Dear Mohammad, How can be possible that the two Si calculations (54 and 2-atom cell) give different band structure ? Aren't they both silicon ? With the same cutoff and equivalent kpoints? Shouldn't they give the same result ?
stefano (sent from my phone) > On 02 Aug 2016, at 04:45, Mohammad Abu Raihan Miah <[email protected]> > wrote: > > Hi Stefano, > > When I simulated for 54 atom Silicon and also for B doped Silicon(53+1), > indirect bandgap was found. Though in the result, the conduction band minima > is not in the correct position (i.e. not in 0.85 X, approximately in 0.65X). > > For the simulation of 2 atoms, indirect bandgap is also found. The result is > correct (i.e. CBM is at 0.85X) > > As underestimation of the bandgap value is well-known problem, it is also > found in both cases. But we are more interested in direct bandgap issue now. > > Any suggestion regarding this will be a great help for me. > > Thank you very much. > >> On Mon, Aug 1, 2016 at 11:40 AM, stefano de gironcoli <[email protected]> >> wrote: >> Dear Mohammad, >> >> I suggest you do the following exercise: >> Compute the band structure of 54 atom of Si with a similar setting to the >> one of your present calculation. Does it have direct gap ? compute the DOS. >> Compute the band structure of Silicon in the 2-atom unit cell >> with celldm(1) = 10.2623467, k_points (automatic) set to 15 15 15 0 0 0 >> Does it have a direct gap ? compute the DOS >> Compare total energy, band structure and DOS in the two cases. >> Which calculation is correct ? >> >> stefano >> >> >> >>> On 01/08/2016 20:19, Mohammad Abu Raihan Miah wrote: >>> Hi all, >>> >>> I am currently simulating band diagram for Phosphorus doped Silicon. I am >>> using 53 Si atoms and 1 P atom for computation. When I computed the band >>> diagram, it resulted in direct band gap, which is not true case for >>> Silicon. I have not relaxed the system. >>> >>> My input file is as follows, >>> >>> &control >>> calculation = 'scf', >>> restart_mode='from_scratch', >>> verbosity = 'high', >>> prefix = 'Silicon_P_54_relax', >>> tstress = .true., >>> tprnfor = .true., >>> pseudo_dir = '.', >>> outdir = 'out_P_rel', >>> wf_collect=.true., >>> max_seconds=10800, >>> / >>> >>> &system >>> ibrav=2, >>> celldm(1) = 30.787040, >>> nat=54, >>> ntyp=2, >>> ecutwfc=50, >>> ecutrho=400.0d0, >>> input_dft='PBE', >>> nbnd=120, >>> occupations='smearing', >>> degauss=0.05, >>> / >>> >>> &electrons >>> diagonalization='david', >>> mixing_mode = 'plain', >>> mixing_beta = 0.7, >>> conv_thr = 1.0d-6, >>> / >>> >>> ATOMIC_SPECIES >>> Si 28.085500 Si.pbe-n-rrkjus_psl.0.1.UPF >>> P 30.973800 P.pbe-n-rrkjus_psl.0.1.UPF >>> >>> ATOMIC_POSITIONS (alat) >>> Si 0.000000 0.000000 0.000000 >>> .................................. >>> P 0.083333 0.083333 0.083333 >>> ............................. >>> Si 0.750000 0.750000 0.750000 >>> >>> K_POINTS (automatic) >>> 5 5 5 0 0 0 >>> >>> Could anyone please explain the reason behind this change from indirect to >>> direct bandgap? Or any suggestion to solve this problem? >>> >>> Thank you very much. >>> >>> Best, >>> -- >>> Mohammad Abu Raihan Miah >>> PhD Student >>> Department of Electrical and Computer Engineering (ECE) >>> University of California, San Diego >>> >>> >>> _______________________________________________ >>> Pw_forum mailing list >>> [email protected] >>> http://pwscf.org/mailman/listinfo/pw_forum >> >> >> _______________________________________________ >> Pw_forum mailing list >> [email protected] >> http://pwscf.org/mailman/listinfo/pw_forum > > > > -- > Mohammad Abu Raihan Miah > PhD Student > Department of Electrical and Computer Engineering (ECE) > University of California, San Diego > _______________________________________________ > Pw_forum mailing list > [email protected] > http://pwscf.org/mailman/listinfo/pw_forum
_______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
