Cerium ADMM generation

[Michele Loriso]

Hi everyone, 

following a previous discussion about the lack of ADMM basis for U (see: https://groups.google.com/g/cp2k/c/BlnhW_qrS2g/m/HRAfYuZtAAAJ), I'd like to ask you if there is a way to generate an ADMM basis set (which in my case is intended for the Ce element) using &ATOM. 

Maybe this is a wrong approach, but I'd like to start from DZVP/TZVP basis sets and LnPP1 pseudopotentials, which are already available for cerium.

Thanks in advance! <3

[Jürg Hutter]

Hi
Yes, you can use the ATOM code in CP2K to optimize a basis set for use in an ADMM calculation.
Short run down of the procedure:
1) From your primary basis set, decide on the structure (number of primitives and contraction pattern)
for your ADMM basis. Take hints from other elements near by.
2) Optimize the exponents using the atom code
3) Generate the contraction using response basis option in the atom code.

regards
JH

________________________________________
From: cp...@googlegroups.com <cp...@googlegroups.com> on behalf of Michele Loriso <michele...@unibas.it>
Sent: Monday, February 16, 2026 11:02 AM
To: cp2k
Subject: [CP2K:22101] Cerium ADMM generation

--
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...@googlegroups.com<mailto:cp2k+uns...@googlegroups.com>.
To view this discussion visit https://groups.google.com/d/msgid/cp2k/56f8a76f-94c8-4ada-a5a3-3cbf5760a728n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/56f8a76f-94c8-4ada-a5a3-3cbf5760a728n%40googlegroups.com?utm_medium=email&utm_source=footer>.