AFM NdNiO2 tutorial: Issue with Replicating the spin polarized state

[YUBODRZHANG]

Hello TRIQS/solid_dmft Community,

I am currently experiencing difficulties in achieving the spin polarized state while attempting to replicate the tutorial on "AFM state of NdNiO2." Starting with a non-magnetic DFT simulation, the magnetic moments produced by the DMFT are virtually zero, only showing minor numerical fluctuations.

Dr. Prof. Hampel previously mentioned the possibility of initiating spin polarization at the DFT level. This approach was effective for me, but I would prefer to introduce spin solely at the DMFT level, in line with Prof. Hampel's recommendation.

I suspect there may be a control flag to guide the DMFT towards the spin-polarized state, but my attempts to manipulate, for example g0_mix, have been unsuccessful. The log file is attached for your reference. result.soliddmft.txt

Thanks,
Yubo Zhang
Minjiang University

[the-hampel]

Dear @YUBODRZHANG,
Recently we changed the way the magmom flag works in the solid_dmft input file. The current value to initiate the magnetic state might be insufficient. Maybe you can try to increase the value:

magmom = -0.001, 0.001

to something like:

magmom = -0.4, 0.4

I wil update the tutorial shortly, after trying the parameters myself.

Best,
Alex

[the-hampel]

Hi @YUBODRZHANG ,

I just updated the tutorial with the new config file and parameters: 24b859e . The documentation page should reflect this shortly.

• Alex

[YUBODRZHANG]

Dear Dr. Prof. Hampel,

I greatly appreciate your help! I've successfully reproduced the tutorial by increasing the number of sweeps before measurement to length_cycle = 3000.

Best regards,
Yubo