ON THE CHEMICAL COMPOUNDS CONTAINED IN ALLOYS. 



135 



phenomena show themselves on the branch ghi ; the summit h occurs 

 at the formula AuAlo, and large crystals of this body are found embedded 

 in mother substance in all alloys between g and i. 



These criteria taken together — ^(1) the occurrence of a summit at a 

 formula percentage, (2) the presence of large crystals of the same kind, 

 decreasing in amount as we descend the branch on either side — are, I 

 believe, an absolute proof of the reality of a compound. The two bodies 

 AuqAI and AuAL, as certainly exist as do the two chlorides of copper. 

 But there are other compounds more obscurely indicated by the curve ; 

 the branch fq has its summit, x, below the branch gh, hence the x 

 body which crystallises at points on gp never occurs alone in a solid 

 alloy. The microscope shows that solid alloys between G and H contain 

 large crystals of AuAl,, surrounded by a coating of the x body, and that 

 oatside this coating there are large independent crystals of x embedded 

 in a minute conglomerate of x and e. The fact is that in the first stage 

 of freezing, while the large crystals of AuAlj, or h, are forming, the 

 liquid part necessarily gets richer in gold until it reaches the composition 

 G. From this moment the H 



crystals cease to form, and the Fig. 3. 



existing ones become coated 

 with the x body, which at 

 lower temperatures crystallises 

 independently in large crys- 

 tals, that in a certain sense 

 are primary. Finally, the 

 residual liquid reaches the state 

 F, and the eutectic conglo- 

 merate forms which is com- 

 posed of crystals of AU2AI 

 and of X. Thus an alloy may 

 contain a compound although 

 that compound does not occur 

 as the sole constituent of any 

 particular alloy. This appears 

 to be a very common case ; for 

 example, Mr. Stead finds that 

 in bronzes very rich in tin the 

 crystals of Cu-^Sii are coated 

 with CuSn, and M. Charpy (9) 

 has recorded a similar feature 

 in the bronzes very rich in 

 copper. In both these cases the 

 eutectic lies outside the coating. • 



MM. Le Chatelier ('*), Gautier, and Gosselin ('°) have traced a number 

 of remarkable freezing-point curves for pairs of metals, of which 

 examples are given here. Unfortunately the composition is stated in 

 percentage by weight, not in atomic per cents. 



Copper- Aluminium (fig. 3). — Here we find two well-marked summits, 

 one very exactly at CugAl, the other near CuAla. 



Copper-AntimonT/ (fig. 3). — Here there is a well-marked intermediate 

 summit which, if it were at a formula point, would almost certainly 

 indicate a compound. M. Le Chatelier attributes the formula CuoSb to 

 it, but measurement on the curve seems to give the summit the formula 

 Cu^Sbj. These two curves, and one of SnCu also due to M. Le Chatelier, 



1000 



900 



800 



700 



Al 



Sb 



6OO 



soo 



400 



0% 10 20 30 40 SO 60 70 80 00 100 



