What’s Not “Clean” about Solar Panels

Industrial processes can't be very green

Solar panel array in Manchester, Vermont. [Credit: MarkBuckawicki, public domain, via Wikimedia Commons.]

I recently saw someone online wax poetic about how solar panels are so benign because they don’t leak oil or emit air pollution or make noise like machinery running on fossil fuels. “They just sit there,” he said, doing their thing. While I agree that we must reduce fossil fuel use for the good of the planet, I must point out that solar panels are not benign.

Like any technological product manufactured by industrial processes from raw materials extracted from the earth, solar panels have an ecological footprint that negatively impacts the more-than-human world.

Currently, the subject of these impacts is most often ignored or when it’s not, is usually hand-waved away. After all, the carbon-centric narrative goes, with the climate crisis being such an existential threat, we must do anything we can to “decarbonize” and that means scaling up solar, wind, etc., as fast as possible. As regular readers will know, I have long advocated for cutting overall energy use and consumption rather than trying to sustain current levels with alternate means. To emphasize: my critique of “green” or “clean” energy is from an environmental perspective, and be assured I’m far from a climate denier.

I’ll also add that I personally appreciate solar power in my own life. As someone who doesn’t have a permanent home and who has regularly ended up in off-grid situations, I own portable solar equipment to keep my gear charged. It’s amazing technology that has allowed me to do my writing and photography in remote places and I’m grateful for that. Especially since I visit some of these locales because they are under threat from expanding development and I want to document them in the interest of their defense.

Desert habitat being wiped out for a huge solar operation, southern California, 2022. [Photo by the author.]

My main concern is with the utility-scale photovoltaic plants (colloquially known as solar farms) because of the large amount of wildlife habitat they wipe out. I love the western deserts and their flora and fauna, and I’m opposed to them being sacrificed. Rooftop, brownfield or parking lot installations are preferable in this way, though there is still the impact of manufacturing and disposing of the panels themselves, which is not trivial, and which I aim to highlight here.

The three main types of solar panels used in utility-scale plants are monocrystalline, polycrystalline, and thin-film. The crystalline types are by far more common. Monocrystalline panels are the most efficient, last the longest and have the highest cost. Polycrystalline panels are more affordable, but are no longer the standard in utility-scale operations. Thin-film is less efficient (and less expensive) than both but can tolerate higher temperatures, which is advantageous in desert regions. Currently, thin-film panels comprise only ~5% of those in use, so I’ll be skipping over them here.

Quartzite to sand to silicon

Silicon is the key material needed for crystalline panels. (Thin-film panels may or may not use silicon, so more on those later.) Silicon is made from quartzite sand, which is in turn from quartzite ore. Quartzite ore is extracted from open-pit quarries or underground mines. As far as habitat degradation goes, mining is a nightmare. Besides the literal loss of land, there’s all the pollution including toxic dust and fumes, chemicals, emissions, noise, etc. Local water sources are often depleted or tainted. Restoration of such spaces to their original states is impossible. Yes, another mix of flora and fauna can thrive there in time—and I’m the last person to throw shade on novel ecosystems—but the loss of the original is permanent. The “green” and “clean” monikers applied to technology like solar panels ignore the mining step, even though it’s absolutely essential.

Benichi quartzite quarry in Russia. [Credit: Сіверян, CC BY-SA 3.0, via Wikimedia Commons.]

Transforming the quartzite ore into sand is a multi-step process involving specialized industrial equipment, high temperatures, lots of water and of course copious energy.

First the ore is crushed, screened, washed, and “calcined” (heated to 1800-2000°F to purify it).

The Keestrack B3 jaw crusher is used for the initial stage of crushing and can handle quartzite; a machine this size could be sufficient for a small-to-medium operation. [Credit: Fiver, der Hellseher, CC BY-SA 4.0, via Wikimedia Commons,]

Next steps include magnetic separation (to remove ferrous impurities), air classifying (which separates the particles by size), and surface treatment (to improve various properties like water repellency).

Above: “The process and results of separating quartz sand using a dry magnetic separator” in Malaysia.

