Abstract
Cu-Sn and Cu-Zn alloys with different solute contents were cast, and spectroscopic surface reflectance of the alloys was analyzed. In a range of the wavelength from 450 nm to 650 nm, sub-spaces composed of 2 and 3 eigen vectors were designed using the algorithm of principal component analysis. The spectroscopic distribution was reconstructed in the sub-space with high accuracy. In the Cu-Sn alloy, a projection onto the primary eigen vector took the maximum at a tin content of 7 mass % , and the projection onto the secondary eigen vector decreased monotonically with tin content. In the Cu-Zn alloy, the color changed complicatedly with zinc content in the binary sub-space. The surface reflectance of copper alloy castings was quantitatively predicted from the solute content in the alloy castings, and the solute content of the alloy was estimated from the spectroscopic surface reflectance.
