- Email: [email protected]
Effect of wet grinding on carbothermic reduction of ilmenite concentrate
MINPRO-02727; No of Pages 6 International Journal of Mineral Processing xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Mineral Processing journal homepage: www.elsevier.com/locate/ijminpro
Effect of wet grinding on carbothermic reduction of ilmenite concentrate Bing Song a, Xuewei Lv a,⁎, Jian Xu a, Huijun Miao b, Kexi Han b a b
School of Materials Science and Engineering, Chongqing University, China Panzhihua Iron & Steel Research Institute, Panzhihua, China
a r t i c l e
i n f o
Article history: Received 6 August 2014 Received in revised form 11 February 2015 Accepted 18 February 2015 Available online xxxx Keywords: Wet grinding Ilmenite concentrate Carbothermic reduction
a b s t r a c t High titania slag has become a good raw material in the development of titanium. Ilmenite is reduced first and then smelted in an electric arc furnace to separate iron from the high titania slag. Therefore, enhanced reduction of ilmenite concentrate can reduce smelting time and energy consumption. Wet grinding method, which is often used to reduce particle size and improve the balling ability of ilmenite, was proposed to enhance the reduction. The effect of wet grinding on the carbothermic reduction of ilmenite was investigated. The wet grinding treatment was found to result in a high reduction rate and improve metallisation. The metallisation degree of reduced ilmenite concentrate increased as the time of wet grinding increased. Under the constant reduction conditions, the degree of metallisation improved from 68.58% to 87.32% while the FeO content decreased from 10.62 to 5.28% as the wet grinding time increased from 10 to 60 min. © 2015 Elsevier B.V. All rights reserved.
1. Introduction As natural rutile and high-grade titania mineral resources decrease worldwide, ilmenite has become one of the important raw materials for the titanium industry because of its titania content (Wu and Zhang, 2006; Pistorius and Coetzee, 2003). Grades of ilmenite contain 45 to 65.8 mass percent TiO2 and ilmenite is regarded as a significant resource in the production of rutile, which can be used directly as a pigment to manufacture titanium (Deng and Luo, 1998; Miller, 1957). However, ilmenite is a type of low-grade titania ore. Therefore, ilmenite is commonly preferred to enrich titania as high titania slag and is processed in an electric arc furnace (EAF) to fully utilise it (Zhao and Guo, 2005; Pourabdoli et al., 2006). Separating iron from ilmenite is an expensive process (Chen et al., 1997) because of the long reduction and smelting time, which results in high power consumption of usually 2000 kW h to 2500 kW h for each tonnage of titania slag. In the past decades, a considerable number of studies (Tripathy et al., 2012; Wang et al., 2008; Kucukkaragoz and Eric, 2006; Gupta et al., 1989) have reported on ilmenite reduction. However, enhancement of ilmenite reduction is less studied. The possible technical routes for enhancing ilmenite reduction can be classified into two methods,
⁎ Corresponding author. E-mail addresses: [email protected] (B. Song), [email protected] (X. Lv), [email protected] (J. Xu), [email protected] (H. Miao), [email protected] (K. Han).
namely, addition of additives to raw materials and mechanical activation treatment methods. For example, Gupta et al. (1989) examined the effect of ferric chloride (FeCl3) addition to the reduction of ilmenite. Adding FeCl3 to ilmenite–graphite mixtures significantly increased the reduction rate. No reduction occurred in the absence of FeCl3, whereas the reaction occurred rapidly at 1273 K with 10% FeCl3. Run et al., 2013 and Ranganathan et al. (2012) reported on the effect of ferrosilicon addition to the reduction of ilmenite. The metallisation ratio and the particle size of iron in the reduced samples increased as Fe–Si amount and reduction time increased. If molten Fe–Si alloy was present during reduction, it could enhance the rate of reduction and the agglomeration of iron. Huang et al. (2004) studied the effect of wet grinding on the pelletising process and the pellets. The strength of wet, preheated and fired pellets was all improved. Chen et al. (1997) reported the results of carbothermic reduction of ilmenite by mechanical activation. After ball milling an ilmenite–carbon mixture at room temperature, the ilmenite was reduced to rutile and metallic iron during subsequent low-temperature annealing. A long milling time lowed to the reduction temperature and increased the reduction rate further. High milling intensity also lowed to the reduction temperature further. Welham and Williams (1999) reported that milling was a non-equilibrium process, in which physical energy was transferred into a powder by crystalline damage, the formation of defects and localised heating during impact, thereby increasing the enthalpy and entropy of the system. The low strength of ilmenite pellet, long reduction time, slow reduction rate and high energy consumption are the major problems in the
http://dx.doi.org/10.1016/j.minpro.2015.02.014 0301-7516/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
2
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
Table 1 Main chemical components of ilmenite concentrate (wt.%). TiO2
Fe2O3
FeO
CaO
MnO
MgO
SiO2
Al2O3
V2O5
45.64
6.53
36.45
1.12
0.855
3.22
3.65
1.02
b0.10
smelting of Panzhihua ilmenite with the EAF processes. Few studies have reported the effect of wet grinding on the carbothermic reduction of ilmenite concentrate. This study investigated the effect of mechanical activation on carbothermic reduction of ilmenite concentrate. 2. Experimental procedures 2.1. Raw materials Ilmenite concentrate was supplied by Panzhihua Iron and Steel (Group) Co. The chemical composition and particle size distribution of the concentrate are listed in Tables 1 and 2, respectively. The X-ray diffraction pattern of ilmenite concentrate showed that FeTiO3, Fe3O4 and MgTiO3 were the main minerals, as showed in Fig. 1. Coke was used as a reducing agent; its composition is shown in Table 3. The average particle size of coke was approximately 74 μm. 2.2. Experimental apparatus and methods 5000 g of ilmenite concentrate was mixed with excess amount of coke (14%mass% of ilmenite) in each experiment to ensure sufficiently reduction of iron oxides in the concentrate. Previous finding (Huang et al., 2012) suggested that coke was important to provide sufficient carbon for full reduction of iron oxides in the ilmenite concentrate. The grinding machine was a laboratory scale mill made of rubber liners, with an inner diameter of 0.5 m and an inner volume of 0.1 m3. Wet grinding was performed by steel balls, which average size was approximately 30 mm. The ball movement was controlled by adjusting the rotation speed. The rotation speed was set at 48 rpm in the experiments. The mixtures of ilmenite concentrate and coke with 4% moisture were ground for 4, 10, 30 and 60 min at room temperature. After wet grinding, the ilmenite concentrate powder was dried at 393 K for 2 h. The size distribution, specific