• Ultraviolet Photography
  •  

Potential UV diffuser - roughened PMMA plates

9 replies to this topic

#1 JMC

    Member

  • Members(+)
  • 349 posts
  • Location: London, UK

Posted 10 March 2018 - 14:47

I've seen a few threads on here where people have been trying to sort out a suitable UV diffuser, so wanted to share these. These plates are used for in vitro sunscreen testing. They are PMMA and have a texture on them, and are 5cm square and a couple of mm thick. I have a couple of different types - referred to as HD6 and Frosted - here's what they look like.
Attached Image: HD6 and frosted plates.JPG

I assumed they are UV transparent (given what they are used for), but to be sure I measured transmission with my Ocean Optics setup.
Attached Image: Trans full PMMA.jpg

Attached Image: TransUV PMMA.jpg

They've got consistent transmission down to about 300nm, with Frosted one letting through less than the HD6.

As they are plastic they should be pretty simple to cut with a dremel to fit a torch. I've not got a UV torch to try these out, but if anyone wants some drop me a PM and I can let you have some for the cost them plus postage.

#2 Mark

    Member

  • Members(+)
  • 428 posts
  • Location: Massachusetts, USA

Posted 10 March 2018 - 23:16

That acrylic has surprisingly good UV transmission. I got a UV fused silica diffuser, and even that doesn't have as good transmission (https://www.edmundop...tical-diffuser/). And more to the point, it has a significant impact (decrease) on UV transmission from my torches. I've since resorted to light painting instead to get better coverage of my subjects. I still get hot spots and strong shadows from time to time though - it works, but its not exactly an easy fix.

#3 JMC

    Member

  • Members(+)
  • 349 posts
  • Location: London, UK

Posted 11 March 2018 - 11:15

Mark, the diffused silica in the link has much higher transmission the PMMA plates I have - the PDF on the Edmunds sight has transmission up around 70%. I was surprised that the measured transmission with my PMMA plates was so low, as in the photograph they look to be letting quite a lot of the light through. I'm guessing it's a quirk of how I'm measuring it - I don't use an integrating sphere, and given they are scattering a lot of the light I can see how this would drop the measured transmission with my setup. They key thing from my point of view is that they look to let as much UV through as visible (down to about 300nm anyway). Which is why they could be useful. As an indication, they are about 5USD each as opposed to 80USD for the UV silica.

#4 JCDowdy

    JCDowdy

  • Members
  • 1,119 posts
  • Location: Arlington, TN

Posted 11 March 2018 - 17:05

Having worked in the area of in-vitro sunscreen testing since the 1980's I am very familiar with the various PMMA plates. I have actually tried both the Schönberg and Helioplate substrates as UV diffusers for UV photography. The thinner Helioplate is much easier to cut and I have used it to make a diffuser for a a small flash. I melt-cut it with a flame heated Exacto blade. These plates are to small to be cut to fit larger strobes but perhaps could be cemented edgewise or fit into an adapter frame.

You are right, correct measurement of transmittance is best done with an integrating sphere. However, your method may provide a some indication of the degree to which these materials scatter and thereby reduce the intensity.

#5 Mark

    Member

  • Members(+)
  • 428 posts
  • Location: Massachusetts, USA

Posted 13 March 2018 - 18:02

@JMC: I see. I originally misread your charts. I assumed your y-axis of 0-10 was scaled 0-100%, not actually 0-10%. Now I understand the PMMA plates have much lower transmission! Thanks for pointing that out.

#6 JMC

    Member

  • Members(+)
  • 349 posts
  • Location: London, UK

Posted 13 March 2018 - 19:47

No problem Mark. I think the lower transmission for the PMMA plates is, at least in part, an artifact of my measurement though. I can't measure the scattered part of the light. I have a collimator on the incoming light source, and the another the same size on the collecting fibre. As I don't know how the Edmund folks got that transmission curve (quite probably they used an integrating sphere to get total transmission) I'm not certain the absolute values can be meaningfully compared.

#7 Cadmium

    Member

  • Members(+)
  • 1,547 posts

Posted 13 March 2018 - 21:57

The UTV and SUTV plastics work well for this sort of thing also.
https://www.polymerp...ents_uvta.shtml

#8 JMC

    Member

  • Members(+)
  • 349 posts
  • Location: London, UK

Posted 14 March 2018 - 14:53

Just as a quick aside, and to demonstrate how scattering does reduce the measured transmission through theses plates with my setup, I did a simple experiment. I've re-run the 2 PMMA plates for transmission, along with a Hoya NDx8 ND filter. I then put a drop of water on each one and re-ran for transmission. The idea here was to put something on them with similar refractive index and very low light absorption, to make a smooth surface and get rid of the roughened surface, and reduce scattering.

I'm showing the transmission spectra in pairs - with and without water. Note the transmission scale is the not the same for all of them. Firstly the Frosted plate;
Attached Image: Frosted with and without water.jpg

Adding water to the frosted plate, massively increased transmission.

Secondly, the HD6 plate;
Attached Image: HD6 with and without water.jpg

Again a big increase in transmission when water was added.

Finally, the Hoya NDx8 neutral density filter;
Attached Image: NDx8 with and without water.jpg

Adding water to the Hoya ND filter, dropped the transmission (it will absorb something after all, and the droplet itself being curved will impact the light transmission). What struck me more here though is how 'not neutral' the so called neutral density filter is.... Also the scan without water, gave a similar transmission to the HD6 plate which surprised me. Simply looking at photos of them, says to me that more light is getting through the HD6 plate..
Attached Image: IMG_2534 without water.JPG

Adding water, it is easy to see the extra transmission from the PMMA plates;
Attached Image: IMG_2536 with water.JPG

Overall, adding the water on to roughened PMMA plates significantly increased measured transmission with my spectroscopy setup, and reduced transmission when added to an already flat substrate. I'm not sure how the Edmund ones could be letting as much light through as in the link above, and still be diffusing the light (if they have measured it like me), but unfortunately I don't have one to test against. I think some of that difference may come down to how I measure though, and the fact that I have a collimated light source and a detector with the same size collimator, so I cannot capture scattered light.

#9 Cadmium

    Member

  • Members(+)
  • 1,547 posts

Posted 15 March 2018 - 00:29

All such diffusers I have tried reduce light 'amplitude'. Sand blasted, and sanded. Not only does it scatter the light more broadly toward the target, but it also reflects light back at the light source, and how much loss there is in the reverse, I have no idea.

#10 Andy Perrin

    Member

  • Members(+)
  • 1,377 posts
  • Location: United States

Posted 15 March 2018 - 02:46

I have actually researched how to find the reflectance of a rough surface, and the answer is found in H.E. Bennett and J.O. Porteus's article, "Relation Between Surface Roughness and Specular Reflectance At Normal Incidence" (Vol 51, No 2, February 1961, Journal of the Optical Society of America). They comment in the introduction,

Quote

Although this theory was developed in connection with the scattering of radar waves from rough water surfaces, it is equally valid in the optical region.

Despite the title, they give both the diffuse and specular parts. The full expression is,
Attached Image: Screen Shot 2018-03-14 at 10.40.11 PM.png
where R0 is the reflectance for a smooth surface, lambda is wavelength, sigma is the RMS roughness of the surface (RMS distance from the mean surface), m is the RMS slope of the profile of the surface, and ∆theta is the acceptance angle of the measuring instrument.

For more you can read the full article here:
https://www.osapubli...i=josa-51-2-123

Edited by Andy Perrin, 15 March 2018 - 02:48.