Self-Cleaning Tech Could Help Solar Panels Keep Efficiency Up

Solar panels are a special kind of magic — turning light into useful electrical energy. However, they don’t work nearly as well when they’re covered in dust, dirt, and grime. Conventional solutions involve spraying panels down with pure water, which is expensive and wasteful, or dry scrubbing, which can cause efficiency loss through scratching the panels. However, innovative new methods may offer useful solutions in this area, as shared by EETimes.

Researchers at MIT have explored the use of electrostatic methods to remove dust from solar panels. By creating a sufficiently strong electrostatic field, dust particles can be compelled to leap off of solar panels. The cleaning method requires no water and is entirely non-contact. It uses a motor system to pass a charged electrode past the surface of the panels, with the opposite charge applied to the panels themselves. This repels the dust from the panels and onto the moving electrode.

Other methods include the use of special “self-cleaning” glass manufactured with a laser etching technique. The method, referred to as Direct Laser Interference Patterning, or DLIP, creates microscopic features on the order of 300 nm to 30 um on the surface of the glass. The pattern creates a so-called “functional surface” from which dirt simply slides off. The laser-etched pattern has no negative impact on transparency.

If these ideas prove practical, expect them to take off quickly. Commercially viable technologies spread fast in the renewable space, as we’ve seen with the rapid uptake of floating solar farms in recent years. 

58 thoughts on “Self-Cleaning Tech Could Help Solar Panels Keep Efficiency Up

    1. Part of me just wants to point at the windshield wipers and sprayers on a car and say “where is my MIT degree?” Parts are readily available and as the solar panels are angled the water or liquid could easily be collected if necessary. However I would likely just calculate the cost vs efficiency uptick of the cooling effect. You would think that solar farms would have built this in as a feature during the hottest hours to increase efficiency, the water does not have to be potable so theoretically it could be recycled minus evaporation. Furthermore the coating will eventually stop working or hinder the photons. The water use could be offset by closing one half of one golf course. Part of me wants to say that the water could easily be sourced using solar desalination techniques, but for some reason that is a zero sum game, nobody talks about it and there are zero projects that i am aware of of any scale. Solar deaalination is effective and cheap to run after the initial cost of producing the equipment. The mechanism for solar desalination is so simple that it can be done by picking up a bit of trash almost anywhere in the world and is a staple survival technique. Granted the chosen location for solar farms is not ideal as they were only thinking maximum sun and the cheapest land/place to install.

      1. As ever – if it was that easy / obvious I suspect there must be a few good practical reasons it’s not being done, and solving the real-world problems is the hard part.

        1. Cost is always the biggest obstical. When the job is bid upon, upkeep is not often factored. This is more likely the practical reason that such mechanisms are not deployed. A pre installed staticly charged coating is likely cheaper in the short term and can be rolled into cost as a “feature”. A Jose with a hose is cheaper up front than an automated spray and collection system. The other ironic part about solar desalination is that energy can be generated from the process… So literally another form of solar energy where the only byproducts are clean water and the gunk left behind. Correct me if i am wrong but this was a tech that was proposed to be deployed right around the time that coal was on the uprise. Coal power was chosen for lower intial investment cost if i recall.

    “As interest in clean energy surges, used solar panels are going straight into landfill.”

    “By 2035, discarded panels would outweigh new units sold by 2.56 times. In turn, this would catapult the LCOE (levelized cost of energy, a measure of the overall cost of an energy-producing asset over its lifetime) to four times the current projection. The economics of solar — so bright-seeming from the vantage point of 2021 — would darken quickly as the industry sinks under the weight of its own trash.”

    1. Point being, solar panel recycling
      a) costs a lot of money
      b) isn’t being done

      The economic incentives given to solar panels mean that since people got their investments back early, they can also replace the panels early to leverage the upgraded technology, so people may be dumping their solar panels much earlier than projected and generating a whole torrent of trash that nobody can deal with.

        1. “solid waste landfills typically charge $1 to $2 to accept a solar panel, rising to around $5 if the material is deemed hazardous waste. By contrast, his company charges $18 per panel.”

          If it was useful, they’d pay for the panels instead of charging to recycle. The other article points out the same thing: $20-30 to recycle, $1-2 to dump.

          1. That is a problem, sure, but I have 20 panels on my roof, if I was replacing them with a whole new install for $20K, I would be fine with $360 to make sure the old ones got recycled. That might be a harder sell for a big commercial install with 100s of panels, but they’d be less likely to replace ’em all at once I imagine.

      1. The step before recycle is reuse, seems like we could do more of that. Sure an old solar panel may be 15% efficient not 25% but buy them for $10, mount and connect them as cheaply as possible (ground or at most ladder height) and they’re still useful. Plaster them in every less-than-optimal location like garden fences, house walls etc.

