'man, and V. M. Ust'yantsev UDC 666.762.453 Magnesium chromite (magnesia--chromite spinel) is the most chemically resistant compound in relation to molten iron~ilicate slags [i, 2]. Magnesium chromite is one of the basic components of the spinellide of chromite ore and is present in it in the form of a solid solution with other spinellides.The presence of impurities, particularly silicates, in chromite ore has a significant detrimental influence on the quality characteristics of the refractories obtained.Therefore, for the production of refractories with increased resistance the use of purer magnesia--chromite raw material is necessary.Such raw materials include chromite obtained by thermal decomposition of magnesium bichromate.The method of its production was developed by the Ukrainian Scientific-Research Institute for Chemistry [3].We have conducted investigations on the use of synthesized magnesium chromite for the production of magnesia--spinellide refractories.As the result of the fact that the corrosion failure of refractories by slags occurs primarily in the binder, magnesium chromite was introduced into the finely ground component of the charge.To improve sintering of the refractory compositions, titanium magnetite concentrate was added to the composition of the magnesium chromite.Kempirsaisk deposit chromite ore of 3-1 mm fraction and sintered periclase powder of 2-0 mm fraction were used as the granular materials.The chemical and grain-size compositions of the original materials are presented in Tables i, 2.Refractory mixtures containing magnesium chromite sinter with difficulty and therefore the main problem was searching for a method of compacting the refractories.The search for the optimum conditions for obtaining dense parts was made with the use of the methods of mathe matical experiment planning.The open porosity P in %, which must be minimized, was adopted as the parameter of optimization.The independent variables influencing the density of the refractory are: ~, wt. % of sintering addition relative to the basic mixture; x2, temperature of preliminary firing of the magnesium chromite, deg C; x~, firing temperature of the samples, deg C; x4, wt. % of magnesium chromite in the charge; and xs, wt. % of granular chromite ore in the charge.The area of determination was the following: xl = 0.7-1.7%; x2 ~ 700-1500~ x3 = 1650-1850~ x4 = 25-35%; xs = 15-25%.To attain the extreme area the method of steep ascent was used [4]~ In the area close to the assumed extreme the subarea for linear approximation of the function of response was selected.In this subarea there was accomplished a 1/4 replica of the full facto~al experiment 2 s with the generating relationships;The samples were pressed in the form of cylinders with a diameter of 36 mm and a height of 40 mm under a pressure of 150 MPa and fired in the tunnel kilns of Magnesite Combine.Three parallel tests were made at each point of the plan.The regression coefficients with the use of the full factoral experiment or fractional replicas of it are dete...

One of the indices for evaluating the quality of GOST 13236-73 electrical periclase is the particle-size composition [i]. The standard covers production of two types of periclase, "fine" and "coarse," differing in the ratio of the fractions. The periclase produced by foreign firms differs significantly in particle-size composition from domestic.* A comparative analysis of the periclase produced by firms in Japan, West Germany, and England shows that its particle-size composition is close to the requirements of GOST 13236-73 for the "fine" type (Table i).In investigations conducted earlier the influence of the particle-size composition of periclase on its electrical resistance was studied. However, attempts to find a correlation relationship between particle-size composition, electrical resistance, and dielectric strength did not lead to any conclusion, The statistically treated lots were obtained from different raw materials and differed in chemical composition, To reveal the influence of particle-size composition among a large number of relationships was impossible.In Eastern Refractory Institute an investigation has been made of the influence of particle-size composition on the electrical resistance of periclase with the use of methods of experiment planning. For the investigation periclase of a single lot but differing in fraction in accordance with GOST 13236-73 was taken, The influence of the content of fractions coarser than 0.4, 0,25, 0.16, 0,063, and 0,04 mm and finer than 0.04 mm was studied. Since 6 the weight portions of the fractions are not independent values (~=I), methods for investii=J gating multicomponent mixtures similar to those presented in [2] were used, A simplex-lattice plan of the second order was taken as the experiment plan. In accordance with the matrix of this plan periclase samples of individual fractions and combinations of them were prepared and the electrical resistance at 600, 800, and 1000~ was determined.The highest values of electrical resistance were obtained for compositions containing the fractions from 0.4 to 0.063 mm. Compositions of fractions finer than 0.063 mm are characterized by lower values of electrical resistances. This may be explained by the content in the fine fractions of a large quantity of iron and calcium oxide impurities, which have a negative influence on the electrical resistance of periclase,

A comparison of the density of sintered alumina specimens processed in the ball mill and planetary mill shows that the difference in the density is especially marked at low temperatures of firing, and is reduced with a rise in the firing temperature. An increase in the density of the blanks made from GLMK alumina as a result of planetary milling means an increase in the density of the specimens, after sintering at 1650~ from 98 to 9970.The density of the blanks made of calcined alumina pressed at 200 MPa is only 49.570; after processing in a planetary mill (specific surface 3.3 m2/g) it is increased to 57.570, which proves to be adequate for densilying the specimens made from this alumina to 99.570 after they have been sintered at 1650~The results obtained indicate that this form of alumina is much more active than GLMK, since even at the low density levels for the blanks it is sintered to a higher density.Thus, the use of finely dispersed alumina obtained by the calcination of finely dispersed aluminum hydroxide with additions of NH4C1 , H3BO3, and MgO, in, combination with processing in a planetary mill, enables us to obtain corundum ceramics with a density of 99.5~0 at a firing temperature of 1650~ At present the standard corundum ceramics produced industrially (density 99.570) from VK 100-1 material are usually fired at 1750~ [2]; therefore, the use of the technology examined above will permit us to extend the service of high-temperature furnaces and reduce the energy capacity of this process.

The investigation of microgranular constituents of magnesia concretes should be made in order to obtain further improvements in the quality. Microgranular constituents from periclase powder, sintered dunite, and chromite are the most common materials for magnesia concretes. The authors therefore studied the thermomechanical properties of pressed and cast TABLE 1 Chemical Composition of Original Materials Material ~ ~madd Periclase powder fractions 3-0 mm 18~2, 42 1 !,55 [ !,44 1 -[ o,sa Sintered Solov evogorsk dunite fractionsl-0 rnm 149,02140,28 I 0,17 I 0,24 I 9,79 ] 0,48 ~Nt. inc. Kempiraiskchromiteorefractions10mm I.lOl.Ol0Oo Io7o 1,33o1 9ol ,78 *Here and subsequently mass proportions are indicated.

Components whose properties are shown in Table 1 were used to study the milling processes of periclase~ dunite, chromite powders, and their combined mixtures~ Mathematical planning of the experiment was carried out according to the D-optimal simplex-solving plan of the third order.This plan enabled us to determine the regression coefficient in the following mathematical model, consisting of a polynomial of the third degree:In the work the number of components q = 3, the order of the model was n = 3, the number of experiments N= C~+n_ I = C~+3-I = i0; and C~ was the number of combinations from the elements with respect to k~The regression coefficients in the given model were calculated on the Mir-i electronic computer.The compositions of the mixtures of the original components were established with respect to the incomplete cubic matrix for the (3,3)-lattice.The powders and mixture were milled in a ballmill for 4 h, and with a mass ratio of balls to powders of 3:1.