New user needing help with NWChem


Clicked A Few Times
Hi, my name is Henrique, I’m from Rio de Janeiro/Brazil.

I am finishing my graduation as a chemist and I have developed a great interest in Inorganic Chemistry. That is why I decided to write a thesis comparing various characteristics of [Fe(OH2)6]2+ and [Fe(OH2)6]3+, mostly using Ab Initio softwares and comparing results with the literature available.

Most of my calculations where performed using MOPAC2012, but I’d love to use NWChem to generate UV-vis spectra and, in this matter, I’m having a really bad time.
Here is my input file:

title "H12O6Fe"

echo

charge 2

geometry
Fe 0.00000000 0.00000000 0.00000000
O -0.37300000 0.22600000 1.90100000
H 0.23200000 -0.27800000 2.43400000
H -0.34900000 1.14100000 2.15600000
O 0.38000000 -0.22200000 -1.90100000
H 0.97200000 0.44300000 -2.23400000
H -0.40700000 -0.23400000 -2.43400000
O -0.00600000 -1.93600000 0.23200000
H -0.77500000 -2.22400000 0.71200000
H 0.04700000 -2.38900000 -0.60200000
O 0.00400000 1.93800000 -0.22500000
H 0.87400000 2.31400000 -0.15000000
H -0.38300000 2.21000000 -1.05000000
O -1.90900000 -0.04700000 -0.39400000
H -2.16500000 -0.79100000 -0.92800000
H -2.24000000 0.76400000 -0.76300000
O 1.91200000 0.03800000 0.38800000
H 2.15700000 0.70900000 1.01500000
H 2.45300000 0.08900000 -0.39100000
end

basis noprint
 * library 6-31G*
end

dft
  noprint "final vectors analysis"
  mult 5
  xc hfexch
end

tddft
  cis
  nroots 10
  notriplet
end

task tddft energy


And this is the end of my output:

 Superposition of Atomic Density Guess
      -------------------------------------
 
 Sum of atomic energies:       -1716.63448140

 Renormalizing density from      86.00 to     84
 
      Non-variational initial energy
      ------------------------------

 Total energy =   -1692.948019
 1-e energy   =   -3715.536900
 2-e energy   =    1301.767054
 HOMO         =      -1.021910
 LUMO         =      -0.951845
 
   Time after variat. SCF:     14.6
   Time prior to 1st pass:     14.6

 Integral file          = /scratch/webmodemo/webmo-4350/163087/input.aoints.0
 Record size in doubles =  65536        No. of integs per rec  =  43688
 Max. records in memory =      0        Max. records in file   = 492162
 No. of bits per label  =      8        No. of bits per value  =     64


 #quartets = 2.722D+06 #integrals = 5.547D+07 #direct =  0.0% #cached =100.0%


           Memory utilization after 1st SCF pass: 
           Heap Space remaining (MW):       13.06            13057712
          Stack Space remaining (MW):       13.11            13105058

   convergence    iter        energy       DeltaE   RMS-Dens  Diis-err    time
 ---------------- ----- ----------------- --------- --------- ---------  ------
 d= 0,ls=0.0,diis     1  -1712.5508075294 -2.43D+03  6.68D-02  1.30D+01    58.0
                                                     4.33D-02  1.80D+01
Last System Error Message from Task 0:: Inappropriate ioctl for device
forrtl: error (78): process killed (SIGTERM)
0:Terminate signal was sent, status=: 15
(rank:0 hostname:buchner.chem.hope.edu pid:8155):ARMCI DASSERT fail. signaltrap.c:SigTermHandler():463 cond:0


I'm using http://www.webmo.net/ to perform my calculations with NWChem (nwchem-5.1.1) because it is faster.

I’d really appreciate any help with this.
Thank you in advance.
Henrique C. S. Junior
Inorganic Chemist - UFRRJ - Rio de Janeiro

Gets Around
It's possible you are hitting resource limits, though this job didn't appear especially difficult when I ran it locally. All I had to do was raise the allowed SCF iterations to get normal completion, though that can't be your problem because the job dies far too early. 5.1.1 is a pretty old version of the software. In any case here is a completed output for your job, run with the latest 6.3 snapshot:

http://pastebin.com/JEaemhwc

Just Got Here
Thank you, that worked!
I was able to reproduce your results.
What software do you use to read your results? At my end Gabedit was not able to generate a UV-vis graphic for the data contained in this output.

Gets Around
You can use contrib/parsers/nw_spectrum.py to generate data for plotting like this:

python nw_spectrum.py < nwchem-logfile.log > data.dat


There are a number of adjustable options if you read the code in nw_spectrum.py, but the defaults will produce something that looks OK. Then just plot the output data with your favorite plotting system. Here's a quick crude image from gnuplot:

http://www.sciencemadness.org/cc/uvvis.png

Clicked A Few Times
I'm very sorry for my delay, your hint worked.
Thank you very much.


Forum >> NWChem's corner >> General Topics