Mlt. R. RAILTON ON NITROGLYCERTNE AND XX3.-On iiitroglyceriue and the Products of its jr)ecmzposit.ion by Potash. By ROBERTRAILTON,F.C.S. Sobero discovered nitroglycerine some time ago but did iiot de-termine its composition,-probably from the fact of its being a highly explosive substance. There is however no danger to be appreheiided from its analysis if sufficient care be taken. The compound from which the following results were obtained was thus prepared. Equal volumes of concentrated nitric and sulphuric acids were mixed and cooled by immersing the flask which contained them in ice-cold water. Syrupy glycerine was then introduced into the flask a few drops at a time the mixture being well shaken and cooled eacli time. After some time the product floated on the top as an oily substance; it was then poured into cold distilled watcr and repeatedly washed in that menstruum by decantation it was afterwards frced as much as possible from water by blotting-paper and an attempt was made to dry it more completely under the bcll- jar of the air-pump but without success as it was rapidly decom- posed on exhausting the jar.From this circumstance the hydrogen could not be estimated but the relative amounts of carbon and nitrogen were satisfactorily deter- mined by Liebig’s process. About a gramme of the conipound was mixed with as much oxide of copper as half filled a combustion-tube thirty-six inches in length. About eight inches of the remainder of the tube was filled with oxide of copper and the tube theii wholly filled with rediiced copper.Tbe tube was then enveloped in copper foil the combustion proceedcd with and the mixed gases collected in graduated tubcs ovci-rriercury. THE PRODUCTS OF ITS DECOMPOSTTION BY POTASB. 223 The first portion of the gases was allowed to escape; four tubes were then filled and the following are the results in the order in which they were collected :-Number of tubes . . . I. 11. 111. TV. Volumes of mixed gases . 101 91.5 99 97 Ditto absorbed by KO . 69 61.0 65 64 I-I_ Ditto remaining gas . . . 32 30.5 34 38 There being slight discrepancies in these results I obtained tubes more finely divided when a second experinient gave- Number of tubes . . I. 11. In. IV. V. VolumesofmixedcS'ases.178 194 193 173 194 Ditto absorbed by potash 117 128 127 115 1.29 Ditto residual nitrogeii . 61 66 65 58 65 Now as the composition of glyceririe is well known and as the volt-une of nitrogen is proved by these experiments ti! be half that of the GO absorbed by potash wc may safely assume the following to be the change which the glycerine undergoes :-Nitroglycerine is heavier than water in which it is brit slightly soluble; it is soluble in alcohol and in ether. When boilccl for sornc time with an aqueous solution of catktic potash it is dccomposed nitrate of potash and glycerine heing formed. A solution of caustic potash in water was made of nearly the same specific gravity as the nitro-glycerine. The potash solution and thc nitroglycerine were introduced into a flask and boiled for several hours ; at the end the liquid became completely homogeneous when it was carefully neutralised with pure sulphuric acid.The siilphate of potash was crystallised out and on further evaporation the nitrats of potash was deposited. This salt was purified by repeated re-crystallisations and afterwards analysed. The nitrate of potash was dried until it ceased to lose weight and wa$ thcn converted into sulphate of' potash. arm. Wcigbt of the salt used . . 0.979 Ditto -KO SO forimd . -0*8$1 Calculated percentage of KO in KO NO,= 46.54 Found in the above-namcd salt . . =46*41 224 MR. G. KAY ON I may add that the salt obtained by the decomposition of nitro-glycerine exhibited all the reactions of nitrate of potash.For the purpose of obtaining the glycerine the solution from which the nitrate of potash had becu crystallised was evaporated to a syrupy consistence and treated with absolute alcohol. The solut,ion of sly-eerine in alcohol thus obtained was evaporated in a water-bath and the residue treated with ether in which the nitroglycerine had there been any undecomposed mould have been dissolvecl. The ether was removed without having dissolved anything and a portion of the residue heated with bisulphate of potash when thc pungent odour of acrolein was evolved. The decomposition of nitroglycerine by means of hydrate of potash may be represented by the equation Nitrobenzole was boilcd with concentrated solution of caustic potash for several hours ;the resulting solution evaporated nearly to dryness water added and again evaporated the lattcr process being repeated as long as any odoiir was pcrceived. The so?ution was then examined for nitric acid and distinct evidence of its presence was obtained.