Influence of TiO 2 and basicity on viscosity of Ti bearing slag

Abstract

The viscosities of CaO–SiO₂–7MgO–TiO₂–12Al₂O₃ (wt-%) slags (CaO/SiO₂ = 0·5–0·9, wt-% TiO₂ = 15–30) were investigated to promote understanding of the effect of TiO₂ and basicity (CaO/SiO₂) on the viscous behaviour of slags containing TiO₂. In practice, these experimental results are significant for the blast furnace (BF) processing of vanadium–titanium magnetite ore and for controlling the selective crystallisation of Ti bearing BF slags. The slag viscosity was found to decrease with increasing TiO₂ content at a fixed basicity. The degree of viscosity decrease with increasing TiO₂ content diminishes in the slag with higher basicity or higher TiO₂ content. An increase in the basicity lowers the viscosity of slags. The apparent activation energy of viscous flow of slags decreased with additions of TiO₂ and increasing basicity. The Urbain model was empirically modified based on the present experimental data, which can be used to predict not only quinary Ti bearing slags but also quaternary and low Ti bearing slags.

Introduction

China is rich in titanium resources, and 95% of them exist as vanadium–titanium magnetite ore.1 Generally, there are two metallurgical processes, i.e. blast furnace (BF) ironmaking and electric furnace process, which are adopted to extract metal from the vanadium–titanium magnetite ore. In the BF process, most of the Ti component is concentrated into the BF slag (Ti-BF slag) (22–25%TiO₂).1 ^– 3 During this process, the viscosity of slag containing TiO₂ is an important parameter that influences the gas permeability, the ability to tap the slag from the reactor, the efficiency of separation of slag from metal, the extent of slag foaming, etc.4 The viscosity of the BF slag has a wide range, depending on the temperature and slag composition. In previous works, numerous viscosity measurements have been carried out for binary, ternary and quaternary slags;5 ^– 13 however, the viscosity of Ti bearing slag was measured comparatively less, and the measurements were mainly focused on the low TiO₂ (<10 wt-%) containing slags.14 ^– 17 Thus, viscosity measurements for Ti bearing slags are required to enhance the performance of the BF process of vanadium–titanium magnetite ore reduction.

Additionally, China has accumulated >70 million tonnes of Ti-BF slag, and it is still increasing at a rate of >3 million tonnes every year.1 ^– 3 Because of the dispersed distribution of the Ti component in various mineral phases, with very fine grains (<10 μm) and complex interfacial combinations, it is difficult to efficiently recover the Ti component through traditional separation processes, such as flotation separation, melting reduction and hydrometallurgy.¹ 1,3 Li et al. 18 and Zhang et al. 3 proposed that selective precipitation was the most promising method to recover the Ti component from the Ti bearing slag. In this method, Ti bearing BF slags were modified and cooled down in a controllable cooling rate to enrich the Ti component into anosovite or perovskite, and these Ti enriched phases were separated using the gravity separation method. The selective precipitation of slags is determined by many variables, such as chemical composition, cooling rate, transport mechanism for both mass and heat and interfacial kinetics.19 Among these variables, the transport phenomenon has a decisive effect on the precipitation event, which is mainly controlled by the viscosity of the slag. However, to the knowledge of the present authors, the viscosity containing high TiO₂ content has not been systematically studied so far. From above, it is necessary to measure the viscosity of slags containing high TiO₂ content and to create a suitable viscosity model for Ti bearing slags at a wide range of temperatures.

In the present study, CaO–SiO₂–7MgO–TiO₂–12Al₂O₃ (wt-%) quinary slag systems with varying basicity and TiO₂ content were designed based on the industrial BF slags from Panzhihua Iron & Steel Corporation (as shown in Table 1).20 The viscosities of these slags were then measured. Based on these experimental data, a model suitable for predicting the viscosity of Ti bearing slags was proposed.

Table 1.

