Artificial Sunlight (official topic)

I can't seem to find the link to the mirror vinyl film you used. Can you give me a source to buy it from?

Have a look for "reflective vinyl mirror" on Amazon or Ebay - there should be a few options! 🙂

Hi, I am really interested in this. I work Overnights and stay mostly nocturnal. My home is built into ground with only the roof and front exposed, meaning my bedroom has no windows, which is great for my sleep schedule but not so much for biology.

Can you point me towards more information on the principals that make this effect possible, the shape and size specifically. My goal would be to make a slimmer package, like i mentioned, my house is in the ground so space is a premium in terms of how large i can make something like this, and expect to hide it somewhere convincing as a window or portal to the outside. Thanks for your awesome work and explanations. 

If i can work out something convincing, i might be interested in building them into faux skylights even. 

I'm designing a house to be built in maybe 2 years, and I have the perfect spot for this to light up most of a 3,000 sqft space.  Granted, the light you use looks a bit small for what I need, but itd be great if you upgrade/update this in a year or so.  Especially if you find a good supply of nanoparticles and make an epoxy setup.

I'm working on building one of these now, but I don't have space for something so deep so I'm trying to make due with an array of focused LEDs.

My biggest question is actually about the diffusion layer, as I'm hoping for something simpler than the soap solution that doesn't require building a water-tight two-pane construction.

What do you guys think about using metallic glitter flakes (say 100 microns) dissolved in clear paint and spread in a thin layer on glass? If that makes sense, would some metals work better than others?

Hi friends of sunshine 🙂

After almost 6 months and tons of trials and errors I finally managed to produce a Rayleigh scattering plastic sheet made out of epoxy and real TiO2 nanoparticles. 

This is the result (prototype with a few bubbles in the epoxy): 

So you can do it at home with a few tools and dont need to use those haevy glass sheets and soap water anymore.

I have to thank DIY Perks (Matt) for his video from last year which got me excited for this project.

If there is any interest I can write which tools you need and how to make it.

WOW! I've got to say, that's an incredible breakthrough for this project! Excellent work Nolo!!! Would you mind explaining how you managed it in detail? I'd really like to have a go myself - if it works as well as it looks like it does in your picture it'll be a big deal!

@diyperks Hi Matt, thank you ! It indeed looks like on the picture. I even measured the Color temperature with my spectrophotometer and it reads about 20000 Kelvin, which is in the range of a blue summer sky 🙂

I used a Ultrasonic Probe device to disperse the TiO2 nanoparticles with the hardener of the epoxy (epoxy comes with component A and B - and B is the hardener).

After 30 Minutes of sonification the particles are fully dispersed in the hardener and then will be mixed with component A of the epoxy.

I built a mold out of Multiplex wood, so I can pour in rectangular or square shapes of epoxy sheets.

BTW, on the picture you see the epoxy sheet with a fresnel Lens sheet behind so the light gets collimated.

The Ultrasonic device I use looks like this:

I got it for 330 Euros, so was fairly cheap for such a device.

But maybe you can also use a high shear mixer like this which costs maybe 100 euros (but I havent tried if it works):

A Ultrasonic Cleaner like this sadly doesnt work, so dont try with these:

The thing is, that nanoparticles always stick together when you order them, it is called agglomerates.

These nanoparticles have to be separated from each other which is called dispersion.

And this is what these machines do, they separat the cluster of particles in individual particles.

The Titanium Dioxide nanoparticles which I have here are about 21 Nanometer and work perfectly.

They also need to be hydrophobized which means they are compatible with epoxy.

You can believe me if I say I read 100 of websites about chemistry and nano particles dispersion etc. online till I understood all this stuff 😬 

If you are willing to buy a machine for dispersion I can also send you some of the nanoparticles for free.

I live in germany and I think shipping to Great Britain will not be a big problem.

I ordered 50 Grams and I have plenty left. You need only very small amounts. Maybe 0.1 to 0.5 grams for a big 100 x 50 cm epoxy sheet.

I ordered my ultrasonic probe on Aliexpress, but maybe you also find a used one for cheap.

High shear mixers like mentioned above (if they work) are much cheaper and are available easily on Ebay and Amazon.

