ALKALIMETRY 8? 



ing a little neutral cliromate of potash to the alkaline solution, whence results sul- 

 phate of chromium, -water, and a separation of sulphur, which will not affect the accu- 

 racy of the alkalimetrical process. 



87. Whether the sample to be analysed contains any sulphuret, sulphite, or hypo- 

 sulphite, is easily ascertained as follows : If, on pouring sulphuric acid upon a por- 

 tion of the sample of soda ash under examination, an odour of sulphuretted hydrogen 

 that is, an odoiir of rotten eggs is evolved, or if a portion of the soda ash, being 

 dissolved in water, and then filtered, produces a black precipitate (sulphuret of 

 load) when solution of acetate of lead is poured into it, then the sample contains a 

 sulphuret. 



88. And if, after adding to some dilute sulphuric acid as much bichromate of 

 potash as is necessary to impart to it a distinct reddish-yellow tinge, and a certain 

 quantity of the solution of the soda ash under examination being poured into it, 

 but not in sufficient quantity to neutralise the acid, tho reddish-yellow colour 

 becomes green, it is a proof that tho sample contains either sulphite or hyposulphite 

 of soda, tho green tinge being due to the transformation of the chromic acid into 

 sesquioxide of chromium. 



89. And if, muriatic acid being poured into the clear solution of the soda ash, a 

 turbidness supervenes after some time if left at rest, or at once if heat is applied, it 

 is due to a deposit of sulphur, an odour of sulphurous acid being evolved, and hy- 

 posulphite of soda is probably present. We say ' probably,' because if sulphurets and 

 sulphites are present, the action of muriatic acid would decompose both, and liberate 

 sulphuretted hydrogen and sulphurous acid ; but as these two gases decompose each 

 other, a turbidnoss due to a separation of sulphur is also formed; thus 2HS + S0 2 

 = 2HO + 2S. 



90. As wo have already had occasion to remark, the soda ash of commerce 

 frequently contains some, and occasionally a largo quantity of caustic soda, the pro- 

 portion of which it is at times important to determine. This may be done, according 

 to Mr. Barreswill, by adding a solution of chloride of barium to the aqueous solution 

 of tho soda ash, by which the carbonate of soda is converted into carbonate of baryta, 

 whilst the caustic soda, reacting upon the chloride of barium, liberates a quantity of 

 caustic baryta proportionate to that of the caustic soda 



in the soda ash. After this addition of chloride of 

 barium, the liquor is filtered in order to separate tho 

 precipitated carbonate of baryta produced, and which 

 remains on the filter, on which it should be washed with 

 pure water. A few lumps of chalk are then put into a 

 Florence flask, a, and some muriatic acid being poured 

 upon it, an effervescence due to a disengagement of 

 carbonic acid is produced ; the flask is then closed with 

 a good cork, provided -with a bent tube, b, reaching to 

 the bottom of tho vessel c, and the stream of carbonic 

 acid produced is then passed through the liquor c, filtered 

 from the carbonate of baryta above mentioned. Tho 

 stream of carbonic acid produces a precipitate of car- 

 bonate of baryta, which should bo also collected on a 

 separate filter, washed, dried, and weighed. Each grain of this second precipitate of 

 carbonate of baryta corresponds to 0'3157 of caustic soda. 



91. As the soda ash of commerce almost invariably contains earthy carbonates, the 

 sample operated upon should always be dissolved in hot water, and filtered in order 

 to separate the carbonate of lime which otherwise would saturate a proportionate 

 quantity of the tost acid, and thus render the analysis worthless. 



92. The quantity of water contained in either po"tash or soda ash is ascertained by 

 heating a weighed quantity of the sample to redness in a covered platinum capsule or 

 crucible. Tho loss after ignition indicates the proportion of water. If any caustic 

 alkali is present, 1 equivalent = 9 of water, is retained, which cannot be thus elimi- 

 nated, but which may, of course, bo determined by calculation after the proportion of 

 caustic soda has been found, as shown before, each 31 grains of caustic soda contain- 

 ing 9 grains of water. 



93. Besides the alkalimetrical processes which have been explained in the preceding 

 pages, the proportion of available alkali contained in the sample may be estimated 

 from tho amount of carbonic acid which can bo expelled by supersaturating the alkali 

 with an acid. The determination of tho value of alkalis from tho quantity of car* 

 bonic acid thus evolved by tho supersaturation of tho carbonate acted upon has long 

 been known. Dr. Uro, in tho 'Annals of Philosophy,' for October, 1817, and then 

 in his 'Dictionary of Chemistry,' 1821, and more recently in his pamphlet 'Che- 

 mistry Simplified,' described several instruments for analysing earthy and alkaline 



