I'm trying to compute the energy of the benzyl bromide with orca. I need it to be at ωB97x-d4/6-31g(d) level. Orca throws me the following error.

** There are no main   basis functions on atom number  14 (Br) **

I also tried the ! NORI keyword, but nothing changed.

can somebody help me?

  • 4
    $\begingroup$ Do you absolutely need to use 6-31G(d)? I’d highly recommend using def2-SVP instead. About the same quality, but faster and includes a much wider range of elements due to built-in effective core potentials. $\endgroup$ May 2, 2022 at 13:31
  • $\begingroup$ unfortunately, I'm taking up an old and already started work and converting back to def2-SVP all calculation is quite impossible. $\endgroup$ May 2, 2022 at 13:40

2 Answers 2


You can download it from this site: Basis Set Exchange: https://www.basissetexchange.org/

Here is the requested basis set for Br in ORCA format:

! Basis Set Exchange
! Version v0.9
! https://www.basissetexchange.org
!   Basis set: 6-31G(d,p)
! Description: 6-31G + polarization on all atoms
!        Role: orbital
!     Version: 1  (Data from Gaussian 09/GAMESS)


S   6
1         0.1137182000E+06       0.1717696000E-02
2         0.1707444000E+05       0.1316744000E-01
3         0.3889576000E+04       0.6504553000E-01
4         0.1097096000E+04       0.2269505000E+00
5         0.3520624000E+03       0.4768357000E+00
6         0.1207002000E+03       0.3583677000E+00
L   6
1         0.2471138000E+04       0.2243687000E-02       0.3790182000E-02
2         0.5893838000E+03       0.2994853000E-01       0.2995979000E-01
3         0.1918738000E+03       0.1256009000E+00       0.1318228000E+00
4         0.7295339000E+02      -0.9832786000E-03       0.3432708000E+00
5         0.3005839000E+02      -0.6013141000E+00       0.4642345000E+00
6         0.1252927000E+02      -0.4913983000E+00       0.2079387000E+00
L   6
1         0.1096411000E+03      -0.5975683000E-02      -0.6907483000E-02
2         0.3858948000E+02       0.5542122000E-01      -0.3041432000E-01
3         0.1637818000E+02       0.2681200000E+00       0.4602725000E-01
4         0.7221836000E+01      -0.1543606000E+00       0.3650689000E+00
5         0.3263697000E+01      -0.7206306000E+00       0.4949232000E+00
6         0.1465499000E+01      -0.3316437000E+00       0.2090394000E+00
L   3
1         0.2103651000E+01       0.3029029000E+00      -0.2826714000E-01
2         0.7547050000E+00      -0.2152659000E+00       0.3503065000E+00
3         0.3005140000E+00      -0.9633941000E+00       0.7182446000E+00
L   1
1         0.1090710000E+00       0.1000000000E+01       0.1000000000E+01
D   3
1         0.6225514000E+02       0.7704229000E-01
2         0.1731284000E+02       0.3707384000E+00
3         0.5607915000E+01       0.7097628000E+00
D   1
1         0.1746486000E+01       1.0000000
D   1
1         0.4366000000E+00       1.0000000


The reason you're getting the error

** There are no main   basis functions on atom number  14 (Br) **

is that there are no basis functions on atom 14. Take a look at the orca_manual_5_0_3.pdf, p. 486 (the manual can be downloaded from the orcaforum). You will notice that it lists H–Zn for the basis set 6-31g(d) under the availability column. That is because this basis set is available for elements 1 (H) through 30 (Zn) by default in orca. As far as I can tell, https://www.basissetexchange.org/ does not have 6-31g(d) (it only lists 6-31g(d,p)), but you can check the availability there visually, since the ESML basis set exchange lists places a little yellow triangle in the bottom left corner. It looks like the ESML version of 6-31g(d,p) goes up to Krypton (element 36), so this one would include bromine (element 35).

Basis sets are listed for specific ranges of atomic numbers because basis sets designed for modelling the valence shells of heavier elements will be contracted. That means that the core shells basis functions will be replaced with something mathematically simpler, like an Effective Core Potential (ECP). This avoids the computational cost associated with very large numbers of basis functions.

In your case, the basis set 6-31g(d) is only available up to Zn, so when you try to run your calculation, it fails, because there are no basis functions on your Br atoms. !NORI won't do anything about this because all it does is turns off the RI approximation, but you can't even run an SCF cycle if you have atoms with no basis functions on them at all. You could use the ESML version of 6-31g(d,p), which goes up to Kr, but the 6-31g(d,p) set available by default in ORCA also only goes up to Zn. Why don't you try 6-311G? Alternatively, you could use something with H-Kr availability. As @Geoff Hutchison pointed out, you could simply use Def2-SVP. If you're concerned about whether the basis set has the same number of basis functions as your desired 6-31g(d) set, you can take a https://en.wikipedia.org/wiki/Basis_set_(chemistry) or Jensen, section 5: file:///home/joseph/Downloads/Jensen_F.-Introduction_to_Computational_Chemistry_(2007).pdf (will download book).

Happy computing!


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .