Cool tech: a switchable solar window

Demonstration device dynamically responds to sunlight by transforming from transparent to tinted while converting sunlight into electricity

Thermochromic windows capable of converting sunlight into electricity at a high efficiency have been developed by scientists at the U.S. Department of Energy’s National Renewable Energy Laboratory (NREL).

Relying on such advanced materials as perovskites and single-walled carbon nanotubes, the new technology responds to heat by transforming from transparent to tinted. As the window darkens, it generates electricity. The color change is driven by molecules (methylamine) that are reversibly absorbed into the device. When solar energy heats up the device, the molecules are driven out, and the device is darkened. When the sun is not shining, the device is cooled back down, and the molecules re-absorb into the window device, which then appears transparent. A video showing the device switch can be seen below.

The NREL-developed demonstration device allows an average of 68 percent of light in the visible portion of the solar spectrum to pass through when it’s in a transparent, or bleached, state. When the window changes color—a process that took about 3 minutes of illumination during testing—only 3 percent is allowed through the window. Existing solar window technologies are static, which means they are designed to harness a fraction of the sunlight without sacrificing too much visible light transmission needed for viewing or the comfort of building occupants. “There is a fundamental tradeoff between a good window and a good solar cell,” said Lance Wheeler, a scientist at NREL. “This technology bypasses that. We have a good solar cell when there’s lots of sunshine and we have a good window when there’s not.”

Lance Wheeler (front) developed a switchable photovoltaic window along with (from left) Nathan Neale, Robert Tenent, Jeffrey Blackburn, Elisa Miller, and David Moore. (Photo by Dennis Schroeder/NREL)

The proof-of-concept paper published in Nature Communications established a solar power conversion efficiency of 11.3 percent. “There are thermochromic technologies out there but nothing that actually converts that energy into electricity,” Wheeler said. He is the lead author of the paper, “Switchable Photovoltaic Windows Enabled by Reversible Photothermal Complex Dissociation from Methylammonium Lead Iodide.”

His co-authors, all from NREL, are David Moore, Rachelle Ihly, Noah Stanton, Elisa Miller, Robert Tenent, Jeffrey Blackburn, and Nathan Neale.

In testing under 1-sun illumination, the 1-square-centimeter demonstration device cycled through repeated transparent-tinted cycles, but the performance declined over the course of 20 cycles due to restructuring of the switchable layer. Ongoing research is focused on improving cycle stability.

The path to commercialization of the technology was explored last year during a two-month program called Energy I-Corps. Teams of researchers are paired with industry mentors to learn what customers want of the technology and develop viable ways to reach the marketplace. Lance Wheeler and Robert Tenent, the program lead for window technology at NREL and co-author on the paper, teamed up to develop a market strategy for a product they called SwitchGlaze. The effort was funded by the Emerging Technologies program within the Department of Energy’s Building Technologies Office.

Wheeler said the technology could be integrated into vehicles, buildings, and beyond. The electricity generated by the solar cell window could charge batteries to power smartphones or on-board electronics such as fans, rain sensors, and motors that would open or close the windows as programmed.

NREL is the U.S. Department of Energy’s primary national laboratory for renewable energy and energy efficiency research and development. NREL is operated for the Energy Department by The Alliance for Sustainable Energy, LLC.


105 thoughts on “Cool tech: a switchable solar window

  1. Can’t wait. Technology looks very interesting. I hope that it translates to relatively inexpensive commercial use, easily.

    • If it was being done with other than exotic materials, I would say it would be interesting. Exotic materials don’t usually “get cheap.” This sounds more like research being funded by a special grant for a special purpose, and released in a PR release so as to gain enough interest to get the grant renewed, extended, or enlarged. I can’t see anything practical about this in the real world of the average human being. Maybe the new robotic civilization that every one keeps talking about will have a use for it.

      • The transistor of 70 years ago might have been thought of in the same way, until it wasn’t.
        Also, consider that things change and sometimes a second or third idea is necessary for the first one to become really useful. When the Aluminum cap was made for the Washington Monument the metal was rare and expensive. Now it is not.
        50 years from now this “cool tech” may be in common use, or forgotten.