You could first order a cheap high shear mixer and try it out and if it doesnt work send it back and then order a Ultrasonic probe. If a high shear mixer works, that would open the door for many more people, as they are not as expensive as Ultrasonic probes.

Fascinating. I believe you when you say you read 100s of websites to learn about the process. Sounds very complex.

Do you think you could achieve good enough results with a homemade ultrasonic probe device? Or is the precision of a bought one necessary to succeed?

@phenomanii I think a homemade device could work. The prebuilt ones only have the advantage to control them more precisely (time and power output etc.). I used the probe always at full power, so you dont need to regulate it in this regard. Also it has the advantage that the box can block the noise from the probe, as it can get pretty annoying for the human ears. But for this you can leave the room or wear headphones 😆 

And you only have to make sure, that the hardener with the nanoparticles dont get too hot while sonicating. So you can stop the device manually when it gets too hot and let it cool down for a while and then sonicate again. My probe has 20 KHz, so the probe should output at this frequency. Give it a try and I can send you some nano particles if you are resident in europe.

@nolo that looks great. Can I ask what form you obtained your nanoparticles? I have powdered titanium dioxide but I had read that it wouldn't be possible to completely separate the particles for proper dispersion from powder.

@jordiemc Thank you! My TiO2 particles are called P25 Type (hydropobized with octysilanes and 21 nm size) and come as powder as well. Its always some sort of powder as far as I know, that is why sonication is necessary to separate the particle agglomerates again into individual particles.

Its like 100's of nanaoparticles are stuck together when you receive them in powder form and the force of the ultrasonic waves separate them again from each other. 

@nolo thanks again for sharing your discoveries! I definitely want to give this a go myself, so I've been having a look on AliExpress for an ultrasonic mixer (or homogeniser). They do seem a lot more expensive than yours, however. Is there any chance you could share a link to the one you bought?

https://www.aliexpress.com/item/1005004100877317.html

https://www.aliexpress.com/item/32969123694.html

I may take you up on that nanoparticle offer! The only stuff I can find online is here  https://www.acsmaterial.com/titanium-dioxide-tio2.html and here https://uk.vwr.com/store/product/2384144/titanium-iv-oxide-aeroxide-p25 but it's very expensive and has a large nm variation window. Any thoughts? I'm not sure how to find out if they're hydrophobized (or is that a process I can do myself??)

@diyperks I looked into my orders on Aliexpress and realized that I ordered it in Germany from a Company called Vevor. They import the stuff from China. I had Aliexpress in my mind because I order much stuff there, but it was wrong. Here is the link to my device I ordered:

https://www.vevor.de/ultraschall-homogenisator-c_11060/ultraschall-homogenisierer-zelltrennung-500ul-50ml-tragfaehigkeit-kosmetik-p_010204703651

But sadly they dont sell it anymore, as you can see.

It was indeed a good offer for the price.

The cheapest way I can now think of is like the user phenomanii mentioned, to build an Ultrasonic probe by yourself.

But I have to search on the internet how to do it and where we can get the parts. Please give me some time.

Regarding the nano particles, they come already hydrophobized from the manufacturer.

To explain what this means is simple. TiO2 parcticles are normally compatible with water which is called hydrophile (hydrophile means "loving water" and comes from greek language).

Hydrophobic means the opposite, like it does not like water and cannot be mixed with it.

Epoxy is a substance which also does not mix with water.

Its almost the same like Oli and water, they do not mix together and need an emulsifier for this.

In cooking we use eggs etc. for example to mix fatty substances with water.

So the manufacturers know this and use machines to create an outer shell to the particles (called silanes) to make the particles compatible with other substances and let them be mixed together without separating after the mix. Otherwise if you mix the particles, after a minute they would float on top of the epoxy instead of having a homogenous mixture which does not separate.

I think it is important to explain this topic here, as it will be more easy for everyone to understand what is going on 🙂 

Regarding the nanoparticles, I bought these:

https://shop.plasmachem.com/titanium-oxide/138-592-titanium-dioxide-nanopowder-p25-grade-hydrophobized.html#/34-weight-50g

They cost 71 euros for 50 grams.

