E. Coli used to produce Hydrogen
Researchers at Texas A&M have modified E. Coli to produce hydrogen with extremely high throughput compared to the natural strain.
By genetically modifying the bacteria, Thomas Wood, a professor in the Artie McFerrin Department of Chemical Engineering, has “tweaked” a strain of E. coli so that it produces substantial amounts of hydrogen. Specifically, Wood’s strain produces 140 times more hydrogen than is created in a naturally occurring process, according to an article in “Microbial Biotechnology,” detailing his research.
This is very interesting development as it could lead to a new usage for hydrogen. E. Coli is a very well-known and utilized throughout industry. So this type of development for a production of hydrogen makes a lot of sense from a technology perspective.
The question I have is around the cost to produce. The press release implies a distributed power use-case:
“Take your house, for example,” Wood said. “The size of the reactor that we’d need today if we implemented this technology would be less than the size of a 250-gallon fuel tank found in the typical east-coast home. I’m not finished with this yet, but at this point if we implemented the technology right now, you or a machine would have to shovel in about the weight of a man every day so that the reactor could provide enough hydrogen to take care of the average American home for a 24-hour period.
Let’s think about this from a cost stand point. If the image here is a “standard” size house with a “HydroEColi Reactor” black box in the back of the house that provides power. To run this reactor, you need to feed it…I don’t know… ReactorFood (some bag you buy at WalMart or something like that). I want to find out how expensive ReactorFood needs to be.
Let’s assume both cases: (1) 80 kg usage per day and (2) 80 kg / day.
My house in Palo Alto, CA - a 4 BR, 3 BA, two story home with four adult males - cost around $190/mo for gas and electric (not water). California tends to have high energy costs, but it also doesn’t have harsh winters or extremely hot summers like in Cleveland or Houston. So we can assume that while the consumption load may vary, the pricing should be about the same. This is an important point to note as consumption amount is a better index to use.
We’ll ignore the cost of the reactor, but we should expect it would cost only a % of the material cost of a house (say, like $15,000 or less).
What we’re really finding here is an upper limit to how much the input into this process needs to be. If it costs as much as gold then it’s bad. If it’s dirt cheap, then we could have something.
We’ll use linear equations:
(1) 80 kg / day:
80 kg/day * X$/kg = $190/mo * 12 mo
80X = 2280
X = $28.50 /kg
(2) 8 kg / day:
X = $285 / kg (10X).
These are good numbers. For comparison, Corn Flakes costs around $15/kg at retail.
So if everything spoken in this article really pan out as spoken, then we’re certainly in the ball park of being able to afford such a device. The 8KG model would be a great opportunity as it would leave a lot of room to make a relatively high cost feedstock.
The next obvious question would be the application for a car. This type of system seems to be of interest to Honda. So it’s development would be a great opportunity for their FCX program.
In all, this is very encouraging research and I’m hopeful to see where it goes.