      • “John F. Hultquist on November 28, 2017 at 1:23 pm
        The transistor of 70 years ago might have been thought of in the same way, until it wasn’t.
        Also, consider that things change and sometimes a second or third idea is necessary for the first one to become really useful.”

        You don’t know how right you are as it was the planar process developed by Jean Hoerni (the second most important invention of the 21st century) that made modern day integrated chips possible.

      • re: “Exotic materials don’t usually ‘get cheap.'”

        While true as a general rule, that’s because rare earths are, well, rare. In this case, the “exotic materials” appear to be composed of common elements. That makes their production an engineering problem, not a prospecting problem. And engineering problems are generally susceptible to efficiency solutions. No guarantees, of course, but it’s not implausible.

    • Sounds like a good scheme.

      I’m working on an insulation that will obviously insulate the warm interior of a dwelling when it is cold outside -and when it is hot outside, insulate the air conditioned interior. However in transition zones where the house interior has cooled but it is warmer outside, it will allow the heat to enter until the desired interior temp is reached – where it will insulate. The opposite when the dwelling is over warm inside but the temperature has dropped outside.
      Will be great for climates with extreme temps during day and nights.



    • Gmak, you don’t have to wait, there’s a low-tech solution. Just use passive solar design for your next house. It costs nothing and it works. My wife and I designed and built a passive solar mud-brick house with tile floors and a carefully-designed wrap-around verandah, and with slow-combustion (wood fuel) heaters and geothermal under-floor heating for winter, and it was great to live in. Black solar roof panels (not PV) provided hot water. Deciduous trees and a grape vine on the equator side of the house were helpful too without blocking the view. Mud bricks provide both insulation and thermal mass and look great. Double brick is as good but more expensive. Brick veneer is perfectly good if you reverse it (bricks on the inside) and put in good insulation. There’s no airconditioning for summer, and it isn’t needed. When you walk into the house in summertime it feels like there’s aircon. It’s at 34S so summer days are hot outside, but evenings are cool because it’s well above sea level, so you just open the windows on a summer evening and close them in the morning. At low altitudes you would need a sea breeze or aircon in summer. Even so, I suspect that the low-tech would be just as good as the high-tech windows, be more fun, and save a bucket-load of money.

      • Mike, most of us have little or no opportunity to build a house, let alone a “next” house. We have to deal with the features and potentials of the house we have. But I’d sure love to have a house like you describe. I live in a small log cabin in the NE corner of Wyoming.

    • I think its fantastic. As the windows darken we can use the generated electricity to drive the light-bulbs inside the room!.

      Sheer genius!

      Next is a wood burner powered domestic generator that will enable you to run an electric fire and an electric cooker!

      And finally by burning diesel in power stations, we will generate enough electricity to run electric cars!

      I am giddy with excitement at all this new green tech!

    • I thought about this 20 years ago; my limited imagination came up with one good use.

      Put it on car windows … run it backwards when car running to keep windows clear … turn off car and windows “tint”.

      Tell people they will get 0.4 mpg less to have it as an option and almost buyers would want the option.

      • DonM, they have been looking at this technology for many applications, but it seems to fail the longevity/practicality threshold. Then there is the fail safe issue. If you require power to keep your windows transparent, any power loss will cause the windows to black out, which would be an issue during driving.
        As I recall there was a sexy looking sports car during the ’70s that had almost everything electrically powered, doors. windows, etc. A minor fender-bender that caused the battery to disconnect would trap the occupants, as could not exit the vehicle.

      • Rocket, that’s exactly what happens time after time when a car gets washed off the road. The car gets upside down in the flood water, and the windows can’t be wound down manually.

      • “Greg Cavanagh November 28, 2017 at 1:24 pm”

        No. That’s differential pressure and it matters not wether the car is upside down or not.

      • Patrick; I’m actually referring to electric window not being usable in a submerged vehicle. A mechanical window works underwater. The second part of Rocket’s post was about the electrics trapping people in cars.

    • Clear glass has a transmittance of 80-90% depending on the exact composition. General purpose sunglasses have a transmittance in the 20-40% range. Special purpose sunglasses (such as might be used in high-glare environments like on glaciers and at sea) go below 20% – sometimes well below.

