Dear Nguyen Chuong It puzzled me for quite a long time. But I have checked the relation of the output constraint energy and the output magnetic moment, and experienced several calculations with QE5.0.2. It seemed that that when the starting magnetizations are interpreted as the target values of magnetization moments in a constained calculation, they are in unit of Bohr not Bohr/e. And their meaning is magnetic moment rather than magnetic moment over charge. "Values range of starting_magnetization at -1 (all spin down for the valence electrons of atom) to 1 (all spin-up)" is only true when they are used to generate trial wave functions. You can check this by run the inputfile. If I have misunderstandings, Please tell me. -- PostDoc Department of Physics, South University of Science and Technology of China