But 50 grams will at least be enough to make 100 big sized Rayleigh scattering epoxy sheets, so its very cheap at the end. 

Anyway I would send you some for free, but we have to solve the issue regarding a proper and not expensive tool to disperse the particles first. 

@nolo If you were interested in selling some premixed hardener nano particle solutions, you would have quite a bit of interest on here I'm sure. 

I found two inexpensive machines on the Alibaba website:

https://www.alibaba.com/product-detail/Laboratory-Ultrasonic-Cell-Disrupter-Ultrasound-Homogenizer_1600172163100.html?spm=a2700.pcfacebook.normalList.2.7e86XX0cXX0cQp

https://www.alibaba.com/product-detail/BIOBASE-Portable-Ultrasonic-Cell-Disruptor-for_1600094477476.html?spm=a2700.galleryofferlist.normal_offer.d_title.68615baf2iLvr0

The first one seems to be better as it has sound proofing box included.

I ordered a couple of times from Alibaba and the site is pretty similar to Aliexpress.

I had no problems to order as private person.

thats really impressive, been trying to do the same thing myself but not got results remotely as good as that. ive got as far as sourcing an ultrasonic probe and have some zinc oxide nanoparticles, but they're hydro and oeophilic rather than phobic and thats probably why im not getting anywhere, so ive ordered from the source you kindly linked to.

i was wondering if you had an tips you would mind sharing, like what was the approx ratio of np powder to the hardener? and does the thickness of the finished window matter? cheers in advance.

If I were able to make one of these Rayleigh scattering sheets, could I use it to create the sky mimicking effect on the recycled LED TV lighting project?  I am interested in a compact lighting solution. 

..

@diyperks No problem, was just wondering whats going on here 🙂 

@mxultra First you have to calculate how much Epoxy you need for your final plastic sheet.

Here is good calculator I use for this:

https://www.epodex.com/de/medien/epoxidharz-rechner/?gclid=CjwKCAjw3K2XBhAzEiwAmmgrAm_8HuHu1t5iurq8MWRU_6WmJ-LBEAGfLyJzcAd1a9fwCmn7NbLQGRoCPsQQAvD_BwE

For example, if I want to make a 50 cm x 50 cm sized sheet with a thickness of 1 cm, I would need 2750 grams of Epoxy (1375 grams is component A and 1375 grams is component B / hardener).

When your calculation is done you need approximately 0.01% of TiO2 Nanoparticles from the 2750 grams which would result in 0.275 grams.

You need a very fine scale for this which can measure under 1 grams. I got mine from Amazon for about 15 euros.

You need really very small amounts like this, which is a good thing. 50 grams would last you for your whole life.

You can then take for example the 0.275 grams of nanoparticles and mix them with the ultrasonic probe in 50 grams of hardener to get the nano particle dispersion.

After this is completed you mix the 50 gram nanodispersion with the rest of the hardener which in this case would be 1325 grams (1375 - 50 = 1325 grams).

Then mix the hardener with the 1375 gramms of component A of the epoxy. All the mixing can be done by hand with a stirring stick like a woodstick or anything you would like to use.

Because once you mixed the nanoparticles with the small part of the hardener via the ultrasonic probe the particles stick no longer together.

I recommend to make very small samples of 10x10 cm first to get a feeling for the process.

I did all tests like this by myself until I knew how its done and how much to use.

Also I have to mention that the amount can vary because for example if you have a LED with 6500K and you want the sunlight to have 4500K, then the amount of nanoparticles can be slightly different than if you want the resulting sunlight to be 3500K. The more nanoparticles you use, the more orange the sunlight will get and vice versa.

I hope you can benefit from this 🙂

@slojams Yes you can put the epoxy sheet with the nano particles in front of the TV. The epoxy sheet will scatter the blue light from the LEDs. The light will be more diffuse though because the light is not collimated as good as if you use a parabolic mirror etc. But I think it will work anyway.

I deleted my previous question about whether its possible to replicate a fresnel lens sheet by casting. It is. 