      So, yes, 3% transmittance is low. But not implausibly low if you lived in Arizona or other very bright environments.

      My question is whether it is a binary switch. That is, is the choice 68% or 3% OR does this material exhibit graduated darkening? Granted, you probably wouldn’t get 11.3% efficiency at a lesser darkening. But it’s an interesting curiousity.

  2. Am I suffering from a lack of imagination? What is the point of a window that is only transparent when it is dark out? That is when people close the curtains.


    • Guess that people want to be in the dark all the time.

    • On 2nd thought, this might pave the way for a sun visor that appeared in Larry Niven’s Known Space series. The visor would only darken in the spot directly in the line of sight between a too-bright sun and the wearer’s eye.


      • I would like to have active goggles using for example lcd technique that would darken the head lights of cars driving towards me but only the headlight. The dark vision would be unimpaired … Of course the wind screen of a car could also do the same trick using a camera as sensor and a computer calculating the are to darken when the position of the driver’s eyes are known.

      • You’re only allowed to tint the windshield above the marking line on the window (AS-1) which is mostly above normal line of sight when driving. Assuming same percentage as side windows which is 35% transmission.

      • Lars, you should have talked with Edwin Land.
        He unsuccessfully tried to sell major automakers on polarized headlight lenses and windshields. They strung him along for 20 years but dropped the idea in the late ’40s, supposedly because of a lack of an economic incentive. By then he had a lot of investors and had created the Polariod Corporation.
        It was only after that that he turned to sunglasses and “instant” film.

    • Especially since for automotive applications you have to allow at least 35% light through the side and rear windows (at least here in North Carolina). So no blocking of the sun through the windows to prevent heating of the interior.

    • SR – I also wondered about that. Am I missing something here? I want my windows to be transparent during the day. Isn’t transpatrency the whole point of having a glass window in the first place – so that you can look out of it and so that natural light from outside can come into the room? And if you have to turn on the lights in the room because you aren’t getting enough natural light coming through the window, isn’t the light bulb in your lamp going to consume more electricity than the window glass produces? I’m so confused!!

      • Don’t double windows with a vacuum or inert gas in between work fairly well? I don’t see this window changing solution as viable when placed against the current technology. At least not now.

      • There have actually been test done on the effect of different gasses within double glazing

        They found that CO2 provided LESS insulation than normal air.

        CO2 TRANSFERS energy better than normal air.

    • Or the smart iWear app that darkens the glasses so you can’t send stupid email/blog/txt/sms/twitter messages?

    • I had a pair of prescription eye glasses that tinted up like sunglasses when it was bright outside and that became clear when I went inside. Of course they didn’t produce electricity. If they had produced electricity my hair would have always been standing on end.

      • Actually you’d have to keep these things away from your eyes because of the leakage of MeNH2. It stinks so you’d have to keep them away from nose too. Sounds almost as stupid as the magic CO2 to fuel thingy.

  3. How about car windows that charge the battery and are opaque – Buffet would like that (Geico).

      • Steve Ta,
        Here in the Soviet republic of Scotland, the law requires you to have clear(ish) car Windows.
        This has nothing to do with the driver being able to see out – its to do with the millions of spy cameras and police being able to see in and persecute motorists.

      • Steve Ta

        Agreed re tinting car windows. In many jurisdictions the maximum permissible tint is 15%.

      • police being able to see in and persecute motorists.

        I thought you meant prosecute, but then I decided you probably had it aright.

  4. “The proof-of-concept paper … established a solar power conversion efficiency of 11.3 percent…..but the performance declined over the course of 20 cycle” More models. Hope the real thing works out.

    • After 20 cycles does it stop producing energy as well or just fade to black? In which case it wouldn’t even be a good solar cell, as those last about 20 years.

  5. to learn what customers want of the technology and develop viable ways to reach the marketplace.

    How about a product that lasts more than “20 cycles?”

    Announcing a product before it is viable is no different than algore screaming from his private jet how “all us little people are destroying the planet”.

    Finish the research and get back to us.