        We’ll still have to figure out recycling for the broken panels.

        1. There is no “cheap” when it comes to mounting solar panels, the wind will pick them up and carry them away or they will collapse under the weight of wet and sloppy springtime snow. Local zoning laws and electric utility rules will require you to have a proper installation.

          1. Those zoning laws dramatically increase the value of residential property and create safe and quiet neighborhoods where people can raise their children and they can be secure in the value of their property. This stuff is what they are talking about when they use the phrase “civilized society”.

          2. >This stuff is what they are talking about when they use the phrase “civilized society”

            These days HOAs tend to be their own fascist states within the state – which has nothing to do with keeping a civilized society.

            But, all the more reason not to live in a “nice” sub-urb and instead move out of the city entirely. Plenty of job opportunities elsewhere as well – as long as you’re not educated in social media or arts.

          3. > This stuff is what they are talking about when they use the phrase “civilized society”.
            Yes, I’m aware. Civilized society is what is destroying the environment, our personal freedom, and the human experience in general, so we need to roll it back a little bit. Sorry if this offends your sensibilities.

        2. Agreed; those with the means will pay a lot to put high peak efficiency panels on the roof in compact suburban neighborhoods and buy a powerwall or similar to even out the daily production. But anyone with a barn, garage, or field may have plenty of area for lower efficiency panels, and what they really need is for the panels to be cheap.

          1. The panels are not where the cost is. The cost is in the mounting and the wiring and the inspections and the insurance and you will incur those costs even if you get the panels for free. You won’t recoup those costs unless your panels are generating sufficient revenue.

          2. The whole point I was trying to convey is that making the few square feet of roof in an expensive suburb host a bunch of expensive panels on expensive mounts with expensive wiring and expensive inspection and expensive insurance is… expensive. And it only works if your power company favorably supports home power generation, which isn’t always the case.
            People who don’t have the means for all that expense but do have more area available with less restrictions could avoid much of the expense by not trying to do it on top of a house, and not trying to store or sell every possible bit of energy – but only if the panels were a lot cheaper to make up for wasting power at noon and buying it at night.

            A basic starting-out example is if someone had a panel (with a buffer) that powers something that’s commonly used (e.g. their lights, a freezer, some computers) and there is a backup grid power supply that kicks in when there’s not enough solar available. It doesn’t scale past the point where your inverter can power every load you have, and it isn’t a way to be paid to invest in solar, but that’s kind of the point – this is for *residential* use. The way to make a return on investment in a solar farm and the way to make use of solar panels residentially are two different things.

          3. >and not trying to store or sell every possible bit of energy

            If you have more than couple square meters of panels, the only way to use them is to sell the power. Peak exceeds demand and every panel after that is just waste.

          4. >only way to use them is to sell the power. Peak exceeds demand and every panel after that is just waste.

            Entirely the wrong way of thinking about it – peak is irrelevant really, it happens maybe 1-30% of the daylight hours over a year, almost certainly nearer the 1% end – so the useful metric is the average output, or perhaps even the 5-20% low output. You want to meet your demands, or be close to it with one of those as that is when you are no longer having to buy in expensive all the time, any excess you might be able to sell is a bonus. And if you really have the big solar farm there are lots of industry that can be energy hogs but don’t require continuos power who will be happy to pay you more than the grid will no doubt.

          5. > Peak exceeds demand and every panel after that is just waste.
            If you wanted to harvest every single joule of energy, you might rather build a solar farm or invest in one and pay your power bill from the yield.

            If you instead want local power and want to ensure that your supply curve is above your demand curve for most of the day, you might want to overprovision the amount of peak supply you actually have. Say you needed at least 24kWh per day for a whole home system, much of it in the afternoon and evening. If a standard system at your location generates six hours times your system’s rating per sunny day, then a rated size for your system might be 4kW if you’re aiming roughly to break even, which might be three thousand dollars of panels alone. If for the same price you could get panels double or triple the nominal capacity, with all else equal, you might begin generating greater than your demand a couple hours sooner in the day, and between storage and this excess generation you might have an easier time meeting the evening demand spike before returning to a certain level of nighttime load. On your days off you might charge an EV with the midday power; on workdays you might do laundry or run hvac / dehumidifiers / automated appliances that can be unattended.

            This isn’t even completely a residential thing; there’s papers researching the economic choices of solar plants that recommend overprovisioning the panels versus the size of inverters in order to clip the peak but meet demand. And it all depends on having cheap enough panels that you don’t need to squeeze out as much as possible from them. In a fair market with a high percentage of solar on the grid, there would be a period of time near noon where it should cost money to produce power because it requires someone to spend resources in order to sink that power in order to stabilize the grid. As such no-one with solar would ever want to output too much power, in case they’d either not get anything for it or have to pay to get rid of it when they could instead throttle back for free. And because of that, they might as well not buy inverters that are sized to be able to output the peak level of generation, because they’ll never make their money back until someone comes along who’s able to use all that power and willing to buy it for a worthwhile price. So they should buy more panels.

