GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted
Green Building News

Researchers Warn of Tsunami of Obsolete Solar Panels

Currently, there are no standards in place for dealing with solar panels that have reached the end of their service life

A boom in photovoltaic modules has vastly increased global power output, but the industry now faces the prospect of an 80-million-ton waste problem by the middle of the century. Photo courtesy Dennis Schroeder / National Renewable Energy Laboratory.

Discarded solar panels could add up to 80 million metric tons of waste globally by mid-century yet there currently is no common plan for managing the problem or recycling the valuable materials that the modules contain, researchers at the National Renewable Energy Laboratory have found.

NREL said the report, which was published at Nature Energy earlier this month, is the first global assessment of how photovoltaic panels could be managed at the end of their 30-year service life. Researchers found there also is no standard for how to recycle the valuable materials the modules contain or how to handle toxic materials and minimize threats to the environment.

Much of the study was focused on the recovery of crystalline silicon, which is used in more than 90% of all modules. It accounts for about half of the energy and carbon footprint of a solar panel, but only a small amount of its bulk.

Lead author Garvin Heath, a senior scientist at NREL who specializes in sustainability science, said in a news release, “PV is a major part of the energy transition. We must be good stewards of these materials and develop a circular economy for PV modules.”

The article, “Research and development priorities for silicon photovoltaic module recycling to support a circular economy,” was written by a 15-person team. The assessment comes at a time of explosive growth in the capacity of PV panels globally, from 1.4 gigawatts (GW) in 2000 to 512 GW in 2018. Solar modules now produce about 3% of electricity worldwide.

Solar panels are responsible for no direct greenhouse gas emissions, and low lifecycle emissions, and provide both local-cost power and a lot of jobs. “However,” the authors said, “as large-scale global PV deployment continues, the challenge of how to handle large volumes of PV modules at the end of their approximately 30-year lifetimes is emerging.”

By 2030, outdated panels are expected to total 8 million metric tons, which will grow 10-fold by 2050 to exceed 10% of all electronic waste globally.

Stranding valuable materials

Reuse or repair of older modules is one option that researchers examined. They predicted that approach would lower the environmental impact of solar modules while increasing their lifetime electricity production. “These are worthwhile circular economy strategies for the PV industry to investigate,” the report notes, “though numerous business model, economic and regulatory challenges must be addressed.”

Among them are the environmental costs of shipping spent panels to repair or reuse facilities, and the need for manufacturers to establish testing and repair procedures as well as an inventory of spare parts. The high cost of the solar business not directly related to the modules—what the report calls the “balance of system costs”—also is a concern.

No matter how good the industry gets at keeping aging panels functioning, they eventually must be replaced. Researchers found that in Europe a priority has been to make sure that valuable materials in the modules are not lost in landfills. The cumulative potential value of materials that could be recovered from obsolete panels will be an estimated $15 billion by the year 2050. Those materials could be used to make an additional 2 billion new modules with a total capacity of 630 GW.

Regulations to guide module recycling are slowly being developed, but adoption appears spotty. The European Union added a PV category to its Waste Electrical and Electronic Equipment regulations eight years ago, but Washington State is the only U.S. jurisdiction so far to require that panels be recycled. Efforts to address the problems are underway in Korea, Japan, Australia, and India, but the recycling process so far isn’t especially effective.

“Even in the EU, where PV recycling is mandatory, very few recycling facilities designed to handle PV modules exist and information about their recycling efficacy and economics is not publicly available,” the NREL report says. Existing recycling lines are good at retrieving bulk materials, such as the aluminum frame, copper wire, and glass. But silver, copper, silicon, and lead, which make up most of the module’s potential value and potential environmental impact, are not recovered.

To date, there have been plenty of small-scale attempts at separating and salvaging valuable materials from modules, the authors said, adding, “We note that little or no consensus about the commercial viability of these approaches exists.”

