[CP2K-user] [CP2K:17941] Possibly incorrect GTH-PBE pseudopotential for Tl

Nicholas Winner nwinner at berkeley.edu
Wed Oct 26 23:14:07 UTC 2022


Thanks professor, I see the mistake now. I'll retry with POTENTIAL_UZH and 
see if it gives improvements. I was doing some benchmarking and found Tl 
compounds had very poor property predictions, maybe q21 will resolve that.

On Wednesday, October 26, 2022 at 4:26:29 AM UTC-7 jgh wrote:

> Hi
>
> your input for the PP optimization misses the 4f electrons
>
> ELECTRON_CONFIGURATION [Xe] 4f14 6s2 5d10 6p1
> CORE [Xe] 4f14
>
> I don't think the optimization will change the PP parameters significantly.
>
> In POTENTIAL_UZH and BASIS_MOLOPT_UZH you can find alternative PP and 
> Basis sets.
> You could for example try q21.
>
> regards
> JH
>
> ________________________________________
> From: cp... at googlegroups.com <cp... at googlegroups.com> on behalf of 
> Nicholas Winner <nwi... at berkeley.edu>
> Sent: Wednesday, October 26, 2022 5:07 AM
> To: cp2k
> Subject: [CP2K:17935] Possibly incorrect GTH-PBE pseudopotential for Tl
>
> I was optimizing Tl lattice parameter, and found a large discrepancy with 
> published PBE results. On a hunch, I decided to re-optimize the 
> pseudopotential, and based on what I tried, it seems the GTH 
> pseudopotential for Tl that ships with CP2K is not correct. The published 
> potential is:
>
> ```
> Tl GTH-PBE-q13 GTH-PBE
> 2 1 10
> 0.55000000 1 12.29058515
> 3
> 0.51280306 2 7.19017017 -2.86301073
> 3.69613096
> 0.57711505 2 4.76094580 -3.67332496
> 4.34633671
> 0.39323001 2 -11.01268700 6.42159202
> -7.28140093
> ```
>
> But, if you run pseudopotential optimization, then you get the following. 
> Note only 1 number changes: `0.39323001 -> 0.32379979 `
>
> ```
> Tl GTH-PBE-q13
> 2 1 10 0
> 0.55000000 1 12.29058515
> 3
> 0.51280306 2 7.19017017 -2.86301073
> 3.69613096
> 0.57711505 2 4.76094580 -3.67332496
> 4.34633671
> 0.32379979 2 -11.01268700 6.42159202
> -7.28140093
> ```
>
> The input file I used to generate this is is listed at the end of this 
> message.
>
> Can the devs, especially Prof. Hutter, comment on whether this is a real 
> issue? If it is, then I hope it can be updated in the repo.
>
>
>
> &GLOBAL
>
> PROGRAM_NAME ATOM
>
> &END GLOBAL
>
> &ATOM
>
> ELEMENT Tl
>
> RUN_TYPE PSEUDOPOTENTIAL_OPTIMIZATION
>
> ELECTRON_CONFIGURATION [Xe] 6s2 5d10 6p1
>
> CORE [Xe]
>
> ! MAX_ANGULAR_MOMENTUM 2
>
> COULOMB_INTEGRALS ANALYTIC
>
> EXCHANGE_INTEGRALS ANALYTIC
>
> &METHOD
>
> METHOD_TYPE KOHN-SHAM
>
> RELATIVISTIC DKH(3)
>
> &XC
>
> &XC_FUNCTIONAL PBE
>
> &END XC_FUNCTIONAL
>
> &END XC
>
> &END METHOD
>
> &OPTIMIZATION
>
> EPS_SCF 5e-9
>
> max_iter 1000
>
> damping .2
>
> n_diis 15
>
> &END OPTIMIZATION
>
> &PRINT
>
> &BASIS_SET
>
> &END
>
> &END
>
> &AE_BASIS
>
> BASIS_TYPE GEOMETRICAL_GTO
>
> &END AE_BASIS
>
> &PP_BASIS
>
> BASIS_TYPE GEOMETRICAL_GTO
>
> &END PP_BASIS
>
> &POTENTIAL
>
> PSEUDO_TYPE GTH
>
> POTENTIAL_FILE_NAME POTENTIAL
>
> POTENTIAL_NAME GTH-PBE-q13
>
> &END POTENTIAL
>
> &POWELL
>
> ACCURACY 1.e-10
>
> STEP_SIZE 0.5
>
> WEIGHT_PSIR0 0.1
>
> &END POWELL
>
> &END ATOM
>
> --
> You received this message because you are subscribed to the Google Groups 
> "cp2k" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to cp2k+uns... at googlegroups.com<mailto:cp2k+uns... at googlegroups.com
> >.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/cp2k/ca5634d0-1e75-41b6-8821-77b4b0ba76e5n%40googlegroups.com
> <
> https://groups.google.com/d/msgid/cp2k/ca5634d0-1e75-41b6-8821-77b4b0ba76e5n%40googlegroups.com?utm_medium=email&utm_source=footer
> >.
>

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/fd1ee1ad-4768-4839-922f-37f0c2a33749n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20221026/01369d85/attachment.htm>


More information about the CP2K-user mailing list