How to model paramagnetism or disordered local moments (DLM) in VASP?
Dear VASP Community,
I am a beginner in modelling and was trying to figure out modelling a paramagnetic state in VASP for bulk, surfaces and interfaces and exploring ways to implement the Disordered Local Moments (DLM) approach. I understand that a truly paramagnetic state cannot be directly simulated using conventional spin-polarized calculations, but an approximation via the DLM method or similar techniques can be used.
I would like to ask for guidance on the best way to implement this approach in VASP. Specifically, I am interested in:
1. DLM: How can one generate and implement a spin-disordered supercell in VASP to approximate a paramagnetic state? Are there any recommended tools or scripts to generate an SQS for spin disorder? Or can we simply say MAGMOM = 5 -3 +3 -5 etc to represent an overall paramagnetic kind of behaviour?
2. Collinear vs. non-collinear magnetism: Would a simple collinear spin disorder (random up/down spins) be sufficient, or is a non-collinear approach (via LNONCOLLINEAR = .TRUE. and SOC) more appropriate? What would an INCAR look like?
3. Alternative methods: Are there other practical ways within VASP to approximate a paramagnetic state?
4. What would be the important things to consider when changing systems from bulk to surfaces to interfaces?
If anyone has experience implementing such an approach, I would greatly appreciate any insights, example INCAR/POSCAR setups, or references to useful literature.
Thank you in advance for your help!