In the United States, the Solar Energy Industries Association (SEIA) has six firms that are capable of recycling modules and inverters; five of them will accept crystalline silicon panels and one will recycle its own thin-film panels. The busier recyclers process only about 100 tons of silicon per month. “Owing to the low volumes of modules being sent for recycling,” the report says, “recycling lines dedicated to c-Si PV modules have not been developed in the U.S.”

Some modules are being taken to landfills, while others are being stored until better recycling options become available.

Recommendations for improving the system

The authors propose a research and development agenda with three goals for recycling crystalline silicon: make recycling cheaper than disposal; find a way to use recycled materials in new modules; and make recycling an environmentally better option than throwing old panels out and using virgin materials to make new ones.

The report makes these points and suggestions:

  • Because of contamination, the metallurgical grade silicon recovered from panels now is worth about $2 per kilogram, but if the recycling process could be improved so that the silicon was of solar grade, its value would increase to $10 per kilogram or more.
  • Less attention should be focused on trying to recover intact silicon wafers. Newer wafers are much thinner than older ones, making intact recovery difficult.
  • Convince an industry that has relied on virgin silicon supply to use recovered silicon. To do so, complete impurity profiles will be required so manufacturers know exactly what to expect. That process hasn’t started yet.
  • Designing modules so they are easier to take apart or recycled hasn’t caught on, but increasing awareness of the issue and using emerging international standards such as Cradle-to-Cradle certification would help.

Scott Gibson is a contributing writer at Green Building Advisor and Fine Homebuilding magazine.



  1. CraigRo | | #1

    Old panels typically don't fail, they just aren't as efficient as new. Upcycle, reuse before recycle. Put them into community programs instead of buying new.

    1. JC72 | | #2

      Re-purposing is just a delaying tactic and unsustainable. In addition the solar market is heavily subsidized which acts as a disincentive to buy used. Imagine for a moment what would happen to the auto industry if vehicles were built to last 30 yrs with just routine maintenance.

      1. antipode | | #5

        Repurposing is indeed a great way to get use out of older panels. At sufficiently low cost, there is almost certainly a market for panels that produce power. While the technology has changed over the years, there are plenty of 1970's era panels still producing power, and if that trend continues, there will be a very long economic life for panels.

        Even when panels lose any economic value due to breakage, etc., there will be a place in landfills for them, just as there has been a place in landfills for drywall, for example, which by volume will likely always be much larger than any amount of recycled PV panels. Relative to other inputs into the waste stream, PV panels are a rounding error.

        1. JC72 | | #11

          As the article mentions the issue is the type of waste. Primarily heavy metals of lead and cadmium. Silicon is also an issue as we constantly read horror stories about the demand for sand and the impact it's having on the environment. In addition these older panels use cadmium telluride. It's nasty stuff and you don't want it getting into ground water. It's why for example using solar panels in hurricane prone areas isn't necessarily a smart idea.

          1. Expert Member
            Dana Dorsett | | #12

            >"Silicon is also an issue as we constantly read horror stories about the demand for sand and the impact it's having on the environment. "


            The silicon used in PV and semiconductors is obtained from mined rock, not sand dredging in river channels, etc.

            >"In addition these older panels use cadmium telluride."


            CdTe PV a relatively NEW (less than 2 decades of commercial deployment) type of thin-film on glass type of panel, and not used in the much more ubiquitous silicon types. The biggest source of cadmium is a byproduct of refining zinc from particular ore types- it's already a disposal problem before it goes into a PV panel (where it is encapsulated, isolated from the environment for decades. To be sure, you don't want to just landfill used CdTe panels, but that's what environmental regulations are about. First Solar is one of the largest CdTe panel manufacturers (which is still a small part of the overall PV market), and has a lifecycle plan for recycling, with a net negative Cd release into the environment when that power output is replacing fossil-burners or nukes. See Chapter 3 of this analysis:


          2. JC72 | | #15


            Interesting. I always thought it was coming from sand rather than mined.

