Problems with steric analysis

Problems with steric analysis

Postby pipelion2000 » Fri Oct 17, 2014 1:49 pm

Dear NBO staff:

As part of our ongoing research, we decided to perform NBO steric analysis on 1,2-trans-dichlorocyclohexane (both conformers). To this end, we use the DFT-B3LYP 6-311++G** optimized geometry and generate the .47 files with Firefly (NBO 5.G) using a single point RHF/6-311++G**. With these .47 files we perform STERIC analysis using NBO 6 (the latest release from october 13).
Unfortunately, the calculation stops with a message:

NATURAL LOCALIZED MOLECULAR ORBITAL (NLMO) ANALYSIS:

NLMO: OWSO failed

This doesn't happens with NBO 5.G or the February 6 NBO 6 version...however, we need to perform such analysis with the latest version of NBO 6 which corrects some problematic issues when using large basis sets (as announced in the bug fixes section, sept 1-2014).

Thanks in advance for your help

Felipe Otálvaro
pipelion2000
 
Posts: 4
Joined: Mon Oct 13, 2014 7:08 pm

Re: Problems with steric analysis

Postby ericg » Fri Oct 17, 2014 2:16 pm

Felipe,

Can you post the optimized geometry that you are using? I'll try this myself.

Do you encounter the same problem when using 6-31++G**? Linear dependence can be an issue with the 6-311G sets and is likely what is clobbering OWSO, although it is interesting that the Feb-6 version worked.

Eric
ericg
 
Posts: 277
Joined: Sat Dec 29, 2012 9:31 am

Re: Problems with steric analysis

Postby pipelion2000 » Sat Oct 18, 2014 1:49 pm

Dear Eric:

Thanks for your prompt reply and interest.

Please find below the geometry requested (specified in ZMat format).

C
C 1 dis2
C 2 dis3 1 ang3
C 3 dis4 2 ang4 1 dih4
C 4 dis5 3 ang5 2 dih5
C 5 dis6 4 ang6 3 dih6
H 3 dis7 2 ang7 1 dih7
H 2 dis8 1 ang8 6 dih8
H 2 dis9 1 ang9 6 dih9
H 1 dis10 2 ang10 9 dih10
H 1 dis11 10 ang11 2 dih11
H 4 dis12 3 ang12 2 dih12
H 5 dis13 4 ang13 3 dih13
H 5 dis14 4 ang14 3 dih14
H 6 dis15 5 ang15 4 dih15
H 6 dis16 5 ang16 4 dih16
Cl 4 dis17 3 ang17 2 dih17
Cl 3 dis18 2 ang18 1 dih18

dis2=1.5345
dis3=1.5267
ang3=113.078
dis4=1.5332
ang4=113.523
dih4=48.907
dis5=1.5267
ang5=113.523
dih5=-46.414
dis6=1.5344
ang6=113.076
dih6=48.907
dis7=1.0872
ang7=111.416
dih7=173.890
dis8=1.0961
ang8=109.566
dih8=65.387
dis9=1.0924
ang9=111.243
dih9=-176.648
dis10=1.0952
ang10=109.445
dih10=-55.192
dis11=1.0938
ang11=106.704
dih11=-118.098
dis12=1.0872
ang12=110.096
dih12=-172.098
dis13=1.0924
ang13=109.144
dih13=173.297
dis14=1.0961
ang14=106.637
dih14=-71.574
dis15=1.0952
ang15=109.448
dih15=68.052
dis16=1.0938
ang16=109.248
dih16=-175.464
dis17=1.8467
ang17=107.203
dih17=75.794
dis18=1.8467
ang18=110.414
dih18=-71.503

When 6-31++G** was used it ended with the same error message.

FELIPE
pipelion2000
 
Posts: 4
Joined: Mon Oct 13, 2014 7:08 pm

Re: Problems with steric analysis

Postby ericg » Sun Oct 26, 2014 3:57 pm

Felipe,

I am unable to reproduce the OWSO failure. I've run your calculation on a number of platforms (all but Windows). Every calculation has completed successfully.

Do you have the source or binary distribution of NBO6? I'm concerned that there might be some incompatibility between the systems on which I build the binary distributions and yours.

Eric
ericg
 
Posts: 277
Joined: Sat Dec 29, 2012 9:31 am

Re: Problems with steric analysis

Postby pipelion2000 » Tue Oct 28, 2014 6:31 pm

Dear Eric:

We have the binary distribution for PC-windows. In addition, we have been working on Windows vista and Windows 7...without problems until now.
Please let me know if we can help somehow.

thanks in advance for your help

FELIPE
pipelion2000
 
Posts: 4
Joined: Mon Oct 13, 2014 7:08 pm

Re: Problems with steric analysis

Postby ericg » Thu Oct 30, 2014 9:35 pm

Felipe,

There is apparently an issue with the implementation of LAPACK in the WIndows version of NBO6. I've disabled LAPACK and built a new distribution for Windows that resolves the OWSO problem in the NLMO procedure. You can download this updated distribution after 2am CDT tomorrow (Oct 31) morning.

Thanks for calling our attention to this problem. I'll attempt to fix the LAPACK problem in NBO6W in the near future.

Eric
ericg
 
Posts: 277
Joined: Sat Dec 29, 2012 9:31 am


Return to General NBO Discussion

Who is online

Users browsing this forum: No registered users and 2 guests

cron