Question
Asked 13 October 2014

Anybody familiar with the freezing of atoms in gaussian software?

If we use -1 for freeze the atoms all the direction of the atoms are freezes  but I want freeze atoms particularly one Cartesian axis just say Z-direction remaining two direction relaxes. Is it possible in Gaussian?

Most recent answer

Ahmed Eid
Chemistry Department, Faculty of Science, Al-Azhar University, Egypt
I agree with Dr Vithaya's answer.
You can try also Z-matrix, there is an option in Gauss view to convert Cartesian coordinates to Z-matrix for this large number of atoms
1 Recommendation

Popular answers (1)

Vithaya Ruangpornvisuti
Chulalongkorn University
Yes you can freeze atom(s) in molecule as partial optimization using Gaussian. It is very easy to set  gaussian input, as following condition.
(1) Input atoms must be in Cartesian coordinate,
(2) insert code number in between Atom name (i.e. O) and atomic x-value as,
(2.1) assignment code as number  = -1 if you want to freez  atom on that line,
(2.2) number = 0, if you want to let atom move in optimization process.
Pleas learn from following example:
%chk=example.chk
# hf/3-21g geom=connectivity
H2O vs C6H6
0 1
O 0 -2.09620750 1.02317579 1.79143348
H 0 -1.63026934 0.29352612 2.20629850
H 0 -2.41660654 1.62464460 2.46758415
H -1 -4.44172210 1.31649620 0.06952996
C -1 -3.34211810 1.31631320 0.06970996
C -1 -2.64473610 0.10833720 0.07039196
C -1 -2.64451710 2.52451920 0.06871396
C -1 -1.24957610 0.10833720 0.07039196
H -1 -3.19449510 -0.84397980 0.07084196
C -1 -1.24969210 2.52459720 0.06919296
H -1 -3.19463910 3.47680020 0.06776096
C -1 -0.55203810 1.31608820 0.07039196
H -1 -0.70006810 -0.84417580 0.07170696
H -1 -0.69949210 3.47674020 0.06913396
H -1 0.54764190 1.31616820 0.07102596
1 2 1.0 3 1.0
2
3
4 5 1.0
5 6 1.5 7 1.5
6 8 1.5 9 1.0
7 10 1.5 11 1.0
8 12 1.5 13 1.0
9
10 12 1.5 14 1.0
11
12 15 1.0
13
14
15
Goodluck,
VR
18 Recommendations

All Answers (13)

Do you want to freeze one particular variable?
2 Recommendations
If you need to freeze 1 particular variable, I would suggest
converting the structure into Z-Matrix format and then
freeze the variable(s) you want to be constant
1 Recommendation
Chammi Palehepitiya Gamage
University of Texas at Austin
You can freeze the whole atom I don't think you can do it for just a specific coordinate. I would ask them, they are pretty helpful. http://www.gaussian.com/g_tech/em_help.htm
Vithaya Ruangpornvisuti
Chulalongkorn University
Yes you can freeze atom(s) in molecule as partial optimization using Gaussian. It is very easy to set  gaussian input, as following condition.
(1) Input atoms must be in Cartesian coordinate,
(2) insert code number in between Atom name (i.e. O) and atomic x-value as,
(2.1) assignment code as number  = -1 if you want to freez  atom on that line,
(2.2) number = 0, if you want to let atom move in optimization process.
Pleas learn from following example:
%chk=example.chk
# hf/3-21g geom=connectivity
H2O vs C6H6
0 1
O 0 -2.09620750 1.02317579 1.79143348
H 0 -1.63026934 0.29352612 2.20629850
H 0 -2.41660654 1.62464460 2.46758415
H -1 -4.44172210 1.31649620 0.06952996
C -1 -3.34211810 1.31631320 0.06970996
C -1 -2.64473610 0.10833720 0.07039196
C -1 -2.64451710 2.52451920 0.06871396
C -1 -1.24957610 0.10833720 0.07039196
H -1 -3.19449510 -0.84397980 0.07084196
C -1 -1.24969210 2.52459720 0.06919296
H -1 -3.19463910 3.47680020 0.06776096
C -1 -0.55203810 1.31608820 0.07039196
H -1 -0.70006810 -0.84417580 0.07170696
H -1 -0.69949210 3.47674020 0.06913396
H -1 0.54764190 1.31616820 0.07102596
1 2 1.0 3 1.0
2
3
4 5 1.0
5 6 1.5 7 1.5
6 8 1.5 9 1.0
7 10 1.5 11 1.0
8 12 1.5 13 1.0
9
10 12 1.5 14 1.0
11
12 15 1.0
13
14
15
Goodluck,
VR
18 Recommendations
Matthew D. Miller
Semichem Inc.
I have performed calculations with a limited number of atoms frozen using cartesian coordinates (truncated protein with a ligand).  I suggest testing the file format shown by Dr. Ruangpornvisuti above.  The column flagged with 0 for atom optimization and -1 for atom freezing does work.  Upon completion of the calculation, files can opened and the frozen atoms will align correctly on top of each other.  This column is not required in an input file by default and is assumed to be 0 when not included.  This is described in the oniom section of the manual (search for freeze).
It is possible to use z-matrix coordinates to freeze atoms, although I am uncertain why this would be preferable to do.  If it is desirable to use z-matrix (use the readfreeze option), it is very important to list atoms that are frozen at the top because atoms which shift in position during optimization will be used as a reference point for the frozen atoms and the frozen atoms will not actually remain in the same position.  However, if the frozen atoms are listed first, each frozen atom will reference frozen atoms and it works.  The "notatoms" and list of atom numbers to be frozen should be included in the input file.  I had some challenges using the z-matrix and suggest using the cartesian coordinates.  Readfreeze is described in the following section:
I suggest you confirm the atom positions do not move by viewing the input and output files overlaid. 
Good luck,
M
2 Recommendations
Prasanta Bandyopadhyay
The University of Manchester
I don't know about the Z-matrix, but I had some 150 atoms to freeze, so I used the following trick.
1. Draw the atoms in Gaussview.
2. Go to edit > Atom Group, group the atoms so that there exists two groups where group 1 contains the atoms to freeze and group 2 to not to freeze (and vice versa).
3. Go to Group Actions > "Select atoms of Gaussian Group 1" to the top right corner.
4. To the top left corner, click Atom Group Class > Freeze.
5. Start your calculation.
Hope this helps.
6. Suggestions or modifications please....
 