surface area, surface morphology and lattice distortion of ilmenite concentrate were then measured. The size distribution and the average particle size were determined using the OMEC particle analyser (PIP9.1). The specific surface area of powders was determined using a Gemini VII 2390 surface area analyser with N2 gas at liquid nitrogen temperature, in which the sample was degassed at 200 °C under vacuum for 1 h before measurement. A TESCAN VEGA 2 scanning electron microscope (SEM) was employed to study the powder morphology. The mixture of powdered ore and coke was then briquetted by using a mould press; each briquette was approximately 30 mm in size and 50 g in weight. The briquettes were dried before being used in the experiment. The samples were placed in a corundum crucible, which was then placed in a vertical tube furnace and introduced into the isothermal zone of the furnace once the desired temperature was reached. Each sample was reduced for 30 min at 1653 K under an argon atmosphere. A schematic of the reduction experimental apparatus is shown in Fig. 2. After reduction, the briquettes were broken into two halves, and the cross section was observed under SEM (TESCAN VEGA 2 SEM, 15 kV beams) and optical microscopy (50 iPOL). The Table 2 Particle size distribution of ilmenite concentrate (raw material). Particle size (μm)
+150 −150–+110 −110–+75 −75–+44 −44–+37 −37
Content (wt.%)
0.9
2.4
23.8
44.5
12.8
15.6
Fig. 1. XRD pattern of Panzhihua ilmenite concentrate.
cross section surface of the samples is platted and rinsed in acetone by an ultrasonic cleaner and then dried before SEM observation, but the cross section surface of sample needs polishing process for optical microscopy analysis. For optical microscopy, samples mounted in ‘Metset’ mounting plastic were ground on silicon carbide papers to 800 grades and polished successively with 6, 3, and 1 μm ‘Hyprez’ diamond lapping compound. Chemical and thermo gravimetric analyses were performed on powder samples. The TFe, Fe2+ and metallic iron (MFe) contents of reduced sample were analysed using wet chemical method. The metallisation degree of the reduced samples was defined as the following equation:
Metallisation ¼
MFe 100% TFe
ð1Þ
where, TFe and MFe are the contents of total iron and metallic iron in the reduced samples, respectively. The thermo gravimetric analysis was conducted using a Shimadzu TA 50I in flowing high-purity argon dried by passing through a column of magnesium perchlorate. The gas flow rate was 80 mL/min. The samples were supported on a platinum pan suspended in the isothermal zone of the reactor, which was purged with argon for 45 min before being placed in the hot furnace. A sample which weighed 40 mg was heated at a rate of 20 K/min, and the temperature was raised to approximately 1400 °C. The weight of the sample and the adjacent temperature were continuously recorded. The temperature was not constant instead it was ramped up to 1400 °C. 3. Results and discussion 3.1. Wet grinding 3.1.1. Particle size and specific surface area after wet grinding The effect of grinding time on the particle size distribution of ilmenite concentrate is shown in Fig. 3. As the grinding interval increased, the particle size distribution shifted to the left, which means that the particle size decreased gradually. When the grinding time is more than 10 min; a second sub-population peak appears at approximately 100 μm or above, which probably resulted from the agglomeration of fine particles. Table 3 Proximate analysis and S and P contents of coke. Component
Fixed carbon
Ash
Volatile
S
P
wt.%
83.66
14.12
2.22
0.65
0.125
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
3
Fig. 4. Average grain size and specific surface area of test mixtures after grinding.
time, then changed slightly when the wet grinding time was more than 10 min. Fig. 2. Experimental apparatus for the pellet reduction tests.
The volume average particle size and specific surface area of ilmenite concentrate particles at different wet grinding times are shown in Fig. 4. The average particle size decreased as the wet grinding interval increased and then changed slightly when the grinding time was above 30 min. The average grain size was reduced from 50.13 μm to 30.24 μm after 60 min grinding. These results agreed with previous results in the literature (Zheng et al., 1996; Zhu et al., 2007). The grinding media (steel ball) under the combined action between the friction of the damp mill scaleboard and the centrifugal force were lefted to a certain height, falling behind to break the materials. Therefore, the large particles can be broken, and obtained particles with small size are considered reasonable. During the grinding tests, the specific surface area of particles increased from 0.23 m2/g to 0.49 m2/g as the wet grinding time increased. The effect of wet grinding on the surface characteristics of concentrate may affect the surface area more drastically. The increase in the specific surface area corresponded to the decrease in particle size. The increase in the surface area of ilmenite concentrate particles produced by wet grinding was considered to increase the contact area between the two and enable even distribution of coke in the concentrate.
3.1.3. Surface morphology The surface morphologies of the particles after different times of wet grinding are shown in Fig. 6. The micrographs are shown at the same magnification for (a) to (e) and (f) at a higher magnification, and the particle size gradually decreased as the wet grinding time increased. The surface morphology became rougher as the wet grinding time increased, as evidenced in the SEM image of the particles after 60 min grinding. The agglomerates made of fine particles or resulting from the coating of fine particles on coarser particles can be observed when grinding time reached 60 min. The agglomerates are easily identifiable in the SEM photographs because of their brightness given that they present a great number of facets compared with coarse particles (Garcia et al., 2002). 3.2. Carbothermic reduction
3.1.2. Microstrain and crystalline size The microstrain and crystal size under different wet grinding times are shown in Fig. 5. The microstrain increased and the crystalline size decreased from 868 nm to 615 nm with the increasing wet grinding
3.2.1. Metallisation degree of reduced ilmenite samples The MFe and FeO contents of the samples were analysed by using wet chemical methods. The results of the chemical analyses are shown in Fig. 7. Under the constant reduction conditions, the metallization degree of ilmenite sample increased as the wet grinding time increased, and the metallization degree can reach up to 87.32% at 60 min of grinding time. The metallisation is significantly improved when the wet grinding time was increased from 0 min to 10 min. This finding suggests that wet grinding can likely improve the subsequent reduction process of ilmenite concentrate because the higher particle surface area can result in more active sites of the surface. The phenomenon can also
Fig. 3. Particle size for ores after wet grinding.
