Once solar panels are operative, they produce electricity without carbon emissions. But making and installing them does involve some emissions. Most of the worries there have focused on elements that go into the panels themselves, like gallium, cadmium, germanium, indium, selenium, and tellurium. But, according to new research, the massive amount of aluminum needed to house the solar rigs of the future could create its own problems.
“I hadn’t realized just how much aluminum was required for the frames, and the modules, mountings, and inverters,” Alison Lennon, a researcher at UNSW Sydney’s School of Photovoltaic and Renewable Energy Engineering, told Ars. She added that aluminum is often used because it is lightweight and corrosion-resistant.
In 2020, the World Bank released an oft-cited analysis called “Minerals for Climate Action: The Mineral Intensity of the Clean Energy Transition.” In this report, the authors identified aluminum as one of the minerals that would need to have its production scale by a huge amount for the world to meet its climate goals. “PV was a large contributor,” Lennon said. “[This] made me think about the problem a bit more.”
However, Lennon said that the World Bank report assumed an early International Energy Agency clean energy roadmap that predicted only 4TW of photovoltaics would need to be installed by 2050. This is a small sum compared to what many updated roadmaps are now predicting.
In Lennon’s paper, she and her team used the target of 60 TW, set by the most recent International Technology Roadmap for Photovoltaics (ITRPV). This would mean that the world would need to produce 4.5 TW of additional capacity each year until 2050 to reach net-zero emissions and limit global warming to under 2° C. For context, by the end of 2020, just over 700 GW were installed.
Gonna carry that weight
The ITRPV report goes into granular detail about the state of the solar energy field, from the size of the modules and their efficiencies, to which ones have aluminum frames, etc. Lennon’s team extrapolated this data from 2030 to 2050 and used data from the industry to measure factors such as how much aluminum was in the frames and how much recycled aluminum can be used in frames and mountings. The team also looked at how the industry would change over time, such as prospects for increasing the efficiency of construction.
From this data analysis, the team was able to predict the amount of aluminum element the world would need by 2050.
The total came to 486 million tonnes to be used for frames, mountings, and inverter casings. To put this number into context, the world bank had calculated around 100 million tonnes. “Our estimate is a lot larger than the World Bank’s estimates,” Lennon said. “The amount of aluminum we’re going to have to produce is going to have to increase an awful lot from what we have now.”
It’s not so much that there’s not enough aluminum in the world—as it is both quite common and fairly easy to extract. Rather, the extraction and production needed could lead to a lot of greenhouse gas emissions. Producing one tonne of aluminum from bauxite—a common source of the element—results in between 14 and 16 tonnes of CO2 or equivalent (the paper assumes this is done in China), Lennon said. “That’s really high,” she said, adding that the smelting process can be quite energy-intensive. “If your electricity is sourced by coal-fired power or fossil fuels in general, the emissions intensity [can be] huge.”
She noted that decarbonizing a country’s electricity system could cut down on this. The conclusion to their study was that it was possible for the world to get enough aluminum for the photovoltaics but that it will take some changes in how it’s produced. Another solution is to try to use recycled aluminum. Aluminum is “infinitely recyclable,” Lennon said. “[W]e need to think carefully about how the aluminum is produced,” she said.
“I think it’s a good story for the PV industry, provided we can get the aluminum industry working along with us, and helping to reduce those emissions.”
Nature, 2022. DOI: 10.1038/s41893-021-00838-9 (About DOIs)
