Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Power

Stanford Engineers Invent a Solar Panel That Generates Electricity At Night (interestingengineering.com) 72

Research Sid Assawaworrarit and his colleagues have outfitted an ordinary solar panel with a thermoelectric generator, capable of generating a small amount of electricity from the slight difference in temperature between the ambient air and the surface of a solar panel pointed deep into space. Interesting Engineering reports: "During the day, there's a light coming in from the Sun and hitting the solar cell, but during the night, something of a reverse happens," Assawaworrarit says. That's because solar panels -- like everything warmer than absolute zero -- emit infrared radiation. "There's actually light going out [from the solar panel], and we use that to generate electricity at night. The photons going out into the night sky actually cool down the solar cell," he says.

As those photons leave the skyward surface of the solar panel, they cary heat with them. That means that on a clear night -- when there are no clouds to reflect infrared light back toward the Earth -- the surface of a solar panel will be a few degrees cooler than the air around it. That temperature differential is what Assawaworrarit and his colleagues are taking advantage of. A device called a thermoelectric generator can capture some of the heat flowing from the warmer air to the cooler solar panel and convert it into electricity.

On a clear night, the device Assawaworrarit tested on the Stanford rooftop generates roughly fifty milliwatts for every square meter of solar panel (50 mW/m2). "I think that's probably a record number," he says. But Assawaworrarit and his team aren't stopping there. He says that with a couple of improvements (and in a good location) such a device could generate twice that amount of electricity. "The theoretical limit is probably about one or two watts per square meter," he says. "That's not a huge number, but there are a lot of applications" where that kind of energy at night would come in handy.

This discussion has been archived. No new comments can be posted.

Stanford Engineers Invent a Solar Panel That Generates Electricity At Night

Comments Filter:
  • While the amount generated won't be enough to power a torch, it hints at being able to provide enough (when spread over a large enough array of panels) to provide sufficient electricity for any base load monitoring systems.

    50mW /m2 means that a small array of 100 panels could provide enough power for a phone

    • For comparison (Score:4, Informative)

      by Vario ( 120611 ) on Wednesday April 06, 2022 @02:22AM (#62421610)

      This might help in applications where battery storage is hard to implement.

      Still, a good solar panel averages at 150 W/m2. So this provides only 0.0003 times the output power during the day. And thermoelectric materials don't come for free, so I don't know if this will ever break even with the energy that was necessary to make the thermoelectric materials in the first place.

      • Re:For comparison (Score:4, Insightful)

        by jlar ( 584848 ) on Wednesday April 06, 2022 @02:52AM (#62421636)

        This might help in applications where battery storage is hard to implement.

        But you will only need something in the order of the size of a mobile phone battery per 100 m2 of solar panels in order to store the same amount of electricity from the day to be used during night. I don't think it will be easy to find an application where it makes more sense to use this type of solar panels (which I assume will have a certain extra cost and possibly reduced day time efficiency compared to regular solar panels) instead of just having a small battery pack to provide power at night (even in cloudy conditions regular solar panels will provide much more power than this).

        But maybe there are some applications?

        • I can imagine a couple uses, but they are out on the edge. The clearest opportunity is going to be in remote locations for monitoring of environmental conditions (temp, pressure, whatever). If the main goal is to collect the data over years, then a battery could be an issue. With this new panel, you could collect the data at night with 50mW, then when the sun comes up and you're getting 100W+, you've now got the power to transmit all the recorded data. Although I don't see much of a market. I think most rem
          • Actually I changed my mind. 50mW is only 3Ah. If you want it to run for 12 hours, you only need a 36Ah battery (plus some overhead). A 40Ah SLA battery is ~$150, so most monitoring stations that has 1m^2 of solar panels can easily accommodate a battery too. The only thing I can think of now, is very cold locations where most batteries have difficulty operating, maybe at the top of a mountain? but thats a stretch..
            • Also lifespan. But solid state cells promise to solve that, by providing batteries with a working life of decades rather than years. If they ever make it into large-scale production.

          • A problem with using this to power sensors is that it doesn't work on cloudy nights.

      • Even if the reliability and longevity of batteries are a problem, at these powers probably even supercapacitors are competetive.

