[QE-users] Can phonon DOS of an isolated molecule be calculated?

[ykhuang at dicp.ac.cn]

Greeting, QE users!
I meet problems when I want to calculate IR spectrum and free energy of isolated CO molecule using fqha.x code.


At first I run opt with high precision:
    &CONTROL
    calculation = 'relax'
    forc_conv_thr = 3.8e-5
    outdir = './CO/'
    prefix = 'scf'
    pseudo_dir = './'
    restart_mode = 'restart'
    tefield = .FALSE.
    verbosity = 'high'
    wf_collect = .TRUE.
    /
    &SYSTEM
    degauss = 0.002
    ecutrho = 300
    ecutwfc = 50
    ibrav = 0
    nat = 2
    noncolin = .FALSE.
    nspin = 1
    ntyp = 2
    vdw_corr = 'grimme-d3'
    /
    &ELECTRONS
    conv_thr = 1e-10
    electron_maxstep = 100
    mixing_beta = 0.1
    mixing_mode = 'plain'
    scf_must_converge = .TRUE.
    startingwfc = 'file'
    /
    &IONS
    ion_dynamics = 'bfgs'
    /
    ATOMIC_SPECIES
    C 12.0107 C.pbe-n-kjpaw_psl.1.0.0.UPF
    O 15.9994 O.pbe-n-kjpaw_psl.0.1.UPF


    K_POINTS {automatic}
    1 1 1 0 0 0


    CELL_PARAMETERS {angstrom}
    20.00000 0.00000 0.00000
    0.00000 20.00000 0.00000
    0.00000 0.00000 20.00000


    ATOMIC_POSITIONS {angstrom}
    C 10.50000000 10.00000000 10.00000000
    O 9.50000000 10.00000000 10.00000000


then I run ph.x to get dynamic matrix:
    &INPUTPH
    outdir='./CO/'
    prefix='scf'
    tr2_ph=1.0d-12
    ldisp=.TRUE.
    nq1=1
    nq2=1
    nq3=1
    epsil=.TRUE.
    trans=.TRUE.
    asr=.TRUE.
    /


I want to get IR spectrum by dynmat.x:
    &input
    fildyn='./matdyn1'
    asr='zero-dim'
    /


,and q2r.x, matdyn.x to obtain phonon DOS:
input of q2r.x:
    &INPUT
    fildyn='./matdyn'
    zasr='zero-dim'
    flfrc='default.fc'
    /
input of matdyn.x:
    &INPUT
    asr='zero-dim'
    flfrc='default.fc'
    dos=.TRUE.
    nk1=1
    nk2=1
    nk3=1
    amass(1) = 12.0107
    amass(2) = 15.9994
    /


However, 
1. I obtained no IR information in ouptut of dynmat.x, The output file ends with:"
     ...Force constants read
     ...epsilon and Z* read
     Acoustic Sum Rule: || Z*(ASR) - Z*(orig)|| =    1.057617E+00
     Acoustic Sum Rule: ||dyn(ASR) - dyn(orig)||=    8.730790E-01
 ", and I have no idea about it. :(


2. The DOS value, second column of DOS file (obtained from matdyn.x) is full of 0.000000E+00, although I've got correct vibration frequencies in the output of ph.x. After consideration, I doubt whether it is meaningful to obtain "regular" DOS pattern of isolated molecule, cuz according to my understanding, DOS of isolated molecule is discrete and may contain several delta functions, rather than a continuum function of frequencies of bulk material. Is this the reason why I get full-of-zero phonon DOS? If so, should I smooth DOS manually? 
There's also a relevant question: I've calculated another system in which CO is absorbed. Vibration information of CO is obtained as usual, however, DOS value of frequencies near CO stretching mode frequency are zero, which is similar with what I get in my isolated CO case. DO these two problems have the same reason? Or I made the same mistake when I calculate these two systems?


Answers will be appreciated. :)


Best regards!


Yike Huang, PhD candidate.
Dalian Institute of Chemical Physics.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200325/8bdf9728/attachment.html>