Fig. 5. Microstrain and grain size after wet grinding.
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
4
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
Fig. 6. SEM of ilmenite concentrate after wet grinding at different times (times for the top set, L to R, are: 0, 4 and 10 min from (a) to (c). Times for the lower set, L to R, are 30 and 60 (d) to (e), respectively.).
Fig. 7. Metallisation and the FeO content of the reduced sample after different times of wet grinding.
serve as explained by the high surface energy which can decrease the activation energy for the reduction. Fig. 7 also demonstrates that the content of FeO decreased from 10.62% to 5.28% as the wet grinding time increased from 0 min to 60 min. With the increasing wet grinding time, the degree of reduction was increased slowly. Fig. 7 also shows that when the wet grinding time is over 30 min, the metallization and FeO content without larger change trend.
The optical microscope images of the sample reduced from the concentrates ground at different wet grinding times are shown in Fig. 8. More than 5 continuous pictures of each sample were taken with an optical microscope to reduce error as much as possible. The observation areas were randomly distributed in the top, bottom, left, middle and right of a sample section. The particle size of the pictures was automatically analysed using the Image J software. The white spots are metallic iron, which clearly indicates that the average iron particle size increased as the pretreatment time of wet grinding increased under the constant reduction conditions. The average iron particle size in the reduced samples is shown in Fig. 9. The iron particle size in the samples reduced from the concentrate which underwent wet grinding for 60 min was significantly larger than that in the other wet grinding times, which showed no obvious change when the wet grinding time was 4 min or 10 min. The metallic iron particle size can reach 104 μm at 60 min of wet grinding because the carbothermic reduction of ilmenite concentrate was accelerated after wet grinding. The grey region is the oxide phase, and the black region shows the pores. 3.2.2. Scanning electron microscope The SEM images of ilmenite concentrates which were reduced 30 min at 1653 K are shown in Fig. 10. The MFe after reduction from ilmenite concentrate increased as the wet grinding time is increased. The metal iron aggregated on the surface of oxide phases and the particle size of metal iron increased its range from 20 μm to 110 μm. These results are attributed to the increase of the specific surface area of ilmenite concentrate and the contact area of reduction reaction of
Fig. 8. Optical microscope images of the reduced samples (pretreatment times, from L to R, are 0, 4, 10, 30 and 60 min, respectively).
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
Fig. 9. Average particle size of iron in the reduced samples with different wet grinding times.
ilmenite concentrate. Therefore, the carbothermic reduction of ilmenite concentrate was accelerated, and the amount of reduction of metallic iron increased. When wet grinding time was 0 min, the metallic iron after reduction formed few and smaller aggregates on the surface of the oxide phase in Fig. 10(a). However, Fig. 10(b) to (d) shows that after reduction, most of the iron particles which are present are spherical, a few exist in the form of strips and the iron grain surface is smooth. When wet grinding time was 60 min, part of the iron phase already existed in the form of an area. The oxide phase softened and the liquid phase part was produced. This finding indicated that the ilmenite concentrate was pre-treated by wet grinding, and the temperature of the reduction reaction of ilmenite concentrate reduced during
5
Fig. 11. TG curves of reduced ilmenite concentrate at different wet grinding times.
subsequent carbothermic reduction processes. The reduced temperature is further discussed in the next section. 3.2.3. Thermoanalysis Thermal Gravity analysis (TG) and Differential Thermal Gravity (DTG) curves of carbothermic-reduced ilmenite concentrate following different wet grinding times are shown in Figs. 11 and 12, respectively. Fig. 11 shows that the mass loss was nearly unchanged from the startup to about 1100 °C. This result is considered because the reduction reaction is a slow solid–solid reaction in this temperature range, as given by Eq. (2). The Boudouard reaction proceeded at higher temperatures, with carbon monoxide being the main constituent in the gas
Fig. 10. SEM of reduction pellets. (Times for the top set, L to R, are 0, 4 and 10 min; times for the lower set, L to R, are 30 and 60 min. Points 1 and 2 are the metal iron phase and oxide phase, respectively.).
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
6
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
decrease from 50.13 μm to 30.24 μm after 60 min. The specific surface area of the material can increase from 0.23 m2/g to 0.49 m2/g as the wet grinding time increased. (2) Wet grinding can help improve the reduction rate of ilmenite concentrate during subsequent carbothermic reduction processes. The metallization of iron in the solid state reduced sample generally increases by increasing the wet grinding time. The metallisation of iron with grinding can reach 87.32% when the wet grinding time is 60 min, whereas the content of FeO decreased from 10.62% to 5.28% by increasing time of grinding. (3) The maximum mass loss rate can increase as the wet grinding time increased. The temperature that is needed to achieve a given maximum mass loss can decrease by grinding; the extent of the temperature reduction is 3, 8, 16 and 45 °C for grinding times of 4, 10, 30 and 60 min, respectively. The period of reduction can be shortened after wet grinding. Fig. 12. DTG curves of reduced ilmenite concentrate at different wet grinding times.
