2

I am running into issues preparing an input file for DFT calculations. Currently, I am trying to submit an optical rotation calculation for an alpha-deuterated benzyl bromide (1 deuterium on the benzyl position, not 2). I have the isotopically-substituted structure optimized using various DFAs but am unable to proceed past this point. I have tried this with and without referencing the old .chk file (and removed the geom=check for that submission). If possible, could anyone guide me in how to solve this problem? My assumption is that the Freq=(Readisotopes) is not compatible with the Polar=OptRot specification, but I am unsure how to fix this and still have it calculate the isotopically labeled compound. The variations of files I have tried are:

%nprocshared=128
%mem=180GB
%chk=/work/F32/m06l-optrot-bnbr-6311pgdp.chk
#p M06L/6-311+G(d,p) Polar=OptRot int=ultrafine empiricaldispersion=gd3 nosymm freq=(ReadIsotopes) CPHF=RdFreq

m06l-optrot-bnbr-6311pgdp


0 1
C    -1.153420000000      1.246497000000     -0.027676000000
C     0.218484000000      1.219778000000      0.206372000000
C     0.941392000000      2.405538000000      0.243455000000
C     0.302536000000      3.633029000000      0.048686000000
C    -1.074755000000      3.649411000000     -0.187339000000
C    -1.798641000000      2.464113000000     -0.224083000000
H    -1.718772000000      0.320541000000     -0.055616000000
H     0.725550000000      0.272906000000      0.361967000000
H     2.013053000000      2.384773000000      0.423440000000
H    -1.577463000000      4.600225000000     -0.344160000000
H    -2.868499000000      2.490001000000     -0.404738000000
C     1.075908000000      4.900664000000      0.094575000000
H     1.957993000000      4.831819000000      0.729465000000
H(Iso=2.014)     0.471732000000      5.751722000000      0.405436000000
Br    1.781396000000      5.410164000000     -1.691602000000


589nm

and

%nprocshared=128
%mem=180GB
%oldchk=/work/F32/m06l-isot-bnbr-6311pgdp.chk
%chk=/work/F32/m06l-optrot-bnbr-6311pgdp.chk
#p M06L/6-311+G(d,p) Polar=OptRot int=ultrafine empiricaldispersion=gd3 nosymm CPHF=RdFreq Geom=Check

m06l-optrot-bnbr-6311pgdp


0 1


589nm
MrAmateurChem
  • 21
  • 2
  • Unfortunately this is impossible to debug. We do not have the checkpoint file and we do not know the actual error. Please include an input file that could be run on a computer. On another note: Are you sure you want to request 128 processes and 180 GB of memory? That seems excessive when we're talking about https://en.wikipedia.org/wiki/Benzyl_bromide. Does your compute node even have that much juice? And furthermore: The 6-311G basis set is not a good choice. And what about spin and multiplicity? – Martin - マーチン Nov 08 '23 at 20:10
  • @Martin-マーチン I updated the post to better reflect the issue. To be transparent, I did not consider the request sizes, as I have been working with a template for a palladium catalyzed reaction. I will change the request sizes accordingly though. Please let me know if any other information would be helpful. – MrAmateurChem Nov 08 '23 at 21:01
  • 1
    I don't currently have access to an installation of g16 to test this, so please also include the error with which the calculation fails. I guess though it is a syntax error, because it doesn't make sense to have a non-integer isotope specified. For the calculation resources requested I guess you are not having budget issues. But even with large platinum complexes that seems rather a lot to request. If you're running on more than one node, you might loose quite a bit through overhead. And you'll burn through calculation time in a jiffy. – Martin - マーチン Nov 08 '23 at 22:55
  • 1
    If the second calculation run off of the first, the chk statements need to be reversed. Also, are you not running the first calculation from an optimisation? The result would be rather meaningless if you don't. And again, 6-311G is not a good basis set choice. Try something modern, like def2-tzvpp. – Martin - マーチン Nov 08 '23 at 23:03
  • 1
    Related: https://chemistry.stackexchange.com/q/27302/4945 – Martin - マーチン Nov 08 '23 at 23:05

0 Answers0