        • You could use super caps for this. Today you can get a 2.7V 3000F capacitor for about $100. At those numbers, the capacitor has approximately 3Ah, which is about what this new tech can generate. So you could use caps, and you'd only need 1 per hours of darkness. BUT, for ~$150 you could get a single lead acid battery with enough capacity for 8hrs.
      • by zazzel ( 98233 ) on Wednesday April 06, 2022 @04:44AM (#62421732)

        So, for a typically large 10kWp rooftop installation (about 60 m^2), this would generate a baseload of... 3 W. Or the equivalent of 3-4 USD of electricity per year, before subtracting daytimes and accounting for adverse effects on daytime solar efficiency.

        It's one of these situations where amortization will never happen.

        • For a town with 10000 rooftop installations, that's 30kW of extra free juice during the night.
          • Re: (Score:3, Informative)

            by jlar ( 584848 )

            For a town with 10000 rooftop installations, that's 30kW of extra free juice during the night.

            Free? Installing this on 600,000 m^2 solar panels is not free. The amount of money used for this would probably be better used to buy 60.018 m^2 per installation instead of 60 m^2 per installation and use the extra electricity to charge a small cell phone size battery which can then power "something" during nighttime. Not that I find this very useful either. The best option would in fact be to just save the money or use it to save electricity in another manner (if you want to go greener).

          • For a town with 10000 rooftop installations, that's 30kW of extra free juice during the night.

            But less power during the day by many times that amount. Since daytime power is worth more, you lose.

            The power during the day is less because the Seebeck effect slows the flow of heat out of the panel, and hot panels produce less power.

          • by kenh ( 9056 )

            You are getting lost in the math.

            10,000 10 KW home solar installations would generate an "extra" 30 KW A YEAR, spread out over all 365 days of the year.

            Adding this technology to 10,000 sixty square meter solar arrays would be far from cheap, certainly not "free", and a much easier way to that "extra" 30 KW would be to simply add one more 10 KW panel array to the city and you'll generate that same 30 KW of electricity in less than a day.

        • could be almost enough to run that dim nightlight you just didn't have enough power to run before. probably makes that nightlight cost thousands more but hey at least you aren't paying for that 3W load.
        • Like most solar, amortization never happens. My boss recently spent $50k on solar for his house, and the math said he would start making money in a little over forty years. He wonâ(TM)t live that much longer nor will his solar batteries. It is a scam.

          • by nasch ( 598556 )

            40 years?? Are his panels shaded by trees most of the day or something? That is several times longer than typical, from what I've heard.

      • Well,
        that is because you lack imagination (or knowledge).
        What about solar panels in Alaska or any arctic region?

        There are pretty cold nights and pretty long periods polar winter.

        Or on the Moon?

        Panels that at least can keep some rudimentary computer and bluetooth or similar connectivity alive, might be useful.

        • by jlar ( 584848 )

          Well,
          that is because you lack imagination (or knowledge).
          What about solar panels in Alaska or any arctic region?

          There are pretty cold nights and pretty long periods polar winter.

          Or on the Moon?

          Panels that at least can keep some rudimentary computer and bluetooth or similar connectivity alive, might be useful.

          On the Moon? Did you see the amount of electricity that this generates? NASA's radiative isotope General Purpose Heat Source units each has a size of 4x4x2 inches, weighs 1.5 kg and produce 250 W. If we assume a production similar to reported numbers you would need solar panels covering 5,000 m^2 to produce that. Good luck with that.

          Same goes to a lesser degree in the Arctic regions and on Alaska. There are simply much more efficient ways to produce that power.

          • We did no talk about amount of electricity.
            We talked about "where, when, why" it might be a useful thing.

            Same goes to a lesser degree in the Arctic regions and on Alaska. There are simply much more efficient ways to produce that power.
            At night? From a solar panel?
            I do not think so.

            • > We did no talk about amount of electricity.
              > We talked about "where, when, why" it might be a useful thing.

              And there is no where, no when that it makes sense to use several square meters of panels to get less power than than an AA battery.

              > > There are simply much more efficient ways to produce that power.
              > At night? From a solar panel?

              That's what we're telling you - from a solar panel is bad way. A terrible way, actually. Better to use the shake light. :)

              • And there is no where, no when that it makes sense to use several square meters of panels to get less power than than an AA battery.

                Well, you get the main power during sunlight, or did I miss anything?

                That's what we're telling you - from a solar panel is bad way. A terrible way, actually. Better to use the shake light. :)
                A battery you do not have, but a solar panel you have right there are two different things, or not?

                Why is everyone on /. always so negative about news releases?

                WTF: no one is selling it as

                • > A battery you do not have, but a solar panel you have right there are two different things, or not?

                  Tell ya what, let's try it out and see.
                  You get your solar panels redone with this special coating, I'll grab a AA battery from 7-11. We'll see which approach is gets it done quicker.

                  • Gets done quicker what? :P

                    You think your AA battery works fine at -30C at night? The panel might ...

                    It is new tech, you should be excited instead of being a nay sayer ...

          • "Or on the Moon?"
            This uses the difference between the solar panel surface and the surrounding air. On Moon you have no air. You could implement a kind of reversible Stirling cycle using a ground-based reservoir and outside heat or cold (you heat a working fluid in an outside thing during day and use it as hot source of a Stirling engine, the cold source being deep beneath lunar surface. At night, you use the cold outside source to power the same engine, but with the "hot" source the inside of the Moon.

            It wi

          • by kenh ( 9056 )

            If we assume a production similar to reported numbers you would need solar panels covering 5,000 m^2 to produce that. Good luck with that.

            I can't imagine the number of lunar launches it would require to send a 5KM^2 solar panel array to the moon, let alone the food, water and oxygen needed to sustain the crew installing the array.

            LOL

        • by Vario ( 120611 )

          Well, in Alaska or the artic region these panels would work even worse.

          It is a thermoelectric generator, so it requires a Delta T to operate. Here the cold side is the open sky, the warm side is the panel and ground underneath. If you reduce this temperature difference by going into a cold climate the efficiency drops rapidly. It might provide a slightly higher benefit in the Sahara desert. But as others here calculated even a tiny battery would provide more power than this device.

          • The temperature difference in Alaska polar night, versus Sahara night, is more or less the same.

            And a battery needs to be loaded. You can not load a battery during poler night in winter from a solar panel to replace the missing Sun ... that is plain obvious, or not?

            • by kenh ( 9056 )

              Stop, it's a science fair trick at this point, little more.

              A square meter solar panel can generate 50 mW at night, if you add this technology to it, by exploiting a temperature differential between the surface of a solar panel and the surrounding atmosphere.

              During an extended dark period, the panel would not be exposed to any sunlight, meaning, among other things, the panel wouldn't get warmed up during the day (no sunlight), no 150W of electricity/hour, all you'd have is 50 mW peak, which would likely dimi

              • Agree with you 100% - but would also like to add that they are hiding the cost of the batteries, charge controllers, etc. by paring it with solar - you could get this energy with anything that has a temperature differential but then the true costs would become visible.
              • Sorry,

                are you really that dumb?

                If it is a rooftop solar installation, as 99% in Germany are, of course it gets warmed up.

                No idea what is wrong with /. idiots nitpicking about every bullshit, and then being completely wrong with their stupid nitpicking

                • by kenh ( 9056 )

                  We're talking about exploiting this tech in arctic regions where "long nights" can be quite long indeed, up to three months [timeanddate.com]...

            • by nasch ( 598556 )

              The temperature difference in Alaska polar night, versus Sahara night, is more or less the same.

              Sahara night time temperature: -4 C

              https://www.indiatimes.com/tre... [indiatimes.com]

              Northern Alaska night time temperature in winter: -20 to -30 C

              https://en.wikipedia.org/wiki/... [wikipedia.org]

              May-September Fort Yukon is actually warmer at night, but that leaves the rest of the year.

              • Yes, and regarding the application of those solar panels: -4c and -20 to -30 is: ta ta more or less the same. (* facepalm *)

                Or do you think it is a majour difference between 250ÂK and 268ÂK ??

      • There is proof of concept, and then there is proof of delusion.

    • While the amount generated won't be enough to power a torch, ...

      Too bad, they could have used it to shine light on the solar cells and -- hmm ...

    • Agreed, this, as it stands is of limited utility.

      Thinking bigger, if you put pipes on the back of the panel, then you might be onto something. During the day, those pipes can wick away the heat in the panels to heat water (or the concrete base under your back patio). During the night, those pipes can wick away the heat in your bedroom (via a radiant heat system in the ceiling) and send the heat into outer space (and give you slightly more electricity in the process). If you've got something permanently hot,

      • you could just be sending your heat to outer space all night, while making some electricity to run your computers.

        By putting the thermoelectric generator between the heat and cold, you are slowing the heat transfer. So you would harvest a tiny amount of energy at the cost of far less cooling of your datacenter.

  • by Anonymouse Cowtard ( 6211666 ) on Wednesday April 06, 2022 @03:11AM (#62421652) Homepage

    I wear my sunglasses at night [youtube.com] I wear my sunglasses at night So I can, so I can Watch you weave then breathe your story lines And I wear my sunglasses at night So I can, so I can Keep track of the visions in my eyes While, she's deceiving me It cuts my security Has she got control of me? I turn to her and say Don't switch the blade on the guy in shades, oh no Don't masquerade with the guy in shades, oh no I can't believe it 'Cause you've got it made with the guy in shades, oh no I wear my sunglasses at night So I can, so I can Forget my name while you collect your claim And I wear my sunglasses at night So I can, so I can See the light that's right before my eyes While she's deceiving me She cuts my security Has she got control of me? I turn to her and say Don't switch the blade on the guy in shades, oh no Don't masquerade with the guy in shades, oh no I can't believe it Don't be afraid of the guy in shades, oh no It can't escape you 'Cause you got it made with the guy in shades, oh no Oh no I say I wear my sunglasses at night I wear my sunglasses at night I wear my sunglasses at night I say it to you now I wear my sunglasses at night I wear my sunglasses at night I wear my sunglasses at night I cry to you I wear my sunglasses at night I wear my sunglasses

  • What does it cost? (Score:4, Insightful)

    by mrwireless ( 1056688 ) on Wednesday April 06, 2022 @03:11AM (#62421654)

    What does adding a thermos-electric generator to panels cost? The power generated is so miniscule that it sound to me like it will never break-even.

    If this entire system might just be able to charge a mobile phone during the night, then it sounds cheaper to just charge a $3 battery cell during the day.

  • The simplest method was discovered some years ago in Spain. You just hook up the right kind of mains powered adaptor to the output of your solar panels. You are then credited at the feed-in tariff rate for everything you put out over the system.

    Enterprising fellows took this to its logical conclusion, it works well and discreetly on cloudy days or in installations with a lot of shade. Why not use it at night also? And so they did.

    It will probably work anywhere at a high enough feed-in tariff rate. An

    • no-one will ever know.
      If no one would ever know, no one had noticed it in Spain.

      In Germany it would be flat impossible.

      And since the Spain incident, I assume everyone on the world is paying attention to such rare situations.

      • by Budenny ( 888916 )

        They got greedy and ran it all night, which was a giveaway, and that is how it was discovered.

        If all you do is up the amount being generated by a bit, and extend the time when you are doing max output into the evening a bit, no-one will ever know. Modest amounts, but free money.

        Would it would be obvious? I suppose if your output is dramatically higher than that of your neighbors on the same street, facing the same way, that might raise queries. If the electricity company is checking. Obviously if they c

        • If all you do is up the amount being generated by a bit, and extend the time when you are doing max output into the evening a bit, no-one will ever know.
          In Germany everyone will know.
          Sorry, you have no idea ow a grid works.

          a) everything is recorded - I have your load profile from last year - for each day
          b) everything is forecasted - aka special weather reports, giving the expected amount of sun - per hour - right now
          c) I have your load profile from yesterday, and any other day, that has a similar forecast a

  • by petes_PoV ( 912422 ) on Wednesday April 06, 2022 @04:05AM (#62421692)

    generating a small amount of electricity

    If they are only after a small amount of electricity, why not harness whatever light is reflected off the moon?

    • You have to use power to move the panel so it's facing the moon when it rises. Then some more to track it until it sets. Plus it's waxing crescent at the moment so you'll get more generated from thermo electric anyway. Might be a possibility during waxing to waning gibbous, but I doubt it would be worth the effort.

    • by mattr ( 78516 )

      Sounds like a good idea. IANAP but apparently a full moon is still only about 1/300 the power from noon sunlight. So maybe you could get something out of it. It seems that there is enough energy to light an LED but not enough intensity to make it out of the solar panel or your inverter.

      https://www.cleanfuture.co.in/... [cleanfuture.co.in].

      • a full moon is still only about 1/300 the power from noon sunlight.

        Nope. You are off by a factor of a thousand.

        A full moon is about 1/400000 as bright as the sun at noon.

    • Because that is happening automatically.

      And you know about moon phases, right? Perhaps 5 days per month the moon gives enough light ...

  • by MancunianMaskMan ( 701642 ) on Wednesday April 06, 2022 @04:53AM (#62421738)
    Physics [wikipedia.org] teaches us that everything radiates heat at ta rate proportional to T^4, and receive likewise from our environment.

