Can not find Ho pseudopotential in Pseudo Dojo library

[deyulu]

Ocean calculations involving Holmium failed with the error message: "ERROR! Could not find pseudopotential for Ho". At the dojo github (https://github.com/PseudoDojo/ONCVPSP-PBE-PDv0.4), there is a Ho potential folder, but Ho doesn't appear in either standard.json or stringent.json. Suggestions on how to generate the Ho potential and input to ocean will be helpful.

[jtv3]

For now: If the elements you need aren't in OCEAN's database then you'll need to provide them. See the EXAMPLE/LiF_oncvpsp for how this works. You'll need the UPFs (assuming QE) for all the elements and the pseudopotential input file for the edge you are interested in.

You can download the input files from the PseudoDojo gthub page: https://github.com/PseudoDojo . Then make sure you have the correct format in the input file. On the first non-comment line it'll either say 'psp8', 'upf', or 'both'. [psp8 is for abinit, and upf is for QE, and both makes both]. Run oncvpsp.x < file.in > file.out for your file. The pseudo will be mixed in with a ton of other output. For UPF format, the first line to keep is "<UPF version" and the last line is "".

As a shortcut, make a quick ocean run with all the elements except the missing one, but quit the job once it enters the OPF stage. OCEAN will extract the necessary .in and .UPF/.psp8 files into Common/psp for you.

[jtv3]

Future development. Add in all the missing elements. For those that don't have estimated ecut values require the user to supply it in the input.

[jjkas]

Hi John and Deyu, It looks like all of the libraries out there that provide psps for QE only have PAW and USSP versions for Ho. Do either of these work for OCEAN, or do you need norm-conserving specifically?

…

On Fri, Jun 7, 2024 at 11:36 AM John Vinson ***@***.***> wrote: Future development. Add in all the missing elements. For those that don't have estimated ecut values require the user to supply it in the input. — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/times-software/OCEAN/issues/328*issuecomment-2155332440__;Iw!!K-Hz7m0Vt54!gjp0bECizNojGga38Y37sIksjSjxnBJGTtrQyxCOBsJt1XPBy5aFVxgD8loTHmFyLJqswNw0qSRQHdpnbNxA8Q$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AA2RF24HWSUO66YLLOQGTP3ZGH4RHAVCNFSM6AAAAABI7ALUB6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNJVGMZTENBUGA__;!!K-Hz7m0Vt54!gjp0bECizNojGga38Y37sIksjSjxnBJGTtrQyxCOBsJt1XPBy5aFVxgD8loTHmFyLJqswNw0qSRQHdoNJR7Emg$> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

-- Joshua Kas (He/Him/His) Research Assistant Professor University of Washington

[jtv3]

Norm-conserving is required. PseudoDojo does have two Ho options, with 4f in the valence and with 4f in the core

[jjkas]

I only see the one with the f-states frozen, i.e., 3+. Can you send a link to the one with f in the valence?

…

On Fri, Jun 7, 2024 at 11:42 AM John Vinson ***@***.***> wrote: Norm-conserving is required. PseudoDojo does have two Ho options, with 4f in the valence and with 4f in the core — Reply to this email directly, view it on GitHub <https://urldefense.com/v3/__https://github.com/times-software/OCEAN/issues/328*issuecomment-2155340286__;Iw!!K-Hz7m0Vt54!i5VPmXTKUzxKKKKJkMi8LnkL2k3S3TN_0ix4F5BRSdeTV-c1zCuAeDdbgYXvvxQ2OJRMkqeuWqLYHgKrJAv72w$>, or unsubscribe <https://urldefense.com/v3/__https://github.com/notifications/unsubscribe-auth/AA2RF26K5OMHZ2AAHH4R7CDZGH5I7AVCNFSM6AAAAABI7ALUB6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNJVGM2DAMRYGY__;!!K-Hz7m0Vt54!i5VPmXTKUzxKKKKJkMi8LnkL2k3S3TN_0ix4F5BRSdeTV-c1zCuAeDdbgYXvvxQ2OJRMkqeuWqLYHgLrFbcanA$> . You are receiving this because you commented.Message ID: ***@***.***>

-- Joshua Kas (He/Him/His) Research Assistant Professor University of Washington

[jtv3]

https://github.com/PseudoDojo/ONCVPSP-PBE-SR/blob/master/Ho/Ho-sp.in