OS THE ELECTROLYTIC METHODS OF QUANTITATIVE ANALYSIS, 305 



be noted, however. In several experiments with chloride present it was 

 found that there was deposition of nickelic hydroxide on the anode, 

 although the normal quantity of ammonia was employed. This is pro- 

 bably due to the destruction of ammonia at the anode in this case, over 

 and above th^t lost by neutralisation. If the discharged ohlor-ions act 

 like free chlorine, we can compare the action with sulphate and with 

 chloride, as represented by the equations — 



12NH3 + 6S04 + 6HoO=:6(NH,).S04 + 30., 

 16NH3 + 12C1=12NH4C1-1-2N2 



Assuming that these are the only actions which take place, it is evident 

 that for the same quantity of current the quantity of free ammonia lost is 

 one-third greater with the chloride. It is therefore advisable in this case 

 to increase the amount of ammonia to 6 to 7 grme., unless a weaker 

 current is employed. 



In cases where nickel is to be deposited from a solution originally 

 containing chloride only, ammonium chloride might perhaps be employed 

 with advantage in place of ammonium sulphate. The formation of 

 sulphide would then be quite impossible. 



Nitrates, unless present only in small quantity, should be destroyed 

 previous to electrolysis, as their presence considerably retards the deposi- 

 tion of the metal. This is evident from the much longer time necessary 

 to decolourise the solution. In course of time the nitrate becomes reduced 

 by the current, and the deposit ultimately obtained is perfectly good, being, 

 in fact, exceptionally lustrous in appearance. 



The presence of fhosphates in the solution is immaterial, provided the 

 precipitation of nickel phosphate is avoided when the ammonia is added. 

 To reduce the risk of precipitation, the ammonia should be added last of 

 all, after the ammonium sulphate and water, and it can be taken that it is 

 all quickly mixed with the liquid. Perfectly good results may also be 

 obtained by using ammonium phosphate alone in place of the sulphate. 



What has been said of phosphate applies equally to arsenate, a matter 

 of consixlerable importance in connection with the assay of nickel ores. 

 There is apparently not the slightest reduction to arsenite, which would 

 undergo further reduction to arsenic. The decanted liquid gives no trace 

 of an immediate precipitate on the addition of hydrochloric acid and 

 hydrogen sulphide. 



The presence of arsenite is wholly inadmissible, as in that case the 

 deposit is quite black, powdery, very loosely coherent, and contains large 

 quantities of arsenic. If it is desired to determine nickel in solutions 

 containing arsenic compounds, it is therefore necessary either to completely 

 oxidise them or to remove them previous to electrolysis. 



Chromates have a very marked and striking effect. The presence of a 

 very small quantity completely prevents deposition, even when electrolysis 

 is continued for a long time. When present it would therefore be neces- 

 sary to get rid of them by one of the ordinary methods before proceeding 

 to determine the nickel electrolytically. 



The addition of sulphites to baths for electroplating with nickel has 

 been recommended, and experiments were tried with varying quantities 

 of ammonium sulphite present in solution. The resulting deposits are 

 exceedingly bright and lustrous, but this seems to be the only advantage. 

 An excessive quantity retards deposition somewhat, so that there is no 

 practical benefit attending the use of sulphite for analytical work. This 



1898. X 



