56 



NA TURli 



[May 21, 1896 



values for k — the dift'usivity of certain metals in lead, tin, 

 bismuth and mercury, given in sq. cm. per day, are as 

 follows : — 



Gold in lead 

 ,, bismuth 

 ,, tin... 



Silver in tin 



Lead in tin 



Khoiliiim in lead 



I'latinuni in lead 



('...la in lead 



G<ild in mercury 



k 

 3-19 at 550° 

 452 ,, 

 4-65 .. 

 4'I4 ,. 

 3-'8 „ 

 3 '04 ,. 

 I '69 at 490" 

 j'oj ,- 

 072 at 11° 



The proof that gold had diffused into lead can then be 

 afforded by the fact that the point of solidification of the 

 second portion of lead is, owing to the presence of gold 

 in it, lower than that of the first portion, and this is 

 shown to be the case in the following way. The first 

 sample of lead is placed in a small crucible, u, and a 



In order to afford a term of comparison, it may be 

 stated that the diffusivity of chloride of sodium in water 

 at 18° is ro4. 



It is at present too soon to draw any conclusion as to 

 the evidence which the results afford respecting the 

 molecular constitution of metals, but it is evident that 

 they will be of value in this connection, because, with 

 the exception of the gases, they present the simplest 

 possible case which can occur — the diffusion of one 

 element into another. Thus the relatively slow rate of 

 diffusion of platinum as compared with gold, points to its 

 having a more complex molecule than the latter. 



It is very difficult within the limits of this brief aljstract 

 to show that molten metals actually pass into each other 

 by diffusion. The following method will, however, serve to 

 give a rough qualitati\e demonstration that such is the case. 

 A white sheet of card, U, is, as Fig. i shows, mounted on 

 a frame, and by means of the gearing C c' can be raised or 

 lowered at a definite rate, which can be maintained by the 

 aid of a metronome. The diffusion cell is shown at A. 

 It consists of a clay tube about 30 mm. long, heated from I 

 its upper part so that the top portion of the tube is hotter 

 than the lower portion. This tube is filled with molten 

 lead, and at the bottom of the lead, is a layer of a lead- 

 gold alloy rich in gold. A sample of lead may be with- 

 draw n from the tube at the beginning of a lecture and 



r» 



TEMPERA ruRE 



B 



<tf 







^'^4- 





1-^ 



set aside, while a second sample of lead may be re- 

 moved at the end of an hour, by which time gold will 

 have diffused into the upper layers of lead, and conse- 

 quently this second sample will be auriferous. Other 

 e.xperiments have shown that in this case con\'ection 

 currents do not complicate the result. 



NO. 1386, VOL. 54] 



protected thermo-junction is immersed in the lead when 

 it is fluid. The spot of light E from the mirror M, of the 

 galvanometer connected with the thermo-junction is 

 allowed to fall on the movable screen, and' as the lead 

 cools down the spot of light traverses the screen from 

 right to left. The consecutive positions 

 occupied by the centre of this spot of light 

 -. J are marked by hand with a stroke of char- 



coal. During the solidification of the metal 

 the spot of light remains in one position, 

 and consequently the portion of the curve 

 which represents the solidification of the 

 metal is the vertical line .v,y. If the second 

 sample of lead be treated in exactly the same 

 way, and its " cooling curve " traced, it will 

 be seen that the freezing point is lowered, 

 and a demonstration is thus afforded that 

 diffusion of gold has occurred in the lead 

 contained in the tube .\. 



The results of the diffusion of platinum 

 and gold in fluid lead during twenty-four 

 hours is shown diagrammatically in Fig. 2. 

 The lead was placed in tubes which were 

 arranged side by side in the oven, to which 

 reference has been made. The columns .\ 

 r. represent the actual length and diameter 

 of the columns of fluid lead. The spheres, 

 which are drawn to the left of the diagram, 

 are slightly smaller than the buttons of gold 

 and of platinum extracted from the several 

 sections, shown by horizontal lines, into which 

 the columns of lead were divided after the 

 metal had been allowed to solidify. The 

 curves represent the respective dift'usivity of 

 gold and of platinum. The %ertical ordinate represents 

 the distiincc in the direction in which diftusion takes place, 

 and the horizontal ordinates concentration of the diffusing 

 metal. Each of the metals gold and platinum which 

 diffused into the fluid column of lead occupied in the 

 form of an alloy rich in lead, the length a d, of the 



V* • ' 