So if you have a fresnel sheet that you salvaged from a TV set, and would like to make copies of it, you can first make a mold of the surface of that sheet (both front and back surface if its a linear fresnel sheet), using something like casting silicone, and use that to make as many fresnel panels as you'd like.

If you use casting epoxy, you can add the titanium dioxide nanoparticles, and the panel will then function to both collimate and rayleigh scatter the light source.

Not that anybody seems to care, but in case someone does, there you go. I plan to clad my entire 9ft high ceiling with these sunlight-mimicking panels that can deliver 10,000 lux at 2 feet from the floor, that's why I'm trying to figure out how to cast my own panels.

@Nolo, do you think the TiO2 nanoparticles can also be dispersed in casting silicone or other resins besides epoxy?

How can Rayleigh scattering be utilized to create the amber hue of low sun?

@devfoxrocks This is a very good idea, as it would also save the costs of the fresnel lens and also saves weight because of the 2 in 1 solution. I have never made a mold myself, but have seen videos on Youtube where they use rubber / silicone to mold figures etc. If it is possible to mold fresnel lenses with those very fine lines, that would be awesome. Maybe you could make one first and show us the results 🙂 For your big project it absolutely makes sense.

Regarding the nanoparticles, it should work too to disperse them in silicone or other resins, because these are also hydrophobic materials.The TiO2 particles I linked to, should normally work for my understanding. Just the normal TiO2 particles, which are uncoated and hydrophile will not work, as they only mix with water.

I also recommend, that you use very high CRI LEDs for the project with a color temperature of 5700K. If you find high CRI LED's with 6500K it would be even better, because there is more blue spectrum in them which will make the TiO2 panel to look even more bluish and saturated.

I bought these LED's:

DIY LED U-Home High CRI Ra 97+ 100W 10000lm 28mm COB LED Ultra Bright High Light Flux Daylight White 5600K for DIY Flashlight Home Cinema Projector

https://www.amazon.de/gp/product/B095RN3NB5/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1

@Nolo

There are plenty of guides and Youtube videos on creating molds of diffraction grating sheets, and the texture of those seem to be on a much smaller scale than fresnel lens sheets, so it shouldn't be a problem at all to create molds of fresnel sheets.

The crafting community copy this diffraction grating texture onto casting silicone, epoxy, resins, gelatin/glycerin sheets, even bars of soap. For the purpose of creating a fresnel sheet, though, it seems the refractive index of the casting material needs to be considered.

By the way, just to clarify, like all fresnel lens sheets, the ones found in flat panel TVs have a texture of concentric rings on one side, which creates the fresnel lens effect, but the other side of the sheet will have a different texture. I believe they're linear lines, that somehow further collimates the backlight. I wasn't aware of this second texture, so I was confused as to how some people were sanding and polishing their fresnel sheets, yet the sheet would continue to have a fresnel lens effects. What they were doing was removing the linear lines of one side of the sheet in order to convert the sheet from a "linear" to a "spot" fresnel sheet, for the purpose of creating a more concentrated, hotter point of focused sunlight.

In deciding on a color temperature for a light source, I wonder- Because of Rayleigh scattering, with the color temperature of the Sun appearing yellow, and the sky light appears blue, then is the light that lands on surfaces a combination of the two, recombined to create the original color temperature and spectrum of sunlight as measured in space?

@nolo thanks so much for answering all of our questions in such detail. Vielen dank!

@ Devfoxrocks Yes, if the blue light from the sky and the yellow beams of the sun combined will shine on a surface, then it can be pretty similar to the color temperature of the sun itself. After my extensive research I think that the CCT of the sun is somewhere between 6000 and 6500K, but there are many different datas about it if you search on the internet.

