Технические науки 1

Технические науки 1.Металлургия.

TWO-COMPONENT REDUCTANT USE IN OBTAINING METALLIC CALCIUM

Falin V.V., Research officer “Vac ETO” LLC, Moscow,

e-mail: raven0607@rambler.ru,

Sukharev A.V., Ph.D. of Science, vice CEO of “Vac ETO” LLC, Moscow,

Tarasov V.P., Head of Chair Nonferrous and Gold metals institution, NUST “MISIS”, Moscow

ИСПОЛЬЗОВАНИЕ ДВУХКОМПОНЕНТНОГО ВОССТАНОВИТЕЛЯ ПРИ ПОЛУЧЕНИИ МЕТАЛЛИЧЕСКОГО КАЛЬЦИЯ

Фалин Владимир Викторович,

научный сотрудник, Фирма "Вак ЭТО" г. Москва.

E-mail:raven0607@rambler.ru

Сухарев Артем Викторович,

кандидат технических наук, заместитель генерального директора,

Фирма "Вак ЭТО" г. Москва

Тарасов Вадим Петрович,

профессор, д.т.н., зав. каф. цветных металлов и золота, НИТУ «МИСиС» г. Москва

АННОТАЦИЯ

Проведены исследования возможности повышения выхода кальция при предварительном просеве оксида кальция и алюминия в процессаx алюминотермического восстановления металлического кальция. Получена формула расчета выхода кальция в зависимости от параметров процесса. Был предложен восстановитель, состоящий из смеси 80 ат.% Al и 20 ат.% Fe, для которого микроструктура шихты после восстановления не отличается от микроструктуры шихты после стандартного алюминотермического восстановления. Для реакции металлотермического восстановления оксида кальция двухкомпонентным восстановителем предложен аналитический метод, учитывающий тепловую реакцию компонентов восстановителя, позволяющий оценивать возможный выход кальция по результатам экспериментов, проведенных при одной температуре.

Ключевые слова: кальций; восстановление оксида кальция; вакуумная печь; алюминотермическое восстановление, смачивание.

ABSTRACT

Property to increase the calcium yield as a result of advanced minus sieving (or screen sizing) of calcium oxide and aluminium in aluminothermic reducing processes were investigated. The calcium yield formula was obtained dependent from process characteristics. The mixture with 80at.% of Al and 20at% of Fe was suggested as reductant, the charge microstructure which is indifferent after the reduction from those that been provided with standard aluminothermic reduction. Analytical method is suggested to metallothermic reaction of calcium oxide by the two-component reductant that considers heat reaction of reducing components enable to measure possible calcium yield by experimental results accumulated at single temperature.

Key words: calcium, calcium oxide reduction, vacuum furnace, aluminothermic reduction, wetness.

Calcium oxide aluminothermic reduction is well-known and one of basic employed industrial processes in obtaining metallic calcium [1]. As a cost of this procedure, up to 60% expenses define to the metallic reductant cost, which stimulates to search new, far cheap reductants. This requires analyzing factors influent to the dynamic of the process more attentively.

As the priority, the cost of obtained calcium, not its purity, is needed for the steel industry, a capability to produce calcium by the cheaper reductant was our investigation. The calcium was obtained in the vacuum distilled electric furnace VOzh-16-22 (fig. 1), the heating block which was prepared from the carbon-carbon composite material (CRC) of different density; there was a calcium vapors condenser cooled by water in the furnace; membrane pump that allows to provide water vapors exhaustion (at relatively low temperatures), is available in the vacuum system.

The X-ray phase analysis was performed on X-ray diffraction machine DRON-3 in the Cu_Kα. Scanning electron microscopy was evaluated on the electron microscope FEI Quanta 600 FEG with the energy disperse X-ray microanalysis attachment EDAX. The mass of samples was measured on laboratory electronic scales AJH-620CE.

Powders were mixed in the turbulence mixer C2.0 with 40rpm of frequency within 60 min.

The following materials were employed in this work: (content in % mass):

- Calcium oxide trademark Schaefer Precal from Schaefer Kalk company as the content of the calcium carbonate was ~2% and hydroxide ~1%; MgO – 0.5%; SiO₂ – 0.1%.

