UV transmission microscopy of sunscreens

[JMC]

Over the last year or so, I've been working away on a UV transmission microscope. The idea being could I build a microscope that would work in the visible, UVA and UVB regions to allow me to look at the structures in topical sunscreen products (creams, lotions etc). The structure of a product and the distribution of the sun filters influences how it spreads and how efficient it is as blocking the UV. A few days ago, I presented some initial images from it at the 15th Sun Protection Conference in London as part of my UV imaging talk, and wanted to share them here too, as the UVP has been a principle part of my learning over the last few years. My talk at the conference won an award from the International Journal of Cosmetic Science - no money for that, but they will cover the open access fees for a paper on this, which will allow me to reach a wider audience with the work.

 

For these images I took two different creams - both oil in water emulsions (like cream) one with a UVA absorbing filter in the oil phase, the other with a UVB absorbing filter in the oil phase. I then dispersed both of these in water, and mixed the two dispersions together. The end result is a mix of oil drops from the two products suspended in a water phase. A droplet of this mixed dispersion was then placed on a fused silica slide and a fused silica coverslip placed on top.

 

First, how the dispersion looks in visible light (mainly 546nm from a filtered mercury xenon lamp).

In normal visible light, you can see the oil droplets in the water phase. Droplet size around a 3 micron and smaller. The droplets all look similar as would be expected based on how they were made and their optical properties in visible light.

 

Now the same sample in UVA (filtered 365nm).

In UVA some of the oil droplets now look black, while others remain clear. The black ones are the ones containing the UVA absorbing filter.

 

And finally, in UVB (313nm).

In the 313nm image, the droplets which are black were clear at 356nm and vice versa.

 

It's early days, and there is still plenty of work to do with it. These are sample images I've got permission to share, as I'm working on a publication of a larger data set with a client. Movement of the samples is an issue - the droplets absorb quite a lot of light in the UV and warm up, and Brownian motion becomes a real problem and limits shutter speed. I can't stack images for the same reason, as there is too much random movement, so these were single images with a 32x objective.

 

The build is based on an Olympus BHB, and I've extensively modified it to remove any glass and replace it with fused silica. I'm currently able to image in the visible light and at 365nm and 313nm, although the microscope itself should be good down to around 250nm (with different light source, filters and camera). At the moment it is using a monochrome converted SLR, which is not ideal especially at higher magnifications, but in future I hope to get hold of a different camera for it.

 

The aim is to use this for helping product development of sunscreens. Skin cancers are on the rise unfortunately, so there is a real need to develop better products, but also to communicate their benefits in a way that consumer can visualise and understand. This is where UV imaging really can help, so is the approach I am taking.

[colinbm]

Congratulations & thanks for sharing.

[nfoto]

My immediate suggestion, which you presumably already have used, is to run two separate sequences for the formulae. Thus avoiding the ambiguity of mixing the droplets together.

[JMC]

Thanks. This was purely a demo of the method. When looking at the products for research purposes, they'll be looked at individually rather than as mixes.

[dabateman]

Can you use a capillary trap slide?

You are basically trying to do single particle imaging.  To aid in that researchers have used light traps with optical tweezers or capillaries slides with large starting reservoirs that funnel down to single channels. 

One made of fused silica might be possible. 

 

You kind of want to do something like this: 

https://pubs.acs.org/doi/10.1021/acs.langmuir.5b04246

 

There average particle size is 3um and looking at mixing in aqueous vs oil solution. 

[JMC]

Possibly David, thanks for the link. The biggest issue is that the oil droplets are very prone to shear and they break down very easily, resulting in just an oily mess. Anything that is done to them needs to be done very carefully, especially when they are put under a coverslip.

[Andy Perrin]

JMC, oil viscosity is highly temperature sensitive (exponential in fact). Perhaps cooling the mixture with Peltier plate underneath might harden  the drops AND slow down the Brownian motion? Obviously not all the way to the freezing point, but a drop of 15C might make a visible difference to the behavior. 

[JMC]

Indeed Andy. Thanks.

[OlDoinyo]

You could use some sort of micro-cuvette device--two panes of glass with an adjustable, very narrow gap between them so that you can create a liquid film of known thickness. Perhaps a slide and cover slip separated by calibrated shims would do.

[JMC]

Cheers Colin, much appreciated, and thank you everyone for the advice - certainly some new directions for me to think about. I will work to improve the method in the background - while the clients I work with want results, they don't always want to fund method development, so this is something I fund myself. As such I need to balance potential improvements against cost.

[JCDowdy]

9 hours ago, JMC said:

... the clients I work with want results, they don't always want to fund method development, so this is something I fund myself. 

Wow, don't I know it!

 

Good job!!

[JMC]

16 hours ago, JCDowdy said:

Wow, don't I know it!

 

Good job!!

Cheers John. Indeed. What I find funny is when someone says to me "I can offer 5 days work. But I don't want you working with any of my competitors". Oh, yes I can survive for a year on your 5 days work, thank you. This is one of the reasons I try and publish what I can about new methods or new techniques. I'd rather have it out there for people to try and see if it works for them, or to adapt and build on.