But if you measure the sunlight at places where the sun shines from the Zenith to the earth with a very clean sky and the least amount of nanoparticles (at Zenith the sunlight has the shortest way and the least amount of atmosphere in the way) the Color temperature of sky and sunlight combined is around 6000K, almost pure white (which is 6500K), like on this picture

And also some of the blue spectrum is scattered outside of the atmosphere in the direction of space and does not reach the earth, thats why I think the sun is above 6000K CCT. But many websites say something around 5800-5900K. Daylight lights for film etc. are often in the 5600-5700K range, which also should mimic sunlight and skylight together, but I think its the intermediate Color temperature, because most of the time the sun is not at the Zenith right above our heads (90 degree angle) and therefore the sun normally goes through more atmosphere which makes the Color temperature of the sunlight lower and also the blue part and CCT of the sky vary, depending on factors like air pollution etc... There are many factors to consider. If someone goes above the atmosphere and measure the light of the sun with a spectrophotometer, then we will have a definitive answer 🙂 I have one picture, where we can see the sun sitting between 6000 and 7000K, looks like about 6400K

@slojams No problem ! Gern geschehen. 

BTW, yesterday I discovered, that the company Yuji LED Ligthing sells now artificial skylights with the blue rayleigh scattering plastic sheets. They begin at about 600 Dollar, which still is a lot of money. But that they are available now is a big surprise for me and I think it is a good alternative to Coelux skylights which cost way more. I think if they become cheaper, then it will be better to buy them instead of messing around with epoxy and Ultrasonic probes. It is a finished product and has everything (LEDs, Lenses, housing, power supply, frame etc.) we otherwise would have to buy to make our own. And I have also have to say, that it is not easy to get the epoxy sheets bubble free, it wil take some practice.

Anyway, here you can see the new skylights from Yuji LED:

https://www.yujilighting.com/skyline/

It comes also in various sizes.

I think its a good alternative to DIY and I will order one for myself to see how the quality is.

I will let you know what I think about it when I ordered and received it.

But if its not ok in this thread, I dont have to do it, because it may be a little bit Off topic.

You might know coelux from 7 year old youtube videos:

They are selling artificial skylights. I'm not sure, how they build it, but there are some drawings on this website:

https://www.360lamps.com/en-europe/products/coelux%C2%AE-45-hc

I guess they are using a lens in front of the light source, two mirrors and a (blue) scattering window.

If someone could lower the price, that would be a really cool thing for every basement. Your videos about artificial windows are perhaps also popular because our winters in northern Europe are very dark and this (but also because of Covid) causes many of us to suffer from winter depression.

Here some more models: https://www.360lamps.com/en-europe/search?type=article%2Cpage%2Cproduct&q=coelux*

I will also try to make the rayleigh scattering sheets with PMMA instead of epoxy. I discovered that it has many advantages and maybe I can get the sheets bubble free due to the very low viscosity of PMMA. Epoxy has a pretty high viscosity, which makes it difficult to get the trapped air bubbles completely out. 

This is the stuff I just discovered:

https://www.harzspezi.de/Acrylharz-SKresin-1702-unser-glasklares-Giessharz-mit-Haerter-2-x-1-kg-mit-2-x-10-g-Haerter

It is the same stuff which is used to create Plexiglass / acrylic glass in the industry.

And this particular product can be used without heating and has a quick curing time of about 1 hour instead of many hours to even days (epoxy).

As soon as I have time to test it, I will keep you updated.

And I think also to built a projector with the LED COB and a long focal length glass lens (to keep the divergence of the light beam low) instead of the huge parabolic mirror.

The projector can be much smaller and hidden in the room in a white unobstrusive case for example.

Then shoot the Light over a distance of about 3 meters through door with a frame and the rayleigh scattering sheet.

A door like this would be very nice I think:

If everything works out, I see three advantages:

- Small sized LED projector instead of huge parabolic mirror

- Acrylic glass sheet with nano particles instead of the glass / water / soap solution

- A prebuild room door looks nice and can be opened and closed

I think a good Biconvex lens for the projector is much better than a fresnel lens, because it has much less chromatic abberations.

I have much fun with the project and thanks again to Matt (DIY Perks), who first started with the idea.

There is really nothing better than collimated sunlight and the blue sky effect.

Any room can feel so much different with it.

If the finished project can look something like this, I would be very happy

I forgot, that it might be possible even to have no chromatic abberations at all, if an achromatic lens is used.

And the projector is much easier to build than a video projector, because the only thing it has to do, is to project the light from the LED COB through a lens to the wall. I have no experience with building one, but I think it should be doable. But the first thing is to produce a bubble free Rayleigh scattering PMMA sheet.