15 
1911-12.] On the Reduction of Ferric Iron. 
reduction is that the solution should contain little free acid, but the mixed 
solution should not be alkaline to litmus. In practice one can ensure rapid 
reduction by addition of sufficient alkali to cause partial precipitation of 
ferric hydroxide, followed by treatment with sulphurous acid. 
II. Note on the Reduction of Ferric Iron by Zinc Dust 
AND BY SOME ZlNC ALLOYS. 
Reduction by zinc, even in the granulated form, is slow, and in many 
respects unsatisfactory. Devada’s alloy (an alloy of zinc, copper, and 
magnesium) has been recommended, and appears to reduce more quickly 
than zinc. Aluminium in sheet form has been found to reduce quickly and 
completely, but leaves an objectionable amount of dross. Some experiments 
in this laboratory indicated that an alloy of zinc with 5 to 10 per cent, of 
aluminium was an efficient reducing agent for iron, and we have therefore 
carried out some comparative tests. 
All experiments on reduction with alloys were performed in test-tubes 
of equal size placed in a thermostat at 60°. The rate of reduction was 
tested in two ways : firstly, by titration of the amount reduced in ten 
minutes ; and, secondly, by finding the time necessary for complete reduction. 
The alloys were all cast in the same mould, and pieces of as nearly equal 
size as possible were employed. Three alloys of zinc and aluminium were 
tried with 5, 10, and 25 per cent, of aluminium respectively. For com- 
parison, commercial “ pure ” zinc was used. A silver-zinc alloy containing 
1 per cent, of silver, and a cadmium-zinc alloy containing 2 per cent, of 
cadmium were also tested. 
The results obtained were somewhat discordant, but from a considerable 
number of experiments the following conclusions have been reached : — 
(1) The reduction appears to be little influenced by the amount of acid, 
but excess of acid reduces the rate of reduction. 
(2) No difference could be detected when hydrochloric acid was used 
instead of sulphuric acid. 
(3) No difference could be detected between the rate of reduction by 
pure zinc and by the alloys mentioned above. 
(4) Individual results varied greatly from the mean, but the variations 
appeared to be due to difference in the surface exposed. It appeared that 
the area exposed had an effect which was overwhelmingly large compared 
with all other factors. 
Zinc dust was accordingly tried, and our results confirm those of 
D. J. Carnegie ( Journ . Ghem. Soc., 53, 468, 1888), that this is an extremely 
useful and rapid method of production. Carnegie recommends the use of a 
