Nano-scratching in solution was performed to the single-crystal iron (100) surface passivated at 0.0-1.0 V (SHE) in pH 8.4 borate buffer solution to evaluate the friction coefficient of the iron (100) surface kept in the passive state and its potential dependence. The friction coefficient obtained with nano-scratching for the passive iron surface depended on normal force, i.e., normal displacement, which resulted mainly from the geometry of the diamond tip. In order to avoid the effect of the tip geometry on friction coefficient, the normalized friction coefficient was newly defined with dividing friction coefficient by geometrical factor. The normalized friction coefficient obtained with nanoscratching in solution for the iron (100) surface kept in the passive sate was significantly larger than those obtained with nano-scratching in air after passivation. The normalized friction coefficient obtained with nano-scratching in air after passivation was almost independent of potential in the passive region. On the other hand, the normalized friction coefficient obtained with nano-scratching in solution increased with increasing potential in the passive region. The difference between normalized friction coefficients obtained with nano-scratching in solution and in air was discussed by taking into account a series of mechano-electrochemical reaction (film rupture, active dissolution and repassivation) which would take place at the moving front of the diamond tip during nano-scratching in solution. The large potential dependence of the normalized friction coefficient obtained with nano-scratching in solution was explained in terms of the increase in repassivation rate at the film rupture sites with increasing potential in the passive region. (C) 2004 Elsevier Ltd. All rights reserved.