2 Recommendations
Mallika Khare
Michigan Technological University
I can help with freezing of individual atoms in z-matrix:
Z-matrix: It includes the bond length, angle and dihedral angle of a particular atom in reference to other atom.
So, when you prepare your molecule in z-matrix format in Gaussian and open the file in notepad, it will show you something like this: I have taken example of NH3 molecule:
N1
H2 1 r2
H3 1 r3 2 a3
H4 1 r4 2 a4 3 d4
r2=1.0
r3=1.0
r4=1.0
a3=110.
a4=110.
d4=120.
This is a z-matrix of ammonia. Here r2, r3, r4, are the bond length, a3, a4 are the angles and d4 is a dihederal. Suppose I want to freeze atom N (numbered 1) and H atom (numbered 3) in this example. So, my z-matrix will look like this:
N1
H2 1 r2
H3 1 r3 2 a3
H4 1 r4 2 a4 3 d4
r2=1.0
r4=1.0
a4=110.
d4=120.
r3=1.0
a3=110.
Here you can see that to freeze H3 and N1 atoms I picked up the associated bond length and angle/dihederal and separated it from the main list. This is how we freeze atoms in z-matrix. While freezing there should always be ONE BLANK line between the main list of bonds & angles and the bonds & angles you want to freeze. Hope this helps. Good Luck! 
4 Recommendations
Khulood Obaid
University of Kerbala
I agree with answer of Mr Vikaya
Ahmed Eid
Chemistry Department, Faculty of Science, Al-Azhar University, Egypt
I agree with Dr Vithaya's answer.
You can try also Z-matrix, there is an option in Gauss view to convert Cartesian coordinates to Z-matrix for this large number of atoms
1 Recommendation

