Cheap Silicon Solar Cell Startup: Sunpreme Gets $50M

Can dirty, cheap silicon make its way into bankable solar cells?

Photo Credit: Image courtesy of Manybits via Creative Commons license

Silicon comes in lots of different grades, and while the higher-end stuff used by the solar industry is getting cheaper, it’s still a lot more expensive than its coarser cousin, so-called “metallurgical-grade” silicon. Turning this material into solar cells could be a profitable business model -- and Sunpreme just got $50 million to pursue it.

The Shanghai and Sunnyvale, Calif.-based startup plans to spend the money building a factory in Jiaxing, China to churn out 30 megawatts per year of Sunpreme’s “SmartSilicon” cells. The cells, based on patents filed in the United States (one of three has been granted), are to be cheap, high efficiency and made on “largely industrial standard production equipment,” according to Sunpreme’s Monday statement, which named the World Bank’s International Finance Corp., Capricorn Investment Group and Tsing Capital’s China Environment Fund as investors.

Sunpreme didn’t highlight its use of upgraded metallurgical-grade silicon (UMG) in its press release, but its patents detail a process for upgrading the stuff in a furnace without gasification, by melting and forming it into ingots, slicing those into multiple wafers, and…well, I’ll let this patent application describe it: “depositing an aluminum layer on backside of each wafer; depositing a layer of hydrogenated silicon nitride on front surface of each wafer; annealing the wafers at elevated temperature; removing the hydrogenated silicon nitride; and removing the aluminum layer.”

The expected efficiency of solar cells made in such a manner isn’t specified, though the patent application does claim they’ll be “comparable [to] and even surpass thin-film solar cells.” In other words, pretty good, though still not as efficient as the best polysilicon solar cells out there.

Other companies have been trying to bring UMG into the solar market. CaliSolar has been working on UMG solar cells for some time, raising $102 million in 2008 to build a factory in California, back when UMG silicon was about one-sixth the price of the higher-grade polysilicon going into the solar industry.

But the equation has changed since then. Solar-grade polysilicon prices have fallen by nearly half over the past six months, with no sign of stopping anytime soon. That’s driven down the prices of silicon solar cells and panels, the dynamic that has made cheap, Chinese-made solar panels the price leaders in the market and put increasing pressure on makers of thin-film (non-silicon) solar companies like Solyndra.

UMG solar cells and panels will be facing the same price pressures, with the added burden that they’ll be bringing an even newer product into a market where bankability is considered the supreme virtue. Sunpreme’s high-profile investors could help in that regard.

In the meantime, I’m curious to see how Sunpreme’s no-gasification method pencils out vis-à-vis the efficiency-to-cheapness metric for UMG. “Upgraded” is the key term here -- companies tackling the UMG solar field still have to clean up their raw material before it’s ready to serve. In February, CaliSolar bought Canadian silicon purifying startup 6N Silicon, which had been supplying its product, and has since declined a $275 million federal loan guarantee to accept a cheaper, $75 million offer from the state of Mississippi to bankroll its facility.

Sunpreme’s reliance on industry-standard manufacturing equipment could also help it drive down costs. In that regard, it sounds a bit like startup AQT Solar, which is making CIGS cells at a 15-megawatt manufacturing line using automated dry sputtering equipment from Intevac. Sunpreme didn’t specify what kind of industry-standard gear it planned to use.