October 31, 1919] 



SCIENCE 



419 



tion depends upon the oxidizing properties of 

 the anion. In solutions of alkali salts with 

 anions of still greater oxidizing power, e. g., 

 bichromates and permanganates, passivity is 

 preserved indefinitely. These salts, as well 

 known, are strong passivating agents; a sat- 

 urated solution of KMnO^ passivates active 

 iron wires with one or two seconds exposure 

 or even less. 



The nature of the cation is also an im- 

 portant factor in the activating or passivating 

 action of a salt. Passive wires retain their 

 passivity longer in solutions of heavy metal 

 palts (especially nitrates) than in correspond- 

 ing solutions of alkali and alkali earth salts. 

 This is well illustrated by copper salts; in 

 m/20 CuCl, activation is rapid (1 to 2 sec- 

 onds), as in all solutions of chlorides; in 

 m/20 CuSO^ it usually requires from 5 to 10 

 minutes, and in m/20 CuCISTOg), more than 30 

 minutes. A similar relation is seen with 

 acids; m/10 HCl activates in less tlian 1 sec- 

 ond; m/10 HjSO^ requires several minutes; 

 while in m/10 HlSTOg passivity is preserved 

 indefinitely. The nitrates of metals with a 

 lower solution-tension than passive iron 

 (AgNOg, HgCNO^)^) also preserve passivity 

 indefinitely and their solutions act as strong 

 passivating agents. 



In solutions of copper salts the precise 

 activation-time can be measured with ac- 

 curacy in a single wire if the wire is kept 

 under close observation, for at the moment 

 when activation is complete metallic copper 

 begins to be deposited upon the bright surface 

 of the iron and in a few seconds the original 

 steel-white luster changes to a dull copper-red. 

 A large number of observations have been 

 made on the influence of different conditions 

 (especially presence of organic compounds, 

 oxidizing agents, and salts of noble metals) 

 upon the rate of activation in m/20 CuSO^ 

 and Cu(N0,)2; in such mixed solutions 

 antagonism-effects are well-marked and will 

 be described below. 



An interesting feature in the behavior of 

 passive wires in these solutions is that activa- 

 tion (involving as it does the deposition of 

 copper on the iron surface) has a marked in- 

 fluence upon the subsequent behavior of the 



same wire when it is repassivated and re- 

 turned to the same solution. Unless the 

 copper coating is removed very completely 

 from such a wire before repassivation, the 

 latter is always found to be distinctly more 

 reactive in the copper salt solution than be- 

 fore, i. e., activation occurs in a much shorter 

 time. My usual procedure in repassivating 

 wires after activation in solutions of copper 

 salts has been to dissolve off the adhering 

 copper with dilute HNO3 and then wipe the 

 wire carefully with a coarse cloth until every- 

 where bright and clean before placing in 

 strong HlSrOj for repassivation. But if the 

 copper is imperfectly removed before repassi- 

 vation (e. g., by simply rubbing the wire with 

 the cloth), and especially if the stay in strong 

 acid is brief, the wire is typically found to 

 become active in the salt-solution within a 

 briefer period than normally. Thus with 

 wires carefully cleaned before passivation, the 

 average time of activation in m/20 CuSO^, 

 with 36 trials with different wires, was 11 

 minutes; while with wires which were passi- 

 vated with some copper still adhering the 

 average of 31 trials was 3 minutes. In m/20 

 Cu(N03), the respective times were: care- 

 fully cleaned wires, 34.2 minutes; imperfectly 

 cleaned wires, 19.7 minutes (average of 20 

 trials in each case). Although the two sets 

 of wires hipe the typical steel luster after 

 passivation and are then indistinguishable in 

 appearance, it is probable that the sensitizing 

 effect of the previous activation is to be 

 referred to the presence of minute particles 

 of copper which remain adhering to the 

 surface of the iron and by acting as anodal 

 regions or nuclei exert upon the general film- 

 covered surface an influence favorable to 

 activation. Under normal conditions the con- 

 tact of metallic copper activates passive iron; 

 presumably, therefore, the presence of this 

 adhering copper would have the same in- 

 fluence in promoting activation as the simulta- 

 neous contact of numerous particles of the 

 metal at different points of the surface; the 

 activation-process is facilitated and occupies 

 a shorter time than before. Activation in a 

 solution of a copper salt thus renders the 

 metal for a time more sensitive to a second 



