DFT taking so long

KamalEldeen Kamal

unread,

Sep 9, 2020, 2:12:31 PM9/9/20

to NWChem Forum

Hi I'm a beginner with Nwchem and I'm trying to calculate the binding energy of Silver Ion on some quantum dot (g-c3n4) by performing optimization step then a single point energy. but the job is taking so long. I am using an instance on google cloud platform with 52gb Ram and the jib is still taking so long. any help?!

attached my input file for the job

Edoardo Aprà

unread,

Sep 9, 2020, 2:16:34 PM9/9/20

to NWChem Forum

Please post input and output files

KamalEldeen Kamal

unread,

Sep 9, 2020, 2:51:03 PM9/9/20

to NWChem Forum

echo

start molecule

title "Heptazine-Ag"

charge 0

memory global 52 gb

geometry units angstroms print xyz autosym

N 5.62100 -2.20900 -2.16900

N 4.45400 -4.24800 -1.63100

N 4.57100 -0.15300 -1.82200

N 3.45600 -2.14600 -1.23300

N 2.24600 -4.14200 -0.88100

N 2.51000 -0.08300 -0.59700

N 1.30700 -2.03000 -0.27000

N 3.50700 1.86300 -1.35900

N 2.41200 3.77100 -0.64000

N 1.20500 2.01400 -1.73700

N 1.27600 5.66000 0.11300

N 0.05800 3.82800 -0.75200

N -1.12000 2.03400 -1.73700

N -1.13000 5.68000 0.11300

N -2.29800 3.81100 -0.64000

N -0.00900 -3.96600 -0.43200

N -1.29100 -2.00800 -0.27000

N -2.26800 -4.10300 -0.88000

N -2.46100 -0.04000 -0.59800

N -3.44300 -2.08700 -1.23200

N -4.47600 -4.17100 -1.62900

N -4.52300 -0.07500 -1.82200

N -5.60900 -2.11300 -2.16800

N -3.42500 1.92200 -1.36000

N -5.62800 2.35900 -0.66000

N -4.65100 3.60900 -2.43200

N -6.88300 4.06700 -1.74400

C 5.49600 -3.53300 -2.05700

C 4.58200 -1.48100 -1.74700

C 3.38300 -3.55100 -1.23500

C 3.53100 0.46600 -1.25200

C 2.40200 -1.40400 -0.67600

C 1.23900 -3.33200 -0.51700

C 2.30200 2.58500 -1.24400

C 1.28200 4.43800 -0.42900

C 0.04700 2.59300 -1.42900

C 0.07800 6.20200 0.33100

C -1.15600 4.45900 -0.42900

C -2.20800 2.62400 -1.24400

C -1.24600 -3.31100 -0.51700

C -2.37600 -1.36300 -0.67600

C -3.39400 -3.49300 -1.23400

C -3.47300 0.52600 -1.25200

C -4.55800 -1.40300 -1.74700

C -5.50700 -3.43900 -2.05500

C -4.63500 2.66700 -1.49000

C -6.73200 3.08600 -0.84900

C -5.80100 4.28400 -2.49800

C 4.72900 2.58700 -1.48900

N 5.71800 2.26100 -0.66000

N 4.76100 3.52800 -2.43100

C 6.83300 2.97000 -0.84800

C 5.92300 4.18300 -2.49700

N 7.00100 3.94800 -1.74300

C -0.02100 -5.40000 -0.66300

N -0.02700 -6.15800 0.42300

N -0.02600 -5.80600 -1.92800

C -0.03900 -7.47100 0.15500

C -0.03700 -7.13400 -2.06100

N -0.04400 -8.02100 -1.05900

H 6.36400 -4.11200 -2.36700

H -6.38500 -4.00300 -2.36500

H -5.86300 5.07400 -3.24200

H -7.58000 2.86600 -0.20400

H 0.08600 7.20600 0.75300

H 5.99900 4.97300 -3.24100

H 7.67800 2.73500 -0.20400

H -0.04400 -8.14500 1.00900

H -0.04100 -7.52800 -3.07500

Ag 0.78200 0.32000 -0.80700

#Symmetry C1

end

geometry ligands units angstroms print xyz autosym noautoz

N 5.62100 -2.20900 -2.16900

N 4.45400 -4.24800 -1.63100

N 4.57100 -0.15300 -1.82200

N 3.45600 -2.14600 -1.23300

N 2.24600 -4.14200 -0.88100

N 2.51000 -0.08300 -0.59700

N 1.30700 -2.03000 -0.27000

N 3.50700 1.86300 -1.35900

N 2.41200 3.77100 -0.64000

N 1.20500 2.01400 -1.73700

N 1.27600 5.66000 0.11300

N 0.05800 3.82800 -0.75200

N -1.12000 2.03400 -1.73700

N -1.13000 5.68000 0.11300

N -2.29800 3.81100 -0.64000

N -0.00900 -3.96600 -0.43200

N -1.29100 -2.00800 -0.27000

N -2.26800 -4.10300 -0.88000

N -2.46100 -0.04000 -0.59800