Similar questions and discussions

How can I circumvent Bend failed for angle errors in Gaussian?
Question
15 answers
  • Anthony NashAnthony Nash
Hi all,
I have a compound with a particular functional group surrounded by water molecules. I am using Gaussian G09 to optimise, first with HF/6-31+g** and then I will progressively move higher per stable structure. Unfortunately, of the twenty different model arrangements (4 functional groups I am interested in and 8, 10, 12, 14, 16) water molecules neighbouring) only three of the twenty ran successfully. The others suffered from the following error (atom numbers will have been different obviously):
GradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGradGrad
Berny optimization.
Using GEDIIS/GDIIS optimizer.
Bend failed for angle    39 -    70 -    68
Tors failed for dihedral    15 -    39 -    70 -    68
Tors failed for dihedral    51 -    39 -    70 -    68
Tors failed for dihedral    69 -    68 -    70 -    39
FormBX had a problem.
Error termination via Lnk1e in /usr/local/g09_d01/l103.exe at Tue Mar 24 10:12:24 2015.
When I investigate this further, the first failure (Bend on the angle 39, 70, 68) is an angle between the oxygen and hydrogen of a water, followed by the hydrogen of a hydroxyl group. It turns out to be exactly 180 degrees. I understand the optimisation algorithm in gaussian struggles with inline angle. 
Surely, restarting with a different arrangement to the water molecules isn't the best or only solution? Any tips or suggestions are very grateful.
Thanks
Anthony
The output file (Gaussian) showing End of file reading connectivity. Error termination via Lnk1e in /usr/local/g09_b01/g09/l101.exe.
Question
15 answers
  • Chethan BkChethan Bk
I am running a DFT calculation using Gaussian 05,,, but the output / log file shows the above mentioned statement. I tried with all possible changes which can be made in input file and checked the space b/w the lines and keywords, but unable to get the converged structure. Can any one help me to solve this convergence problem?
My Input file is like this..
%mem=2GB
# opt freq b3lyp/6-31g geom=connectivity
Title card
0 1
C -6.02619527 9.44527291 -0.71253003
C -4.97668356 9.89555985 -1.74430456
C -5.41433211 9.54127357 -3.18798470
C -7.65683239 10.03305367 -2.44048523
C -7.30718981 10.22873199 -0.96683834
H -5.27098963 9.94279746 -4.16774469
H -4.82659707 10.95398877 -1.66069636
H -6.20196164 8.40208993 -0.83104254
H -7.85894719 9.00324965 -2.64166873
H -7.13425811 11.26665372 -0.78372848
O -6.53256968 10.46693586 -3.22290224
O -5.56381472 9.69389357 0.61729905
O -8.36655799 9.76938263 -0.10288899
C -8.90267268 10.86635140 -2.79984380
H -8.71454864 11.90018067 -2.60784077
H -9.72858740 10.54498701 -2.20802438
O -9.20429830 10.68541419 -4.18551448
H -8.46665363 10.98542331 -4.71919452
H -9.17399192 10.26804893 -0.26957573
H -5.39756593 10.62148689 0.73301079
O -4.05417354 9.58839301 -3.16755348
O -3.70377791 9.16457303 -1.47569403
H -3.58492919 9.53697555 -3.89463091
H -2.87727273 9.25288215 -2.08709783
B -1.71685742 8.59334576 -4.26508851
O -2.97405212 9.47014766 -4.68353092
H -3.33306313 9.06071177 -5.47655795
O -0.37714720 8.54894999 -5.03191713
H 0.22561508 7.94829750 -4.59308374
O -2.00900133 9.22803064 -2.58256445
H -1.38453846 8.69191602 -2.08439557
O -2.24696092 7.14354050 -4.13098207
H -1.61273913 6.59985936 -3.63238399
H -2.36881319 6.77772795 -5.00860295
And output file terminates as....
End of file reading connectivity.
Error termination via Lnk1e in /usr/local/g09_b01/g09/l101.exe
Can anyone explain a consistent link 9999 error during TS with Gaussian?
Question
29 answers
  • Anthony NashAnthony Nash
I seem to be consistently producing link 9999 errors during a TS search of a structure I generated from a mod-redundant scan geometry. As an example, I scan geometry modifying the bond length (fundamentally the position of the hydrogen) of hydrogen coming off a protonated water molecule (first frame) and finally bonding to oxygen of a carbon based molecule (e.g. enolization of the carbonyl group in glucose). I take the scan geometry of where the hydrogen is halfway between both molecules, and there is a slight adjustment of the O=C double bond (as this will change to a single bond upon associating with hydr0gen HO-C). Note: During the scan geometry, there are no major changes to the over all structure e.g., big steric clashes causing a large dihedral rotation of part of the glucose.
I have then tried a couple of basis sets.
# opt=(calcfc,ts) wb97xd/3-21+g* scrf=(solvent=water)
# opt=(calcfc,ts) wb97xd/6-31+g(d,p) scrf=(solvent=water)
With the smaller basis set I was hoping to approximate a TS structure and use it as a starting structure for a larger basis set. But in either case I produce the error:
Error termination request processed by link 9999.
Error termination via Lnk1e in /usr/local/g09/l9999.exe at Wed May 21 19:31:46 2014.
With the last optimisation step looking like this:
Item Value Threshold Converged?
Maximum Force 0.086348 0.000450 NO
RMS Force 0.014307 0.000300 NO
Maximum Displacement 0.874616 0.001800 NO
RMS Displacement 0.240702 0.001200 NO
Predicted change in Energy=-1.075204D-02
Optimization stopped.
-- Wrong number of Negative eigenvalues: Desired= 1 Actual= 4
-- Flag reset to prevent archiving.
Any ideas what I can do to rectify this problem. It is becoming a bit of a habit of Gaussian to kick out these errors every time I try a TS using a frame from mod red scan bond length.
I have now turned to QST3 using the first, last and intermediate scan step to determine my TS.

Related Publications

Article
A molecular modeling exercise and an associated exam that have both competitive and cooperative learning aspects are described. Collaborative efforts are facilitated by Web-based information management. The exercise/exam is appropriate for use in undergraduate or graduate quantum chemistry or molecular modeling courses that have access to modest co...
Article
Full-text available
"Soft theories," i.e., "heuristic models based on reasoning by analogy" largely drove chemistry understanding for 150 years or more. But soft theories have their limitations and with the expansion of chemistry in the mid-20th century, more and more inexplicable (by soft theory) experimental results were being obtained. In the past 50 years, quantum...
Got a technical question?
Get high-quality answers from experts.