Chemical Affinity in terms of Electromotive Force. 291 



The earlier value is deduced from observations in which 

 the heat of formation of copper oxide from the metal by 

 combustion is involved, the copper being in a more or less 

 compact state, filings &c. ; the latter involves the determination 

 of the heat-evolution during the precipitation of spongy copper 

 from copper sulphate by iron. Leaving out of sight other 

 sources of difference between the two values, this affords a 

 reason why the former value should be the higher, since heat 

 is evolved in the transformation of spongy into compact 

 copper *. On the whole, it is evident that the net che- 

 mical change taking place in a Daniell cell (i. e. the displace- 

 ment of copper from copper sulphate by zinc) corresponds to 

 an E.M.F. which is a little higher than 1*105 volt by an 

 amount which is the greater the more compact the copper 

 precipitated, and is approximately constant when the zinc 

 plate is surrounded by zinc-sulphate solution of the same 

 equivalent strength as the copper-sulphate solution surrounding 

 the copper plate, but is influenced by the strength of the 

 dilute sulphuric acid when the fluid surrounds the zinc plate. 

 The amount of this influence can be approximately calculated 

 from Thomsen's determinations of the heat developed in the 

 formation of the solutions of strengths indicated by H 2 S0 4 , 

 nH 2 where n varies {JDeut. chem. Ges. Berichte, iii. p. 496). 

 Thomsen finds in gramme-degrees per gramme molecule : — 



ii gramme-degrees. 



9 14940 



19 16248 



49 16676 



99 16850 



n gramme-degrees. 



199 17056 



899 17304 



799 17632 



1599 17848 



From which table the values for any intermediate values of 

 n can be obtained by interpolation. In the experiments 



* That this is so is shown by the circumstance that if a current be sent 

 through a decomposing-cell containing copper-sulphate solution and 

 copper electrodes, of which the positive one is of compact rolled metal, a 

 considerably higher difference of potential is set up, under any given 

 conditions and with a steady current, than is set up when the -f- electrode 

 is replaced by one covered with freshly electro-depositr d metal. The more 

 spongy texture of the latter corresponds to a greater heat-development 

 during solution than that taking place with the compact metal, and hence 

 to a diminution of the work that has to be done by the cm-rent in passing • 

 with not very powerful currents the difference often exceeds -02 or -03 

 volt, corresponding to 450 to 700 gramme-degrees per gramme eqiuvalent. 

 In a somewhat similar way, but using the mercurial calorimeter, Favre 

 found (Comptes Bendus, lxxiii. p. 1258) that electro-copper gave out 

 about 1000 gramme-degrees more heat than rolled metal per gramme 

 equivalent ; this would correspond to an E.M.F. of -044 volt. 



