“How do I solve a Rubik’s cube?” Most of us have googled this at least once, while secretly thinking, can I solve it with random turns? Well, the odds are not so high: 7.844×10^-24% chance! Add the keyword “beginner” to your search, and you’ll start getting ideas: “Divide the Rubik’s cube into layers and solve each layer applying the given algorithm…”

Solving the cube one layer at a time … seems feasible! Until you read the next bit: “Keep in mind that when you solve the one side, you have to solve the adjoining row at the same time.” All of a sudden it sounds like a formidable challenge, like the challenge of displacing the low capacity but stable graphite with the ultrahigh capacity yet unstable silicon in the anode of lithium-ion batteries!

The successful displacement of graphite with silicon could dramatically transform how energy is stored and transported. The solution to it, however, is a known challenge for battery scientists. It seems as if we are often compromising something to fix something else. Having said that, it is very difficult, if possible at all, for a single work published in an academic journal to address the whole set of key elements for achieving the practical use of silicon anodes in lithium-ion batteries.

Silicon on the sidelines

In 2015, Elon Musk of Tesla announced that partial replacement of graphite with silicon in the anode of Model S batteries increased the car’s driving range by 6%, creating a race among battery manufacturers to pack more silicon in.

The second most abundant element in the earth’s crust (about 28% by mass), with an ultra-high capacity to hold lithium (in theory as much as ten times the capacity for a given weight compared to graphite), silicon has been an ideal candidate to explore as a new anode. The maximum specific capacity of silicon at room temperature is 3,579 mAh g^-1, comparable to the highest capacity anode material, lithium metal, due to the formation of Li₁₅Si₄ when silicon combines with lithium. It also offers an advantageous low working potential, promising high energy density when used in full cells and ensuring improved operation safety by inhibiting the growth of lithium dendrite.

The 15:4 ratio for silicon compared to the 1:6 ratio for graphite (LiC₆) promises a unique match for the ultra-high capacity sulfur cathode, while graphite or carbon in any form will never be a match for sulfur. While the “large capacity to store lithium” and “abundant” boxes are addressed, the instability of silicon anode long intrigued researchers around the globe and has limited the amount of silicon in commercial electrodes to minimal quantities, which is rumoured to be at around 6%. Also, silicon is abundant, but does that mean manufacturing battery-grade silicon will be cheap? (Or green?)

Breaking down big problems

Large capacity brings large stress, an inherent problem of high-capacity electrode materials: silicon, aluminium, germanium, tin and sulfur – silicon is the most expansive when it combines with lithium. The expansion of the Si anode is almost 300%, significantly higher than the 10% of the standard graphite electrode.

The large undesirable volume change leads to pulverization of silicon electrode. Not only is the electrical wiring in the electrode destroyed but the continuous pulverization results in the continuous reformation of the solid electrolyte interface (SEI) protective layer. The continuous formation-rupture-reformation results in the undesirable thickening of the SEI and at a cost of consumption of lithium from the cathode.

Thus, the superior capacity performance decays rapidly. The adverse effect of electrode disintegration becomes dramatically more pronounced with any increase in the amount of silicon in the electrode, the key parameter for achieving practical capacities, and has limited the amount of silicon in the electrodes explored to date to minimal quantities.

Two main strategies have been explored extensively to address the inevitable volume-change issue: 1) nano-structuration or nano-scaling of silicon particles (down-sizing), and; 2) providing additional space to accommodate the expansion (room to breathe).

One useful approach in any problem-solving process is taking a big problem and breaking it down into smaller ones. The more effective you can break those big problems down into smaller solvable components, the better. There are numerous scientific reports on the successful downsizing of large-sized silicon into nano-scale, delivering satisfactory performance in the battery. However, quite often, these approaches are as effective as solving one side of the Rubik’s cube. Most methods are questionable when upscaling is involved, and cost matters! In fact, while industrial-grade silicon costs $1-2/kg, micro-silicon and nano-silicon could cost up to $2,000 and $20,000 respectively. These prices are by no means competitive, considering that the low-capacity yet highly stable cousin, graphite, costs only $10-20/kg.

Scrap to nanostructured

When it comes to low-cost silicon sources, recovering the precious solar-grade silicon from the end-of-life photovoltaic panels (PV) and repurposing it for battery applications has been explored for more than a decade now. Nonetheless, a cheap silicon source does not necessarily mean that the final battery-grade silicon is still price-competitive.

A typical crystalline silicon (c-Si) PV module contains about 5% silicon, which is valuable, considering its high purity (99.9999%). Unfortunately, the majority of approaches to recover this pure Si from PV wastes rely on hydrofluoric acid. Hydrofluoric acid assists greatly with removing the unwanted components present in PV panels but it is a potent contact poison. Not only handling it but disposing of it requires extreme care, which contributes to the cost of the silicon recovery process, and yet it will never be risk-free.

The use of caustic hydrofluoric acid can be avoided. My colleagues at the Group of Research in Energy and Environment from Materials at the University of Liege in Belgium run an extensive PV recycling program, led by Dr Frederic Boschini. Moving towards practical recycling solutions, we have identified a substitute for the hydrofluoric acid used in the chemical treatment of the PV wastes.

Nicolas Eshraghi and Abdelfattah Mahmoud, lead researchers of this patented technology, demonstrated that a critically controlled yet industrially adaptable chemical process in the absence of hydrofluoric acid enables the effective removal of the unwanted components, leaving behind the ultrapure silicon particles with the highly desired size range of nanometric to micrometric.

Nanosizing of the Si particles was done by an optimized milling process, certainly an inexpensive route compared to chemical vapor deposition and/or structuration from nano-powder precursors. The ultra-pure and nanostructured silicon, recovered from PV panels in one step, delivers good performance when used as an anode material in lithium-ion batteries, promising for a high-quality electrode material, synthesised from a cheap-priced source via a green, scalable and cost-effective route.

Both the Uliege team and the Monash team believe that quality and cost-effectiveness should be pursued at the same time, as the battery manufacturers don’t like any conflict between the two.

Remember, sometimes there are no shortcuts: “Peeling the stickers off a Rubik’s cube is not the solution, and a centre square will always remain a centre square no matter how you turn the cube.”

Dr Mahdokht Shaibani is an energy storage solution provider with expertise in materials synthesis, engineering and scale-up for next-generation energy storage systems. She has a PhD in mechanical engineering from Monash University.