          6. >peak is irrelevant really,

            It is VERY relevant to your cost per kWh. Affordable cost can only be achieved by utilizing all the power. If you throw away 80-90% then your power will cost 5-10x as much.

          7. >the useful metric is the average output

            “Average output” is only met by utilizing all the power you have available. The year-round average, or the capacity factor of a solar panel is between 7-14% of its nominal power depending on the region. If you toss away the peaks, then it drops to low single percentages.

          8. >you might want to overprovision the amount of peak supply you actually have

            Over-provisioning the peak has diminishing returns on the average because of the shape of the power curve of solar PV. You have to build many many times more peak capacity and it won’t make much of a difference at all for the non-peak hours. You’re just wasting capacity.

          9. >there’s papers researching the economic choices of solar plants that recommend overprovisioning the panels versus the size of inverters in order to clip the peak but meet demand.

            The power curve is steep enough that it won’t matter. By over-provisioning the panels, you’re increasing the ramp between about 8-10 AM which is a relatively tiny portion of the total energy you’re going to gain, and also making the down-ramp between 2 – 16 PM equally steep, so you’re also adding to the problem of finding enough fast ramping capacity to take over the load. That is, you need more gas turbines to pick up early in the afternoon and then rise up to the demand as the sun goes down.

          10. >The peak is often a good 50% and more of the energy you have available throughout the day.

            The daily peak is nothing at all to do with the overall peak, and can be varied easily even on a fixed install simply by having some more east/west facing collections. But still on a fixed mount only very specific times of day and times of year will hit the true peak, and it was against that measure I was I referencing.

            Still on the daily curve you do have a point, though a little bit of levelling that out and timing demand to peaks – position the panels at the right angle for the times of day with a peak, batteries, etc.

          11. Having built a small solar system for an off-grid hunting cabin, just buying golf cart batteries for minimal system storage cost 150% of the panel cost. An inverter for a reasonable AC load cost as much as 1 panel. The solar charge controller was 1.5 times 1 panel. The panels are the least part of the cost, and the longest lasting.

            But, I’m all for reusing them at end of life. And why not set them up to hydrolyze water? Store the hydrogen and use it to power a cook stove. So long as you store it well ventilated and outside buildings you care about, it shouldn’t be worse than the methane from your septic system…

          12. Dude: if you throw away 90%, but the panel costs 1%, your energy cost is 10%. Btw, doing some roughly approximating math before posting reduces power consumption of the internet servers and increases credibility of the poster.

        3. REUSE! So much room for reuse that just gets ignored. I’d love to have solar panels as my deck railing, plastered above my septic field, placed on top of my roadside trashcan/mailbox structure, how about the shed! The “used” market has completely evaporated in the USA Midwest.

          1. Around here we have plenty of junkyards and used furniture stores and consignment clothing stores and used car dealers and used sporting goods stores and Redbox machines and there is a thriving market in used building materials from old houses so I would say that people do indeed reuse a lot of things and they do it when it makes sense. Reusing old solar panels does not make economic sense and that is why it is not done.

            If you really want to save a LOT and not just small potatoes then you should focus on reusing consumer waste where you can really make a dent and n the landfill if altruism is your goal.

          2. If you want the panels in working condition, you have to pay more for the crew to handle them carefully instead of just throwing them off the roof into the back of a dump truck. After all, everyone wants a living wage and every minute counts.

            Module cost in a residential installation was 40% in 2010. It has fallen down to 10-20% in 2021. The labor, permitting, taxes, planning, insurance, financing, etc. take almost all of the money, so there’s not a lot of savings to be had from re-using panels. That’s the irony of solar power: it IS very cheap, but actually building and using it is not.

          3. That said, the business around solar power has grown this fat because solar power was and still is highly subsidized. The laws and regulations, and the prices, were made up with the purpose of increasing the cost to take private advantage of the generous public handouts.

            Work tends to expand to fill the time allotted, and green technology increases in price to claim all the tax money you give it.

    2. Oh man I adore these solar/wind BuT wHaT aBoUt tHe GaRbAgE arguments.

      I mean, the first and most obvious thing that makes this argument hilarious is that climate change and waste disposal are only tenuously connected issues. We can continue dumping trash for decades without seriously harming our species. We can’t keep producing GHG at the current rates.

      The second is… it’s a drop in the bucket. The waste produced by lifecycle replacements of solar panels and windmills is an order of magnitude less than plastic grocery bags, and they aren’t even the biggest form of plastic waste.

      The third is, it’s temporary. Both the technology used in production and the technology used in recycling are advancing. It’s the same with lithium battery packs. Every generation the loop is drawn tighter.

