elements from Bayan Obo tailings, Metall. Res. Technol. 116, 1–11 (2019) [Google Scholar]
  3. N. Nowak, T. Cardinal, F. Adamietz, M. Dussauze, V. Rodriguez, R.L. Durivault, C. Deneuvilliers, J.E. Poirier, Influence of niobium and titanium introduction on optical and physical properties of silicate glasses, Mater. Res. Bull. 48, 1376–1380 (2013) [Google Scholar]
  4. H. Park, J.Y. Park, G.H. Kim, I. Sohn, Effect of TiO2 on the viscosity and slagstructure in blast furnace type slags, Steel Res. Int. 83, 150–156 (2012) [CrossRef] [Google Scholar]
  5. J.L. Li, Q.F. Shu, K.C. Chou, Structural study of glassy CaO-SiO2-CaF2-TiO2 slags by Raman spectroscopy and MAS-NMR, ISIJ Int. 54, 721–727 (2014) [CrossRef] [Google Scholar]
  6. K.X. Jiao, J.L. Zhang, Z.Y. Wang, C.L. Chen, Y.X. Liu, Effect of TiO2 and FeO on the viscosity and structure of blast furnace primary slags, Steel Res. Int. 87, 1–9 (2016) [CrossRef] [Google Scholar]
  7. C. Feng, M.S. Chu, J. Tang, J. Qin, F. Li, Z.G. Liu, Effects of MgO and TiO2 on the viscous behaviors and phase compositions of titanium-bearing slag, Int. J. Miner., Metall. Mater. 23, 868–880 (2016) [CrossRef] [Google Scholar]
  8. C.B. Shi, D.L. Zheng, S.H. Shin, J. Li, J.W. Cho, Effect of TiO2 on the viscosity and structure of low-fluoride slag used for electroslag remelting of Ti-containing steels, Int. J. Miner., Metall. Mater. 24, 18–24 (2017) [CrossRef] [Google Scholar]
  9. H.X. Gao, Z.C. Wang, S.Z. Wang, Properties and structure of niobosilicate glasses, J. Non-Cryst. Solids 112, 332–335 (1989) [CrossRef] [Google Scholar]
  10. J.H. Lopes, A. Magalhaes, I.O. Mazali, C.A. Bertran, Effect of niobium oxide on the structure and properties of melt-derived bioactive glasses, J. Am. Ceram. Soc. 97, 3843–3852 (2014) [Google Scholar]
  11. I.P. Alekseeva, V.V. Golubkov, O.S. Dymshits, A.A. Zhilin, V.I. Petrov, M.Y. Tsenter, Spectroscopic and X-ray diffraction investigations into the specific features of crystallization of potassium niobium silicate glasses, Glass Phys. Chem. 30, 311–319 (2004) [CrossRef] [Google Scholar]
  12. L.F. Santos, L. Wondraczek, J. Deubener, R.M. Almeida, Vibrational spectroscopy study of niobium germanosilicate glasses, J. Non-Cryst. Solids 353, 1875–1881 (2007) [CrossRef] [Google Scholar]
  13. D.Y. Wang, B.J. Yan, S.C. Du, Determination of activities of niobium in Cu-Nb melts containing dilute Nb, Metall. Mater. Trans. B 46, 533–536 (2015) [CrossRef] [Google Scholar]
  14. S.J. Jeong, T.S. Kim, J.H. Park, Relationship between sulfide capacity and structure of MnO-SiO2-Al2O3-Ce2O3 system, Metall. Mater. Trans. B 48, 545–553 (2016) [CrossRef] [Google Scholar]
  15. Z.W. Zhao, X.X. Chen, B. Glaser, B.J. Yan, Experimental study on the thermodynamics of the CaO-SiO2-Ce2O3 system at1873 K, Metall. Mater. Trans. B 50, 395–406 (2019) [CrossRef] [Google Scholar]
  16. J.H. Park, Structure–property correlations of CaO–SiO2–MnO slag derivedfrom Raman spectroscopy, ISIJ Int. 52, 1627–1636 (2012) [CrossRef] [Google Scholar]
  17. X.Y. Fan, J.L. Zhang, K.X. Jiao, R.Z. Xu, K.D. Wang, Influence of B2O3 on viscosity and structure of low MgO slag containing titanium, Metall. Res. Technol. 115, 1–8 (2018) [Google Scholar]
  18. Z.J. Wang, Q.F. Shu, S. Sridhar, M. Zhang, M. Guo, Z.T. Zhang, Effect of P2O5 and FetO on the viscosity and slag structure in steelmaking slags, Metall. Mater. Trans. B 46, 758–765 (2015) [CrossRef] [Google Scholar]
  19. J.X. Gao, G.H. Wen, T. Huang, B.W. Bai, P. Tang, Q. Liu, Effect of slag-steel reaction on the structure and viscosity of CaO-SiO2-based mold flux during high-Al steel casting, J. Non-Cryst. Solids 452, 119–124 (2016) [CrossRef] [Google Scholar]
  20. U. Balachandran, N.G. Eror, Raman spectrum of the high temperature form of Nb2O5, J. Mater. Sci. Lett. 1, 374–376 (1982) [Google Scholar]
  21. Y.G. Zhao, P. Zhang, Effects of lanthanides on structural and dielectric properties of NdNbO4-LnNbO4 ceramics, Ceram. Int. 44, 1935–1941 (2018) [Google Scholar]
  22. B.O. Mysen, P. Richet, Silicate glasses and melts, properties and structure, Elsevier, Amsterdam, 2005 [Google Scholar]
  23. J. Qi, C.J. Liu, C. Zhang, M.F. Jiang, Effect of Ce2O3 on structure, viscosity, and crystalline phase of CaO-Al2O3-Li2O-Ce2O3 slags, Metall. Mater. Trans. B 48, 11–16 (2016) [CrossRef] [Google Scholar]
  24. Z.Y. Cai, B. Song, L.F. Li, Z. Liu, X.K. Cui, Effects of CeO2 on viscosity, structure, and crystallization of mold fluxes for casting rare earths alloyed steels, Metals 9, 333 (2019) [Google Scholar]
  25. M.T. Wang, J.S. Cheng, M. Li, F. He, Raman spectra of soda–lime–silicate glass doped with rare earth, Physica B 406, 3865–3869 (2011) [CrossRef] [Google Scholar]
  26. Y.H. Lin, L.Q. Zhang, X.L. Huang, Influence of CaF2 on the apparent viscosity of CaO-SiO2-MgO-Al2O3-TiO2 slags, Metall. Res. Technol. 114, 1–5 (2017) [Google Scholar]
  27. L.B. Deng, X.F. Zhang, M.X. Zhang, X.L. Jia, Effect of CaF2 on viscosity, structure and properties of CaO-Al2O3-MgO-SiO2 slag glass ceramics, J. Non-Cryst. Solids 500, 310–316 (2018) [CrossRef] [Google Scholar]
  28. J.H. Park, K.Y. Ko, T.S. Kim, Influence of CaF2 on the viscosity and structure of manganese ferroalloys smelting slags, Metall. Mater. Trans. B 46, 741–748 (2014) [CrossRef] [Google Scholar]
  29. B.O. Mysen, D. Virgo, C.M. Scarfe, Relations between the anionic structure and viscosity of silicate melts – A Raman spectroscopic study, Am. Mineral. 65, 690–710 (1980) [Google Scholar]
  30. I. Sohn, D.J. Min, A review of the relationship between viscosity and the structure of calcium-silicate-based slags in ironmaking, Steel Res. Int. 83, 611–630 (2012) [CrossRef] [Google Scholar]
  31. Y.L. Zhen, G.H. Zhang, K.C. Chou, Influence of Al2O3/TiO2 ratio on viscosities and structure of CaO-MgO-Al2O3-SiO2-TiO2 melts, ISIJ Int. 54, 985–989 (2014) [CrossRef] [Google Scholar]
  32. C. Feng, J. Tang, L.H. Gao, Z.G. Liu, M.S. Chu, Effects of CaO/SiO2 on viscous behaviors and structure of CaO-SiO2-11.00 wt.% MgO-11.00 wt.% Al2O3-43.00 wt.% TiO2 slag system, ISIJ Int. 59, 31–38 (2019) [CrossRef] [Google Scholar]

