Black Silicon Makes Solar Cells Cheaper
A one-step process creates a highly antireflective layer for photovoltaics.
A simple chemical treatment could replace expensive antireflective solar cell coatings, bringing down the cost of crystalline silicon panels. The treatment, a one-step dip in a chemical bath, creates a highly antireflective layer of black silicon on the surface of silicon wafers, and it would cost just pennies per watt, say researchers at the National Renewable Energy Laboratory (NREL). They’ve used it to create black silicon solar cells that match the efficiency of conventional silicon cells on the market.
The crystalline silicon wafers used to make today’s solar cells are treated to create a textured surface, then coated with an antireflective layer, usually silicon nitride, using high-vacuum processes. This additional layer increases the value of a solar cell by improving its efficiency–it suppress reflection so that more photons actually enter the silicon wafer instead of bouncing off its surface, increasing the flow of electricity off the cell. But the extra layer also adds to the expense. “We believe it can be cheaper,” says Howard Branz, principal scientist in silicon materials and devices at NREL. Even with a coating, the best-quality silicon solar cells typically reflect 3 percent of the light that hits them. Branz’s lab is developing inexpensive ways to create black silicon, which reflects almost no light.
Prototype solar cells made at NREL have the best efficiency ever reported for black silicon cells. Monocrystalline silicon cells with the black surface, and no additional antireflective coating, convert 16.8 percent of the light that hits them into electricity, about the same efficiency offered by a typical crystalline silicon solar cell coated with antireflective material. The previous record for black silicon cells was 13.9 percent.
To replace the vacuum-deposition processes used to treat the surface of a silicon wafer, Branz’s lab developed a chemical process that can be performed at ambient temperature and pressure using equipment already on site at solar-panel factories. A wafer is submerged in a bath containing a water solution of hydrogen peroxide, hydrofluoric acid, and chloroauric acid, which is made up of hydrogen, chlorine, and gold. The small amount of gold in the acid bath acts as a catalyst for chemical reactions. It’s not clear exactly what the chemical reactions are, but they lead to the formation of gold nanoparticles that drill nanoholes at varying depths into the wafer. Branz says the gold can be reused again and again.
This etching process takes three minutes at room temperature, and less than a minute at 40 ºC. The result is a black, highly absorbent silicon wafer whose surface is riddled with tiny tunnels of varying depths. The pores create a surface with no sharp edges to reflect the light, and the variation in their depth is key, because the length of the tunnels determines which wavelength of light it will interact with. Variation in tunnel length allows the surface to trap a broad spectrum of light.
Several groups are developing black silicon for solar cells and other optical devices. Others have used a multistep process to create the silicon-tunneling reactions, starting by placing metal particles over the surface of a wafer and then adding the acids. “We’re doing it in a single, liquid step that requires no vacuum processes at all,” says Branz. Beverley, MA-based startup SiOnyx uses laser pulses to generate tiny cones on silicon surfaces. Their formula for black silicon creates a different material than NREL’s, but with similar properties. Company cofounder and principle scientist James Carey says the company hopes to enter the market within a year. Carey would not disclose what photovoltaic efficiencies SiOnyx has achieved using its process.
The baths used in today’s factories to clean solar cells between etching steps could be used to create the black silicon coating, says Branz, which means this process could be adopted by manufacturers at very low cost. The agency has applied for patents on the process.
Couldn't make it to Cambridge? We've brought EmTech MIT to you!Watch session videos