Acknowledgment phase. Therefore, the reaction was considered mainly a gas–solid reaction at this stage and the mass loss rate significantly improved. The process is reflected in the equations represented in Eq. (3) (El-Guindy and Davenport, 1970). The effect of wet grinding time was more noticeable when the temperature approached 1100 °C; the mass loss increased as the wet grinding time increased. The reduction reaction can proceed quickly because the small particle size increases the contact area between ilmenite concentrate and coke. 2FeTiO3 þ C ¼ 2Fe þ 2TiO2 þ CO2 ;
ð2Þ
C þ CO2 ¼ 2CO and FeTiO3 þ CO ¼ Fe þ TiO2 þ CO2 :
ð3Þ
Fig. 12 shows the relationship between the mass loss rate (%/min) and the temperature during the reaction process. The temperature that is needed to achieve a given mass loss rate can be reduced by wet grinding; the extent of the temperature reduction was 3, 8, 16 and 45 °C, respectively, for the four grinding times given in Fig. 12. At a given temperature, the mass loss increased as the wet grinding time increased as shown in Fig. 11; the change in the trend of the weightlessness of ilmenite concentrate is consistent with the findings shown in Fig. 11. This result suggests that wet grinding can improve the rate of the reduction reaction; the reduction process of ilmenite concentrate is accelerated and the reduction time shortened. Compared with the results of TG and the reduction experiment, the mass loss increased as the wet grinding time increased from 0 min to 4 min, as shown in Fig. 11. The metallization of reduced samples is also significantly increased, as illustrated in Fig. 7. In contrast, the maximum mass loss occurred in 60 min of wet grinding time, during which the metallisation also reached its maximum. The mass loss and metallisation mainly depend on Eqs. (2) and (3). Therefore, the results of TG agree with the experimental results. 4. Conclusions After the wet grinding of ilmenite concentrate at room temperature, ilmenite concentrate was reduced during subsequent carbothermic reduction processes with coke. The enhanced reduction process in the pre-treatment of wet grinding samples was attributed mainly to thorough mixing of coke with disordered ilmenite nanocrystallites during grinding. The conclusions can be summarized as following: (1) Wet grinding can reduce the particle size and increase the specific surface area of ilmenite concentrate. The average grain size can
The authors are especially grateful to the Major Program of National Nature Science Foundation of China (Grant no. 2013CB632604) for providing financial support for this work. The chemical compositions of all the samples were analysed by Panzhihua Iron & Steel Research Institute, Pan Gang Group. References Chen, Y., Hwang, T., Marsh, M., Willuams, J.S., 1997. Mechanically activated carbothermic reduction of ilmenite. Metall. Mater. Trans. A 28, 1115–1121. Deng, G.Z., Luo, F.C., 1998. Titanium Metallurgy. Metallurgical Industry Press, Beijing, pp. 56–57. El-Guindy, M.I., Davenport, W.D., 1970. Kinetics and mechanism of ilmenite reduction with graphite. Metall. Mater. Trans. B 1 (6), 1729–1734. Garcia, F., Le Bolay, N., Frances, C., 2002. Changes of surface and volume properties of calcite during a batch wet grinding process. Chem. Eng. J. 85, 177–187. Gupta, S.K., Rajakumar, V., Grieveson, P., 1989. The influence of weathering on the reduction of ilmenite with carbon. Metall. Trans. B 20 (5), 735–745. Huang, Z.C., Li, Q., Yang, Y.B., Jiang, T., 2004. Effect of wet grinding of mix on quality of iron ore concentrate pellet. J. Cent. South Univ. Sci. Technol. 35 (5), 753–758. Huang, R., Lv, X.W., Bai, C.G., Zhang, K., Qiu, G.B., 2012. Solid state and smelting reduction of a Panzhihua ilmenite concentrate with coke. Can. Metall. Q. 51 (4), 434–439. Kucukkaragoz, C.S., Eric, R.H., 2006. Solid state reduction of a natural ilmenite. Miner. Eng. 19 (3), 334–337. Miller, J.A., 1957. Titanium: A Materials Survey. Bureau of Mines Information Circular, pp. 202–210. Pistorius, P.C., Coetzee, C., 2003. Physicochemical aspects of titanium slag production and solidification. Metall. Mater. Trans. B 34 (5), 581–588. Pourabdoli, M.A.H.D.I., Raygan, Sh., Abdizadeh, H., Hanaei, K., 2006. Production of high titania slag by electro-slag crucible melting (ESCM) process. Int. J. Miner. Process. 78 (3), 175–181. Ranganathan, S., Bhattacharyya, K.K., Ray, A.K., Godiwalla, K.M., 2012. Investigations on the reduction and growth of particles of iron from ilmenite ore. Miner. Process. Extr. Metall. 121 (1), 55–63. Run, H., Lv, X.W., Bai, C.G., Zhang, K., Qiu, G.B., 2013. Enhancement reduction of panzhihua ilmenite concentrate with coke and conglomeration of metal with ferrosilicon. Steel Res. Int. 84 (9), 892–899. Tripathy, M., Ranganathan, S., Mehrotra, S.P., 2012. Investigations on reduction of ilmenite ore with different sources of carbon. Miner. Process. Extr. Metall. 121 (3), 147–155. Wang, Y.M., Yuan, Z.F., Guo, Z.C., Tan, Q.Q., Li, Z.Y., Jiang, W.Z., 2008. Reduction mechanism of natural ilmenite with graphite. Trans. Nonferrous Metals Soc. China 18 (4), 962–968. Welham, N.J., Williams, J.S., 1999. Carbothermic reduction of ilmenite and rutile. Metall. Mater. Trans. B 30 (6), 1075–1081. Wu, X., Zhang, J., 2006. Geographical distribution and characteristics of titanium resources in China. Titan. Ind. Prog. 23 (26), 8–12. Zhao, P., Guo, P.Y., 2005. New technology of producing titanium slag by reducing ilmenite at low temperature. Iron Steel Vanadium Titan. 26 (2), 1–4. Zheng, S.L., Hao, B.H., Mao, J.F., 1996. Study on mechanochemical effect of powdery quartz in dry and wet grinding. J. Wuhan Univ. Technol. 11 (2), 1–6. Zhu, De-Qing, Xu, Xiao-feng, Ou, Ying-qin, Zhou, Wei, Pan, ian, 2007. One-step direct reduction of damp milled magnetite concentrate pellets. Iron Steel (10), 6–11.
