We are working on some sintered ceramic samples that have different surface colours. Energy dispersive X-ray spectroscopy does not yield surface compositions that would explain a change in colour. We can see that some samples have different surface roughness that is associated to surface porosity due to incomplete sintering and grain growth. Some of the samples appear to have have grain/pore size on the same scale as the red portion of the visible spectrum. Thus I am thinking that some samples appear blue as the longer wavelengths may be being absorbed, whereas the shorter blue wavelengths are being reflected to give a blue appearance under natural light. I have searched literature for surface roughness induced colour shift but have not found anything useful. Can anybody point me to some relevant references? Thank you.