Dear Wien2k users!
I am trying to minimize the GaP supercell with Nitrogen-Nitrogen interstitial pair using min -i 100 -s 10 -j 'run_lapw -p -I -i 40 -ec 0.001 -fc 1.0 -cc 0.001' I constructed the supercell with 64 atoms and replaced one P atom by the pair of interstitial N atoms. After 16 cycles the program stopped with following message: . . . . . . . . . . . in cycle 16 ETEST: .0000099600000000 CTEST: -.0009901 ec cc and fc_conv 1 1 1 > stop Fallback to compatibility mode with "old" save_lapw broyden files deleted, clm*, dmat*, vorb*, vresp*, eece*, scf and struct files saved under GaPNN_1 27.977u 0.412s 0:28.89 98.2% 0+0k 0+20648io 0pf+0w 0.844u 0.336s 0:01.66 70.4% 0+0k 1272+16896io 8pf+0w clmextrapol_lapw has generated a new GaPNN.clmsum and calculations were renewed. Further, after 12 cycles, the program stopped again with the message: . . . . . . . . in cycle 12 ETEST: .0000040600000000 CTEST: -.0009842 ec cc and fc_conv 1 1 1 > stop :WARNING: Step size reduced due to overlapping spheres -- check RMT 28.185u 0.408s 0:29.11 98.1% 0+0k 0+20648io 0pf+0w 0.872u 0.320s 0:01.53 77.7% 0+0k 0+16896io 0pf+0w clmextrapol_lapw has generated a new GaPNN.clmsum and so on without ending. Initially the RMTs for Ga and P atoms were RMT(Ga,P)=2.0 and for N atom ? RMT(N)=1.3 In isolated nitrogen molecule the interatomic distance is approximately 1.295 Angstrom. I guessed that the interatomic distance between the interstitial N atoms in GaP supercell would be larger then 1.295 A. So I reduced the RMT for N atom and took RMT(N)=1.0, aiming to prevent the overlapping of spheres for N-N pair. (2*RMT(N)=1.06 A < 1.295 A). In spite of that, the result of ?mini? remains the same ? the interatomic distance between the interstitial N atoms becomes even smaller (!!!) then that in isolated nitrogen molecule and I still have the problem with overlapping spheres.