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
Contents lists available at ScienceDirect
International Journal of Mineral Processing journal homepage: www.elsevier.com/locate/ijminpro
Effect of wet grinding on carbothermic reduction of ilmenite concentrate Bing Song a, Xuewei Lv a,⁎, Jian Xu a, Huijun Miao b, Kexi Han b a b
School of Materials Science and Engineering, Chongqing University, China Panzhihua Iron & Steel Research Institute, Panzhihua, China
a r t i c l e
i n f o
Article history: Received 6 August 2014 Received in revised form 11 February 2015 Accepted 18 February 2015 Available online xxxx Keywords: Wet grinding Ilmenite concentrate Carbothermic reduction
a b s t r a c t High titania slag has become a good raw material in the development of titanium. Ilmenite is reduced first and then smelted in an electric arc furnace to separate iron from the high titania slag. Therefore, enhanced reduction of ilmenite concentrate can reduce smelting time and energy consumption. Wet grinding method, which is often used to reduce particle size and improve the balling ability of ilmenite, was proposed to enhance the reduction. The effect of wet grinding on the carbothermic reduction of ilmenite was investigated. The wet grinding treatment was found to result in a high reduction rate and improve metallisation. The metallisation degree of reduced ilmenite concentrate increased as the time of wet grinding increased. Under the constant reduction conditions, the degree of metallisation improved from 68.58% to 87.32% while the FeO content decreased from 10.62 to 5.28% as the wet grinding time increased from 10 to 60 min. © 2015 Elsevier B.V. All rights reserved.
1. Introduction As natural rutile and high-grade titania mineral resources decrease worldwide, ilmenite has become one of the important raw materials for the titanium industry because of its titania content (Wu and Zhang, 2006; Pistorius and Coetzee, 2003). Grades of ilmenite contain 45 to 65.8 mass percent TiO2 and ilmenite is regarded as a significant resource in the production of rutile, which can be used directly as a pigment to manufacture titanium (Deng and Luo, 1998; Miller, 1957). However, ilmenite is a type of low-grade titania ore. Therefore, ilmenite is commonly preferred to enrich titania as high titania slag and is processed in an electric arc furnace (EAF) to fully utilise it (Zhao and Guo, 2005; Pourabdoli et al., 2006). Separating iron from ilmenite is an expensive process (Chen et al., 1997) because of the long reduction and smelting time, which results in high power consumption of usually 2000 kW h to 2500 kW h for each tonnage of titania slag. In the past decades, a considerable number of studies (Tripathy et al., 2012; Wang et al., 2008; Kucukkaragoz and Eric, 2006; Gupta et al., 1989) have reported on ilmenite reduction. However, enhancement of ilmenite reduction is less studied. The possible technical routes for enhancing ilmenite reduction can be classified into two methods,
⁎ Corresponding author. E-mail addresses: [email protected] (B. Song), [email protected] (X. Lv), [email protected] (J. Xu), [email protected] (H. Miao), [email protected] (K. Han).
namely, addition of additives to raw materials and mechanical activation treatment methods. For example, Gupta et al. (1989) examined the effect of ferric chloride (FeCl3) addition to the reduction of ilmenite. Adding FeCl3 to ilmenite–graphite mixtures significantly increased the reduction rate. No reduction occurred in the absence of FeCl3, whereas the reaction occurred rapidly at 1273 K with 10% FeCl3. Run et al., 2013 and Ranganathan et al. (2012) reported on the effect of ferrosilicon addition to the reduction of ilmenite. The metallisation ratio and the particle size of iron in the reduced samples increased as Fe–Si amount and reduction time increased. If molten Fe–Si alloy was present during reduction, it could enhance the rate of reduction and the agglomeration of iron. Huang et al. (2004) studied the effect of wet grinding on the pelletising process and the pellets. The strength of wet, preheated and fired pellets was all improved. Chen et al. (1997) reported the results of carbothermic reduction of ilmenite by mechanical activation. After ball milling an ilmenite–carbon mixture at room temperature, the ilmenite was reduced to rutile and metallic iron during subsequent low-temperature annealing. A long milling time lowed to the reduction temperature and increased the reduction rate further. High milling intensity also lowed to the reduction temperature further. Welham and Williams (1999) reported that milling was a non-equilibrium process, in which physical energy was transferred into a powder by crystalline damage, the formation of defects and localised heating during impact, thereby increasing the enthalpy and entropy of the system. The low strength of ilmenite pellet, long reduction time, slow reduction rate and high energy consumption are the major problems in the
http://dx.doi.org/10.1016/j.minpro.2015.02.014 0301-7516/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
2
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
Table 1 Main chemical components of ilmenite concentrate (wt.%). TiO2
Fe2O3
FeO
CaO
MnO
MgO
SiO2
Al2O3
V2O5
45.64
6.53
36.45
1.12
0.855
3.22
3.65
1.02
b0.10
smelting of Panzhihua ilmenite with the EAF processes. Few studies have reported the effect of wet grinding on the carbothermic reduction of ilmenite concentrate. This study investigated the effect of mechanical activation on carbothermic reduction of ilmenite concentrate. 2. Experimental procedures 2.1. Raw materials Ilmenite concentrate was supplied by Panzhihua Iron and Steel (Group) Co. The chemical composition and particle size distribution of the concentrate are listed in Tables 1 and 2, respectively. The X-ray diffraction pattern of ilmenite concentrate showed that FeTiO3, Fe3O4 and MgTiO3 were the main minerals, as showed in Fig. 1. Coke was used as a reducing agent; its composition is shown in Table 3. The average particle size of coke was approximately 74 μm. 2.2. Experimental apparatus and methods 5000 g of ilmenite concentrate was mixed with excess amount of coke (14%mass% of ilmenite) in each experiment to ensure sufficiently reduction of iron oxides in the concentrate. Previous finding (Huang et al., 2012) suggested that coke was important to provide sufficient carbon for full reduction of iron oxides in the ilmenite concentrate. The grinding machine was a laboratory scale mill made of rubber liners, with an inner diameter of 0.5 m and an inner volume of 0.1 m3. Wet grinding was performed by steel balls, which average size was approximately 30 mm. The ball movement was controlled by adjusting the rotation speed. The rotation speed was set at 48 rpm in the experiments. The mixtures of ilmenite concentrate and coke with 4% moisture were ground for 4, 10, 30 and 60 min at room temperature. After wet grinding, the ilmenite concentrate powder was dried at 393 K for 2 h. The size distribution, specific surface area, surface morphology and lattice distortion of ilmenite concentrate were then measured. The size distribution and the average particle size were determined using the OMEC particle analyser (PIP9.1). The specific surface area of powders was determined using a Gemini VII 2390 surface area analyser with N2 gas at liquid nitrogen temperature, in which the sample was degassed at 200 °C under vacuum for 1 h before measurement. A TESCAN VEGA 2 scanning electron microscope (SEM) was employed to study the powder morphology. The mixture of powdered ore and coke was then briquetted by using a mould press; each briquette was approximately 30 mm in size and 50 g in weight. The briquettes were dried before being used in the experiment. The samples were placed in a corundum crucible, which was then placed in a vertical tube furnace and introduced into the isothermal zone of the furnace once the desired temperature was reached. Each sample was reduced for 30 min at 1653 K under an argon atmosphere. A schematic of the reduction experimental apparatus is shown in Fig. 2. After reduction, the briquettes were broken into two halves, and the cross section was observed under SEM (TESCAN VEGA 2 SEM, 15 kV beams) and optical microscopy (50 iPOL). The Table 2 Particle size distribution of ilmenite concentrate (raw material). Particle size (μm)
+150 −150–+110 −110–+75 −75–+44 −44–+37 −37
Content (wt.%)
0.9
2.4
23.8
44.5
12.8
15.6
Fig. 1. XRD pattern of Panzhihua ilmenite concentrate.
