How to recycle solar panels: the cheap and cheerful approach

A recent journal paper suggests a new approach to recycling solar panels [1]. This is very much a cheap and cheerful approach (my words) compared to the gold standard ‘full recovery end of life’ approach. A comparison of the two methods is illuminating. The main author of the paper, Pablo Dias, has since set up a company called Solar Cycle to commercialise his process. It is great to hear of technology progressing from research in a university into the real world. Dias is from the School of Photovoltaic and Renewable Energy Engineering, University of New South Wales, (Sidney, Australia) and some of the contributors to the paper were from other universities in Brazil but Solar Cycle has been set up in Texas, USA. 

The cheap and cheerful approach:

• Involves only mechanical process and electrostatic separation; no high temperatures and no acids
• Produces less pure products – more like metal ‘ore’ than refined metals 
• Is less capital intensive
• Can be economic at small scales

What are solar panels made of?

Solar panels are largely made up of glass. By weight, they are (roughly) 76% glass, 8% aluminium frame, 5.8% polymer encapsulating the cells, 3.6% mainly silicon solar cells, plus about 3% metals (including cables), mainly copper and aluminium but with small amounts of silver (0.032%), tin and lead [2]. 

Assuming a full recovery process, in terms of economics the most value is from silver and aluminium followed by silicon, copper and copper cables as shown in the chart below.  (The costs of the cheap and cheerful process are not represented in this chart which comes from another study.) The glass, although it is the largest proportion by weight has little value.

Chart showing the economic value of products from solar cell recycling comparing ASU (a process developed by Arizona State University to maximise recovered metal purity), FRELP (a full recovery end of life approach) and a hybrid method. From [2]. 

Why is the glass worth so little?

Glass is valuable providing it is pure and the right sort of glass [3]. This is why our council recycling collections take only glass bottles, not window glass. Other glass may have no value at all. Glass recovered from general waste which is not very pure costs up to £30/tonne to dispose of (2021 prices) [4].

How are panels recycled?

Reference [2] describes a reference plant for FRELP recovery. This is what it says:

…First, the module frames, cables, and junction boxes are removed and sold for further recycling. The glass with backsheet attached is heated in an infrared belt furnace to weaken its encapsulation, removing the glass from the remaining layers of polymer and cells by knife cutting. The separated glass is then sieved and optically sorted into clean glass to be sold and contaminated glass (containing more than 2% by mass of impurities [20]) to be disposed of. The encapsulant-coated cell strings are mechanically comminuted and incinerated at an off-site incineration plant to remove organic compounds. The remaining bottom ash after incineration is then treated via sieving, leaching, and filtration. Nitric acid is used to dissolve copper, silver, and other metals, leaving behind primarily silicon wafer fragments. Vacuum filtration is used to separate silicon from the leaching solution. In the most expensive step, the leaching solution is electrolyzed in three steps to refine metals, and the spent electrolyte is neutralized and filtered prior to disposal.

The cheap and cheerful process is a great deal simpler than this. The solar panel is detached from its aluminium frame and the panels are shredded to pieces smaller than 2mm. These granules are then separated using an electrostatic process into two streams: one with a high proportion of metal but no glass, the other mostly glass. Neither stream is pure. The metal stream (3.3% by weight) captures 95% of the silver, 98% of the copper and 74% of the aluminium. There is silicon in both metal and glass streams.

Overall, the cheap and cheerful approach is more economic and the environmental impacts are similar.

The lower purity of these streams halves the economic value of the materials compared to FRELP. However, the lower process cost more than make up for this. The main part of the cost is labour to remove the frame – hopefully this can be automated. The energy cost is half that of the FRELP approach and the capital cost is less than a tenth. This chart from [1] compares costs.

Chart from [1] showing processing costs. Process A reclaims the frames only, with manual separation. A1 has automated separation. B1 and B2 are ‘cheap and cheerful’, with manual or automated separation. C is the full recovery option – with much greater capex (pink) and also higher electricity cost (green). 

The overall life cycle analysis reported in [1] suggests that overall the two approaches have similar overall environmental impacts. The benefits of material recovery far outweigh the environmental costs.

The cheap and cheerful approach is less risky.

There are two significant advantages to the low capital cost of the cheap and cheerful approach. Firstly, while we have low volumes and uncertain supply chains it is very risky to make a large investment in equipment. Running a large plant well below capacity for lack of raw material can quickly drive a company to bankruptcy. Secondly, the low capital cost makes it possible to build many small factories and reduce the travelling distance for the whole panels. Given the mass of the panels, much of which is low value glass, this is important too, both for economic cost and environmental impact.

The overall recycled fraction by mass, is very low with the cheap and cheerful approach.  The glass has low economic value, if any, and could well end up in landfill or used for aggregate to make roads. However, since it is inert the environmental impact is low. Dias suggests that his process will be useful mainly as an interim solution until more complete recycling processes become economically possible. However, the complete solutions need more energy as well as more equipment to run. This trade-off between completeness on the one hand and energy use and cost in the other is a recurring theme in recycling.

The cheap and cheerful process proposed for PV panels is quite similar to that currently used for cars.

• High value components that are easy to separate are removed manually (cf removing the aluminium frames from the panels) 
• Polluting oil is drained (this step does not apply to PV panels) 
• Remaining parts are shredded and the pieces are sorted, with separate streams for steel and aluminium for recycling. The remainder usually ends up in waste incinerators, with energy recovery.

There is a nice video of car recycling here.

Using these strategies, EU countries report reuse/recycling rates for cars higher than 90% [5].

In summary - in recycling as always, we should not let the perfect be the enemy of the good. Cheap and cheerful is at least a good start. However, once supply chains are established, more comprehensive processes with higher capital costs that could be profitable become less risky. 

[1] High yield, low cost, environmentally friendly process to recycle silicon solar panels: Technical, economic and environmental feasibility assessment, Pablo R. Dias et al (Renewable and Sustainable Energy Reviews) 2022 

[2] Technoeconomic analysis of high-value, crystalline silicon photovoltaic module recycling processes , (Solar Energy Materials and Solar Cells) 2022 

[3] Breaking down the factors behind scrap glass prices (Recycling Product News) 2017 

[4] Glass Prices 2021 (letsrecycle.com) 2022

[5] End-of-life vehicle statistics (Eurostat)