Hi again...
Firstly, thanks Ibrahim for your reply.
Well, after remaking the imput (in this case in cartesians to avoid the ghost atoms), I have fixed the problem but I don't have clear why it was the solution.
Here is the problematic imput file:
-------------------------------------------------------------------------------------------------------------------------
%chk=/data1/diego/GAU/metals/palladium/Pd1hcpx.chk
%mem=20GB
%nproc=8
# MP2/GENecp geom=connectivity scf=(maxcycle=500, qc) guess=read counterpoise=2 maxdisk=50GB
Title Card Required
0 1
Pd(Fragment=1) -0.08140700 0.03787500 -0.00549500
O(Fragment=1) -1.97461900 -0.99338100 -0.06543100
O(Fragment=1) 1.81180500 1.06913100 0.05444000
O(Fragment=1) 0.86489800 -1.70544900 0.09890800
O(Fragment=1) -1.02771300 1.78119900 -0.10989800
H(Fragment=1) -2.38699000 -1.26562000 -0.87355800
H(Fragment=1) -2.46575500 -1.21329400 0.71409400
H(Fragment=1) 2.30294000 1.28904400 -0.72508400
H(Fragment=1) 2.22417600 1.34137000 0.86256800
H(Fragment=1) 1.18280000 -1.98870600 0.95001100
H(Fragment=1) -1.42728100 2.11870900 0.68515200
H(Fragment=2) 3.82319200 -1.61126300 -0.66318300
H(Fragment=2) 3.09668700 -1.62449400 -0.52137500
1 4 1.0 5 1.0
2 6 1.0 7 1.0
3 8 1.0 9 1.0
4 10 1.0
5 11 1.0
6
7
8
9
10
11
12
13
O H 0
6-31G**
****
Pd 0
lanl2dz
****
Pd 0
lanl2dz
----------------------------------------------------------------------------------------------------------------------------
As mentioned, the calculation starts without problems but after the first SCF cycle the process fail with "Error termination in NtrErr"
The problem was fixed by removing the "qc" keyword from the keywords line. But Why?
Are there any inconsistency with the use of the QC algorithm and the counterpoise procedure?
Thanks again.
Regards.
Diego*************UPV-ITQ
Spain.
From: Mahmoud A. A. Ibrahim m.ibrahim|compchem.net <owner-chemistry|a|ccl.net>
To: "Gomez, Diego Armando " <darkego21|a|yahoo.com>
Sent: Fri, April 23, 2010 11:43:39 PM
Subject: CCL:G: Error termination in NtrErr at MP2 calculation.
Dear DiegoI read your complain about a NtrErr error. This type of errors happens in different cases, one of them as Dave mentioned due to space issue, and in this case NMax information should be mentioned in last few lines, which is not available in the supported information you gave. Another possibility is the checkpoint file, what happened is your checkpoint file has not be created, or optimization data (for first cycle) has not be saved in your checkpoint file. that is what you have got.There is no technical solve for this error. Just, try to build up your input file again, save checkpoint file (%chk=filename.chk) and change its path.If you got the error again, you would send your input file and we can test it.Sincerely;M. IbrahimOn Thu, Apr 22, 2010 at 9:24 PM, DIEGOI GOMEZ darkego21]_[yahoo.com <owner-chemistry,+,ccl.net> wrote:
Hi Dave,
thanks for your reply!!
I'm using Gaussian09 and my system is really small, I'm doing a test with one Pd(II) atom in square planar stereo-chemistry with two H2O molecules and two OH- groups as ligands interacting with one H2 molecule.
--------------------------------------------------------------------------------------------------------------------
118 basis functions, 217 primitive gaussians, 124 cartesian basis functions
29 alpha electrons 29 beta electrons
-------------------------------------------------------------------------------------------------------------------
Previously I did one MP2 optimization of the system (without counterpoise) with no problems. So I don't believe that the Error is related with storage capacity (actually, I have done calculation on bigger systems with more than 300 basis functions). I speculate that It seems to be related with the reading of some file or keyword when the second step starts.
I forgot to mention that in this particular case the Pd cluster is fixed (opt=z-matrix) an the optimization is only done on the H2 coordinates.
However I have not tried that option (%rwf), I'll give a look...
Thanks again..Diego
UPV-ITQ
Spain.
From: "Close, David M. CLOSED_._mail.etsu.edu" <owner-chemistry]~[ccl.net>
To: "Gomez, Diego Armando " <darkego21]~[yahoo.com>
Sent: Thu, April 22, 2010 8:02:19 PM
Subject: CCL: Error termination in NtrErr at MP2 calculation.
Diegoi:
This is a common error that means you have exceeded disc storage space. This happens with MP2 calculations that requires a great deal of scratch disc space for storing integrals. There are common fixes to this problem, but it is necessary to tell us about the size of the calculation you are doing. Like how many heavy atoms, and what program you are using. Basically a 32 bit machine is limited to 16 Gbytes of storage. Have you tried simple fixes like %rwf= …. ?
Regards, Dave Close.
Hello there.
I'm doing an optimization of a pair of interacting molecules at MP2 level with simultaneous counterpoise correction. I'm using two basis set (lanl2dz and 6-31G**) through the "GEN" keyword. The calculation starts without problems, so the first SCF cycle and the second-order perturbation are accomplished. However, when the next step starts the follow Error is reported:
---------------------------------------------------------------------------------------------------------------------------------------------
***** Axes restored to original set *****
Cartesian Forces: Max 0.199455160 RMS 0.055013514
Operation on file out of range.
FileIO: IOper= 2 IFilNo(1)= -522 Len= 118 IPos= 118 Q= 47965310131088
dumping /fiocom/, unit = 1 NFiles = 98 SizExt = 524288 WInBlk = 512
defal = T LstWrd = 233431552 FType=2 FMxFil=10000
Number 0 0 0 0 0 0 0 0
Base 12203972 390144 495616 321536 12752384 10698752 201216 365568
End 12204032 402432 527360 322048 12818944 11614208 201728 372736
End1 12204032 402432 527360 322048 12818944 11614208 201728 372736
Wr Pntr 12203972 388608 495616 321536 12533760 10698752 201216 365568
Rd Pntr 12203972 388608 495616 321536 12533760 10698752 201216 365568
Length 60 12288 31744 512 66560 915456 512 7168
(...)Error termination in NtrErr:
NtrErr Called from FileIO.
----------------------------------------------------------------------------------------------------------------------------------------------
Thanks in advance for any sugestion.
Diego
UPV-ITQ
Spain.
--
Mahmoud A. A. Ibrahim
Current Address
7.05, School of Chemistry,
The University of Manchester,
Oxford Road, Manchester, M13 9PL,
United Kingdom.
Home Address
Chemistry Department,
Faculty of Science,
Minia University,
Minia 61519,
Egypt.
Contact Information
Email: m.ibrahim,+,compchem.net
Website: www.compchem.net
Fax No.: +20862342601