Researches on the Chemical Equivalents of Certain Bodies .
By RICHARD PHILLIPS , F.R.S. , Sc. Received January 17 , -Read February 14 , 1839 .
THE late Dr. TURNER , in an elaborate memoir on atomic weights , contained in the Philosophical Transactions for 1833 , observes , that " Dr. PROUT 's hypothesis as advocated by Dr. THOMSON , that all atomic weights are simple multiples of that of hydrogen , can no longer be maintained , " and he further asserts that hypothesis " to be at variance with the most exact analytic researches which have been conducted .
" Although the experiments of Dr. TURNER , and the inferences which he has drawn front them , agree very nearly with those of BERZELIUS , it still appeared desirable further to investigate this subject , and it occurred to me that the inquiry might be conducted in a mode not liable to some of the objections which may be urged against the processses usually adopted .
For the purposes of this investigation , I deem it peculiarly fortunate that Dr. TURNER has adopted a whole number as the equivalent of silver , and on this account , as well as for other obvious reasons , I selected it as the basis for an inquiry respecting the equivalents of chlorine and some other elementary gases .
From Dr. TURNER'S experiments ( Philosophical Transactions , 1829 ) , it appears that 108 parts , or one equivalent of silver , yield 143A424 parts of fused chloride , which result coincides very nearly with the determination of BERZELIUS .
Dr. PROUT , however , has objected to the fusing of the chloride of silver , that during the operation , it yields hydrochloric acid , which is admitted by Dr. TURNER to be the case , although he could not discover that chloride of silver , that had been dried at 300 ?
, lost so much as a thousandth of its weight by subsequent fusion .
It seemed to me that the chance of error arising from the fusing of the chloride of silver might be entirely removed , and other advantages gained , by making an experiment with silver on a large scale , with such proportions of the substances employed as were deemed to be equivalents ; and instead of calculating from the whole product of the fused chloride , to do it merely from the weight of such small portion only as might arise from the difference between theoretical views and experimental results .
With this purpose I purified some silver by dissolving it in nitric acid , precipitating by a chloride , dissolving the precipitate in ammonia , again throwing down the chloride and reducing it to the metallic state .
I could not discover any impurity in the silver thus obtained : I therefore dissolved 216 grains ( 2 equivalents ) in nitric acid , and decomposed the nitrate formed , by adding 108 grains of pure hydrochlorate of ammonia dissolved in water .
I need hardly state , that the weight of the hydrochlorate was taken on the assumption , that this salt contains one equivalent of chlorine 36 , one of azote 14 , and four equivalents of hydrogen 4 , equal 54 .
The chloride of silver precipitated was collected and washed on a filter , but instead of drying and fusing it , it was neglected , for a reason already assigned .
Having ascertained that the solution from which this chloride was pr ecipitated contained excess of hydrochlorate of ammrnronia , I added nitrate or silver to it and to the washings ; the chloride of silver thus obtained was collected on a double filter of Chinese paper and washed and dried ; it weighed 2*58 grains , a quantity too small to admit of any appreciable error from deficient drying in not subjecting it to fusion .
Assuming that 144 of chloride of silver indicate 36 of chlorine , 2-58 grains will give 0'645 grain as the weight of the excess of chlorine in the two presiumed equivalents of hydrochlorate of amrmlonia , and one half of this , or 0 322 grain , subtracted from 36 , will reduce the equivalent of chlorine from that number to 35*678 .
On tepeating this experiment I obtained 2-56 grains of chloride of silver , which brings the equivalent of chlorine to 35'680 .
In order to bring under discussion the equivalents of other elementary bodies , I prepared some nitrate of silver from the pure metal ; it was twice crystallized , and a portion of it , fused in a glass capsule , weighed 271 57 grains ; adopting the equivalent weights above stated , and 8 for oxygen , 54 of hydrochlorate of amionia should decompose 170 of nitrate of silver , and 86 263 of the hydrochlorate would therefore be required for the decomposition of the fused nitrate .
The solutions of these quantities of the salts were accordingly mixed , the precipitate was , as before , separated and washed , and on adding nitrate of silver to the filtered solution and washings , 1P74 grain of chloride of silver was obtained , indicating an excess of 0*435 grain of chlorine ; now according to the assumed equivalents , 54 of hydrochlorate of ammonia contains 36 of chlorine , 86 263 , the quantity employed in the experiment , therefore contained 57'508 , from which , if we subtract 0,435 , the excess , there will remain 57'073 ; if then 86,263 give 57'073 , 54 , the equivalent of hydrochlorate of ammonia , will contain 35'727 of chlorine instead of 36 .