cross section surface of the samples is platted and rinsed in acetone by an ultrasonic cleaner and then dried before SEM observation, but the cross section surface of sample needs polishing process for optical microscopy analysis. For optical microscopy, samples mounted in ‘Metset’ mounting plastic were ground on silicon carbide papers to 800 grades and polished successively with 6, 3, and 1 μm ‘Hyprez’ diamond lapping compound. Chemical and thermo gravimetric analyses were performed on powder samples. The TFe, Fe2+ and metallic iron (MFe) contents of reduced sample were analysed using wet chemical method. The metallisation degree of the reduced samples was defined as the following equation:
Metallisation ¼
MFe 100% TFe
ð1Þ
where, TFe and MFe are the contents of total iron and metallic iron in the reduced samples, respectively. The thermo gravimetric analysis was conducted using a Shimadzu TA 50I in flowing high-purity argon dried by passing through a column of magnesium perchlorate. The gas flow rate was 80 mL/min. The samples were supported on a platinum pan suspended in the isothermal zone of the reactor, which was purged with argon for 45 min before being placed in the hot furnace. A sample which weighed 40 mg was heated at a rate of 20 K/min, and the temperature was raised to approximately 1400 °C. The weight of the sample and the adjacent temperature were continuously recorded. The temperature was not constant instead it was ramped up to 1400 °C. 3. Results and discussion 3.1. Wet grinding 3.1.1. Particle size and specific surface area after wet grinding The effect of grinding time on the particle size distribution of ilmenite concentrate is shown in Fig. 3. As the grinding interval increased, the particle size distribution shifted to the left, which means that the particle size decreased gradually. When the grinding time is more than 10 min; a second sub-population peak appears at approximately 100 μm or above, which probably resulted from the agglomeration of fine particles. Table 3 Proximate analysis and S and P contents of coke. Component
Fixed carbon
Ash
Volatile
S
P
wt.%
83.66
14.12
2.22
0.65
0.125
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
3
Fig. 4. Average grain size and specific surface area of test mixtures after grinding.
time, then changed slightly when the wet grinding time was more than 10 min. Fig. 2. Experimental apparatus for the pellet reduction tests.
The volume average particle size and specific surface area of ilmenite concentrate particles at different wet grinding times are shown in Fig. 4. The average particle size decreased as the wet grinding interval increased and then changed slightly when the grinding time was above 30 min. The average grain size was reduced from 50.13 μm to 30.24 μm after 60 min grinding. These results agreed with previous results in the literature (Zheng et al., 1996; Zhu et al., 2007). The grinding media (steel ball) under the combined action between the friction of the damp mill scaleboard and the centrifugal force were lefted to a certain height, falling behind to break the materials. Therefore, the large particles can be broken, and obtained particles with small size are considered reasonable. During the grinding tests, the specific surface area of particles increased from 0.23 m2/g to 0.49 m2/g as the wet grinding time increased. The effect of wet grinding on the surface characteristics of concentrate may affect the surface area more drastically. The increase in the specific surface area corresponded to the decrease in particle size. The increase in the surface area of ilmenite concentrate particles produced by wet grinding was considered to increase the contact area between the two and enable even distribution of coke in the concentrate.
3.1.3. Surface morphology The surface morphologies of the particles after different times of wet grinding are shown in Fig. 6. The micrographs are shown at the same magnification for (a) to (e) and (f) at a higher magnification, and the particle size gradually decreased as the wet grinding time increased. The surface morphology became rougher as the wet grinding time increased, as evidenced in the SEM image of the particles after 60 min grinding. The agglomerates made of fine particles or resulting from the coating of fine particles on coarser particles can be observed when grinding time reached 60 min. The agglomerates are easily identifiable in the SEM photographs because of their brightness given that they present a great number of facets compared with coarse particles (Garcia et al., 2002). 3.2. Carbothermic reduction
3.1.2. Microstrain and crystalline size The microstrain and crystal size under different wet grinding times are shown in Fig. 5. The microstrain increased and the crystalline size decreased from 868 nm to 615 nm with the increasing wet grinding
3.2.1. Metallisation degree of reduced ilmenite samples The MFe and FeO contents of the samples were analysed by using wet chemical methods. The results of the chemical analyses are shown in Fig. 7. Under the constant reduction conditions, the metallization degree of ilmenite sample increased as the wet grinding time increased, and the metallization degree can reach up to 87.32% at 60 min of grinding time. The metallisation is significantly improved when the wet grinding time was increased from 0 min to 10 min. This finding suggests that wet grinding can likely improve the subsequent reduction process of ilmenite concentrate because the higher particle surface area can result in more active sites of the surface. The phenomenon can also
Fig. 3. Particle size for ores after wet grinding.