Chemical composition of BF slag from Panzhihua Iron and Steel Corporation20

CaO	SiO₂	MgO	Al₂O₃	TiO₂	Others
22–25	22–26	7–9	13–17	22–25	FeO, V₂O₅

Experimental

Preparation of materials

The materials used in the present work are listed in Table 2. The samples were prepared by mixing analytical reagent grade chemicals of CaO, SiO₂, MgO, TiO₂ and Al₂O₃. The slags were designed as shown in Table 3. The basicity was set as 0·5, 0·7 and 0·9 respectively. TiO₂ was fixed at 15, 20, 25 and 30 wt-% respectively. The mixtures, ∼160 g for each sample, were premelted in a vertical tubular furnace (GSL-06-16LA; Precondar) for 30 min at 1773 K. The liquid slag was subsequently quenched and crushed for further examination. The chemical compositions of the slag samples were analysed before the experiment by X-ray fluorescence (XRF) spectroscopy (S4-Explorer; Bruker) to confirm the composition. The results are also included in Table 3. It can be seen that the measured values are very close to the designed chemical compositions.

Table 2.

Oxides used in present study

Material	Purity/%	Supplier
Calcium oxide (CaO)	⩾99·9	Sigma Aldrich
Silicon oxide (SiO₂)	⩾99·8	Beijing Yili Fine Chemical Co., Ltd
Alumina (Al₂O₃)	⩾99·5	Xilong Chemical
Magnesium oxide (MgO)	⩾98·5	Sinopharm Chemical Reagent Co., Ltd
Titania (TiO₂)	⩾99·0	Sinopharm Chemical Reagent Co., Ltd

Table 3.

Experimental compositions and measured values from present experiment

No.	Initial composition/wt-%						Final composition by XRF/wt-%
	C/S	SiO₂	CaO	Al₂O₃	MgO	TiO₂	C/S	SiO₂	CaO	Al₂O₃	MgO	TiO₂
1	0·5	44·00	22·00	12	7	15	0·50	45·00	22·05	12·21	7·06	13·68
2	0·5	40·67	20·33	12	7	20	0·50	41·28	20·79	12·33	6·94	18·66
3	0·5	37·33	18·67	12	7	25	0·52	37·33	19·49	12·33	7·29	23·56
4	0·5	34·00	17·00	12	7	30	0·52	34·05	17·92	12·32	7·27	28·44
5	0·7	38·82	27·18	12	7	15	0·72	38·62	27·87	12·31	7·20	14·00
6	0·7	35·88	25·12	12	7	20	0·74	35·31	26·33	12·04	7·13	19·19
7	0·7	32·94	23·06	12	7	25	0·74	32·60	24·34	12·80	7·34	23·44
8	0·7	30·00	21·00	12	7	30	0·76	29·36	22·53	11·93	7·13	29·05
9	0·9	34·74	31·26	12	7	15	0·97	33·63	32·81	11·92	7·16	14·47
10	0·9	32·11	28·89	12	7	20	0·97	31·28	30·36	11·92	7·18	19·26
11	0·9	29·47	26·53	12	7	25	0·97	28·75	27·94	12·13	7·12	24·06
12	0·9	26·84	24·16	12	7	30	0·97	26·39	25·62	12·05	7·36	28·58

Apparatus and procedure

The experimental apparatus for viscosity measurement is shown in Fig. 1. The Brookfield digital viscometer head was connected to the working spindle by an alumina shaft. The crucible and spindle used in the experiments were made of molybdenum in order to prevent slag contamination. The dimensions of the Mo crucible and spindle are listed in Table 4. An electric resistance furnace with MoSi₂ heating elements was employed for viscosity measurements. The temperature was monitored by the Pt–10Rh/Pt thermocouple, which was positioned below the crucible. It should be mentioned that the temperature was carefully measured to obtain the zone of constant temperature before the viscosity measurement, and the crucible and thermocouple were seated at this zone during the experiments. The viscometer was calibrated using standard oil of known viscosity at room temperature.

Figure 1.

Experimental apparatus for measurement of slag viscosity

Table 4.

Dimensions of each component adapted in present study

Suspension wire	Material	Alumina
	Wire diameter/mm	4·5
	Wire length/cm	45
Spindle	Material	Molybdenum
	Shaft diameter/mm	5
	Shaft length/mm	15
	Bob diameter/mm	10
	Bob length/mm	20
	Degree of tapper/°	120
Crucible	Material	Molybdenum
	Inner diameter/mm	38
	Inner depth/mm	64

The slag (140 g) was held in a Mo crucible. The Mo crucible containing the slag sample was placed in the reaction chamber of the resistance furnace. Argon gas (0·1 NL min⁻¹) was flown into the reaction chamber to prevent the oxidation of the crucible and the spindle. The furnace was heated to 1823 K at the rate of 5 K min⁻¹ and held for >100 min to completely melt the slag and stabilise the temperature before the viscosity measurement. Each measurement was performed during the cooling cycle with an equilibration time of 30 min at each temperature to ensure sufficient thermal equilibration within the liquid slag. Three different rotating speeds (150, 175 and 200 rev min⁻¹) were employed at each temperature to ensure that the viscosity was independent of the rotating speed. It should be mentioned that the variation in the viscosity values due to the changes in the rotation speeds was <3%. The average value was adopted as the viscosity value in the present experiment. The XRF analysis of the slag after the experiment was also explored for slag no. 1, and it was found that there was no pick-up of Mo elements during the experiment.

