io 4 SCIENCE PROGRESS 



PHYSICAL CHEMISTRY. By Prof. W. C. McC. Lewis, M.A., D.Sc, 



University, Liverpool. 

 Disperse systems. Colloids. — Although the method of pre- 

 paring colloidal solutions of metals in water and other media 

 by sparking the metal under the surface of the liquid has been 

 a familiar one for many years past, it nevertheless offers some 

 problems in regard to the actual mechanism of the process 

 which are by no means cleared up. The general consensus of 

 opinion is that the first stage of the process, namely the dis- 

 integration of the metal, is brought about by purely thermo- 

 mechanical means. This is, however, only a small part of 

 the problem. The resulting system may be either stable or 

 unstable, that is the metal may remain more or less permanently 

 in suspension in the finely divided state, or it may rapidly 

 agglomerate and deposit itself from the medium. We shall 

 only possess a satisfactory view of the sparking method of pro- 

 ducing colloids when we have solved the problem of resulting 

 stability. Since the existence of colloidal solutions depends 

 upon the realisation of these stable conditions it is natural to 

 expect that the problem has been the subject of considerable 

 investigation. One of the most interesting contributions is 

 that of Beans and Eastlack (/. Amer. Chem. Soc. 37, 2667, 

 191 5). They approach the subject by means of the electrical 

 conductivity exhibited by the colloidal solutions prepared by 

 sparking. A preliminary calculation based upon the known 

 size, charge, and mobility of the particles in a colloidal solution 

 of gold in water showed that the specific conductivity of the 

 colloid itself should be of the order io -10 mhos. This is only 

 o*oi per cent, of the conductivity of the water itself, so that if 

 the colloidal solution simply consists of the finely divided 

 metal in suspension in unchanged water, the conductivity of the 

 solution should not be distinguishable from that of the water. 

 This, however, is not the case. The solution exhibits a larger 

 conductivity than that of the water. Further, there is a 

 marked difference in the enhanced conductivity according to the 

 metal employed. Thus, in the case of colloidal platinum the 

 increase in specific conductivity is eight times that observed 

 in the case of gold. Beans and Eastlack connect this larger 

 conductivity with the fact that colloidal platinum is more stable 

 than colloidal gold. The idea is, that the greater stability of 

 the platinum particles is due to the presence of small amounts 