- Aluminium of AVP; with the aluminium content such as: Mn – 0.38%, Cu – 0.063%, Si – 0.063%, Fe – 0.31%, Mg – 2.1%, Zn – 0.06%, C – 0.023%, Ti – 0.65%; the particle size ≤ 2mm, the active aluminium content less than 98.3%

- Iron shot number 0.5 with the content of: C~4%, Si – 1.2 – 2.0% and Mn – 0.4 – 0.7%/

- Aluminium wire AD1 with diameter 1.6mm

Fig. 1 Vacuum distilled furnace VOzh-16-22

Experiments with calcium oxide of different particle content were performed by us and with aluminium wire (aluminium is chosen as explored metal-reductant).

In the famous work it was found that milling of calcium oxide increases the yield of the calcium in aluminothermic reduction [2].

The dynamics of reduction determines by interphase boundary sizes for CaO-Al(melt)-Ca(vapor) heterogeneous system whereas other parameters being equal. Within this framework, analogical experiments were conducted to study the influence of advanced sieving (or screen sizing) particles of CaO powder with Al mark of AVP using to the calcium yield value as well as the influence of administrated aluminium size to the calcium yield. The reduction was carried out in the VOzh-16-22 under following terms: the temperature is 1350⁰C, the curing time (holding) is 3 hours, residual pressure is 1-10Pa.

The dependence between the calcium yield (relative yield) and average CaO powder size (after particle separation on sizes) as a result of reduction at 1300⁰C, holding within 3 hours and residual pressure of 1-10Pa is given on the fig. 2. Contrast to these reports [2], it is found a nonmonotonic dependence of the yield on the particle size of the calcium, which is a reliable approximation value of R2 = 0.93 can be recorded as a polynominal of the 2 stage. Maximum yield was correspondent to the particle size x: 0.63 ≤ x ≤ 0.8mm.

The inverse problem to determine the influence of the aluminium yield to the calcium yield in the aluminothermic CaO reduction was suggested while comparing calcium oxide reduction with Al mark AVP and its wire AD1. The table 1 demonstrates particle size of the AVP powder and its distribution along the aluminium wire length.

Относительный выход кальция .jpg

Fig. 2 Relative calcium yield in the aluminothermic oxide reduction in dependence of particle size of CaO; the confidential interval is correspondent to the “diameter” point in the yield determination.

Table 1. Mesh-size distribution (sizes in mm) AVP, besides distribution along the wire AD1 length L, used as reductants.

	Percentage particle content,%
	1,6≤x<2,5	1,25≤x<1,6	1≤x<1,25	0,63≤x<1	0,16≤x<0,2	L≤2	2<L≤3	3<L≤5
AVP	33,3	33,6	20,9	12,1	0,1	-	-	-
AD1	-	-	-	-	-	12	35	53

Despite apparent differences in the value of aluminium reductant surfaces (powder and wire), the calcium yield was above 7% according to the mentioned mode in the case of using AVP powder (the confidence interval determination was ±1.2%) as compared to those where the same quantities of aluminium is used on wire.

Using the primary sieving (screen sizing) operation of calcium oxide is influent to the yield value thereby, as well as the size of used reductant (in the studied range).

It was demonstrated previously that the yield value depends in the wetting ability of the oxide surface by reducing agent, besides the difference in the electronegativity of the metal-reductant [3]. Since the one of basic expenses during the aluminothermic reduction is the cost, it is necessary to minimize the cost of the reductant. Such requirements are satisfied by the cast-iron, for instance.

To answer the question whether it is possible to reduce calcium oxide by the cast-iron, series of experiments were performed in the VOzh-16-22 with residual pressure of 1Pa, and the temperature of 1350 ±200C with holding within 3 hours. The sample of conventional calcium oxide reduction by aluminium was used as “evident”. The briquette content (%mass): calcium oxide 72% - 28% of the iron shot (iron); calcium oxide 76% - aluminium of 24%. The calcium yield in provided experiments was as follows: using Fe – 0.20; Al – 0.57 as reducing agents.

Following the calcium oxide reduction by the cast-iron, spherical particles till 1.5mm were found on the briquette surface. The same were located on the depth of the briquettes. The shape of these particles indicates a negligible amount of wetting in the system Fe-CaO. The cast-iron lines are visible on the diffraction pattern of briquettes (fig. 4). Consequently, a significant portion of the reducing metal was not involved in the reducing reaction.

$\\Main\homedir\Папка для полного доступа\Олег\9\08-08-2013\Изображение 023.jpg$

Fig. 3 Briquette surface photo after the Fe – reduction (cast-iron)

$\\Main\homedir\Папка для полного доступа\Фалин\Диссертация\Рис для Ивана\Fe1.jpg$

Fig. 4 Diffraction pattern of briquettes after reduction. Non-determined maximums are calcium oxide and hydroxide.

