The effects of pulsed substrate biasing on thin amorphous carbon coatings: a statistical design of experiment study

A statistical design of experiment (DOE) was used to investigate the effects of medium frequency pulsed substrate bias voltage, pulse frequency and pulse width on 150 nm sputtered C coatings doped with Cr. Within the scope of this work, the DOE treatment of nanoindentation/nanoscratch, atomic force microscopy and x-ray photoelectron spectroscopy results were shown to be a successful method for investigating and potentially optimizing very thin sputter ion plated coatings. Bias voltage was shown to be the most significant of the three bias factors for mechanical properties. Bias frequency and pulse width showed effects that partly matched the voltage results and this was discussed with regard to an increase in peak voltage accompanying an increase in frequency and pulse width. For surface topography, the results were more complicated with the significance of each parameter varying according to the property measured. This work also demonstrated the complex inter-relationships that exist between the three bias parameters, meaning that any attempt to optimize the pulse bias condition for a given coating property would require a consideration of all three factors (within the available operating window of the pulse unit).