SECTION B.— CHEMISTRY. 



RECENT INVESTIGATIONS IN THE 

 CHEMISTRY OF GOLD 



ADDRESS BY 



Prof. CHARLES S. GIBSON, O.B.E., M.A., Sc.D., F.R.S., 



PRESIDENT OF THE SECTION. 



' How is the gold become dim ! how is the most fine gold changed ! ' 



(Lam. iv. i) 



By the resuks of investigations in which workers in this country have 

 played an impoitant part during the last ten years it is now realised that 

 fewer anomalies exist among the metals of sub-group iB, copper, silver 

 and gold, than was formerly believed to be the case. The only funda- 

 mental property which these metals have in common with the alkali 

 metals is that they are all capable of being univalent. The metals of the 

 sub-group differ from the alkali metals in their atomic structure ; the 

 former have eighteen electrons in their penultimate electronic group 

 whereas the latter have eight electrons in that group. While there are 

 differences in their multivalency, the multivalency of copper, silver and 

 gold must be correlated with the eighteen electronic group of these 

 metals. The univalency and bivalency of copper and silver are well 

 established ; the tervalency of silver must still be regarded as doubtful. 

 On the other hand, while the bivalency of silver has only been established 

 comparatively recently, modern investigations have shown that it is ex- 

 tremely unlikely that gold can exist in the bivalent condition and this 

 metal continues to exhibit the anomaly, distinguishing it from copper and 

 silver, of existing only in the univalent and tervalent conditions. 



As far as the existence of normal salts is concerned, argentous silver 

 differs greatly from cuprous copper and aurous gold. There is no 

 evidence for the existence of any normal aurous salt and, for example, 

 cuprous sulphate is at once decomposed by water with separation of 

 metallic copper, and cuprous nitrate does not exist. On the other hand, 

 in the solid state cuprous and silver halides have non-ionic lattices in 

 their crystals which are isomorphous. Since cuprous chloride is bi- 

 molecular, it is reasonable to assume that in its halides the cuprous atom 

 is 2-covalent. Recently, chemical evidence has indicated that this is also 

 true of aurous gold in the analogous compounds and therefore under 