The results showed ineffective reduction of the calcium oxide by “pure” cast-iron therefore, it was analyzed a possibility to produce two-component reductant as aluminium-cast iron.

The following systems like calcium oxide – aluminium oxide – ferric oxides were studied well enough, for instances, in [4,5]. However, interface wetting issues remained outside researcher’s interests.

In the present study calcium oxide powders and reductant (R) were mixed in ratio 80%at of CaO: 20%at of R, and then pressed to briquettes. The iron shot DChL and aluminium powder AVP mixture was employed as the reductant R in numerous proportions (% at): 80Al-20Fe, 60Al-40Fe, 40Al-60Fe, 20Al-80Fe, 100Fe. The reduction mode was analogical to the aluminothermic reduction (temperature 1350⁰C, holding time within 3 hours, at residual pressure 1Pa).

The table 2 exposes data upon the calcium yield, defined as the number of produced calcium to the calcium amount in the oxide of the primary charge.

Table 2. The charge phase content after reduction and the yield from the calcium in the dependence of the reductant composition.

Reductant composition R, at.%	The charge phase content after reduction	Calcium’s yield, % (1350 ⁰C, 180 min.)
100Al	12CaO*7Al₂O₃, Al₂Ca	57
80Al – 20Fe	12CaO7Al₂O₃, 4CaO7Fe₂O₃, Al₂Ca	53
60Al -40Fe	3CaOAl₂O₃, 12CaO7Al₂O₃, CaO, CaAl_1,9O₄C_0,4, Al_0,4Fe_0,6	37
40Al – 60Fe	3CaOAl₂O₃, 12CaO7Al₂O₃, CaO, Fe₂O₃	33
20Al-80Fe	12CaO7Al₂O₃, CaAl_1,9O₄C_0,4, CaO, (Fe_0,899Al_0,101)(Al_0,899Fe_0,101)O₄	28
100Fe	CaO, Fe, Fe₂O₃	20

As the Table 2 shows (deviation in determination is ±2%), a monotonic decrease of calcium yield is observed while there is an increase of iron-cast in the reductant composition. Moreover, an exclusion of iron cast from the composition, i.e. aluminium reduction at R=16%at, gives 42%.

Under various costs of aluminium and iron-shot nearly by 5 times, employing the reductant 80%at of Al – 20%at of Fe is able to low it less than 25%, when the reductant is acquired, practically without decreasing the yield in comparison to the aluminothermic reduction.

SEM analysis of data with successively decreasing amount of iron-cast demonstrates that withdrawal and disappearance of areas are occurred (at 20 atomic % of iron and 80 atomic % of aluminium), despite the presence of local variations in the composition of the charge, where only iron or only aluminium are located, which characterize the component with low Al content.

The charge microstructure, containing 80 at.% Al and 20 at/% iron-cast (fig. 5) is similar with the structure of the charge after aluminothermic reduction (fig. 6).

рис.3.bmp

Fig. 5 Microstructure and linear structure elements composition of the charge after reduction with the 80at% Al and 20% iron-cast mixture

$C:\Documents and Settings\Minkov\Local Settings\Temporary Internet Files\Content.Word\#12_p1_17.tif$

Fig. 6 The charge microstructure after aluminothermic reduction

Previously [6], in order to explain discrepancies between the energy activation value of the reductant, defined in the yield analysis with temperature, besides heat vaporization values of the calcium and diffusion activation energy in structures similar to CaO, the formula, the conclusion which is based on dependence between the surface tension and temperature, is suggested:

(1)

Y, T, TLiq, t, denote the yield value of the calcium, reducing temperature, deterioration temperature, reducing time, respectively. n- is the constant, dependent from the stage of the process: at n = 0.5 the process is controlled by the chemical reaction, at n=0.33 the diffusion is control stage [4], if n=0, it is correspondent to depletion of reducing areas. K, Q, A denote coefficient of proportionality, heat vaporization of the calcium and the area over which the calcium evaporation is presented, respectively (the crucible area).

