Bloom Energy vs. Cogeneration Reality

Cogeneration can achieve better CO2 reductions than the Bloom Energy fuel cell when operating on the same type of fuel.

That's the short version, according to Bob Spitzka, President of Water & Energy Management (WEM) of Danville, California, in a white paper he's written with colleague James Hall. Both are licensed Professional Engineers in northern California who have collaborated on the feasibility and design of nearly 100 cogeneration projects in California and Hawaii during the past 36 years. 

When you splash on the scene like the Kleiner Perkins and NEA-funded Bloom Energy, making bold pronouncements about power costs and energy security, one has to expect a bit of a spotlight to continue to shine on the hype. Especially when you're working on a 150-year-old technology that has yet to yield a profitable fuel cell company.

Greentech Media editor-in-chief Michael Kanellos was the lone voice of restraint during the 60 Minutes lovefest early last year. And Rich Hilt welcomed John Doerr to the world of energy in his Dear John letter.

Since then, Bloom has made announcements about their stellar line-up of customers and their Power Purchase Agreement program. The firm has raised about $500 million in venture funding.
 
WEM's paper presents a comparison of CO2 emission reductions derived from the installation of a 100-kilowatt Bloom Energy ES-5000 fuel cell and a 100-kilowatt natural gas fired cogeneration system (cogen). Thousands of cogen systems have been installed over the last few decades and are in use across the world, providing heat and electricity to thousands of facilities.

According to Spitzka, Cogen, also described as combined heat and power (CHP) or distributed generation (DG), is not perceived by the general public as a system that can decrease CO2 emissions.

Instead, in his estimation, enormous amounts of public funds are being allocated for grants, subsidies, and special financing for installing Bloom Boxes because the Solid Oxide Fuel Cells are ostensibly "green" with a low-carbon footprint.

Spitzka concedes that while the application of the Bloom unit can provide benefits to facilities where the use of thermal energy (hot water) is minimal or nonexistent, there are many facilities, such as hospitals, schools, research facilities, factories, senior residences, etc., that use hot water on a continual basis where cogen can achieve better CO2 reductions than the Bloom Box. 

Background on Cogeneration

The cogen process involves natural gas to power an engine coupled to an electrical generator to supply some or all of the facility’s power while capturing the thermal energy from the engine and using it to preheat the facility's hot water boiler. This reduces natural gas consumed in the facility’s boiler and reduces operating cost and CO2 emissions.

The cogen process results in overall thermal efficiencies of over 80 percent; most utility power plants are around 35 percent efficient, according to the author. The cost of cogen systems is estimated to be $3,000 per installed kilowatt versus Bloom at more than $8,000 before subsidies. (Editor's note: Bloom competitors Panasonic, ClearEdge Power and Ceres Power have made cogen heat/electricity fuel cells that hit the 80 percent level seen in cogen plants refrred to in the paper. Bloom maximizes electricity and minimizes heat because electricity is a higher value form of energy, it says, and possible credit issues. The small amounts of heat go toward running the fuel cell itself.)

System Performance Comparisons

The white paper performs a direct comparison between installing a single 100-kilowatt Bloom unit and a 100-kilowatt cogen system, both operating on natural gas at a facility currently purchasing its electricity and natural gas for its boilers from Pacific Gas & Electric (PG&E), and evaluates each system’s effect on net site CO2 emissions.

One of the benefits that Bloom cites in its literature is that its unit can provide a “carbon neutral” generation solution when operating on renewable fuels such as biogas. While this claim is valid, Cogen has the identical ability to operate on a biogas. However, Spitzka's analysis looks at the operation of each system using the same natural gas currently being purchased for the facility’s boilers.

The following table is a summary of the comparison between the Bloom unit and a Cogen unit, each rated at 100 kilowatts and operating for 90 percent of the year (7884 hours/year). 

Site CO2 Emissions Comparison Summary

* Avoided Site Boiler Fuel = 7.0 therms/hr. / 75 percent boiler Eff = 9.33 therms/hr.

Although the Bloom unit produces significantly less CO2 compared to cogen in the production of electricity only, the significant reduction in boiler site CO2 emissions due to cogen’s thermal energy output results in a net decrease in CO2 emissions for cogen as compared to a net increase for the Bloom unit.

Conclusions

According to Spitzka, in "many facilities the application of cogen will reduce site CO2 emissions, while the Bloom system, in geographic areas such as California, may actually increase site CO2 emissions. These conclusions are unknown to many politicians and the media, so they enthusiastically push the Bloom technology without acknowledging the monetary cost and performance of the Bloom system may not be the sole technology worthy of public support."

Spitzka writes, "The American people deserve to have their public funds spent on projects that have been properly vetted by independent and technically competent engineers, not by media hype. Let’s invest in the most effective technology available to maximize cost and environmental benefits for the user before all of the public money is all spent."

A colleague, also with experience in this business, said,"I think [WEM is] being a little unfair to Bloom insofar as cogen purposefully tries to use the waste heat for 'process.' It would be better to compare a 100-kW system to a Bloom system that added the necessary 'space heat' to whatever the kWh output is."

***

The white paper from WEM goes into far greater detail on the calculations performed to reach these numbers. We've asked Bloom to comment on these findings, and we are awaiting their response.