new version

[chemshd]

Hi Stefan,
when do you plan to publish the new features of the 5.0.0 version?

[Stefan Stoll]

You can download EasySpin 5 now - it's still in beta status, but essentially complete. Have a look there!

[chemshd]

Well, I see the new version, and I like it very much, especially the curry function.
Actually I wrote a function to achieve this before you release this 5.0 version.

I had a rough look of this function and realized that, maybe I am wrong, it is to simulate powder data, rather than single crystal one.
Is this correct?

I would suggest include the single crystal simulation as well.
and...maybe also include the susceptibility tensor option.

regards,

ShangDa

[chemshd]

To change cgs-emu to SI for magnetic susceptibility, why do you use
chizz_SI = chizz*10;

I would say use
4pi*10(-6)*chi-mol(SI, m3/mol) = chi-mol(cgs, emu/mol=cm3/mol)

one can refer to
J. Chem. Edu. 1972, 49, 114
Pure Appl Chem, 2005, 77, 497

[Stefan Stoll]

The curry function can calculate both powders and single crystals. See the examples, there is one for single crystals. It essentially works the same as for other EasySpin functions. The documentation is incomplete at the moment, since it doesn't mention single crystals.

Regarding units, thanks for pointing this out. We will fix it.

[Stefan Stoll]

You can now download 5.0.1, where the units of the molar magnetic susceptibility are now correct SI units (m^3 mol^-1). All the documentation and examples are updated to reflect this change.

Also, I included the single-crystal capabilities of curry in the documentation.

Calculating the full susceptibility tensor is now on our list for future enhancements - thanks for the suggestion!

[chemshd]

Thank you very much for this update.
Only a small suggestion, maybe you can also include another function calculating the partition function of a system.
Since you need this to calculate curry in any case.

I am also curious how do you calculate the numerical derivative ?
I did this by finite difference method, calculating the partition function of two(for magnetization) or three(for susceptibility) very closed field(1 Oe), then do the difference. While I come with a problem that the result is somehow related to the difference of the field I choose, e.g., 1 Oe and 10 Oe will be difference, and sometimes, the 1 Oe result reaches the matlab precision limit.
How do you do this derivative?

[Stefan Stoll]

For the numerical derivative, ES uses the procedure from Press et al, Numerical Recipes in C, chapter 5.7.