METHOD OF ANALYSIS FOR FIRE-CLAYS, FELDSPARS, KAOLIN, AND FIRE-SANDS.
The method of chemical analysis adopted and pursued in the examinations of the fire-clays, feldspars, kaolins, and fire-sands, which are given on pages 70-76 was as follows : One gramme of the air-dried pulverized material was digested in sulphuric and hydrofluoric acids until the silica was completely dissipated ; the residue was dissolved in hydrochloric acid (a few drops of nitric acid being added to oxidize the iron), and the alumina, sesquoixide of iron and titanic acid were precipitated by am monia (in some cases by acetate of ammonia). In the filtrate the lime was precipitated by oxalate of ammonia, and weighed as carbonate. The filtrate from the lime was divided into two equal portions. In one of these the magnesia was determined by precipitation by phosphate of ammonia. The second por tion was evaporated to dryness and heated to drive off the am moniacal salts. The residue was dissolved and chloride of barium was added to remove the sulphuric acid, and, then caustic lime to remove the magnesia. The liquid was boiled and then fil tered. To the filtrate, ammonia and carbonate of ammonia were added to remove the chloride of barium and lime ; the liquid was filtered, evaporated to dryness, and the ammonia salts driven off by ignition. The potash was precipitated by bichlo ride of platinum and weighed as potassio-bichloride of plati num. The alcoholic filtrate was evaporated to dryness, the platinum compound decomposed by heating to redness with oxalic acid ; treated with water ; filtrated ; a few drops of hy drochloric acid added ; 'evaporated to dryness, and weighed as chloride of sodium.
A second sample (one gramme) was treated with hydro fluoric and sulphuric acids, as before, and then ammonia added to precipitate titanic acid, alumina and oxide of iron. This precipitate was reserved for the determination of the titanic acid. The filtrate was treated as before for the determination of the potash and soda, as duplicates.
A third sample was fused with carbonates of potash and soda ; the fused mass treated with water ; hydrochloric acid added in excess ; evaporated to dryness to render the silicic acid insoluble ; treated with dilute hydrochloric acid ; heated, and then filtered for the total silicic acid. This weighed determi nation was checked by the difference in the analysis by hydro fluoric acid. The alumina, oxide of iron and titanium were precipitated by ammonia as in the first sample. Lime and magnesia were also determined as before (duplicate determi nations).
The precipitate in the second sample, reserved for the titan ium determination, was treated with a solution of caustic pot ash and heated, to remove the alumina. The insoluble portion, consisting of oxide of iron and titanic acid, was collected on a filter, burned, fused with bisulphate of potash, dissolved in water, and saturated with hydrosulphuric acid gas, to reduce the iron oxide. The liquid was filtered and boiled ; the titanic acid was precipitated and collected on a filter, then burned and weighed.
For the determination of the quartz the clays were digested in the sulphuric acid, and the liquid filtered. The insoluble matter on the filter was burned and weighed as a duplicate of the total silica. This insoluble matter was then boiled in a so lution of potash, and the undissolved residue weighed as quartz. These determinations were duplicated by the same method.
The moisture or hygroscopic water was determined by heat ing over a water-bath, and the loss at 212° F. (1 oo° C.) taken as its amount. The samples were then heated to redness, ignited, and the loss noted as combined water. In most of the dark-colored clays there was some organic matter.
In a few analyses this was estimated ; in others the combined water includes very small amounts of organic matter.
The iron was determined by volumetric analysis, using the method by photo-chloride of tin.