October 31, 1919] 



SCIENCE 



417 



were first passivated by esposui'e to 1.42 

 HNOg; they were then washed thoroughly in 

 distilled water^ and transferred to the solution 

 under examination ; after the lapse of a meas- 

 ured time of exposure they were again washed 

 in water and placed in dilute HlSTOg {s. g., 

 1.20). If activation has occurred in the solu- 

 tion, darkening of the surface and efferves- 

 cence begin instantly; while if the exposure 

 has been insufficient, no effect is seen and the 

 wire remains passive. 



In most solutions of salts and other electro- 

 lytes a spontaneous return of activity occurs 

 after a g-reater or less interval, the length of 

 which depends upon the nature and concentra- 

 tion of the compound. In the total effect 

 produced by the solution both classes of ions 

 are concerned; and no relation of activating 

 or passivating influence to the sign of ionic 

 charge is apparent. All ions of strong oxi- 

 dizing properties tend to stabilize the passive 

 condition (e. g., Cr^Oj, MnO^, Ag, Au, Pt), 

 while others (especially CI, Br, I) have a 

 strong activating influence, referable possibly 

 to a colloidal (aggregative or dispersive) 

 action upon the film. In pure solutions of 

 alkali and alkali earth salts (except those with 

 strongly oxidizing anions like MnO^ and 

 Cr^Oy) passivity is always soon destroyed, 

 within periods ranging from a few seconds or 

 minutes to half an hour or sometimes longer. 

 The presence of oxygen in the anion (espe- 

 cially if in a terminal position) appears 

 always to retard the process of activation. 

 In solutions of nitrates, chlorates, phosphates, 

 sulphates, carbonates, tartrates, citrates, ace- 

 tates (»i/2 to m/20) passivity remains for 

 usually several minutes; activation is more 

 rapid with hydrates; while halides in' much 

 lower concentration activate within a few 

 seconds. 



Different specimens of iron vary consider- 

 ably in the time required for activation in a 



1 Paspive wires retain their passivity in distilled 

 water for a considerable period — ranging in vari- 

 ous experiments from 15 to 23 minutes at room 

 temperature — but not indefinitely, a fact again 

 indicating that the passive state is preserved only 

 when there ds the possibility of continued rapid 

 oxidation. 



given solution, apparently because of varia- 

 tions in the finer structure of the metal. AU 

 of the following observations were made with 

 the same kind of wire (no. 20 piano wire, ca. 

 1 mm. in diameter) ; the metal of this wire is, 

 however, not homogeneous; in particular it 

 was always found that freshly cut and sliglitly 

 used wires were decidedly more sensitive to 

 activation than wires which had been used 

 for some time and in which the outer layer 

 had been dissolved away. Apparently the 

 surface-layer of a drawn wire is less homo- 

 geneous than the core, and hence forms a less 

 uniform and stable passivating surface-film. 

 To obviate this soui-ce of irregularity, in the 

 following experiments wires were used which 

 had been reduced by the acid to two thirds or 

 less of the original diameter. Such wires 

 when passivated exhibit a relatively uniform 

 behavior in salt" solutions, although in the 

 more slowly activating solutions the exact 

 time required for activation still shows con- 

 siderable variation. This variability has prob- 

 ably the same basis as the variability in the 

 rusting properties of different specimens of 

 iron. In each trial of a particular solution in 

 any series of experiments several independent 

 determinations with different wires are there- 

 fore necessary; as a rule these show good 

 agreement, with occasional well-marked varia- 

 tions, due presumably to accidental variations 

 of structure or composition in the exposed 

 surface of the metal. 



Halides. — Solutions of metallic chlorides, 

 including those of noble metals (Hg, Au, Pt), 

 activate passive iron with great rapidity. 

 The following series of experiments with 

 four chlorides are typical and illustrate the 

 dependence of time of activation upon con- 

 centration; they also show the relative unim- 

 portance of the cation in such solutions. In 

 each solution seven trials with seven separate 

 wires were made with each length of exposure; 

 time was marked by a metronome; the wires 

 were transferred by glass hooks from the dis- 

 tilled water to the tubes containing the solu- 

 tions. The time given is the usual minimal 

 exposure required to render the passive wire 

 reactive to dilute HNO3 ; with briefer ex- 

 posures the wires remain passive. 



