On Mon, 2009-03-30 at 21:59 +0800, lfhuang wrote: > Any original papers or good papers on the parameter "use_masses" ?
There is no paper on use_masses (as far as I know). Normaly one does not need to use use_masses. This parameter was added out of experience with some systems for the quick-min NEB minimization (i.e. quick-min is a kind of damped dynamics). If some bond is much stiffer than the others then this bond will prevent using larger ds (=time step). Say for example that you want to break C-H bond of adsorbed C2H2 molecule. The triple CC bond is much stiffer than the others, and in this case one may assign to C atom larger "fictitous" mass compared to other atoms. For example you can assing to C a mass of, say, 1.5 and to H the mass of 1.0. If use_masses aids in faster convergence in a given case, and what is the optional ratio between masses, have to be tested. It is difficult to tell it in advance. Regards, Tone -- Anton Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822) Please, if possible, avoid sending me Word or PowerPoint attachments. See: http://www.gnu.org/philosophy/no-word-attachments.html
