SpinDynamics.org User Support Forum

I am trying to simulate the resonance conditions for my EPR/DNP based on field-swept experiments.

The EPR program 'fieldsweep' seems to be very accurate as I change the mw_freq and field sweep parameters.

For solid-effect DNP, I am using 'dnp_field_scan', and I notice several very odd things especially compared to the fieldsweep simulation.

1) sys.magnet seems to NOT set the center of the DNP field sweep, but DOES set the 1H frequency needed for the SE resonance.
2) parameters.mw_frq seems to only start to make sense if specified in rad/s, rather than Hz (which is counter to the description)
3) parameters.mw_frq seems to in fact set the center of the field sweep, but still does not totally match the 'fieldsweep' resonance conditions,
which seems counter-productive to have the center of the field swept experiment defined by a frequency.

In fieldsweep I can use the following 'key parameters':

%%%%%%%%%%
% Electron g-tensor
inter.zeeman.eigs=cell(1,1);
inter.zeeman.euler=cell(1,1);
inter.zeeman.eigs{1}=[2.003 2.003 2.003]; %BDPA tensor taken as isotropic
inter.zeeman.euler{1}=[0 0 0];

....

parameters.mw_freq=2.0e9; %%freq
parameters.window=[0.070 0.076]; %%field

...
%%%%%%%%%%%


Then in dnp_field_scan I can use the following but NOT observe the zero-crossing of the solid effect at the same field value:

% Magnetic field
sys.magnet=0.071; %field sets 1H larmor but not field sweep?

% Electron g-tensor
inter.zeeman.eigs=cell(2,1);
inter.zeeman.euler=cell(2,1);
inter.zeeman.eigs{1}=[2.003 2.003 2.003]; %take BDPA as isotropic
inter.zeeman.euler{1}=[0 0 0];

% Coordinates (Angstrom)
inter.coordinates={[0.00 0.00 0.00];
[3.00 0.00 0.00]};

......

parameters.mw_frq=2.0e9/(2*pi); %in rad/s ?? %sets center of field sweep?
parameters.fields=linspace(-20,+20,256)*1e-3; %mT

....

%%%%%%%%%%%%%

Any help would be appreciated, thank you.

Okay, in the order of the original post:

1. The sys.magnet parameter does set the centre of the sweep - this is the field that goes into the background Hamiltonian in Line 70 of dnp_field_sweep.m function. The microwave offset is then added in Line 78, and the field offset inside the loop. The code is well commented, could you take a look? Maybe my conventions for what is meant by offset is different (by a sign?) from yours.

2. Definitely Hz, see the 2*pi factor in Lines 76 and 79 dnp_field_sweep.m function.

3. This I think is the core of the matter. I would be happy to update that function to work in the way experimentalists see as more natural. Is there some simple picture that you would like that function to reproduce, and a simple description of how you would like it parameterised? It's a quick job for me to update the API.

Aha, I see the confusion. In dnp_field_scan I treated the mw_freq as the irradiation frequency rather than an offset to the frequency determined by sys.magnet. Again, this seems counter to the convention of the 'fieldsweep' function for EPR, which is where I got confused, though the difference is in fact commented.

I suppose for an experimental approach I would like to be able to set the center field and applied e- irradiation frequency independently, rather than an offset to frequency.

Thank you very much for helping clear this up.

It would actually be very useful to know what syntax fees more natural - I can make alterations of course. How would you wish that parametrisation to work?

Well the 'fieldsweep.m' function syntax all makes a lot of sense / feels natural. The sys.magnet becomes obsolete and you're left with independent parameters for irradiation frequency and the fields, which ultimately is how the experiment would be parameterized. It would make a lot sense to me if the field-swept DNP functionality matched this EPR one.

Alright - I think I can do the syntax matching. Working on it...