Abstract
The complexity of simulating indoor corrosion conditions has been elucidated by investigating a series of representative laboratory corrosion tests for electrical contact materials. The chemical composition of corrosion products formed during laboratory exposures of Cu substrates and Ni‐, Au‐ and Sn‐electroplated samples has been analyzed by x‐ray photoelectron spectroscopy and compared with results after field exposures in three telephone central offices. None of the evaluated laboratory tests is able to reproduce fully the corrosion products formed during actual field tests. The greatest similarities in chemical composition have been achieved in a laboratory atmosphere containing 
(0.25 ppm) and 
(1.8 ppm) which is able to produce several major phases also identified after field exposures in the telephone central offices. The results indicate that further improvement in similarities may be obtained by using a lower level of
than was used in this work, and by introducing 
as a possible additional pollutant which may partly replace 
. The conclusions drawn from this work are restricted to environments similar to those found at present field sites.