  6. 11.3% efficiency? What’s current solar cells? Credit where credit is due, Progress is progress. Lots of room for improvement.

    • Best solar cell to date are about 22% with some promising as much as 28%.
      As solar cells go these are poor performers.
      But….they only last 3 weeks.

      Heck, they may be crappy, and they will be expensive, but they won’t last long….[facepalm]

      • I must have misread the opening paragraph “Thermochromic windows capable of converting sunlight into electricity at a high efficiency have been developed by scientists at the U.S. Department of Energy’s National Renewable Energy Laboratory (NREL).” I thought “…converting sunlight into electricity at a high efficiency” meant some kind of breakthrough in solar conversion technology. Guess not.

    • Well NREL itself reports already efficiencies of four times that for triple junction; triple band-gap cells. And Nakamura’s group at USB are working on a better top layer for that to raise the record number.


  7. Trying to envision where such a product would have a market to make mass production profitable for a company. Not seeing it.

    Toyota Prius models a few years had an option with a solar PV roof panel that ran an internal cooling fan to circulate outside air in order to keep interior temps down when the car was parked in the sun. No longer offered. What does that say about market demand?

    The whole solar power PV thing marketed as a consumer good is just virtue signaling IMO.

    I’m not dissing PV research. Solar PV research is needed though as solid applications for space vehicles and off-grid applications will benefit from higher efficiency and longer life PV’s.

    • Haven’t checked the 2018 models but that option was still available for the 2017 Prius.

      It was limited by geography, though, and I don’t know why. Maybe it’s just not available in your region? It’s one of the favorite features of my 2015 and it runs pretty much all summer even here in NE Ohio.

  8. The photo darkening technology has been around for decades…with the same results. The ability to change degrades with every cycle. Eyewear has been using this for decades. Eventually the lenses fade to completely dark and won’t revert to transparent.

    “When is a window not a window?”
    “When you can no longer see through it and it becomes an expensive wall.”

    “But what good is window that isn’t a window.

    • So now in the summer time with the sun out, the windows darken and I’m turning the lights on in the middle of the day so that I can see while inside? Almost sounds like a solution looking for a problem imo.

      • California’s energy code requires “day-light harvesting” as part of the building design. Not seeing how this could comply, even with infinite life and higher efficiency.

  9. Too bad more climate research grants weren’t used on more research in advanced carbon materials instead.

  10. Interesting.

    “When solar energy heats up the device, the molecules are driven out, and the device is darkened.” What outside conditions are required to result in sufficient heat? Here in Canada, it is often sunny (and you would like to generate solar electricity), but it is not very warm (sometimes downright cold).

    • I see where those 20 cycles set the limit: “””””….. “When solar energy heats up the device, the molecules are driven out, and the device is darkened.” …..”””””

      Eventually you lose so much glass molecules it just shatters.

      g (G would never buy a self eradicating product)

  11. You can get a pedal machine that charges batteries, watch tv , keep fit and charge a battery, makes more sense.

    • Only one small flaw in that system. All that exercise will increase your respiration resulting in a measurable increase in the CO2 you’ll be releasing into the atmosphere. It would be far more ecologically responsible to park your butt on the couch and read a good book by sunlight of course. Oh, and no beer – each bubble is pure CO2 dooming baby polar bears to slow death.

      • hardly. a few billion more
        people will make no difference
        at all — there have been trillions
        of animals breathing on the planet
        throughout the holocene, while
        atmo co2 stayed nearly constant
        until we started digging up
        and burning fossil fuels.

  12. Having been involved in advanced energy systems (solar cells, energy storage materials) since the mid 1990’s, I can say with some certainty that this is one of hundreds of NREL announcements, almost always around congressional budget time, that ultimately amounts to nothing. Their several other PV spinouts all failed within a couple of years over similar issues, or cost.

  13. Heat energy causes the activation, “When solar energy heats up the device, the molecules are driven out, and the device is darkened.” So, on a cold winter night the interior surface will be warmed by the structure’s heating system. Will this cause a partial or full cycle to take place thereby diminishing the number of cycles (a whopping 20 in all ) by one when there’s no sunlight to generate the electricity? I wonder if a partially cloudy day with a strong wind driving the cloud cover through rapid alternating periods of full sun and partial shade will impact its ‘life expectancy’? One thing for sure, with so short a life span they will be renewable – by replacement over and over again. Sounds like it would be perfect for most government buildings and contractors.

