For a 405nm laser diffracting through a 10nm spaced grid, how to compute the interference pattern? Or more generally, is there a relatively simple formula to compute the diffraction interference pattern when slit width is an order of magnitude smaller than the wavelength?

How I tried first to solve the problem before posting:

I attempted to use the general diffraction formula which works fine if slit-width is greater than the wavelength, but this equation doesn't appear to apply as it's not possible to compute the sine of a number greater than 1.

I understand that if slit width is so relatively tiny, like 1/40th of wavelength, the wave that gets through becomes a point source of a spherical wave at either side of the slit, and I should use Huygens-Fresnel principle, which I learned from the second answer (Jess Brewer) here What-happens-if-the-gap-width-is-smaller-than-the-wavelength. Further research lead me to a general application of Kirchhoff's diffraction formula here Kirchhoff%27s_diffraction_formula which is significantly more difficult for me to understand and therefore apply.

A similar question has been asked on various places, but even accepted answers do not actually suggest a simple formula  - other than suggest Kirchhoff's formula will help -  most instead re-explain for situations where slit widths approach the wavelength like here the-equation-of-diffraction/206189 or touch glancingly on the issue i.e. by suggesting that EM simulation programs can help.

I think that there also would be an evanescent wave, which appears to be a form of standing wave that once established has net zero energy transmision unless a particle interacts with it. This (standing) evanescent wave attenuates exponentially, beyond the tiny slits, perhaps detectable only to about 100nm on the other side. However, if we look at water waves going through a slit much smaller than wavelength, we observe that the waves do in fact go through and continue on as point source circular waves (ie distinct to the evanescent variety).

I also researched that the grid should be much thinner than the wavelength, otherwise the grid will act with light as a Faraday cage attenuates other EM spectra, but in my example the grid thickness (10nm) is much smaller than the wavelength so this should not be a problem.