  2. john_m1 | | #3

    Can the engineers, physicists, and material scientists who design these things design recycleability into the panels? We've pretty much reached the limits of small. There is a new metal I saw being used in an expensive bike like that makes it very difficult to cut. It wasn't exotic materials but how they put it together. I'm not a scientist of engineer, but we reduce the waste stream by accounting for it up stream in its lifecycle.

    I also agree we don't always do a great job trading one problem for another.
    The first R is reduce, we use fewer panels we get reduced waste. There is waste with oil and coal and nuclear. We can however ask the question up front, how will break this down once it fails.

  3. DavidfromPNW | | #4

    green like an EV. LOL. Typical green new deal scenarios. Fairy dust and unicorns dancing at the end of the rainbow.

    What? We have to do something with spent batteries. And useless old solar panels. We didn't know.

    Oh boy. The hits just keep on coming.

    1. Expert Member
      Peter Engle | | #6

      Or, we did or do know. And we're looking at ways to address the issue 30 years before it's a big issue. That's what the article is about. It's called planning. The process uses science.

      1. john_m1 | | #7

        Completely agree, you have to ask the question. When (you pick) coal, oil, nuclear etc were being developed we didn't do that. Cars were a mass market product 30-40 years we passed pollution controls. There were charging stations in Manhattan 100 years ago when it wasn't clear what type a motor would prevail. Imagine where battery tech would be today had we invested in that at the level we invested in ICE's.

  4. Expert Member
    RICHARD EVANS | | #8

    "Discarded solar panels could add up to 80 million metric tons of waste globally by mid-century"

    Well, the United States alone burns 7-8 times that amount of coal EACH YEAR for electricity generation*.

    But yea David B, let's call PV "Rainbows and Unicorns" and stick with the dirty status quo. :-)

    Thinking about this 30 years in advance gives me hope that we will address this. It also goes to show that super-insulation and air-tightness should come before PV.


    1. john_m1 | | #9

      There really isn't much in the luck or accidents, that is not how progress works.
      Early in the Obama admin the DOE targeted the reduction in cost of solar panels via pushing the research. They didn't reach their goal but they came close. I understand where David B is coming from, but we never asked these sorts of questions before or if we did we didn't do much about it. Yes PV causes new problems, and not using PV exacerbates existing problems. Consequences either way.

  5. lance_p | | #10

    The industry has invested its time researching how to reduce the cost of solar panels, not how to make them more recyclable. The problem with solar WAS that it cost too much. Now that it's cheap and generating clean power there's a problem - they'll be in landfills.

    Or will they? I can't see acres of PV panels being discarded if they're still generating power. What I can see is a huge movement of people using reclaimed panels not unlike reclaimed insulation or any other reusable building product. DIYers and low income community projects are teeming with people who are smart and have more time than money and would surely re purpose older panels and maybe even fix panels with dead cells etc. similar to what's possible with dead battery packs (replacing the few damaged cells to restore the pack).

    Panels that are still generating 50% of their new power levels will surely be worth more to someone than their scrap value. Companies who deal in new solar equipment could surely harvest a lot of these old panels to be sold for modest profits.

    1. lance_p | | #16

      The article linked to at the bottom of this page:

      ...suggests we won't have much to worry about for a long time as far as degradation of solar panel performance is concerned. According to those studies, worst case scenario for modern panels is less than 1%/year, and best case is only 0.2%/year. Averaging that out to about 0.6%.year, a 50 year old panel will still be producing 70% of its rated power.

      How cheap would PV need to be to make a business case out of tearing down an existing installation still operating at 70% and replacing it with new panels? I'm sure the answer will vary based on each situation, but I bet that number is very low. And this is after 50 years.

      Another thing not being discussed is whether the decay in performance is linear? If the greatest losses are in the first years of service and the rate of performance decay slows down over time, we could see panels having useful lives many decades longer than what these worst case scenario studies are estimating.

      Of course, if space is a limiting factor and the only way to meet a new requirement is to switch to a newer more efficient (power dense) panel, there won't be much choice.

  6. burninate | | #13

    “PV is a major part of the energy transition. We must be good stewards of these materials and develop a circular economy for PV modules.”

