Formation of hydrated calcium silicates at elevated temperatures and pressures

Abstract

An X-ray investigation of the crystalline hydrated calcium silicates which occur naturally as minerals indicated that the following are distinct com- pounds: . Okenite, CaO.2~iO,.2H'06· gy,rolite, 2CaO. ~S~O,.2H,O; cre~tmoreite, 2CaO.2S10,.3H,O; xonothte, 5Ca .5S10,.H,O; afwllhte, 3CaO.2S10,.3H,O; foshagite, 5CaO.3SiO,.3H,O; and hillebrandite, 2CaO.SiO •. H,O. Riversideite is apparently not a distinct mineral and is probably the same compound as crest- moreite. Crystalline preparations having X-ray patterns identical with those of gyrolite, xonotlite, and foshagite were synthesized by hydrothermal treatment of calcium silicate in the form of the anhydrous crystalline compounds, glass, and amorphous hydrate. Other compounds formed were cristobalite, wollaston- ite, pseudowollastonite, 4CaO.5SiO,.5H,O, CaO.SiO,.H,O, 10CaO.5SiO,.6H.O, 2CaO.SiO,. H,O, and 3CaO.SiO,.2H,O. A low-temperature modification of anhydrous tricalcium disilicate, which inverts to the usual form at 1,024° ± 5° C, resulted from the action of water vapor on tricalcium disilicate at 500° C, 380 atmospheres. Gyrolite was obtained at temperatures between 150° and 4000 C, 4CaO.5Si02 .5H20 at 150° to 275°C, CaO.Si02 .H,O at 150° C, xonotlite at 175° to 390° C, wollastonite at 400° C and higher, foshagite at 3000 to 350° C, 10CaO.5SiO,.6H,O at 100° to 200° C, and 3CaO.SiO,.2H,O at 200° to 450° C. Attempts to prepare okenite, crestmoreite, afwillite, and hiUe- brandite were unsuccessful. Treatment of tricalcium and dicalcium silicates with water at room temperature for several years gave crystalline altera- tion products which were different from other known hydrated calcium sili- cates. That derived from dicalcium silicate had a composition approaching 3CaO.2SiO,.l.5H,O. Attempts to isolate and identify the hydrated calcium silicate binding material in commercial sand-lime brick and in laboratory-prepared brick were unsuccessful, but a partial separation of the cementing material was apparently obtained. Applications of the results to the autoclave testing of cements are discussed.