  14. Here come the politicos with startup venture firms. At least they will not have Uncle O backing them.

  15. In 1986 (I may be off a couple of years) Alvin Marks, the inventor of Polaroid Lenses developed Lumiloid.
    It utilized the same principles as polaroid but diferent mateials and a much finer line spacing. Lumiloid achieved 80 to 85% photovoltic efficiency. It has since disappeared down a rabit hole and if I were a serious conspiracy nut I am sure someone doesn’t want it brought to market. Google Lumiloid!!

    • This has NOT “disappeared down a rabit hole”, and it works but cannot still be produce as a decent marketable device, and is still been researched.
      duckduckgo “Optical rectenna” or “Solar Nantenna Electromagnetic Collectors”.
      It could even harness energy from IR, meaning direct convertion of heat into electricity (There are even claims it could be turned into an aircon device that produce energy instead of consuming it, but this collides with 2nd law, and would need some sort of diode working like a Maxwell’s demon as far as I understand it).

  16. They must overcome the degradation of the film resulting from switching . . . maybe they can do it, maybe not.

    11% PV efficiency is not bad for a window, particularly large office buildings that are most glass exteriors.

    There is an energy advantage just from the shading (darkening) effect alone in warmer sunny climates like Florida and California — blocking heat gain from semi-tropical sun.

  17. Not very impressive all around. Conversion efficiency is less than a solar panel and the window
    area on a car that would receive appreciable sunlight is quite small. The device also has no lifespan
    of any consequence. An office building would have a lot of window area and, if these things have any practical application, this would seem to be it.

  18. [Comment by Rocket Scientist]
    “The photo darkening technology has been around for decades…with the same results. The ability to change degrades with every cycle. Eyewear has been using this for decades. Eventually the lenses fade to completely dark and won’t revert to transparent.”

    Not true. I have glasses that darken to sunglasses when outdoors in sunlight, then become transparent when not exposed to ultraviolet light (indoors or outdoors at night or on a cloudy day). I have had them for over a year, which means that they have gone through several hundred darkening / lightening cycles without losing their effectiveness.

    The difference between my glasses and the National Renewable Energy Laboratory’s window is probably in the mechanism used to darken the lens, which in NREL’s case is used to generate electricity, but my glasses are only intended to improve my vision, not generate electricity.

    The technological challenge is not in making the glass darken or lighten, but in enabling it to generate electricity when sunlit and be transparent in the shade.

    Still, I agree with other commenters that a 21-cycle lifetime is much too short to be practical.

      • For years I needed glasses stronger by 1 diopter. I reached the end of my patience and stayed with my current prescription. Twenty years later, I still need glasses stronger by 1 diopter.

      • Rhoda R
        November 28, 2017 at 1:46 pm

        Your eyes were stable for five years?! Wow. My prescription changes every year or so.
        Curious George
        November 28, 2017 at 5:09 pm

        For years I needed glasses stronger by 1 diopter. I reached the end of my patience and stayed with my current prescription. Twenty years later, I still need glasses stronger by 1 diopter.

        Your eyes are controlled by muscles, they weaken with age unless exercised.
        I, and others had started to notice in my early forties that reading was getting more difficult (people were teasing me about my arms not being long enough) and recalled reading a method for training the eyes that I had read twenty years before. By chance, I saw the same booklet a few months after, bought it, read it again and have never had to wear glasses. I am now nearly seventy.

        The book was about the Bates method, written by him (as it happens it is a difficult read – brilliant insight, but not a writer – thankfully a booklet, not a huge tome).

        Obviously this will not work if there is a disease affecting the eye (have a checkup), but as noted by many, anybody connected with the ophthalmic trade will tell you it is rubbish (no conflict of interest there!)

        Once the principle is understood the exercises can be done anytime anywhere, in my case usually in odd moments when waiting for a bus etc. These days I am not always aware of doing them and they take no planning or time from other activities.
        The theory behind it provides the explanation for the gradual deterioration year by year described above.