      We need to do better, and we need to use less power in the first place. But these arguments are so incredibly flimsy, and come from such an intensely disingenuous position.

      1. >climate change and waste disposal are only tenuously connected issues

        The point about LCOE is not. if it actually costs a whole lot more because of unaccounted externalities, then the whole thing is a dead duck in the water.

        > We can continue dumping trash for decades without seriously harming our species. We can’t keep producing GHG at the current rates.

        Sure, but if you commit to THIS solution, then you’re simply kicking the can down the road and creating problems further along. That line of thinking is no better than saying “We can keep burning fossil fuels because we have CCS!”.

        >the technology used in recycling are advancing

        Sure. Call me when it’s ready. Otherwise, the cards are still in the deck and not in your hand.

        1. Let’s be real here. There isn’t a big PV panel recycling or reusing industry because there are very few panels coming out of service. There are startups and research groups working on developing PV recycling. History suggests this is an engineering challenge, not an insurmountable obstacle.

          And “unaccounted externalities?” Get back to me when you have paid attention to the unaccounted externalities of coal and oil in your cost comparison.

      2. >we need to use less power in the first place

        No, we don’t. Energy is something we need more of, not less, because we want to and need to do things as individuals and as societies that require a whole lot more energy than what we have at our disposal right now. Even recycling stuff needs more energy than extracting what already exists – which is the point of why we’re mining stuff instead of reclaiming dilute waste streams.

        You can’t save out of a loan to pay the debt, you have to make more.

      3. If humans were really having a significant impact on global CO2 levels then the signal from the covid lockdowns would have shown up on the Keeling curve dataset, but it didn’t. If you can’t even detect a 10% reduction despite the measurements being in parts per million then it is reasonable to assume that there is no provable correlation using that key dataset. This is why I do consider demonstrable pollution from landfill dumping to be the most pressing issue. As for plastic bags, what first world nation still favours those? Where I am we have brown paper bags, even the fake biodegradable plastics etc. have been rejected.

        1. “If humans were really having a significant impact on global CO2 levels then the signal from the covid lockdowns would have shown up on the Keeling curve dataset, but it didn’t. If you can’t even detect a 10% reduction despite the measurements being in parts per million then it is reasonable to assume that there is no provable correlation using that key dataset.”


          Verification, Validation, and Confirmation of Numerical Models in the Earth Sciences
          Naomi Oreskes; Kristin Shrader-Frechette; Kenneth Belitz
          Science, New Series, Vol. 263, No. 5147. (Feb. 4, ***1994***), pp. 641-646


          Verification and validation of numerical models of natural systems is impossible. This is because natural systems are never closed and because model results are always non-unique. Models can be confirmed by the demonstration of agreement between observation and prediction, but confirmation is inherently partial. Complete confirmation is logically precluded by the fallacy of affirming the consequent and by incomplete access to natural phenomena. Models can only be evaluated in relative terms, and their predictive value is always open to question. The primary value of models is heuristic.

          Numerical models are increasingly being used in the public arena, in some cases to justify highly controversial decisions. Therefore, the implication of truth is a serious matter. The terms verification and validation are now being used by scientists in ways that are contradictory and misleading. In the earth sciences-hydrology, geochemistry, meteorology, and oceanography-numerical models always represent complex open systems in which the operative processes are incompletely understood and the required empirical input data are incompletely known. Such models can never be verified.

          1. It is a bit like using AI “enhanced” images as scientific evidence and claiming that you are presenting a fact rather than a derived analysis that can’t be justified as it was done using a black box process.

        2. CO2 has a half life of ~120 years in the atmosphere. So even a complete shutdown of human activity (there wasn’t) for one year might lead to a 0.25% drop = 1ppm. (Natural levels being around 280ppm).
          What size of signature might have been detectable, do you think?

          1. The change in the _rate_ of CO2 rise should have dropped by about 10% but you’d be lucky to find _any_ change so you can no longer correlate human activity with CO2 rise rates.

      1. I do that on the easier panels – it can easily clean one full panels long edge with the battery pressure washer and the water bottle (so a lower pressure than mains works well enough). Then I have a window cleaning cloth on a long stick to scrub the rest as needed – but the dirt settles on the lower easy to get to panels so much more that it is really easy really. Should do it more often than I do though.

    1. NASA has the slight problem that there’s almost no air on Mars so you can’t just blow the dust off, and electrostatic brushes have the issue that the charge won’t dissipate so cleaning the brush becomes difficult. Dust just sticks to it and won’t come off, just as with the solar panels themselves.

  2. Charge your Roomba, remove the internal brushes and let it go on the panel. It will not fall off the panels and use the iRobot phone app to bring it back to you. This will get the debris and dust off. But watch it carefully just in case!

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.