Fig. 5. Microstrain and grain size after wet grinding.
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
4
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
Fig. 6. SEM of ilmenite concentrate after wet grinding at different times (times for the top set, L to R, are: 0, 4 and 10 min from (a) to (c). Times for the lower set, L to R, are 30 and 60 (d) to (e), respectively.).
Fig. 7. Metallisation and the FeO content of the reduced sample after different times of wet grinding.
serve as explained by the high surface energy which can decrease the activation energy for the reduction. Fig. 7 also demonstrates that the content of FeO decreased from 10.62% to 5.28% as the wet grinding time increased from 0 min to 60 min. With the increasing wet grinding time, the degree of reduction was increased slowly. Fig. 7 also shows that when the wet grinding time is over 30 min, the metallization and FeO content without larger change trend.
The optical microscope images of the sample reduced from the concentrates ground at different wet grinding times are shown in Fig. 8. More than 5 continuous pictures of each sample were taken with an optical microscope to reduce error as much as possible. The observation areas were randomly distributed in the top, bottom, left, middle and right of a sample section. The particle size of the pictures was automatically analysed using the Image J software. The white spots are metallic iron, which clearly indicates that the average iron particle size increased as the pretreatment time of wet grinding increased under the constant reduction conditions. The average iron particle size in the reduced samples is shown in Fig. 9. The iron particle size in the samples reduced from the concentrate which underwent wet grinding for 60 min was significantly larger than that in the other wet grinding times, which showed no obvious change when the wet grinding time was 4 min or 10 min. The metallic iron particle size can reach 104 μm at 60 min of wet grinding because the carbothermic reduction of ilmenite concentrate was accelerated after wet grinding. The grey region is the oxide phase, and the black region shows the pores. 3.2.2. Scanning electron microscope The SEM images of ilmenite concentrates which were reduced 30 min at 1653 K are shown in Fig. 10. The MFe after reduction from ilmenite concentrate increased as the wet grinding time is increased. The metal iron aggregated on the surface of oxide phases and the particle size of metal iron increased its range from 20 μm to 110 μm. These results are attributed to the increase of the specific surface area of ilmenite concentrate and the contact area of reduction reaction of
Fig. 8. Optical microscope images of the reduced samples (pretreatment times, from L to R, are 0, 4, 10, 30 and 60 min, respectively).
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
Fig. 9. Average particle size of iron in the reduced samples with different wet grinding times.
ilmenite concentrate. Therefore, the carbothermic reduction of ilmenite concentrate was accelerated, and the amount of reduction of metallic iron increased. When wet grinding time was 0 min, the metallic iron after reduction formed few and smaller aggregates on the surface of the oxide phase in Fig. 10(a). However, Fig. 10(b) to (d) shows that after reduction, most of the iron particles which are present are spherical, a few exist in the form of strips and the iron grain surface is smooth. When wet grinding time was 60 min, part of the iron phase already existed in the form of an area. The oxide phase softened and the liquid phase part was produced. This finding indicated that the ilmenite concentrate was pre-treated by wet grinding, and the temperature of the reduction reaction of ilmenite concentrate reduced during
5
Fig. 11. TG curves of reduced ilmenite concentrate at different wet grinding times.
subsequent carbothermic reduction processes. The reduced temperature is further discussed in the next section. 3.2.3. Thermoanalysis Thermal Gravity analysis (TG) and Differential Thermal Gravity (DTG) curves of carbothermic-reduced ilmenite concentrate following different wet grinding times are shown in Figs. 11 and 12, respectively. Fig. 11 shows that the mass loss was nearly unchanged from the startup to about 1100 °C. This result is considered because the reduction reaction is a slow solid–solid reaction in this temperature range, as given by Eq. (2). The Boudouard reaction proceeded at higher temperatures, with carbon monoxide being the main constituent in the gas
Fig. 10. SEM of reduction pellets. (Times for the top set, L to R, are 0, 4 and 10 min; times for the lower set, L to R, are 30 and 60 min. Points 1 and 2 are the metal iron phase and oxide phase, respectively.).
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014
6
B. Song et al. / International Journal of Mineral Processing xxx (2015) xxx–xxx
decrease from 50.13 μm to 30.24 μm after 60 min. The specific surface area of the material can increase from 0.23 m2/g to 0.49 m2/g as the wet grinding time increased. (2) Wet grinding can help improve the reduction rate of ilmenite concentrate during subsequent carbothermic reduction processes. The metallization of iron in the solid state reduced sample generally increases by increasing the wet grinding time. The metallisation of iron with grinding can reach 87.32% when the wet grinding time is 60 min, whereas the content of FeO decreased from 10.62% to 5.28% by increasing time of grinding. (3) The maximum mass loss rate can increase as the wet grinding time increased. The temperature that is needed to achieve a given maximum mass loss can decrease by grinding; the extent of the temperature reduction is 3, 8, 16 and 45 °C for grinding times of 4, 10, 30 and 60 min, respectively. The period of reduction can be shortened after wet grinding. Fig. 12. DTG curves of reduced ilmenite concentrate at different wet grinding times.
