Received 18 January 2016; accepted 17 April 2016; published 20 April 2016
The development of the economic basis of Armenia is directly related to both the creation and wide introduction of new types of materials with special properties and the development of new advanced methods for their obtaining. Such materials belong to the iron monosilicide powders showing magnetic and electric properties which can be used as an alloying additive in the production of electrical-sheet steels and silicon-containing alloys with special physico-chemical properties, as well as in microelectronics  . From that standpoint, organizing a power-saving technology for obtaining such a valuable alloying additive by utilizing the secondary wastes is an urgent problem.
A perspective method for obtaining powders is the mechanically activated self-propagating high-temperature synthesis (MASHS). The given method allows to form unique materials and alloys at a significant decrease in powder consumption by using the exothermal effects of reactions and does not require special complex equipment. By preliminary mechanical activation, it is possible to affect the structure of the reaction mixture and the SHS parameters purposefully, thus ensuring the opportunity of regulating the structure and the phase composition of the powders to be synthesized   . However, the ways of obtaining powder-like iron silicide by changing the chemical route in the mechano-activated reaction mixtures have not been practically studied. At the same time, the combustion regime ensures great possibilities in the structure variation of the synthesized materials, while the SHS implementation in powder systems eases the grinding of the synthesis products and the extraction of the required fractions of powders  -  .
2. The Statement of the Problem and the Goals of Investigation
The goal of the investigation is to obtain monophase iron silicide by preliminary mechano-chemical activation of the burden, containing a mixture of wastes of a certain composition obtained from metallurgical production: the Alaverdi copper-smelting and the Yerevan molybdenum factories, as well as NaNO3 and CaO by combined aluminothermal reduction in the SHS regime, and the study of structurization conditions of the obtained ferroalloys.
A perfect and rational way of obtaining iron silicide is the aluminothermal method, allowing to obtain that alloy by combined reduction of iron and silicon oxides present at the slags. The process of aluminothermal reduction of slags represents a SHS as a result of which, the amount of the produced power satisfies the self-arbitrary proceeding of the process and does not require heat expenditure from outside. That process can be implemented by the out-furnace method, being an advantage of this method   .
In case of need, NaNO3, whose aluminothermal reduction is exothermal and leads to the increase in the heat balance of the reduction process, can be added into the burden. As a slag-forming additive, CaO is also added into the burden.
Due to high temperatures (2500...3000 K), obtained in the SHS process, the burden seems to be boiling. Under these conditions, iron and silicon are reduced and, dissolving, form ferrosilicium, while the mixed oxides (MgO, CaO, Na2O, Al2O3) contained in the burden, form difficult-to-dissolve slags, which, after cooling is easily separated from the metallic phase. To provide such heating conditions in the thin layer of a solid substance by an external heating source of power is quite difficult or almost impossible. The extreme conditions mentioned above determine the advantage of the SHS process   .
3. Materials and Methods of Investigation
To measure the slag parameters, the converter and dump slags of the Alaverdi copper-smelting factory and the molybdenum slag of the Yerevan Pure Iron Plant were used. According to the data of the chemical analysis, the element composition (by metals) of the copper waste is the following: 44.59% Fe and 17.5% Si in the dump slags and 47.95% Fe and 10.10% Si in converter slags. The element composition of the molybdenum slag is 8.75% Fe and 32.11% Si.
Investigation of the phase composition was carried out on the X-ray Diffractometer (DRON-3) in Cukα radia- tion. X-Ray technique was carried out in continuous mode within the interval of angles 2θ = 10˚… 90˚  .
Investigation of the element composition of samples (by oxides) was conducted on the certified X-ray fluorescent spectrometer ED 2000 of “Oxford Instruments Analytical” Company (Great Britain)   .
The morphology of the cover surface was studied on the high-resolution electronic scanning microscope Mira of Tescan Company (Czech Republic) using a microroentgenospectral analyzer INCA Energy 350. To investigate the distribution of the elements over the surface, photography in the characteristic X-ray radiation was applied   .
The lower part is filled with quartz sand; the upper part is an open-top conical cap. In the middle of the experiments were carried out in the reactor (see Figure 1(a)), representing a metallic reservoir, consisting of two parts.
4.The Experiment Results and Their Discussion
Since the dump and converter slags differ from each other by the content of iron and silicon, for comparison, burdens with dump and converter slags were prepared separately in case of previously selected optimal conditions and after 60 minutes of mechanical activation in the vibromill of the type M 30.
(a) (b) (c)
Figure 1. Laboratory SHS installation (a), metallic (b) and slag (c) phases.
Table 1. Numbering of burden for SHS reduction.
The results of X-ray fluorescent analysis of samples No. 1 - 4 are introduced in Table 2.
The element composition of Sample No. 5 is as follows: Si―74.7%, Fe―13.22%, Al―1.1%, Ca―1.5%. The data in the table show that by the content of iron and silicon, almost similar data are obtained in both samples. At that, in the dump slags, the content of iron and silicon in the metallic phase is relatively higher, while in the slag- lower. At comparing the roentgenograms of samples No. 1 - 4 (see Figure 2), it is seen that the phases Si [27-1402] and FeSi Ferrsilicite [35-0822] exist in all samples. The main phase of sample No. 1 is FeSi [33-0020]―40%, then come descending quantities―Si, FeSi, Al2O3, AlFe, CaTiO3, FeTiO3, FeSi2. In Sample No. 2, the main phase is FeSi Ferrsilicite [35-0822], there are also Si [27-1402], Fe [06-0696], FeSi2, SiO2, as well as products of aluminothermal reduction of the type Al8Fe2Si, Al0.7Fe3Si0.3. The phase composition of Sample No. 3 represent FeSi Ferrsilicite [35-0822] and Si [27-1402], as well as FeSi2 Ferrodisilicite [35-0822], SiO2 [14-0654], Al2O3, Al0.5Fe3Si0.5 are also present. The phase composition of Sample No. 4 represents FeSi Ferrsilicite [35-0822] and Si [27-1402], as well as FeSi2 Ferrodisilicite [35-0822], Al0.7Fe3Si0.3. The phase composition of Sample No. 5 represents Si [27-1402], FeSi Ferrsilicite [35-0822], and Ferrodisilicite [35-0822], FeSi2∙SiO2, Al3Fe2Si3.