    I remember this from a physics class at university: our teacher made us calculate how much heat would be lost from a room at 20degC through a rooflight pointing at the clear night sky (at almost ~0K) The calculated amount was something ridiculous, teaching us that the atmosphere is nowhere near transparent, and somewhat disappointing our teacher who had clearly expected that he'd made some reasonable assumptions to get a ballpark figure (within an order of magnitude is good enough to make many a physicist happy).

    In the context of the solar-cell-at-night I assume it makes a huge difference if it's a clear night or a cloudy one.

    • by dargaud ( 518470 )
      Direct application: If you have a solar oven (the old-style parabolic mirror, with a hook for a boiler at the focus point), you can use it to freeze water at night: just point it straight up and place water at the focus point. It'll loose heat faster.
  • ROI? (Score:4, Informative)

    by bradley13 ( 1118935 ) on Wednesday April 06, 2022 @04:54AM (#62421740) Homepage

    The ability to radiate heat into space is known. It only applies to very limited frequencies, and you have to prevent the heat from being replaced by contact with the atmosphere, so the radiating surface is commonly placed in a vacuum. Here's an article with a simple description of the whole process. [sciencenew...udents.org]

    TFA says that - with clear skies - their solar cells will be a few degrees cooler than ambient. A brief look at the efficiencies of thermocouples shows that the expected voltage will (depending on materials) probably be around 0.25mV to 0.5mV. Back of the envelope, with lots of assumptions, they might generate 50mV from an array of thermocouples on a solar cell. It's harder to estimate the available current, but it won't be much, simply because generating the power equalizes the tiny temperature differential.

    It's seriously hard to imagine that a meaningful amount of power can be generated, compared to the added production costs. Another attempt, not combined with a solar cell (so, probably more efficient) produced 25 mW/m2. So a solar plant that generates 1MW at noon might produce 150W at midnight.

    • I don't think this is using thermocouples. With only a few degrees temperature difference, the output is puny, as you have noted.

      A long time ago, I read of practical heat->electric generators using stacks of thermocouples heated by something like an oil or gas lamp. The temperature difference could be a few hundred degrees. For a type K thermocouple, which has a fairly high output, you get about 50 mV at around 1300 deg. C. I don't have the tables to hand. Let's say we can achieve 1300/4 = 325 deg C. Th

  • I've always know that photons cary heat with them but they do not carry it.
  • Sid Assawaworrarit, fresh from the grand success of generating electricity at night using solar panels, celebrated it by binge watching Monty Python. When the scene about King Arthur being snubbed by the French legionnaire, " ... I fart in your general direction ...", he abruptly stopped the video. Whipped out his lap top and dashed off a new research paper.

    Abstract: New method to improve the speed and efficiency of ships

    If one or more passengers and crew members line up at the stern of the ship and collectively fart sternwards the additional momentum imparted on the ship would reduce the fuel consumption. We have applied for a research grant to study it in the field and validate it. Plan calls for about 1 million dollars in funding for the instruments needed to measure the change, study the fart inducing parameters and cruise ship tickets for researchers and their famil^H^H^H^H^H volunteers

    • by nwaack ( 3482871 )
      Thank you for this :)
    • I think more could be done to improve the efficiency of this mechanism. The primary component of flatulence is methane. Simply using this directly as a propellant seems a waste of a good hydrocarbon resource. The new improved version collects the farts, then uses those to drive an internal combustion engine. A good fart is definitely combustible, as a chemist friend demonstrated when I was at uni. It burns with a blue flame. We never investigated whether this processing of flatulence products broke down sul

  • We saw a story here on /. back in 2019 [slashdot.org] about people doing this (not new even then btw, this is an old idea). Did these guys at Standford just pick up the $30 kit and plug it in to their solar panels?
  • At night the solar panels are colder than the ground, so the TE panel generates a little energy. During the day the panels are hotter, so the TE panel might be able to generate a little more if it's bidirectional. But the TE panel insulates the solar panel from air cooling. And solar panels are less efficient when they get hot. So my guess is that the TE panel provides less energy than it makes the solar panel lose.

  • "he slight difference in temperature between the ambient air and the surface of a solar panel pointed deep into space. "

    They suck the heat out of the satellite or station or astronauts.

    So in wearables it could cool you while generating electricity for your phone.

"The following is not for the weak of heart or Fundamentalists." -- Dave Barry

Working...