Dear Lammps users and devs
I’m trying to use the “fix-phonon” to calculate the phonon dispersion for MoS2 Monolayer using a potential i’ve been developing using machine learning and DFT data. I was able to reproduce the phonon dispersion using lammps as a tool in a package called Alamode ( T. Tadano, Y. Gohda, and S. Tsuneyuki, J. Phys.: Condens. Matter 26 , 225402 (2014)) without problems at gamma
I’m having problems next gamma. the phonon dispersion explodes to negative values and positive values of order of >10^3 thz. If i include the option “-r” at “./phana” even if knowing I should not use it because I’m not including charges on my systems the problem of exploding values at gamma dissapear but I end up with highly unnacurate values for frequencies at Gamma.
I’m making simulations using 50x50x1 cell (7500 atoms total) and for timestep = 1 fs for 5000000 steps.
Here’s the input im using:
Blockquote
atom_style atomic
boundary p p p
read_data MoS2.data.huge_phon.pos
velocity all create 300 35131 dist gaussian
group S type 1
group Mo type 2
pair_style aenet
pair_coeff * * v-1 S Mo good_phonon0.ann S Mo
timestep 0.001 #ps
thermo 500 # intervalo para mostrar output instantaneos.
neighbor 2.0 bin
neigh_modify every 3 delay 0 check yes
dump myDump all xyz 1000 dump.MoS2_300K.lang.xyz
dump_modify myDump element S Mo
fix 1 all langevin 300 300 0.1 73504 zero yes
fix 2 all nve
fix 3 all phonon 10 50000 1000000 mos2_huge_map.in mos2Phonon_lang nasr 50
run 5000000
thanks in advance for the help