Results and discussion

Viscosity of CaO–SiO₂–MgO–TiO₂–Al₂O₃ slags

In order to confirm the reproducibility of viscosity measurements, a repeated experiment for sample no. 1 was carried out. Figure 2 shows the comparative results. It can be noted that the experimental results matched very well, and the deviation is relatively small (<2%). Considering the experimental uncertainties usually associated with viscosity measurement, the method of measurement of viscosity was credible. In the present study, the effects of basicity and TiO₂ content on the viscosity of the CaO–SiO₂–MgO–TiO₂–Al₂O₃ quinary slag system were investigated based on this method.

Figure 2.

Comparison of viscosity changes between original experiment and repeated result for sample 1

The results of the viscosity measurements are presented in Table 5. The viscosity data in this table are the average values obtained using three different rotation speeds. The deviation of the experimental data from the mean value is <3% for all the measurements. In Table 5, a clear trend, as expected, was observed, i.e. the viscosity is increasing with decreasing temperature.

Table 5.

Measured viscosity values of various Ti bearing slags

1	Temperature/°C	1510	1480	1450	1427	1408	1400	1385	1357
	Viscosity/Pa s	0·401	0·557	0·642	0·755	0·909	1·108	1·395	1·755
2	Temperature/°C	1506	1479	1447	1424	1408	1397	1385	1357
	Viscosity/Pa s	0·275	0·359	0·421	0·491	0·579	0·692	0·852	1·067
3	Temperature/°C	1505	1477	1447	1426	1408	1400	1385	1357
	Viscosity/Pa s	0·225	0·248	0·291	0·372	0·438	0·517	0·638	0·784
4	Temperature/°C	1507	1479	1456	1437	1433	1408	1384	1357
	Viscosity/Pa s	0·180	0·200	0·240	0·291	0·349	0·412	0·491	0·656
5	Temperature/°C	1507	1477	1457	1448	1424	1408	1383	1357
	Viscosity/Pa s	0·273	0·299	0·337	0·412	0·491	0·596	0·731	0·911
6	Temperature/°C	1505	1475	1449	1434	1428	1408	1382	1357
	Viscosity/Pa s	0·217	0·228	0·261	0·309	0·370	0·440	0·538	0·649
7	Temperature/°C	1510	1478	1462	1451	1433	1408	1382	1357
	Viscosity/Pa s	0·169	0·206	0·231	0·269	0·302	0·353	0·434	0·508
8	Temperature/°C	1500	1483	1476	1455	1430	1408	1378	1357
	Viscosity/Pa s	0·140	0·149	0·170	0·191	0·218	0·254	0·307	0·384
9	Temperature/°C	1500	1470	1448	1434	1425	1408	1382	1357
	Viscosity/Pa s	0·204	0·232	0·273	0·303	0·338	0·395	0·483	0·599
10	Temperature/°C	1506	1491	1472	1456	1431	1408	1379	1357
	Viscosity/Pa s	0·149	0·179	0·202	0·228	0·269	0·326	0·415	0·498
11	Temperature/°C	1517	1500	1479	1455	1430	1408	1378	1357
	Viscosity/Pa s	0·137	0·151	0·168	0·195	0·225	0·272	0·332	0·389
12	Temperature/°C	1508	1473	1454	1444	1433	1408	1383	1357
	Viscosity/Pa s	0·123	0·142	0·141	0·158	0·188	0·210	0·258	0·303

