The present author, in continuing the previous study (This Journal, 1936, 44, 531), reports here the results of further studies on the reaction of alkali solution to kaolin or allied matter. The following is the brief abstract from the original Japanese communication.
(1) The comparative tests were carried out to prepare alkali kaolin or artificial feldspathic matter from kaolin by the hydrothermal reaction of sodium or potassium hydroxide under pressure. This hydrothermal reaction of alkali solution on kaolin was already reported in other papers (S. Nagai: “Studies on Products of Hydrothermal Reactions on Clayey Substances, ” Journal Soc. Chem. Ind., Japan, 1935, 38, 371B, 732B; 1936, 39, 7B, 45B, 96B, 252B). The products are to be called as alkali kaolin, artificial nepheline hydrate or artificial feldspathic matter of zeolite group, having the rational formula R₂O⋅Al₂O₃⋅2SiO₂⋅nH₂O, and these are clear differences between sodium and potassium products. The sodium product Na₂O⋅Al₂O₃⋅2SiO₂⋅nH₂O of nepheline hydrate type is easily prepared even under ordinary atmospheric pressure and has the rational formula Na₂O⋅Al₂O₃⋅2SiO₂⋅H₂O of 1mol, combined water by drying at 110°C. The potassium product K₂O⋅Al₂O₃⋅2SiO₂⋅nH₂O of kali-nepheline hydrate type is, on the contrary, difficultly prepared under low steam pressure (5-10 atm. and 150-180°C), nearly completely prepared under higher steam pressure (20-30 atm. and 210-232°C), and has the rational formula K₂O⋅Al₂O₃⋅2SiO₂⋅1/4H₂O, losing the large part of combined water by drying at 50°C.
(2) These products of alkali kaolin or nepheline hydrate type were tested their base exchange property by treating with calcium or magnesium chloride solution 5-10% uuder steam pressures (30-50 atm. and 232-263°C), and obtained 0.5 (Na, K)₂O⋅0.5(Ca, Mg)O⋅Al₂O₃⋅2SiO₂⋅nH₂O, being base exchanged nearly 50%.
(3) The special clayey raw material “Rosekinendo” for refactory use, containing pyrophyllite Al₂O₃⋅4SiO₂⋅H₂O, was analysed and the comparative tests of the hydrothermal reaction of alkali solution under pressure was applied to this material as above expianined. Pyrophyllite is seen to be in the quite same way reacted by sodium hydroxide solution under pressure (10-20 atm. and 180-210C) to Na₂O⋅Al₂O₃⋅2SiO₂⋅nH₂O and sodium silicate solution from the excess silica, (4-2) SiO₂, and difficultiy reacted by potassium hydroxide solution under higher pressure (20-30 atm. and 210-232°C) to K₂O⋅Al₂O₃⋅2SiO₂⋅nH₂O and potassium silicate solution. These products Na₂O⋅Al₂O₃⋅2SiO₂⋅nH₂O and K₂O⋅Al₂O₃⋅2SiO₂⋅nH₂O are quite equal in every points to those from kaolinite as above explained.
(4) Prophyllite was not completely reacted under atmospheric pressure by the alkali solution, which is the quite different point from kaolinite. This property can be used with success for the method of separation of pyrophyllite from kaolinite, by treating the mixture of pyrophyllite and kaolinite with 20-30 solution of sodium hydroxide under atmospheric pressure for long time, filtering and washing the residue composed from non-reacted