Comparison (1) with experimental data constitutes a quite satisfactory match for the aluminothermic reduction [6]. However, the spread between the calculated and experimental values is far tangible for the reductant that consists of aluminium and iron-cast. We hypothesized to be a necessity in energy consideration due to exothermic reaction in the aluminium and iron-cast mixture:

(2)

Employing aluminium or iron-cast as the reductant, the energetic capability of the CaO-reductant system must compensate the heat vaporization of the calcium.

Using two-component reductant, among which exothermic reaction is followed with reduction, the energy expenses are lower. In the formula (1), Q means the following:

(3)

Qvapor., denote the heat vaporization of the calcium and enthalpy alteration of the exothermic reaction (2) at the temperature T of the process, respectively.

Fig. 7 demonstrates a comparison of the yield upon experimental data and calculated according to (1) – (3), which is dependent from the reducing composition (holding time is 3 hours for all experiments). Melting point in the iron-cast (4% wt) was 1185⁰C.

a) b)

c) d)

Fig.7 Dependence of the calcium yield from the reducing temperature of CaO:

calculated values experimental values a) reductant content 80% at. Al – 20% at. Fe; b) reductant content 60% at. Al – 40% at. Fe; c) reductant content 40% at. Al – 60% at. Fe; d) reductant content 20% at. Al – 80% at. Fe; e) reductant content is 100% of Fe.Total content of briquettes is 80% at. of CaO; 20% at. of the reductant. Deviation of yield dimension is , deviation in temperature dimension is .

As results show, calculated values and experimental data are in good agreement with all types of reducing agents, except iron-cast (without aluminium) with low wetting of the calcium oxide as it was mentioned above. Consequently, ability to estimate analytically the calcium yield is limited by wetting material of the reductant on the surface of the calcium oxide in the metallothermic reduction of the CaO.

Delivered results admit to specify the following:

· Performing reduction the particle size changes of calcium oxide (sieving or screen sizing) are inefficient to the calcium yield in the same terms (average size 0.2-1.5mm). Aluminium size changes are independent to the calcium yield, raising aluminium size till 3mm decreases the calcium yield.

· Ability to replace pure aluminium on to the mixing system Al-Fe-C was studied as the content of the reductant varies from 100% of aluminium to 100% of iron-cast.

· Scanning electron microscopy and X-ray analysis demonstrated the microstructure and phase content of the charge with reductant composition 80% at. CaO and 20% at. iron-cast after reduction (the charge content 80%at CaO and 20% at. reductant) corresponds to microstructure and phase content of aluminates with aluminium content in the charge of 40%at after aluminothermic reduction

· Correction was applied and experimentally confirmed with accuracy of the yield ±3.5% the suggested analytical dependence between calcium yield and temperature in constant presence of two-component reductant during two-composition reductants research.

· Obtained results allow expenses in acquiring the reductant to lower less than 25% in comparison with aluminothermic reduction in similar yield value.

References

1. Obzor rynka kal'tsiya metallicheskogo v Rossii i mire. http://www.infomine.ru. avgust 2011 g. – 102 s.

2. Matvienko I.I., Demenev N.V. Аlyumotermicheskoe vosstanovlenie okisi kal'tsiya, Trudy instituta khimii. Ural'skij filial АN SSSR, 1958, vyp.2, str.121-131.

3. Falin V.V., Minkov A.O., Sukharev A.V., Tarasov V.P. Reductant melt boundary tension effect on metal thermal reduction of calcium oxide// Mezhdunarodnyj nauchno-issledovatel'skij zhurnal: Sbornik po rezul'tatam XXV zaochnoj nauchnoj konferentsii Research Journal of International Studies, 2014, №3, Часть 2, с.72 – 74.

4. Zotzl M., Pollmann H. Stability and properties of brownmillerites Ca₂(Al,Mn,Fe)₂O₅ and perovskites Ca(Mn,Fe)O_3-x in the system Ca₂Fe₂O₅-"Ca₂Mn₂O₅" - "Ca₂Al₂O₅"// J.Am.Ceram.Soc., 2006,v.89, № 11, p. 3491-3497-79.

5. Nikolik S., Genao G., ZHao B., KHajes P.S. Fazovoe ravnovesie v shlakovoj sisteme FeO – Fe2O3 – CaO – SiO2//Tsvetnye metally, 2011, №8/9, s. 73 – 80.

6. Falin V.V., Minkov O.B., Molev G.V. i dr. Vozmozhnost' zameny alyuminiya v proizvodstve metallicheskogo kal'tsiya vosstanovleniem ego oksida//Tsvetnye metally, 2014, №1, s.53 – 58.