Acknowledgment phase. Therefore, the reaction was considered mainly a gas–solid reaction at this stage and the mass loss rate significantly improved. The process is reflected in the equations represented in Eq. (3) (El-Guindy and Davenport, 1970). The effect of wet grinding time was more noticeable when the temperature approached 1100 °C; the mass loss increased as the wet grinding time increased. The reduction reaction can proceed quickly because the small particle size increases the contact area between ilmenite concentrate and coke. 2FeTiO3 þ C ¼ 2Fe þ 2TiO2 þ CO2 ;
ð2Þ
C þ CO2 ¼ 2CO and FeTiO3 þ CO ¼ Fe þ TiO2 þ CO2 :
ð3Þ
Fig. 12 shows the relationship between the mass loss rate (%/min) and the temperature during the reaction process. The temperature that is needed to achieve a given mass loss rate can be reduced by wet grinding; the extent of the temperature reduction was 3, 8, 16 and 45 °C, respectively, for the four grinding times given in Fig. 12. At a given temperature, the mass loss increased as the wet grinding time increased as shown in Fig. 11; the change in the trend of the weightlessness of ilmenite concentrate is consistent with the findings shown in Fig. 11. This result suggests that wet grinding can improve the rate of the reduction reaction; the reduction process of ilmenite concentrate is accelerated and the reduction time shortened. Compared with the results of TG and the reduction experiment, the mass loss increased as the wet grinding time increased from 0 min to 4 min, as shown in Fig. 11. The metallization of reduced samples is also significantly increased, as illustrated in Fig. 7. In contrast, the maximum mass loss occurred in 60 min of wet grinding time, during which the metallisation also reached its maximum. The mass loss and metallisation mainly depend on Eqs. (2) and (3). Therefore, the results of TG agree with the experimental results. 4. Conclusions After the wet grinding of ilmenite concentrate at room temperature, ilmenite concentrate was reduced during subsequent carbothermic reduction processes with coke. The enhanced reduction process in the pre-treatment of wet grinding samples was attributed mainly to thorough mixing of coke with disordered ilmenite nanocrystallites during grinding. The conclusions can be summarized as following: (1) Wet grinding can reduce the particle size and increase the specific surface area of ilmenite concentrate. The average grain size can
The authors are especially grateful to the Major Program of National Nature Science Foundation of China (Grant no. 2013CB632604) for providing financial support for this work. The chemical compositions of all the samples were analysed by Panzhihua Iron & Steel Research Institute, Pan Gang Group. References Chen, Y., Hwang, T., Marsh, M., Willuams, J.S., 1997. Mechanically activated carbothermic reduction of ilmenite. Metall. Mater. Trans. A 28, 1115–1121. Deng, G.Z., Luo, F.C., 1998. Titanium Metallurgy. Metallurgical Industry Press, Beijing, pp. 56–57. El-Guindy, M.I., Davenport, W.D., 1970. Kinetics and mechanism of ilmenite reduction with graphite. Metall. Mater. Trans. B 1 (6), 1729–1734. Garcia, F., Le Bolay, N., Frances, C., 2002. Changes of surface and volume properties of calcite during a batch wet grinding process. Chem. Eng. J. 85, 177–187. Gupta, S.K., Rajakumar, V., Grieveson, P., 1989. The influence of weathering on the reduction of ilmenite with carbon. Metall. Trans. B 20 (5), 735–745. Huang, Z.C., Li, Q., Yang, Y.B., Jiang, T., 2004. Effect of wet grinding of mix on quality of iron ore concentrate pellet. J. Cent. South Univ. Sci. Technol. 35 (5), 753–758. Huang, R., Lv, X.W., Bai, C.G., Zhang, K., Qiu, G.B., 2012. Solid state and smelting reduction of a Panzhihua ilmenite concentrate with coke. Can. Metall. Q. 51 (4), 434–439. Kucukkaragoz, C.S., Eric, R.H., 2006. Solid state reduction of a natural ilmenite. Miner. Eng. 19 (3), 334–337. Miller, J.A., 1957. Titanium: A Materials Survey. Bureau of Mines Information Circular, pp. 202–210. Pistorius, P.C., Coetzee, C., 2003. Physicochemical aspects of titanium slag production and solidification. Metall. Mater. Trans. B 34 (5), 581–588. Pourabdoli, M.A.H.D.I., Raygan, Sh., Abdizadeh, H., Hanaei, K., 2006. Production of high titania slag by electro-slag crucible melting (ESCM) process. Int. J. Miner. Process. 78 (3), 175–181. Ranganathan, S., Bhattacharyya, K.K., Ray, A.K., Godiwalla, K.M., 2012. Investigations on the reduction and growth of particles of iron from ilmenite ore. Miner. Process. Extr. Metall. 121 (1), 55–63. Run, H., Lv, X.W., Bai, C.G., Zhang, K., Qiu, G.B., 2013. Enhancement reduction of panzhihua ilmenite concentrate with coke and conglomeration of metal with ferrosilicon. Steel Res. Int. 84 (9), 892–899. Tripathy, M., Ranganathan, S., Mehrotra, S.P., 2012. Investigations on reduction of ilmenite ore with different sources of carbon. Miner. Process. Extr. Metall. 121 (3), 147–155. Wang, Y.M., Yuan, Z.F., Guo, Z.C., Tan, Q.Q., Li, Z.Y., Jiang, W.Z., 2008. Reduction mechanism of natural ilmenite with graphite. Trans. Nonferrous Metals Soc. China 18 (4), 962–968. Welham, N.J., Williams, J.S., 1999. Carbothermic reduction of ilmenite and rutile. Metall. Mater. Trans. B 30 (6), 1075–1081. Wu, X., Zhang, J., 2006. Geographical distribution and characteristics of titanium resources in China. Titan. Ind. Prog. 23 (26), 8–12. Zhao, P., Guo, P.Y., 2005. New technology of producing titanium slag by reducing ilmenite at low temperature. Iron Steel Vanadium Titan. 26 (2), 1–4. Zheng, S.L., Hao, B.H., Mao, J.F., 1996. Study on mechanochemical effect of powdery quartz in dry and wet grinding. J. Wuhan Univ. Technol. 11 (2), 1–6. Zhu, De-Qing, Xu, Xiao-feng, Ou, Ying-qin, Zhou, Wei, Pan, ian, 2007. One-step direct reduction of damp milled magnetite concentrate pellets. Iron Steel (10), 6–11.
Please cite this article as: Song, B., et al., Effect of wet grinding on carbothermic reduction of ilmenite concentrate, Int. J. Miner. Process. (2015), http://dx.doi.org/10.1016/j.minpro.2015.02.014