3.1 Reduction Mechanism
In Table 7, it can be seen that the reduction effect of iron oxide in pellets was better and the reduction rate was reached more than 95%. The main mineral components in red mud are goethite, hematite, gibbsite, kaolinite, colloidal silica, quartz, diaspora and so on. During the reduction process, the reduced materials can be regarded as a system composed of monomer oxides such as Fe2O3, A12O3 and SiO2. In the reduction process of these materials, not only the reduction phase transition of iron-containing oxides, but also solid-state reaction among oxides. In the reduction process, the main reactions between carbon and iron oxides in waste cathode and anode carbon powder are as follows [5]:
Waste cathode and anode carbon powder were used as reducing agents, and carbon gasification reaction (8) would occur if the carbon amount was excessive. Therefore, the direct reduction of iron oxide was carried out with the participation of CO. During the whole reaction, the solid-solid reaction (1) ~ (3) between green balls and reducing agent was quite small, while the reaction mostly was gas-solid reaction (4) ~ (7), that is, the reducing agent first occurred oxidation reaction to generate CO, which reacted with iron oxide. Finally, the equilibrium control of reduction atmosphere was realized through disproportionation reaction (8). Compounds such as calcium-iron olivine and iron spinel were generated during the reduction process. These compounds could form a series of low-melting mixtures, which would make the reduction process more complicated and difficult. It has been proved that adopting a high reduction temperature of 1180℃ was beneficial to the reduction.
The total iron content in the reduced pellets reached about 55~58% which was about 47% in the raw materials. The metallization rate is above 95% (Figure 3), that is, almost all iron were reduced. According to the weight balance, it showed that only the iron oxide has been reduced to metallic iron. All the results showed that the reduction effect was ideal. The other components in red mud and waste cathode were relatively stable even at high temperature. There was no vapor volatilization, which further verified the rationality and stability of the cooperative treatment of red mud, waste cathode and anode carbon by pyrometallurgy. The technical route is suitable for the treatment of hazardous wastes, such as waste cathode and anode carbon powder, which contains fluoride and other harmful substances [8].
3.2 Blending Method of Carbon
The reduction effect showed that waste cathode and anode carbon powder can be directly used as reducing agent at high temperature. Carbon plays the main role of reducing and has good reactivity at high temperature [6], which was beneficial to the reduction of iron oxides. The content of C and S in the green ball has a great influence on the quality of subsequent products. Fig.3 showed the comparison of the content of C and S after reduction of different carbon blending method.
The internal carbon blending was one kind of green ball made by directly blending red mud with anode carbon with a weight charging ratio of 20%. Fig.4 showed after reduction the contents of C and S were 8.8% and 0.89% respectively, corresponding to external blending, which blending red mud ball with anode carbon were 0.266% and 0.021% respectively. The content of C and S in the metalized pellets was quite different that internal blending was about 40 times higher than external blending.
In the internal blending sample, it showed that anode carbon powder was directly added to the pellet and was the main reason which caused the higher of sulfur content. Since petroleum coke were the main raw material for producing cathode and anode carbon brick, which were basically treating at a high temperature of 1200℃. Thus, in the cathode and anode most of the residue inorganic sulfur were stable and would not volatilize and affect the product quality. After undergoing high temperature reduction, sulfur still stayed in the metallized pellets. So the quality of the internal blending carbon pellets was badly affected the cost of the subsequent process.
However, for the external blending, due to the natural physical space isolation, sulfur was still remained in a separate reducing agent, so the S content in the metallized pellets affected by the carbon blending method was relatively low, which did not affect the use by the subsequent process. This also showed that the sulfur in all raw materials was mainly inorganic sulfur, which has been solidified in the raw materials, i.e., there was unnecessary to add desulfurizer for gas desulfurization.
In addition, although the metallization rate of internal blending was relatively higher, it was badly affected the iron content of about 5% lower than external blending due to the entrainment of ash and other unbeneficed components in the anode carbon. Therefore, considering comprehensively, pellet quality reduced by the external blending was obviously better than the internal blending. It is more reasonable to choose the method of external blending to treat the red mud without additional desulfurizer for solidification and desulfurization.
3.3 Melting Separation Affected by Iron Content
Considering the green pellets, since it has the disadvantage of lower iron content and small bulk density, it was uneasy to conduct electricity and heat in the experimental scale furnace. Therefore, pellets need to be briquetted [7]. Firstly, no extra flux iron block was charging in the melting process. After melting, the block was relatively loose, and the slag and iron were basically not separated (S1). In fact, it was a mixture of slag and iron. The main reason was that the quantity of molten iron was too small form a big molten bath and the quantity of slag was relatively large. Under laboratory conditions, the slag volume was quite large that made it difficult for slag to float up and separate naturally. Therefore, it was necessary to improve the molten pool conditions.
A pure iron block was successfully obtained with 30% mass charging of the iron block (S2). The separation effect of slag and iron was good and there was an obvious slag-iron separation interface. The iron block was compact and the slag block was basically compact, which indicated that the slag phase floated more fully during melting process and achieved the purpose of slag-iron separation (Figure 5).
The composition of the molten iron block was shown in Table 8. From the component analysis, the iron content reached 98.85% and the carbon content is 0.13%. It was actually the composition of steel, that is, molten steel was obtained. In the high metallization rate pellets after direct reduction, it showed that only iron was simple substance, while other components were mainly oxides with high melting point. During melting process, most of iron melted and entered into the molten iron , a few numbers of oxides such as Si and Mn were reduced and melted into molten iron by the carbon brought from the pellets, and other oxides basically entered into the slag. The quality of the molten steel was so good that various target steel products can be obtained after controlling the content of S, P and other components through appropriate refining process, thus greatly simplifying the production process from comprehensive utilization of red mud to end products.
Table 8
Main Components of Melted Iron (wt %)
Sample Number
|
Fe
|
C
|
Si
|
Mn
|
P
|
S
|
S2
|
98.85
|
0.13
|
0.58
|
0.04
|
0.027
|
0.26
|
The composition of the slag after melting and separation was shown in Table 9. The iron content in the slag was less than 3%. It was estimated that the iron recovery rate in sample S2 was 96.5% which indicated the iron loss was few. A large amount of alkali metal and fluoride originated from raw material have been solidified in the slag, that is, the raw materials were stable and safe after being processed by the pyrotechnic process. In addition, the aluminum compound contained in the slag was as high as about 37%, which had good economic recycling value and can be returned to the alumina extraction process through blending, thus to realize the fully use of everything by self-circulation and zero emission within the enterprise.
Table 9
Main Components of Slag (wt %)
Sample Number
|
Al2O3
|
SiO2
|
TiO2
|
CaO
|
Fe2O3
|
MgO
|
Na2O
|
F
|
S2
|
37
|
9.2
|
8.5
|
7.7
|
3.3
|
1.6
|
17
|
13
|