    No, we don't need to do that. 80 million metric tons of MSW by 2050 is a drop in the bucket. We currently produce 2 billion metric tons MSW a year, which represents 60 billion metric tons by 2050 assuming no growth.

    We may choose to implement some recycling in the event that proves economically feasible, but the sustainability case for solar panels is in no way reliant on recycling, and the environmental gain we get from solar panels is orders of magnitude higher than the environmental harm we would suffer if we all literally dumped used solar panels in the nearest river; Any sort of significant impediment to adoption that we introduced to require recycling would be net harmful right now.

    1. maine_tyler | | #14

      "if we all literally dumped used solar panels in the nearest river"

      Yep, you went too far. Doesn't build your case.

      On the production side, we also have mining issues to contend with, which are possibly more significant than the waste stream issues.

      Certainly, we don't need a list 'for solar' and a list 'against solar' as if it's a simple yea or nay proposition. The only way we'll allow ourselves to identify the problems created by these solutions to other problems (AGW) is to admit the complexity and nuance of human/environmental systems coupling.

      1. burninate | | #17

        I left out the part about "and take a dump in a beaver dam while you're there" because I thought it was excessive, but I think it's probably literally true. We did and do all sorts of horrendous shit to the environment, and that ESPECIALLY includes "Continuing to use coal power".

        Sunpower sells something on the order of a billion watts nameplate capacity per year for a corporate total of about 300k tons CO2 emissions per year. Those cells at a 20% duty cycle produce 200 million watts power, or ~2 billion kwh a year. Burning coal emits roughly 1kg/kwh; So these panels Sunpower makes save about 2M tons CO2 per year by displacing coal, and will continue to for >30 years (>60M tons emissions reductions based on this year's production). That's a factor of 200 improvement on the alternative. That means every 400-watt panel, over 30 years of 20% duty cycle, saves 20 TONS of coal from being burned, hundreds of tons of overburden from being dumped into rivers, saves 100 grams of radioactive uranium and thorium from ending up spread over the biosphere by a smokestack, saves *economically relevant* amounts of respiratory illness.

        The cost of that 400W panel sitting still degrading ("slowly becoming gravel") in even our most precious landscapes (Yosemite anyone?), does not compare, it is entirely negligible by comparison to the positive effects. What we'll do instead is dump it in a landfill where nobody will cut their toes, and call it a day.

        This is not a problem that needs solving, and if it adds the smallest asterisk to the purchase of a new solar panel, if it adds even 0.1% to the overall cost, it isn't worth it from an environmental perspective.

        Maybe it ceases to be negligible when our (currently 1 billion kilowatt in size) national power grid has 20 billion kilowatts nameplate of solar photovoltaic instead of 0.04 billion kilowatts, when coal and natural gas and oil and poverty have been eliminated and we can focus on higher-order needs like "trace copper/tin leakage from landfill watersheds", but we're not there yet, and we won't be for a long time.

        1. maine_tyler | | #18

          I understand you have the best of intentions, and some all-in-good-fun hyperbolic humor.
          Dropping a brownie into a beaver lodge sounds harmless, whereas crushing toxic metals into streams and rivers very much does not.
          If attempts to 'recycle' or at least 'manage' the waste issues with solar/wind etc. prevent us from moving forward with the needed innovation, then I think I can generally agree with you.

          I disagree that that necessarily needs to be the case. I disagree that the waste (and related mining issue) is utterly and completely not even a tiny bit of an environmental concern. (I also disagree with your assessment of the magnitude of how beneficial an individual unit of solar is given things like EROI, grid power reliance issue with VRE's, need for storage, etc., but it's too late to unravel that).

          The notion that we can't accommodate other environmental concerns along the road to a low carbon-emitting future seems pessimistic, if not naive or even plunderous. I could see it easily becoming a 'cutting of the nose to spite the face' if left completely unchecked.

          My argument would go that we need to internalize externalities, at which point we may find it affordable to not only install more solar, but to recycle it as well. Or, if there really is NO issue, as you claim, we may not.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |