Bingo. When Solyndra was formed (and to be honest, my company at the time made process equipment for CIGS deposition), solar grade polysilicon was in short supply, and was spiking in price to almost $600 per Kg. Investment in thin film amorphous silicon cells (Applied Materials and Oerlikon bet heavily here too) and CIGS (CuInGaSe) cells as the solution. Until the wild growth of the solar cell industry, spot prices for polysilicon was ~$30 - $50 Kg (from memory, all sites that have historical data seem to demand a subscription to access it).
Of course, making CIGS cells was difficult. Extremely difficult. The deposition processes are tricky, and even small variances in the composition cause great fluctuation of efficiencies. There were probably 100 players chasing the CIGS dream, including 10 or so well funded players. But at the same time, the foundries who had the ability to tool up for producing crystalline silicon saw the $600 per Kg with dollar signs (or, more accurately renminbi signs) in their eyes, and began building capacity.
Starting in 2011, much of this new capacity was coming online, and prices for polysilicon ingots plummeted. Today it is around $26 - $30 per Kg, really close to the price needed for the magical $1/W for a solar installation. Solyndra went tits up, Applied Materials and Oerlikon left the business (I have some interesting stories there as well) and the world is dominated by panels made in China.
Fun fact, in 2008, QCells, a German company was the number one producer of solar panels. Before that Japan's Sharp was #1, now most of the top 10 are Chinese companies.
Solyndra was a bet, a bet that Silicon prices would remain high, but in hindsight, a foolish bet for depending on a source raw material that can be refined from beach sand.