To finally get to pure silicon, the sand is mixed with a carbon source (like coal) and put in an arc furnace. As the oxide burns away, silicon is left behind, though still with some impurities, which are removed using hydrogen and hydrochloric acid. The final result must be greater than 99.9999% silicon to be solar grade.

Submerged arc furnace for silicon production. [Credit: Feriasolar.]

Purified silicon to panels

For monocrystalline panels, this nearly 100% silicon is made into ingots through a fascinating process called the Czochralski Method. A “seed crystal” on a shaft is lowered until it just touches the surface of a vat of molten silicon, and then is slowly raised and rotated. A crystalline structure of silicon forms in a cylinder up to six feet long, vaguely like growing sugar crystals on a string. (For polycrystalline panels, molten silicon is cooled in molds.)

A monocrystalline silicon ingot and several silicon rods on display in the Museum of Mineralogy in Munich. [Credit: ArticCynda, CC0, via Wikimedia Commons.]

The ingot is sliced into thin wafers (180–300 micrometers thick) with a diamond-coated precision saw. The wafers are cleaned in baths of acidic and alkaline liquid and with ultrasound. Then they are treated with an alkaline solution that roughens the surface at the microscopic level, reducing reflectivity so more of the light hitting the wafers is absorbed. Next they are “doped” to maximize their conductivity. “Doping” uses phosphorus oxychloride to infuse the surface with minute impurities, which is what make the wafers functional as electrical components. Yes, after all that complicated refining, the wafers won’t function until purposefully made less pure in a very particular way. The doping step requires temperatures of 1475-1650°F.

A few more coatings are applied to the wafer: on the front, silicon nitride for anti-reflectivity and silver for conductivity, and on the back, aluminum to complete the electrical circuit. The front of the wafer is the positive side and the back is the negative side. At this point, the wafers are finished solar cells, and are tested to ensure efficiency and output..

To manufacture a solar panel, individual cells are strung together with metallic “busbars” and “bus ribbons” to carry the current (lots of soldering at this step), and the resulting grid of cells is sandwiched between layers of encapsulation (usually an ethylene vinyl acetate film) with glass on top and a weatherproof plastic “backsheet” underneath. After being laminated with heat, the now joined layers are affixed in a frame with a junction box on back.

Components and layers of a typical crystalline solar panel. [Credit: César Domínguez, CC BY-SA 4.0, via Wikimedia Commons.]

Recycling

The International Renewable Energy Agency estimates that by 2050, the world will have to deal with ~78 million metric tons of solar panel waste. There’s no coordinated plan or regulations to deal with this. Currently, the rate of recycling is around 10% but the number of solar panels reaching their end of service life now is much lower than it will be in the future due to the great number of panels being manufactured and installed. That is, if the current number of panels being recycled didn’t change, then in a couple decades the percentage would be lower than 10. So if we’re serious about recycling solar panels, we have a lot of work ahead of us.

A part of a field of solar panels near Ogwell in Devon, England. [Credit: Partonez, CC BY-SA 4.0, via Wikimedia Commons.]

The challenges might be primarily logistical and economical. Technically speaking, the glass panes and aluminum are fairly simple to sort out and the silicon wafers can be melted down and re-purified, though dealing with the encapsulation layer is “not straightforward.” Also, as with any industrial processes, recycling will itself require machines and energy and will generate waste.

Logistical challenges include building recycling facilities, setting up systems of collection, and legislating the policies to make it all happen. Economically, whether recycling “pencils out” or not will depend on a number of circumstances, such as whether profit motive is the deciding factor.

At the moment, though, solar panel recycling is barely a thing, and we can’t just count on the hope that “we’ll work that out later.” It really needs to be prioritized right now, if only to clean up the mess we’ve made so far.

Picture the big picture

It’s true that a solar panel does not leak oil or emit air pollution or generate noise. But its manufacture and disposal are not benign. Next time you see a utility-scaled photovoltaic plant in person or otherwise, try to picture the footprint it left elsewhere, from the gaping hole of the quartz-ore mine, to the all factories and industrial machines involved along the way, to the piles of old panels that may or may not be recycled.

I don’t know how big of a role solar energy will play in the years and decades ahead, but I hope it is small because our overall energy consumption ends up declining. My personal best case scenario is no new energy infrastructure because we reduce that much that fast.