{
  "id": 98225,
  "title": "Quantum chemical methods used for generating provided data",
  "url": "/competitions/champs-scalar-coupling/discussion/98225",
  "author_name": "",
  "post_date": "2019-07-02T06:50:16.683290800Z",
  "votes": 4,
  "comment_count": 1,
  "views": 0,
  "content": "<p>What are the actual quantum chemical methods that were used to calculate the potential energies, Mulliken charges etc.? I.e., method, basis set, and all that.\nAlso, what is the unit of the potential energy? I am guessing Hartree, but would like to be sure.</p>",
  "messages": [
    {
      "id": "566432",
      "postDate": "07/02/2019 06:50:16",
      "content": "<p>What are the actual quantum chemical methods that were used to calculate the potential energies, Mulliken charges etc.? I.e., method, basis set, and all that.\nAlso, what is the unit of the potential energy? I am guessing Hartree, but would like to be sure.</p>",
      "rawMarkdown": "What are the actual quantum chemical methods that were used to calculate the potential energies, Mulliken charges etc.? I.e., method, basis set, and all that.\nAlso, what is the unit of the potential energy? I am guessing Hartree, but would like to be sure.",
      "votes": null
    },
    {
      "id": "569914",
      "postDate": "07/07/2019 14:55:31",
      "content": "<p>After a couple of calculations, my current (semi-educated) guess is: B3LYP//6-31(2df,p). That's what was used for the QM9 calculations and the re-calculated total energies are very close to what is given as \"potential energies\" in the data set. The differences where below 0.01 Ha. That is about the difference that I get to the QM9 data set. Considering that there is no information on grid setup, whether cartesian basis functions or spherical harmonics were used, etc..., this is close enough for me to suggest this method. One thing to note, it's the Gaussian B3LYP variant, not the TURBOMOLE variant. The latter was used for the QM9 calculations, AFAICT.</p>",
      "rawMarkdown": "After a couple of calculations, my current (semi-educated) guess is: B3LYP//6-31(2df,p). That's what was used for the QM9 calculations and the re-calculated total energies are very close to what is given as \"potential energies\" in the data set. The differences where below 0.01 Ha. That is about the difference that I get to the QM9 data set. Considering that there is no information on grid setup, whether cartesian basis functions or spherical harmonics were used, etc..., this is close enough for me to suggest this method. One thing to note, it's the Gaussian B3LYP variant, not the TURBOMOLE variant. The latter was used for the QM9 calculations, AFAICT.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 569914,
      "author_name": "fuhlig",
      "author_url": "",
      "post_date": "07/07/2019 14:55:31",
      "content": "<p>After a couple of calculations, my current (semi-educated) guess is: B3LYP//6-31(2df,p). That's what was used for the QM9 calculations and the re-calculated total energies are very close to what is given as \"potential energies\" in the data set. The differences where below 0.01 Ha. That is about the difference that I get to the QM9 data set. Considering that there is no information on grid setup, whether cartesian basis functions or spherical harmonics were used, etc..., this is close enough for me to suggest this method. One thing to note, it's the Gaussian B3LYP variant, not the TURBOMOLE variant. The latter was used for the QM9 calculations, AFAICT.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "566432": "What are the actual quantum chemical methods that were used to calculate the potential energies, Mulliken charges etc.? I.e., method, basis set, and all that.\nAlso, what is the unit of the potential energy? I am guessing Hartree, but would like to be sure.",
    "569914": "After a couple of calculations, my current (semi-educated) guess is: B3LYP//6-31(2df,p). That's what was used for the QM9 calculations and the re-calculated total energies are very close to what is given as \"potential energies\" in the data set. The differences where below 0.01 Ha. That is about the difference that I get to the QM9 data set. Considering that there is no information on grid setup, whether cartesian basis functions or spherical harmonics were used, etc..., this is close enough for me to suggest this method. One thing to note, it's the Gaussian B3LYP variant, not the TURBOMOLE variant. The latter was used for the QM9 calculations, AFAICT."
  },
  "source": "meta"
}