Effect of TiO₂ on viscosity of Ti bearing slags

Figure 3 shows the effect of TiO₂ on the viscous behaviour of CaO–SiO₂–7MgO–TiO₂–12Al₂O₃ (wt-%) quinary slags with varying basicity, where the TiO₂ content was varied between 15 and 30 wt-% at each basicity that increases from 0·5 to 0·9. It can be noted that the slag viscosity decreases with increasing TiO₂ content at each basicity. This is consistent with previous works,^{8
14} 8,14,15 although the chemical compositions are different. Saito et al. 8 have investigated the effect of TiO₂ on the viscosity of 40CaO–40SiO₂–20Al₂O₃ (wt-%) slags. The viscosity of these quaternary slags decreased with an increase in the content of additive TiO₂. Shankar et al. 14 studied the viscosities of CaO–SiO₂–MgO–Al₂O₃ and CaO–SiO₂–MgO–Al₂O₃–TiO₂ slag systems. They found that at high basicity (∼0·8), the slag viscosity decreased with a small amount of TiO₂ (∼2 wt-%) addition in the slag. Handfield et al. 15 also investigated the influence of TiO₂ addition and temperature (1573–1873 K) on BF type slags containing 1–1·2 wt-%TiO₂ in both homogeneous liquid melts and slag melts containing solid particles. These measurements showed that TiO₂ reduces the viscosity of slags at a given temperature. It can therefore be concluded from the above results that TiO₂ may decrease the viscosity of the slag melt.

Figure 3.

Viscosity of CaO–SiO₂–7MgO–TiO₂–12Al₂O₃ (wt-%) system as function of TiO₂ concentration from 1793 to 1623 K with a basicity = 0·5, b basicity = 0·7 and c basicity = 0·9

The temperature dependence of the viscosity is usually expressed by Weymann–Frenkel’s equation (1) where A is a proportionality constant, E _w is the apparent activation energy for viscous flow, R is the gas constant and T is the absolute temperature. The apparent activation energy for slags E _w represents the frictional resistance for viscous flow, and the variations in the apparent activation energy can suggest a change in the structure of the molten slag or more directly a change in the cohesive flow units comprising the slag structure.10 Based on the experimental data, the apparent activation energies for viscous flow of Ti bearing slags were evaluated according to Weymann–Frenkel’s equation. The results are presented in Fig. 4. It can be noted that the apparent activation energy decreases with both increments of basicity and TiO₂ content, which indicates that a simpler silicate network unit for viscous flow was formed, consequently reducing the frictional resistance, i.e. the degree of polymerisation of the slags decreases with increasing TiO₂ content. This is similar to the behaviour of basic oxide in the slag melt. It may suggest that TiO₂ may primarily behave as a basic oxide in the present study.

Figure 4.

Effect of a TiO₂ concentration and b basicity on apparent activation energy of viscous flow of slags

Figure 5 shows the viscosity of slag as a function of basicity and TiO₂ content at 1630 and 1681 K respectively. The slag viscosity decreases by increasing the TiO₂ content at a given temperature. It is observed that the degree of viscosity decrease with increasing TiO₂ addition diminishes with increasing basicity, which is similar to a basic oxide component acting upon a network modifier in the slag melt.21 As the basic components in the slag melt exceed the acidic components, the network of molten slag has already been depolymerised into simpler complexes. Under this condition, further addition of the basic component (such as TiO₂) has a limited effect on the viscosity. This phenomenon was also investigated by previous works.¹⁶ 16,17 Sommerville and Bell16 reported that TiO₂ lowers the viscosity of CaO–SiO₂–MgO–Al₂O₃–TiO₂ slag, and the degree of viscosity decrease is lower at higher basicity. Morinaga et al. 17 investigated the viscosity of the CaO–SiO₂–TiO₂ system, and they suggested that the effect of TiO₂ in lowering the viscosity should be greater at lower basicity levels. Although it needs more evidence to verify whether this behaviour is relevant to the basic property of TiO₂, yet it is sure that the effect of TiO₂ in lowering the viscosity decreases with increasing basicity of the slag. In Fig. 5, it can also be noted that the viscosity decrease diminishes with increasing TiO₂ content.

Figure 5.

