394 On a Quantitative Bloicpipe Assay of Chromium, 



Witli proper care in the execution, the results obtained bj 

 this method are very satisfactory ; generally rather a little too 

 high, but rarely more than O'l milligram from the truth, and 

 remarkably constant. The time required is generally about one 

 and a half to two hours. 



Of the acids above mentioned, I have not had opportunity of 

 experimenting on the titanic and vanadic; but it is not liKely 

 that they would cause any difficulty in the separation. Tungstic 

 acid remains with the green precipitate, and renders it flocculent, 

 so as to make the use of the mercurial salt superfluous. After 

 washing the filter a few times, six to eight drops of strong ammo- 

 nia are poured on the filter, which is subsequently washed with 

 an ammoniacal solution of sal-ammoniac. The separation is thus 

 effected without difficulty. When antimony is present, the solu- 

 tion after the nitre fusion must be effected ivithoxit boiling^ and 

 the residue washed with solution of nitre; if silica has to he 

 precipitated, it must be made to collect in flakes by shaking in 

 a test tube. After the bisulphate fusion, a spoonful or two of 

 iariaric acid are thrown into the beaker before rinsing the cover. 

 In the presence of copper, either silica must be added in the 

 nitre fusion, or more simply, some caustic potash is added to the 

 boiling solution, before precipitating the silica, if any. The 

 chromate of copper then contracts into black scales of pure oxyd. 

 The presence of a large amount of magnesia renders it advisable 

 to add silica in the nitre fusion, lest too much of that earth 

 should be carried over into the bisulphate solution by the ammo- 

 niacal salt. A small quantity does not vitiate the results When 

 the halogens are present, it is necessary to expel them before the 

 nitre fusion. This is done by a preliminary fusion with a small 

 quantity of bisulphate of potash, when chromic oxi/d alone is 

 resent; if there is chromic acid^ it must be previously reduced 

 y heating with KCy or KO, OxOs, otherwise the^chromium 

 will be partially volatilized. If simultaneously reducible metals 

 are present which would endanger the platinum crucible, the 

 fusion must be effected in a porcelain one. Tanfah'c and niohc 



ds 



tliey are not likely to interfere often, I have not devised any 

 method for their separation, which might require another filtra- 

 tion. The rest of the elements are separated without difficultj? 

 in the ordinary course of the assay. 



m 



