by Din » Fri Jul 20, 2018 8:33 am
lobaz wrote: ↑Thu Jul 19, 2018 1:27 pm
Dinesh, do you have any hands on experience how the scalar theory holds with surface relief holograms, such as in photoresists?
It holds quite well if the amplitude is low. As I mentioned elsewhere, if the resist is etched so that the sin wave is clipped, , which usually happens, you get additional Fourier components, but, these also follow the scalar theory. Commercial resists, such as from Towne, have emulsion depths of 1 micron, and they're usually not etched all the way to the glass, so the amplitude is low. So display holography with commercial plates follow the scalar law, and, in fact, you use the grating formula λ = dsin(θ) when we do "pseudo-colour" transmission holography. That is, you have several H1's, each for a different colour image, and copied at a different angle so that, in the final resist H2, you see several colours at the same time.
If the amplitude is high compared to the (inverse) spatial frequency, then you start to go to the vector regime. Now, you begin to get Bragg selectivity, but the reconstruction begins to become dependent on the polarisation of the recon beam. "High" means that h/d~>3, where h is the amplitude and d is the (inverse) spatial frequency.So, if the spatial frequency is 1000 l/mm, the spacing is 1 micron, and if the amplitude is much greater than 3 microns, you begin to see vector effects.
In our case, we coat our own resist. Our resists usually are about 1 micron - 2 micron. However, I did have to make a resist plate that was over 3 microns deep for a specific type of blazed grating, with a period of over 1000 l/mm. This grating did have Bragg selectivity (it was supposed to!), but I didn't notice any polarisation effects. However, I wasn't looking for them, so there may have been polarisarisation effects I didn't notice. The grating was designed to work for white light sources, so, if I wanted to test polarisation, I would have to reconstruct it with a laser, at different polarisations.
[quote=lobaz post_id=69685 time=1532024841 user_id=2217]
Dinesh, do you have any hands on experience how the scalar theory holds with surface relief holograms, such as in photoresists?
[/quote]
It holds quite well if the amplitude is low. As I mentioned elsewhere, if the resist is etched so that the sin wave is clipped, , which usually happens, you get additional Fourier components, but, these also follow the scalar theory. Commercial resists, such as from Towne, have emulsion depths of 1 micron, and they're usually not etched all the way to the glass, so the amplitude is low. So display holography with commercial plates follow the scalar law, and, in fact, you use the grating formula λ = dsin(θ) when we do "pseudo-colour" transmission holography. That is, you have several H1's, each for a different colour image, and copied at a different angle so that, in the final resist H2, you see several colours at the same time.
If the amplitude is high compared to the (inverse) spatial frequency, then you start to go to the vector regime. Now, you begin to get Bragg selectivity, but the reconstruction begins to become dependent on the polarisation of the recon beam. "High" means that h/d~>3, where h is the amplitude and d is the (inverse) spatial frequency.So, if the spatial frequency is 1000 l/mm, the spacing is 1 micron, and if the amplitude is much greater than 3 microns, you begin to see vector effects.
In our case, we coat our own resist. Our resists usually are about 1 micron - 2 micron. However, I did have to make a resist plate that was over 3 microns deep for a specific type of blazed grating, with a period of over 1000 l/mm. This grating did have Bragg selectivity (it was supposed to!), but I didn't notice any polarisation effects. However, I wasn't looking for them, so there may have been polarisarisation effects I didn't notice. The grating was designed to work for white light sources, so, if I wanted to test polarisation, I would have to reconstruct it with a laser, at different polarisations.