To make a silicon solar cell, you start with a slice of highly purified silicon crystal, and then process it through several stages involving gradual heating and cooling. But figuring out the tradeoffs involved in selecting the purity level of the starting silicon wafer — and then exactly how much to heat it, how fast, for how long, and so on through each of several steps — has largely been a matter of trial and error, guided by intuition and experience.
MIT researchers think they have found a better way. An online tool called “Impurities to Efficiency” (I2E) allows companies or researchers exploring alternative manufacturing strategies to plug in descriptions of their planned materials and processing steps. After about one minute of simulation, I2E gives an indication of exactly how efficient the resulting solar cell would be in converting sunlight to electricity.
One crucial factor in determining solar cell efficiency is the size and distribution of iron particles within the silicon: Even though the silicon used in solar cells has been purified to the 99.9999 percent level, the tiny remaining amount of iron forms obstacles that can block the flow of electrons. But it’s not just the overall amount that matters; it’s the exact distribution and size of the iron particles, something that is both hard to predict and hard to measure.
The team found a way to use basic physics and a detailed computer simulation to predict exactly how iron atoms and particles will behave during the wafer-manufacturing process. They then used a highly specialized measurement tool — an X-ray beam from a synchrotron at Argonne National Laboratory — to confirm their simulations by revealing the actual distribution of the particles in the wafers.
Free of charge, the I2E Web site has been online since July, and users have already carried out approximately 2,000 simulations. Without the tool, there are simply too many possible variations to test, so people end up selecting the best from a small number of choices. Now that the simulation tool is available, it helps manufacturers balance product quality against production time.