N -3.44300 -2.08700 -1.23200

N -4.47600 -4.17100 -1.62900

N -4.52300 -0.07500 -1.82200

N -5.60900 -2.11300 -2.16800

N -3.42500 1.92200 -1.36000

N -5.62800 2.35900 -0.66000

N -4.65100 3.60900 -2.43200

N -6.88300 4.06700 -1.74400

C 5.49600 -3.53300 -2.05700

C 4.58200 -1.48100 -1.74700

C 3.38300 -3.55100 -1.23500

C 3.53100 0.46600 -1.25200

C 2.40200 -1.40400 -0.67600

C 1.23900 -3.33200 -0.51700

C 2.30200 2.58500 -1.24400

C 1.28200 4.43800 -0.42900

C 0.04700 2.59300 -1.42900

C 0.07800 6.20200 0.33100

C -1.15600 4.45900 -0.42900

C -2.20800 2.62400 -1.24400

C -1.24600 -3.31100 -0.51700

C -2.37600 -1.36300 -0.67600

C -3.39400 -3.49300 -1.23400

C -3.47300 0.52600 -1.25200

C -4.55800 -1.40300 -1.74700

C -5.50700 -3.43900 -2.05500

C -4.63500 2.66700 -1.49000

C -6.73200 3.08600 -0.84900

C -5.80100 4.28400 -2.49800

C 4.72900 2.58700 -1.48900

N 5.71800 2.26100 -0.66000

N 4.76100 3.52800 -2.43100

C 6.83300 2.97000 -0.84800

C 5.92300 4.18300 -2.49700

N 7.00100 3.94800 -1.74300

C -0.02100 -5.40000 -0.66300

N -0.02700 -6.15800 0.42300

N -0.02600 -5.80600 -1.92800

C -0.03900 -7.47100 0.15500

C -0.03700 -7.13400 -2.06100

N -0.04400 -8.02100 -1.05900

H 6.36400 -4.11200 -2.36700

H -6.38500 -4.00300 -2.36500

H -5.86300 5.07400 -3.24200

H -7.58000 2.86600 -0.20400

H 0.08600 7.20600 0.75300

H 5.99900 4.97300 -3.24100

H 7.67800 2.73500 -0.20400

H -0.04400 -8.14500 1.00900

H -0.04100 -7.52800 -3.07500

end

geometry Ag

Ag 0.78200 0.32000 -0.80700

end

basis spherical

Ag library stuttgart_rsc_1997_ecp

C library 6-311+G*

H library 6-311+G*

N library 6-311+G*

end

ECP

Ag library stuttgart_rsc_1997_ecp

END

charge -1

set geometry ligands

dft

grid xfine

xc M06

mult 2

vectors output Triazine.movecs

end

task dft

charge +1

set geometry ag

dft

mult 2

vectors output Triazine.movecs

end

task dft

charge 0

set geometry all

dft

convergence energy 1d-9

end

DRIVER

EPREC 1e-5

MAXITER 100

end

task dft optimize

task dft frequencies

dplot

Title Homo

vectors Triazine.movecs

gaussian

output lomo.cube

kamal sobhy

unread,

Sep 9, 2020, 3:10:19 PM9/9/20

to NWChem Forum

sorry but my email had some problem with attachments. here are the files

output13.txt

trial.txt

Edoardo Aprà

unread,

Sep 9, 2020, 4:23:25 PM9/9/20

to NWChem Forum

suggestions:

You are using a single process in your run. NWChem's performance benefits by using multiple processes.https://nwchemgit.github.io/Running.html#parallel-execution-on-unix-based-parallel-machines-including-workstation-clusters-using-mpi
Use a more moderate memory line (especially when running in parallel) memory total 2 gb
I would not use m06 and xfine grid for the initial geometry optimization. My suggestion is to use a GGA functional (PBE96), that allows you to use Coulomb fitting, that speeds up the calculation. I would use a smaller basis set, too. Here are the changed lines of the modified input file

basis "ao basis" spherical
* library def2-svp
end
basis "cd basis" spherical
* library "weigend coulomb fitting"
end

ECP
Ag library def2-ecp

END

charge -1
set geometry ligands
dft

xc xpbe96 cpbe96
noio
direct

mult 2
vectors output Triazine.movecs

maxiter 199
end
task dft

KamalEldeen Kamal

unread,

Sep 9, 2020, 7:54:05 PM9/9/20

to NWChem Forum

any further modifications you recommend in order to get better results for this calculation?

Reply all

Reply to author

Forward