Effect of TiO₂ concentration on viscosity of CaO–SiO₂–7MgO–TiO₂–12Al₂O₃ (wt-%) at a 1630 K and b 1681 K

Effect of basicity on viscosity

Figure 4b shows the apparent activation energy for viscous flow versus basicity. It was found that the apparent activation energy decreased with increasing basicity, suggesting that the structural units for viscous flow of slags become smaller and simpler with increasing basicity. In general, it is well known that an increase in basicity results in the decrease in slag viscosity because the silicate structure changes from a three-dimensional network to chains or rings as basic oxides (such as MgO and CaO) increase. For example, the depolymerisation reaction of [Si₃O₉]⁶⁻ ring units is described by the following equations9 (2) (3) (4) where M²⁺ denotes cations such as Ca²⁺ and Mg²⁺. With increasing amount of basic oxides, the depolymerisation reaction such as in equations (3) and (4) results in the formation of [Si₂O₇] dimer (NBO/Si = 3) and [SiO₄] tetrahedral (NBO/Si = 4). It indicates that the size of the silicate flow unit becomes smaller with increasing content of basic oxides, resulting in decreasing slag viscosity.

Estimation of viscosity

A number of models for predicting the viscosity of metallurgical slags have been developed in the past, such as Riboud,22 Urbain,22 NPL23 and Iida et al. 24 These theoretical models are generally classified into four types, i.e. statistical mechanical theory, hard sphere theory, theory of corresponding states and semiempirical and empirical models. Among these models, the semiempirical and empirical models have been widely developed over the past decades, which are generally expressed in the form of the Arrhenius equation or Weymann–Frenkel’s equation (equation (1)). The estimated viscosities of slags containing TiO₂ have been evaluated using these various models, such as Riboud and Urbain models, which are suitable for the prediction of viscosity of slags containing TiO₂. It should be pointed out that these two models are based on Weymann–Frenkel’s equation, and TiO₂ was treated as network former in the Riboud model, while TiO₂ was treated as network modifier in the Urbain model. Thus, the efforts in the present study were focused predominantly on the above two models. Figure 6 shows the comparison of the experimental data with the viscosities calculated by Riboud and Urbain models. As indicated, the agreements between the estimated viscosities by the aforementioned models and measured values show a large discrepancy. The prediction model proposed by Urbain was observed to predict a consistent trend for the slag containing different TiO₂ contents, which may be due to that TiO₂ was classified as a modifier (basic oxide). This is consistent with the previous analysis, as shown in the section on ‘Effect of TiO₂ on viscosity of Ti bearing slags’. It should be pointed out that the Urbain model is based on the CaO–Al₂O₃–SiO₂ system, which is generally given by the following equation (5) where A and B have a relationship that can be described as (6) Here, A and B can be calculated by dividing the slag components into three categories, which are shown in the following:

Figure 6.

Viscosity estimated by models as function of measured viscosity

glass formers

ii.

modifiers

iii.

amphoterics .

The normalised values of , and can be calculated by dividing X _G, X _M and X _A by the term . Urbain has proposed that parameter B was related with and . Parameter B can be expressed as follows (7) (8) In Fig. 6b , it can be noted that there is a multiple relationship between the measured and estimated data by the Urbain model. Therefore, the authors try to modify coefficient A in equation (5). Based on the experimental data in the present study, coefficient A in the Urbain model was re-evaluated. Figure 7 shows the re-evaluated relationship between A′ and B, i.e. (9) The comparison between the experimental data in the present study and the viscosities calculated by the modified Urbain model is shown in Fig. 8. It can be noted that the predicted viscosity by the modified Urbain model was quite reasonable under the present experimental conditions. In order to confirm the accuracy of the modified Urbain model, some measured viscosities from other researchers22 were compared with the values calculated by the modified Urbain model. In Fig. 9, the chemical compositions of slags are similar to the present study. It can be noted that the estimated viscosities of the modified Urbain model and the measured value matched very well. The modified Urbain model was also used to predict the viscosity of slags containing low TiO₂ content. Figure 10 shows a comparison between the measured viscosity values by Shankar et al. 14 and the predicted values using the present Urbain model. It can be noted the modified Urbain model can predict the viscosities of low Ti bearing slags very well, i.e. the modified model can be used to predict the viscosity in quinary low Ti bearing slags. Furthermore, we investigated the applicability of the modified Urbain model in quaternary Ti bearing slags,8 as Fig. 11 shows. It can also be noted that the measured viscosities and estimated value matched very well.

Figure 7.

Relationship between −ln A and B of modified Urbain model

Figure 8.

Viscosity estimated by modified Urbain model as function of measured viscosity

Figure 9.

Viscosity estimated by modified Urbain model as function of measured viscosity

Figure 10.

Viscosity estimated by modified Urbain model as function of measured viscosity

Figure 11.