Cite this article as: Zhuang Ma, Zengwu Zhao, Wentao Guo, Zhi Wang, Influence of Nb2O5 and basicity on the viscosity and structure of CaO-SiO2-Nb2O5-CeO2-CaF2 slag system, Metall. Res. Technol. 117, 307 (2020)

All Tables

Table 1

Chemical compositions of experimental slag samples (wt.%).

Table 2

Assignments of Raman bands in the spectra of CaO-SiO2-Nb2O5-CeO2-CaF2 slag system.

All Figures

thumbnail Fig. 1

XRD patterns of quenched slag samples.

In the text
thumbnail Fig. 2

Schematic diagram of experimental apparatus for viscosity measurement.

In the text
thumbnail Fig. 3

Effect of Nb2O5 addition on the viscosity of the CaO-SiO2-Nb2O5-5 wt.% CeO2-5 wt.% CaF2 slag system (C/S = 1).

In the text
thumbnail Fig. 4

Effect of basicity on the viscosity of CaO-SiO2-8 wt.% Nb2O5-5 wt.% CeO2-5 wt.% CaF2 slag system.

In the text
thumbnail Fig. 5

Natural log of viscosity vs. 104/T of CaO-SiO2-Nb2O5-5 wt.% CeO2-5 wt.% CaF2 slag system.

In the text
thumbnail Fig. 6

Activation energy for viscous flow of CaO-SiO2-Nb2O5-5 wt.% CeO2-5 wt.% CaF2 slag system.

In the text
thumbnail Fig. 7

Raman spectra of as-quenched samples from 1823 K with varying Nb2O5 content (C/S = 1).

In the text
thumbnail Fig. 8

Deconvoluted Raman spectra of as-quenched samples from 1823 K with varying Nb2O5 content (C/S = 1).

In the text
thumbnail Fig. 9

Relative fraction of Qn (0, 1, 2, 3) as a function of Nb2O5 content (C/S = 1).

In the text
thumbnail Fig. 10

Deconvoluted Raman spectra of as-quenched samples from 1823 K with varying basicity (Nb2O5 = 8 wt.%).

In the text
thumbnail Fig. 11

Relative fraction of Qn (0, 1, 2, 3) as a function of basicity (Nb2O5 = 8 wt.%).

In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.