View full abstract

With the same reason of the previous report (part III), the author has studied the reactions between 5CaO⋅3Al₂O₃ and 2CaO⋅Fe₂O₃ at high temperatures by microscopic and X ray methods.
The samples were made by heating, at 1300, 1350°C and over, 16 mixtures of 5CaO⋅3Al₂O₃ and 2CaO⋅Fe₂O₃ with various molar ratios as shown in the next table.
Table 1. Molar ratios of 5CaO⋅3Al₂O₃ and 2CaO⋅Fe₂O₃ in the mixtures
The experimental results are as follows:
(1) 2CaO⋅Fe₂O₃ and 5CaO⋅3Al₂O₃ make each other solid solutions of limited miscibility.
(2) The miscibility in these solid solutions increases with the rise of temperature.
(3) The samples of No. 13-19 which were made by firing at 1350°C and those of No. 13-17 made at 1300°C belong to the same solid solution series, and their X ray diffraction patterns are of the same type with that of 2CaO⋅Fe₂O₃. That is to say, a certain mol of 5CaO⋅3Al₂O₃ seems to be miscible in the lattice of 2CaO⋅Fe₂O₃, The limited miscibility is 0.5mol of 5CaO⋅3Al₂O₃ to 1mol of 2CaO⋅Fe₂O₃.
Reversively, a small amount of 2CaO⋅Fe₂O₃ is miscible in 5CaO⋅3Al₂O₃, and makes a solid solution having a diffraction pattern of 5CaO⋅3Al₂O₃ type.
(4) In the fired samples of No. 20-28 or so, two solid solutions of 2CaO⋅Fe₂O₃ and 5CaO⋅3Al₂O₃ types co-exist.
(5) These two solid soiutions seem to have an eutectic point microscopically.
(6) The lattice spacing of the samples (No. 18-20) having the greatest miscibility in the 2CaO⋅Fe₂O₃ type solid solutions is, on the whole, the same with that of the sample of the molar ratio of 2CaO⋅Fe₂O₃:3CaO⋅Al₂O₃=1.00:0.25-the molar ratio of the utmost limit to which 3CaO⋅Al₂O₃ is miseible into 2CaO⋅Fe₂O₃ in its original form-and also the same with that of the fired sample of 5CaO:Al₂O₃:Fe₂O₃, but never reaches to that of the fired sample of 4CaO:Al₂O₃:Fe₂O₃.

View full abstract

Corhart block is an excellent refractory raterial for glass-melting tank furnace on account of its high resistance against the corrosion of molten glass. The cause of this property is due not only to the chracteristic structure of electro-casted block but also to the special corrosion phenomena which is exhibited during service. In this paper the writer has treated the problem on the mechanism of corrosion and some influence upon the properties of glass. The brief summary is as follows:-
(1) The mode of corrosion was described on some specimens which had been employed in tank furnace. It is noted that the altered layers covered nearly all the surfaces of corroded parts and that a stalactitic erosion developed on some downward faces of block.
(2) The internal relations of corrosing layers were observed by a polarized microscope and the phases newly developed by interaction between the refractory material and glass were determined by optical analysis.
(3) The properties of glassy material accompanied around the blocks were examined and the effect of corrroded material upon molten glasses was discussed.
(4) Summarizing the results obtained, the writer considered the mechanism of corrosion and found that the following three stages had been carried out during a course of corrosion. (i) Formation of altered zone consisting of corundum corystals and some glassy matrix, which have been produced by the dissociation of mullite; (ii) Formation of viscous liquid layer along the boundary between corundum zone and molten glass, and that liquid phase can crystallize out nephelite, Na₂O⋅Al₂O₃⋅2SiO₂, during coooling; and (iii) Disappearence of the above mentioned nephelitic layer into molten bath through diffusion. Thus a corrosion phenomena proceeds toward the inner part of the block, holding a certain equilibrium between layer and layer under given conditions. When these steps proceed satisfactorily, the case may be called a normal corrosion of corhart, and the layer being consisted of corundum or nephelite may be regarded as a propecting layer, because of retarding the velocity of corrosion.
(5) If the above equilibrium is disturbed spontaneously or externally, then the velocity of corrosion would be accelerated locally or selectively in places. Such a case may be distinguished as an abnormal corrosion against before. In some parts of block facing downward, uppermost nephelite layer and the next corundum zone tend to separate and fall gradually into molten glass and consequently, for instance, a stalactitic type of corrosion will be developed.
(6) The normal corrosion of corhart does not give any visible effect upon the property of glass, but the abnormal one exhibits sometimes a tendency to produce stones in glass plate. The stones derived from corthart block are characterized in enclosing nucleus such as corundum, nephelite or rarely carnegieite It is, however, proved that the strains caused by the above stones in surrounding glass are very weak compared with that of chamotte-stones and that there is no risk to induce any crack in the glass plate.

View full abstract