This experinent was repeated with 357*34 grains of fused nitrate of silver , and 113 508 grains of hydr ochlorate of ammonia ; the excess of chllorine was 0*7 grain , which calculating as before , makes the equivalent of chlorine 35 667 ; and taking the mean results of the four experiments , viz. 35-678 , 35b680 , 35 727 , and 35X667 , we have 35 688 as the equivalent of chlorine .
Although Dr. PROUT'S hypothesis requires that the error , or the difference between 35-688 and 36 = 03 12 , should not be divided among the various substances employed in the experiment , it may , nevertheless , be worth while to observe to what the error amounts when the numbers and quantities of the elements included in the operation are summed up : they are 36 equiv. silver ... 108 1 equiv. chlorine. .
36 , in the hydrochlorate of ammonia .
4 equivs. hydrogen. .
4 , one in the acid , three in the alkali .
6 equivs. oxygen .
48 , one with the silver , five in the acid .
2 equivs. azote ... 28 , one in the acid , one in the alkali .
14 equivs. weighing. .
224 We have thus 14 equivalents of various elements , the sum of whose weights is 224 , yet the mean error , or 0'312 , without allowing for circumstances which I shall presently notice , is only about 1-717th part of the whole weight .
It may also be remarked , that omitting the silver , and considering the other elements in their gaseous state , the error in volume will be comparatively less than that in weight ; though it must at the same time be admitted that this difference is derived from the greater density of chlorine than of the other gases .
1 equiv. chlorine ... 36 grains = 46 cubic inches , nearly .
4 equivs. hydrogen .
4 grains = 186 cubic inches .
6 equivs. oxygen. .
48 grains = 139 cubic inches .
2 equivs. azote. .
28 grains 93 cubic inches .
Total .
464 cubic inches .
The weight of chlorine in error being , as already mentioned , 0,312 grain , or about 0'4 cubic inch , is only 1-1160th of the whole volume of the gases .
I have every reason to believe that all the substances which I used in the experiments above detailed , were of the greatest degree of purity ; but I may observe that the adrmixtures most likely to occur in any of them would increase the quantity of the chlorine , or diminish that of the silver ; and ainy substance producing eithter of these effects would erroneously diminish the equivalent of chlorine by increasing the weight of the precipitated chloride of silver , assumed to be derived from the error of the theory ; thus hydrochlorate of ammonia always contains sufficient excess of acid to redden litmus paper , and any moisture or foreign metal which the nitrate of silver might contain would produce corresponding results .
From these experiments and considerations I am of opinion , that no material , and scarcely even any appreciable etror can arise from considering the equivalents of hydrogen , oxygen , azote , and chlorine , as 1 , 8 , 14 , and 36 respectively .
The specific gravity of oxygen and azote may be obtained by comparing their equivalent weights with the composition and density of atmospheric air .
The mean of various experimental results of the weight of 100 cubic inches of air is about 31 grains , and this agrees very nearly with the determination of Dr. PROUT .
Now if pure air consist of 20 cubic inches of oxygen and 80 of azote , or of one equivalent of oxygen and 2 equivalents of azote , the weights of which are respectively 8 and 14 , the 20 cubic inches of oxygen will weigh 688 grains , and the 80 of azote 24*08 grains , 37 giving 30*96 grains as the weight of 100 cubic inches of air ; if , as more commonly admitted , air consists of 21 volumes of oxygen and 79 of azote , we shall have 7*224 grains as the weight of the oxygen , and 23*779 as that of the azote = 31P003 grlains as the weight of 100 cubic inches of air ; either of these determinations is sufficiently near to show , that we cannot be far wrong in estimating the weight of 100 cubic inches of oxygen at 34'4 grains , and the same volume of azote at 30 1 grains ; and the weight of the equivalent of hydrogen being 8 that of oxygen , and its volume twice as great , and the equivalent of chlorine being to that of oxygen as 36 to 8 , while their volumes are equal , it follows that 100 cubic inches of hydrogen and chlorine will weigh respectively 2*15 and 77*4 grains , and the density of the four elementary gases in question , compared with air = 1 , will be Hydrogen ... ... . .
006935 Azote ... ... ... 0'97097 Oxygen ... ... . .
I1 0968 Chlorine ... . .
2*49678 On comparing these densities with those stated by Professors THOMSON , TURNER , and GRAHAM , it will be observed that they are less than those given by Dr. THOMSON , which is accounted for by his having assumed that 100 cubic inches of air weigh 31*1446 grains , while they vary , but not very materially , from the weights assumed by Professors TURNER and GRAHAM , in being in some cases rather lighter and in others somewhat heavier .
THOMSONr .
TURvNE .
GRAHAM .
Hydrogen ... .
0'0694. .
0'0690 ... 0069 Azote ... ... 09722 ... 0'9727 ... 0'976 Oxygen ... . .
11111 ... 11025 1 1026 Chlorine ... .
2-5000 ... 2-4700 .
2*470 38

