From sywang1221 at gmail.com Tue Sep 1 07:42:21 2026 From: sywang1221 at gmail.com (Shiyang Wang (W0728)) Date: Tue, 1 Sep 2026 00:42:21 -0700 (PDT) Subject: [CP2K-user] [CP2K:22418] Re: SCF convergence issue for Fe2O3/Co3O4 heterojunction using diagonalization In-Reply-To: <1ebae9d7-3538-4d2a-a081-1d3258840d2fn@googlegroups.com> References: <1ebae9d7-3538-4d2a-a081-1d3258840d2fn@googlegroups.com> Message-ID: Hi, Currently in develop branch of CP2K, OT with k-points has been available. For diagonalization, you can try Kerker mixing with common settings: ``` &MIXING METHOD KERKER_MIXING ALPHA 0.4 BETA 1.6 ALPHA_MAG 1.0 BETA_MAG 0.0 NBROYDEN 8 &END MIXING ``` Alternatively you can try ADIIS but it usually spends much more steps than common density mixing schemes. Regards, SY ?2026?8?31???? UTC+8 17:09:22 ??? > Dear colleagues , > > I am trying to optimize an Fe?O?/Co?O? heterojunction structure using > diagonalization, but I have been unable to achieve SCF convergence despite > many attempts. > > I have tested several different MIXING METHOD settings and also increased > the electronic smearing temperature from 300 K to 1000 K. I have set &BS > for Fe, Co and O. And I set DFT+U for Fe and Co. However, the SCF > residual still only decreases to around 0.1 and does not converge further. > > Apart from switching to the OT method, are there any other approaches or > SCF settings that I could try to improve the convergence for this system? > > I have attached my input file for reference. Any suggestions or advice > would be greatly appreciated. > > Best regards, > > Milk > > -- 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 visit https://groups.google.com/d/msgid/cp2k/d61a884e-e78b-494f-a996-6a5117b4f4a0n%40googlegroups.com. -------------- next part -------------- An HTML attachment was scrubbed... URL: