Abstract
An attempt has been made to form the ordered phases Cu5Mn, Cu3Mn (and CuMn), by long-term annealing within the proposed phase boundaries. However neutron diffraction experiments provide no evidence for superlattice formation at Cu83.3Mn16.7 and Cu75Mn25, despite the extreme sensitivity of the neutron method to chemical ordering in this particular case. In addition, the changes in the short-range order produced by annealing are found to be insignificant, so these experiments suggest that the phases Cu5Mn and Cu3Mn do not exist. Furthermore there is no evidence of an CuMn ordered structure at the equiatomic composition, and two-phase specimens are observed. This appears to be consistent with either an inhomogeneous solidification or miscibility gap effects of the kind established for manganese-rich alloys.