David Kramer’s report “The cost of solar energy production has plunged, but it needs to fall further” (Physics Today, June 2021, page 27) gives an excellent overview of the present state of solar technology. One point, however, needs clarification. In evaluating cadmium telluride, the author correctly recognizes the toxic properties of cadmium, which is a carcinogen. He cites the claim that CdTe is virtually insoluble in water, but that dismisses the groundwater contamination risks posed if CdTe panels end up in a landfill.
In 2010–11, I worked at the now-defunct Amelio Solar in Ewing, New Jersey. Suspecting that the fluid that oozes through most landfills might be sufficiently acidic (or sometimes basic) to dissolve CdTe, my colleagues and I conducted experiments in collaboration with a team at the Catholic University of America in Washington, DC.1
Both groups cut 50 mm squares from a commercial CdTe module. My team cracked each square with one hammer blow to simulate the breakage that would likely occur when discarded panels were dumped. We then placed each piece in a closed jar containing 150 mL of citric-acid-based buffer solution with a pH value of either 3, 4, 5, or 6. The last square went into deionized water. The Catholic University team members scribed their samples with a diamond saw to avoid the randomness of the hammered breakage. They used rainwater as the reference and solutions with pH values of 4 for acid and 10 for base.
Over the course of six months, my team analyzed the solutions for cadmium using inductively coupled plasma optical spectrometry. The sample in pure water with a pH of 7 showed no additional damage, and even at the end, the cadmium concentration in the solution was only 7 ppm. The results were markedly different for the acidic solutions. The farther the pH diverged from 7, the faster the cadmium leached into the solution. After three months, the most acidic samples had completely delaminated, and the cadmium concentrations in the solutions in which they were immersed had reached over 100 ppm. We also observed that the higher concentrations had leveled off well before the end of the experiment, suggesting that virtually all the cadmium had leached out of those specimens.
The Catholic University team ran their tests for 70 days. They obtained essentially similar results, with about 3 ppm cadmium for rainwater, 54 ppm for pH = 4, and 140 ppm for pH = 10.
In 2017 a team at the University of Arizona published a study with experiments similar to ours in both method and results.2
Even if CdTe-module makers promise to reclaim the products at the end of their useful life and recycle the cadmium, can we be confident that they will actually do so 30 years in the future?
At least two additional tests have shown that CdTe modules present a serious groundwater pollution hazard if they end up broken in a landfill permeated by something other than neutral water with a pH of 7.3,4 If we are to use such technology, we must guarantee a safe disposal and recycling program, and include its full cost and risks in any evaluation of CdTe photovoltaics.