Someone at fuel cell manufacturer Bloom Energy knows the recipe to get journalistic coverage of a product: Claim a heritage from the space program (the company’s founder K.R. Sridhar was involved in Mars mission research), get a big-name venture capital firm (Kleiner Perkins) to invest a lot of money (reportedly $400 million) and then let the whole operation bake in stealth mode for an inordinate amount of time (seven years). Finally, get Google involved somehow (the Mountain View, CA Internet colossus is Bloom’s first customer).  Today, Bloom Energy officially comes out of the oven with its 100 kW distributed generation fuel cells and, right on plan, it has garnered an inordinate amount of the Internet and the media’s attention, including mentions in 60 Minutes and The New York Times. While the Bloom gurus have handled the P.R. process masterfully, they forgot to mention a few other things about their Bloom boxes:

Our Product Does Not Produce Electricity More Efficiently Than Centralized Generation.

Unlike other fuel cell concepts that try to produce both usable heat and electricity, Bloom has purposely avoided the combined heat and power (CHP) market. Its boxes make only electricity. As a result, they run at an efficiency of approximately 50% (in other words, for every two units of energy put into the box, one equivalent unit of electricity comes out the other end). That’s more than twice the efficiency of a diesel generator, which is great. But a centralized combined cycle natural gas power plant is even more efficient than that (sometimes reaching 60% efficiency). Subtract the line losses in transporting the electricity from a central plant to the point of consumption and you end up with approximately the same efficiency.

 Our Product Isn’t Much Cleaner than Centralized Generation.

Bloom boxes turn natural gas into electricity without combusting anything. That seems like it should be much cleaner than old-fashioned centralized natural gas plants. But the carbon dioxide emissions are still the same. And the trace amounts of sulfur found in natural gas must still be removed and disposed of. That leaves NOx emissions and particulate emissions, neither of which are emitted by a fuel cell. But natural gas produces such small amounts of either in traditional combustion plants that they are hardly a major source of environmental concern.

Our Electricity Is More Expensive to Produce than Most Other Forms.

At a capital cost of $7-$8 per Watt, it’s hard to find a competitive technology that’s more expensive than buying a Bloom box. Even photovoltaics cost less when purchased on the scale of 100 kW at a time. When analyzed on a levelized cost of energy (LCOE) basis, the Bloom economics look a little better, coming in at somewhere between 8 and 10 cents per kWh according to the company (IDC Energy Insights believes that 15 to 20 cents per kWh is a more realistic figure).  Even when compared to other forms of distributed generation, which tend to be more expensive than centralized generation, Bloom’s capital cost is two to four times higher than the competition. And a brand new centralized combined cycle natural gas power plant shouldn’t come in over $3 per Watt in capital costs. Bloom boxes are, to put it bluntly, extremely expensive.

We’re Not Ready to Execute on the Most Exciting Aspect of Our Technology: Storage. While Bloom is positioning itself as a straight-up electricity generation play, its technology has a unique twist to it: the process of turning hydrocarbon into electricity can, theoretically, be reversed. In other words, when electricity prices are low (i.e. during the night), the cell stack can stop pushing out electrons and instead take them in while pushing out hydrogen atoms. That hydrogen can be stored in a tank and then be used as a fuel to produce electricity during the day, when prices are higher. This opens up the possibility of having a decentralized source of high-efficiency energy storage, something which the utilities industry and end users would pay dearly for. Nevertheless, the Bloom team told IDC Energy Insights that its product in its current form does not have the capacity to perform the storage feat, although future iterations might have that feature.