By the results of the phase analysis of sample No. 5 (standard ferrosilicium), it is seen (see Figure 3) that the main phases are Si [27-1402] and FeSi [35-0822] which make almost 98%. The products of aluminothermal reduction of the type Al8Fe2Si, Al0.7Fe3Si0.3 are not available.
Figure 2. The result of decoding the roentgenograms of the Sample No. 5, where: -FeSi, -Si.
Figure 3. Comparison of the roentgenograms of the slag fraction of samples No. 1, 2 and 4, where: -A2O3 Corund, -SiO2 Cristobalite, -SiO2 Quartsit.
Table 2. Element composition of samples No. 1 - 4.
The phase composition of the slag fraction of samples No. 1, 2 and 4 are practically identical and represent Al2O3 [46-1212] Corundum, SiO2 Quartz and SiO2 Cristobalite, and small quantities of FeSi, Fe3Si and Fe2O3 (Figure 4).
The phase composition of the slag fraction of samples No. 3 and 4, despite the identical composition of source mixtures, are quite different. The main phase of sample No. 4 is Al2O3 [46-1212] Corundum, as well as phases SiO2 Quartz, SiO2 Qristobalite, Fe and Al are present. In sample No. 3, the phase Al2O3 [46-1212] Corundum is not available, the main phase is FeSi, Si and AlFe3, Al0.7Fe3Si0.3, Al0.7Fe3Si0.3, are present (see Figure 4).
In Figure 5 the morphology of the standard sample No. 5 is shown.
Figure 4. Comparison of the roentgenograms of slag fractions of samples No. 3 and 4 (from bottom to top), where: -A2O3 Corund, -SiO2 Cristobalite, -SiO2 Quartsit.
(a) (b)(c) (d)
Figure 5. Morphology of the standard sample No. 5 (the light grey section is the phase containing FeSi, the dark grey section―the phase containing only Si). (a) ×200; (b) ×500; (c) ×3000 (d) ×10,000.
Figure 6 shows that there are mainly two phases: the phase FeSi (the light grey section), and the phase containing only Si (the dark grey section). It is distinctly confirmed at enlargement (see Figure 7).
In Figures 8-11, the results of investigating the morphology of samples of metallic phases 1 - 4 are introduced.
Figure 6. Morphology of sample No. 5. (a) ×5000; (b) ×20,000.
(a) (b)(c) (d)
Figure 7. Morphology of sample No. 1 (metallic phase) (the light grey section is the phase containing FeSi, the dark grey section: the phase containing only Si). (a) ×200; (b) ×500; (c) ×3000; (d) ×10,000.
(a) (b)(c) (d)
Figure 8. Morphology of sample No. 2 (metallic phase). (a) ×200; (b) ×500; (c) ×3000; (d) ×10,000.
(a) (b)(c) (d)
Figure 9. Morphology of sample No. 3 (metallic phase). (a) ×200; (b) ×500; (c) ×3000; (d) ×10,000.
(a) (b)(c) (d)
Figure 10. Morphology of sample No. 4 (metallic phase). (a) ×200; (b) ×500; (c) ×3000; (d) ×10,000.
(a) (b)(c) (d)
Figure 11. Morphology of sample No. 1 (slag phase). (a) ×200; (b) ×500; (c) ×3000; (d) ×10,000.
In Figures 11-14 the results of investigating the morphology of samples of slag phases 1 - 4 are introduced.
The images show that both the phase composition and morphology of the slag fraction of samples No. 3 and 4, despite the identic composition of the samples No. 1 and 2, essentially differ from each other, however, the main phase of all the samples are Al2O3 and SiO2.
After full recovery of FeO and SiO2, iron and silicon form a FeSi2. At the same time, if the iron is completely spent on education FeSi2, silicon remains in the excess, and forms the short-term phase Fe3Si2, Fe2Si5, and does not affect the formation of the main phase. Finally, as a result of these phases, reaction and ordering of silicon crystal lattice formed the most stable intermetalid FeSi. This is especially true for high temperatures. That’s the reason why after the final cooling is homogeneous monosilicide FeSi, its existence is confirmed by X-ray diffraction (Figures 7-10).
Summarizing the main regularities of formation of structure and phase composition of the iron-silicon powders by the SHS method, we can conclude that the preliminary mechanical activation of the reaction burden containing a mixture of waste of a particular composition obtained from metallurgical production in Alaverdi-
(a) (b)(c) (d)
Figure 12. Morphology of sample No. 2 (slag phase). (a) ×200; (b) ×500; (c) ×3000; (d) ×10,000.
(a) (b)(c) (d)
Figure 13. Morphology of sample No. 3 (slag phase). (a) ×200; (b) ×500; (c) ×3000; (d) ×10,000.
(a) (b)(c) (d)
Figure 14. Morphology of sample No. 4 (slag phase). (a) ×200; (b) ×500; (c) ×3000; (d) ×10,000.
The investigation was carried out due to the financial support of the State Science Committee of RA Ministry of Science and Education in the framework of the Armenian-Belorussian joint scientific project No. 13 RB 049. The experimental data are obtaibed in the Belorussian State Scientific and Production Association of Powder Metallurgy.