Viscosity estimated by modified Urbain model as function of measured viscosity

Conclusions

The influences of TiO₂ and basicity on the viscosity of CaO–SiO₂–MgO–Al₂O₃–TiO₂ slag system were investigated using the rotating method, and the Urbain model was modified based on the present experimental results to predict the viscosity of Ti bearing slags. The obtained results are summarised as follows.

The viscosities of these quinary slags decreased with an increase in TiO₂ content. The effect of TiO₂ on the viscosity diminishes in the slag with a higher basicity. The viscosity was also found to decrease with increasing basicity of slags.

TiO₂ may primarily act as a basic oxide and decreases the viscosity of the slag melts in the present slag system.

The modified Urbain model based on the present experimental data can predict Ti bearing BF slags in a wide component range, including low Ti bearing slags and quaternary slags.

Footnotes

Acknowledgements

The authors would like to thank Dr Y. Y. Zhang and L. M. Hou for their help during the viscosity measurements. Financial supports from the National Natural Science Foundation of China (grant no. 50902003), the National Basic Research Program (programme no. 2007CB613608) and the Key Projects in the National Science and Technology Pillar Program (programme no. 2010BAE00316) are gratefully acknowledged.

References

: ‘Theory of smelting V and Ti-magnetite by blast furnace’; 1996, Beijing, Science Press.

Zhang

Wang

Sui

: ISIJ Int., 2006, 46, 458–465.

Zhang

Sui

: Miner. Eng., 2007, 20, 684–693.

Wang

: ‘Metallurgy of iron and steel’, 2nd edn; 2005, Beijing, Metallurgical Industry Press.

Park

Kim

Min

: Metall. Mater. Trans. B, 2008, 39B, 150–153.

Park

Min

Song

: Metall. Mater. Trans. B, 2002, 33B, 723–729.

Park

Min

Song

: Metall. Mater. Trans. B, 2004, 35B, 269–275.

Saito

Hori

Nakashima

Mori

: Metall. Mater. Trans. B, 2003, 34B, 509–516.

Park

Min

Song

: ISIJ Int., 2002, 42, 344–351.

10.

Kim

Park

Min

: Steel Res. Int., 2010, 81, 17–24.

11.

Park

Min

Song

: ISIJ Int., 2002, 42, 38–43.

12.

Kim

Sohn

: ISIJ Int., 2011, 51, 1–8.

13.

Kim

Lee

Min

: ISIJ Int., 2004, 44, 1291–1297.

14.

Shankar

Lahiri

Seetharaman

: Metall. Mater. Trans. B, 2007, 38B, 911–915.

15.

Handfield

Charette

Lee

: JOM, 1972, 24, 37–40.

16.

Sommerville

Bell

: Can. Metall. Q., 1982, 21, 145–155.

17.

Morinaga

Suginohara

Yanagase

: J. Jpn Inst. Met., 1974, 48, 658–662.

18.

Wang

Zhang

: ISIJ Int., 2011, 51, 1396–1402.

19.

Turkdogan

: ‘Physical chemistry of high temperature technology’, 284–285; 1983, London, Metals Society.

20.

Hai

Wen

Yin

: Express Inf. Min. Ind., 2006, 448, 21–23.

21.

Zhang

Jahanshahi

: Metall. Mater. Trans. B, 1998, 29B, 117.

22.

Mills

: ‘Slag atlas’, 2nd edn, 368; 1995, Düsseldorf, Verein Deutscher Eisenhuttenleute (VDEh).

23.

Mills

Sridhar

: Ironmaking Steelmaking, 1998, 26, 1.

24.

Iida

Sakai

Kita

Murakami

: High Temp. Mater. Processes, 2000, 19, 153.

Influence of TiO 2 and basicity on viscosity of Ti bearing slag

Abstract

Introduction

Chemical composition of BF slag from Panzhihua Iron and Steel Corporation20

Experimental

Preparation of materials

Oxides used in present study

Experimental compositions and measured values from present experiment

Apparatus and procedure

Dimensions of each component adapted in present study

Results and discussion

Viscosity of CaO–SiO2–MgO–TiO2–Al2O3 slags

Measured viscosity values of various Ti bearing slags

Effect of TiO2 on viscosity of Ti bearing slags

Effect of basicity on viscosity

Estimation of viscosity

Conclusions

Footnotes

Acknowledgements

References

Viscosity of CaO–SiO₂–MgO–TiO₂–Al₂O₃ slags

Effect of TiO₂ on viscosity of Ti bearing slags