Error whilst calculating the superconducting gap

[eliephys78]

Dear all,

I am running the Pb example to solve the isotropic Eliashberg equations to obtain the superconducting edge gap and Tc at various temperatures. EPW performs the first run but then crashes with a mysterious error involving several "tasks":

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
task # 57
from mix_broyden : error # 1
factorization
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
task # 23
from mix_broyden : error # 1
factorization
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
.
.
.
.

How do I solve this:. the input file is below:

--
&inputepw
prefix = 'pb',
amass(1) = 207.2
outdir = '/fefs1/physics/eamoujaes/Pb-EPW/epw/OUT/'

elph = .true.
ep_coupling =.true.
kmaps = .false.
epbwrite = .true.
epbread = .false.
ephwrite =.true.
epwwrite = .true.
epwread = .false.
restart=.true.

nbndsub = 4
bands_skipped = 'exclude_bands = 1-5'

wannierize = .true.
num_iter = 300
dis_win_max = 21
dis_win_min = -3
dis_froz_min= -3
dis_froz_max= 13.5
proj(1) = 'Pb:sp3'

wdata(1) = 'bands_plot = .true.'
wdata(2) = 'begin kpoint_path'
wdata(3) = 'G 0.00 0.00 0.00 X 0.00 0.50 0.50'
wdata(4) = 'X 0.00 0.50 0.50 W 0.25 0.50 0.75'
wdata(5) = 'W 0.25 0.50 0.75 L 0.50 0.50 0.50'
wdata(6) = 'L 0.50 0.50 0.50 K 0.375 0.375 0.75'
wdata(7) = 'K 0.375 0.375 0.75 G 0.00 0.00 0.00'
wdata(8) = 'G 0.00 0.00 0.00 L 0.50 0.50 0.50'
wdata(9) = 'end kpoint_path'
wdata(10) = 'bands_plot_format = gnuplot'

elecselfen = .false.
phonselfen = .false.
eliashberg=.true.
liso=.true.
limag=.true.
lpade=.true.
lacon=.true.
muc=0.09
temps(1) = 0.3
temps(2) = 0.9
temps(3) = 1.5
temps(4) = 2.1
temps(5) = 2.7
temps(6) = 3.3
temps(7) = 3.9
temps(8) = 4.1
temps(9) = 4.3
temps(10) = 4.4
temps(11) = 4.5


fsthick = 0.5 ! eV
degaussw = 0.005 ! eV
wscut =0.2

dvscf_dir = '/fefs1/physics/eamoujaes/Pb-EPW/phonons/save'

nk1 = 6
nk2 = 6
nk3 = 6

nq1 = 6
nq2 = 6
nq3 = 6
mp_mesh_k=.true.
nkf1 = 12
nkf2 = 12
nkf3 = 12

nqf1 = 12
nqf2 = 12
nqf3 = 12
/
16 cartesian



Thanks in advance

Elie Moujaes
Federal university of Rondonia
Brazil

[hpaudya1]

Hi Elie,

This is an expected error for the temp. greater than (close to) the Tc.

You can see the WARNING in your output file as follows;

WARNING WARNING WARNING

The code may crash since tempsmax = XXX K is larger than Allen-Dynes Tc = YYY K

I am not sure in which temp the code crashed, and what is the value of the SC gap at that temp. The code crashed because the SC gap is too small and it is difficult to converse. I expect the calculations are fine for the lower temperatures.

Best,
Hari Paudyal

[eliephys78]

Dear hpaudya1,

Thanks for your time to reply to y message. However, I cannot see any warning in the output file before the code crashed. The temperature at which the code crashed is actually 0.3 K . the estimated Tc from Allen Dynes is around 12.37K. I ma trying to reproduce the Pb example in the EPW folder. These are the last lines before the CRASH:

temp( 1) = 0.30000 K

Solve isotropic Eliashberg equations on imaginary-axis

Total number of frequency points nsiw( 1) = 739
Cutoff frequency wscut = 0.1201

iter ethr znormi deltai [meV]
1 8.768002E-01 8.459037E+00 3.256073E+00
2 3.261105E-01 7.829624E+00 4.117034E+00
3 4.049062E-01 6.829744E+00 5.652636E+00
4 2.219816E-01 6.077911E+00 7.090387E+00
5 1.927775E-01 5.506282E+00 8.405549E+00
6 8.931713E-02 5.802137E+00 7.719953E+00
7 9.292297E-02 5.557825E+00 8.303817E+00
8 4.564069E-02 5.676160E+00 8.016786E+00
9 6.694556E-03 5.659049E+00 8.057643E+00
10 1.271638E-03 5.662163E+00 8.050115E+00
11 6.477691E-04 5.660508E+00 8.054085E+00
12 1.116250E-04 5.660201E+00 8.054814E+00
13 3.584403E-05 5.660293E+00 8.054593E+00
14 2.766078E-06 5.660286E+00 8.054610E+00
Convergence was reached in nsiter = 14

iaxis_imag : 0.06s CPU 2.39s WALL ( 1 calls)


Pade approximant of isotropic Eliashberg equations from imaginary-axis to real-axis
Cutoff frequency wscut = 0.1200

pade Re[znorm] Re[delta] [meV]
666 NaN NaN
Convergence was reached for N = 666 Pade approximants

raxis_pade : 0.10s CPU 4.16s WALL ( 1 calls)


Analytic continuation of isotropic Eliashberg equations from imaginary-axis to real-axis

Total number of frequency points nsw = 5000
Cutoff frequency wscut = 0.1200

[eliephys78]

I have increased the first value of the temperature from 0.3K to 0.9 K and it worked.

I was able to resolve the issue.

Thanks for your help