Zero-field splitting induced by Dzyaloshinskii-Moriya interactions

[thanasis]

I am trying to calculate the zfs induced by Dzyaloshinskii-Moriya interactions to the quartet state of a S=1/2 triangle, as predicted here (http://dx.doi.org/10.1021/ic701796q).

The way I have already done this for anisotropic exchange and dipolar interactions was to take the interactions in their local ij-frames, transform each one them to the molecular frame (see previous post), sum them up using Bencini-Gattechi's coefficitents ("EPR of exchange-coupled systems", p. 103), then use Matlab's eig function to find the eigenvalues and eigenvectors of the sum matrix (let's call it D).

Then it is easy to calculate D end E from the matrix in the reference frame of the eigenvectors as:

Code: Select all

Dzfs = 3*D(3,3)/2;
Ezfs = (D(1,1)-D(2,2))/2;

For dipolar and anisotropic interactions that has worked out quite well, but for DMI interactions, the interaction matrix is antisymmetric, i.e. GG = [0 Gz -Gy; -Gz 0 Gx; Gy -Gx 0], and by definition has no real eigenvalues. It seems I cannot use this approach to calculate the zfs-component induced just by DMI.

For certain combinations of Gx, Gy and Gz, pepper does give me a very nice zfs spectrum. Is there any way I can extract the respective D value?

[Stefan Stoll]

There is currently no way of doing this with EasySpin. Have you considered simulating energy levels for a few orientations using the DMI, and then fitting a spin system with D to them?

[thanasis]

Something simlar: I print out the energy levels and directly calculate the energy difference which is directly related to D. Initially, I did that for other interactions even before doing eigendecomposition.

But thanks for the reply, it's good to know I didn't overlook something obvious.