        Good luck


  19. With a little smart tech, it could recharge your electric car’s battery while it’s parked for long periods.

    • If you need to charge your car battery every time you park it you need a new battery.
      We already have reliable inexpensive technology to accomplish this called a trickle charger.

      This is just a reinvented wheel, only they managed to make it more expensive and less reliable.

      • Not so. You charge your battery whenever the engine is running. But, given the total area available for this “solar cell”, it would need some 200 hours of sunshine to completely charge a discharged battery.

      • How much of the electricity generated by the alternator get stored in your battery during driving depends on how many systems on your vehicle are drawing upon this energy. Mostly the battery acts as simple capacitance to the circuit providing spike buffering. Once it is completely charged all the excess current merely flows through.
        If it is loosing power during when not operating it is because of continuous load from alarms, clocks, etc. If you are finding your car battery drained of power after even for a short time as a week, you have a battery issue or a short circuit in your system.

  20. The thinking is all wrong. It’s not windows that generate electricity we want but bricks, concrete and cladding electro-panels.

    • And you Sir, are a wise person indeed. The use of solar PV for NON transparent building materials would make way more sense.
      A brick, which is able to store ‘heat’ and produce energy, might be an ideal combination….don’t you think?
      Mind you, the connections for the thousands of bricks may become a technology limiting issue, but I’m sure that would or could be overcome – but I’m not planning to build my house around any of these technologies, just yet.

  21. The only apparent worthwhile use for energy production from ‘glass’ is in domestic/commercial or industrial windows.
    There has to be an area sufficient to produce energy to pay for the cost added to the ‘window’.
    This likely use would be limited for several simple reasons…. not all windows face the sun….not all windows would want to be ‘tinted’, for practical and aesthetic reasons….but mainly, as normal glass is in the range of 80 to 90+ % transparent, any reduction , like down to 60% as this glass is said to be in its transparent state, would not be acceptable to most occupants/owners of the buildings.
    As a builder/specifier in that industry, I couldn’t and wouldn’t include this feature in any of my buildings. IF I wanted to have sufficient energy production at a ‘reasonable’ price, I would still want a large area which is non-intrusive to the structure and which is specifically designed to produce energy and at as high an efficiency level as is possible, not hamstrung by the need to do another job – be transparent.
    Even that said, I doubt I would have the economic justification to add solar PV anyway, if connected to an easily accessible power supply – ie, grid electricity

  22. “but the performance declined over the course of 20 cycles due to restructuring of the switchable layer”

    Sorry to sound cynical but sort that little glitch out guys and get back to us. It kind undermines the ‘renewable’ thing. Sounds more like ‘disposable’ like a plastic water bottle. I reckon about 20 refils is their limit too.

  23. “but the performance declined over the course of 20 cycles ”

    In engineering speak, it does not work.

    Old and boring is low e glass and proper over hand is how you do it. Direct sunlight provides heating in the winter and blocked in the summer.

    The purpose of is to enjoy the view.

  24. i buy asian items and when can USA Make something without costing to daen much to buy plus AMERICANS items are junk like old Japan items were

  25. This is yet another addition to the long list of nutty ecological schemes that are totally mis-directed. Windows are there to look out of, and not only just at night. Windows that go black to produce electricity are better replaced with windows that you can look out of during the day and solar cells on the roof directed at the sun average location. Whether having solar cells to produce electricity is quite another matter for argument.

  26. The bottom line is how much does it cost? Glass is a very significant cost in a large multi storey building. This electricity generating glass will obviously add a considerable amount to that cost. It would be a complete waste of money if the same amount of power could be generated by everyday cheap solar panels at a fraction of the cost.

  27. I won’t buy this stupid device, and don’t see anyone who would.
    Most windows are vertical, and that’s a terrible orientation for a solar device. Sunshade or curtains do better.
    It could make sense to collect the light into electricity when it is not needed in the building, but this must be in human/building control, not heat/sun control, and there’s no need of this 2-in-1 device when a cheap and more efficient second device can be use with a normal window (either inside or outside).

Comments are closed.