EasySpin forum

Dear Stephan,
I tried the discussed issue of an radical with one Cl couplings (of 12 all equivalent nuclei) : I do get the Error:

Error using isotopologues
Cannot compute isotope patterns for equivalent nuclei.
Error in expandcomponents
Error in garlic
Error in isotest (line 8)
garlic(Sys,Exp);

So I do something wrong, it looked so easy when you showed it to me.
Burkhard.

EasySpin does not support equivalent nuclei for tme mixtures of isotopes. You should define each nucleus separeately. For your case it should be defined like this:

Burkhard, katarkon has it right. In the current ES version, for multiple equivalent nuclei, you have to spell them out if you want natural abundance mixtures. I know it would be nice if this wasn't necessary and one could just specify Sys.Nucs='Cl' and Sys.n=12. We'll put this on our to-do list.

Thank you Stepan and katarkon, with this method I was able to get the whole hyperfine pattern. As you guessed right, the direct way to simulate natural abundance equivalent nuclei would be a nice feature to have.

Thinking about this a little more, this is actually an important feature we should include in the next EasySpin version. The savings can be enormous.

In your case of N=12 Cl sites, there are 2^N = 4096 isotopologues (since we have one of 2 Cl isotopes, 35Cl and 37Cl, per site). If the sites were not magnetically equivalent, we would have to simulate all 4096 spectra and add them up with the appropriate abundance weights. And that's what the ES code with Sys.Nucs='Cl,Cl,Cl,Cl,Cl,Cl,Cl,Cl,Cl,Cl,Cl,Cl' does.

Now, however, the sites are all equivalent. This means, for example, that the isotopologue with 37Cl in sites 1 and 5 gives the same spectrum as the isotopologues with 37Cl in site 6 and 12, or any other pair of sites. Therefore, there are only a total of N+1=13 distinct spectra. Each spectrum is due to a number of equivalent isotopologues: [1 12 66 220 495 792 924 792 495 220 66 12 1], where the first entry refers to the 35Cl-only isotopologue, the second to the isotopologues with one 35Cl, etc. The simulation would therefore run much faster: 13/4096 = 0.003.

It is nontrivial to implement this for the general case of an element with more than 2 isotopes (e.g. Sn), but we will attempt it.

After some initial work, we are optimistic that this feature will be implemented in the next version.

Meanwhile, here is a list of isotopologue spin systems and their abundances that you need to run separately in the current version (5.1.10). Make sure the hyperfine couplings are properly scaled between 35Cl and 37Cl. The last two or three isotopologues can probably be skipped, since they are of very low relative abundance, and will probably not show above the noise in your experimental spectrum.

This is now implemented in version 5.2. The script you posted works now as expected.

Dear Stefan,
thank you
Without simulation all isotope contributions of 12 Chlorine and 4 Silicon separably, I could get the same with one garlic call
Burkhard.

Great! Let us know once you've published on this. This could be a great case to include in the EasySpin examples.

I see that the output of the function isotopologues() was changed. New version does not provide data about hyperfine splitting more. Is any way to rid it out?