April 2008 Archives

Sunday's New York Times caries an article by Michael Fitzgerald, "Home Brew for the Car, Not the Beer Cup", that describes a potential step toward garage production of biofuels, specifically ethanol.

(Update: I wrote to Mr. Fitzgerald in the hopes of getting more information, and he responded with this:

The company currently has only this placeholder site with a form on it:  www.efuel100.com.

It intends to announce its product May 8th, at which time it says it will have more information available.

So we will just have to wait to find out more.)

I have speculated for the last year or so about the feasibility and utility of distributed microbial production of biofuels.  Petroleum refineries and shipping infrastructure are big for a reason; due to both physics and economics it only makes sense to build big, expensive projects.  In contrast, once you have a bug that turns sugar or cellulose into fuel, the production process could in principle look a lot more like brewing beer.

Companies like Amyris are working on building bugs that can churn out a variety of fuels, and they are aiming for production capacity on a scale that is smaller than Big Oil.  Thus far, however, most of these companies seem to be aiming for hundreds of millions of liters rather than a few liters of production capacity (see my previous post "Amyris Launches Cane-to-Biofuels Partnership").  New technology may lead to rethinking this approach.

To give some context, here is a short excerpt from my recent article in Systems and Synthetic Biology, "Laying the foundations for a bio-economy" (see the original for references):

The economic considerations of scaling up direct microbial production of biofuels are fundamentally and radically different than those of traditional petroleum production and refining. The costs associated with finding a new oil field and bringing it into full production are considerable, but are so variable, depending on location, quality, and local government stability, that they are a poor metric of the average required investment. A very straightforward measure of the cost of increasing supplies of gasoline and diesel is the fractional cost of adding refining capacity, presently somewhere between US$ 1 and 10 billion for a new petro-cracking plant, plus the five or so years it takes for construction and tuning the facility for maximum throughput. Even increasing the capacity of working facility is expensive. Shell recently announced a US$ 7 billion investment to roughly double the capacity of a single, existing refinery.          

In contrast, the incremental cost of doubling direct microbial production of a biofuel is more akin to that incurred in setting up a brewery, or at worst case a pharmaceutical grade cell culture facility. This puts the cost between US$ 10,000 and 100,000,000, depending on size and ultimate complexity. Facilities designed to produce ethanol by traditional fermentation and distillation can cost as much as US$ 400 million.

Pinning down the exact future cost of a microbial biofuel production facility is presently an exercise in educated speculation. But, for both physical and economic reasons, costs are more likely to be on the low end of the range suggested above.

This is particularly true for a fuel like butanol. While distilling or filtering alcohol from the fermented mix would reduce the palatability of beer, it is absolutely required to produce fuel grade ethanol. However, unlike ethanol, butanol has only a limited miscibility in water and therefore does not require as much energy to separate. If an organism can be built to withstand the ∼8% concentration at which butanol begins to phase-separate, the fuel could simply be pumped or skimmed off the top of the tank in a continuous process. Costs will fall even further as production eventually moves from alcohols to hydrocarbon biofuels that are completely immiscible in water. Moreover, beer brewing presently occurs at scales from garages bottling of a few liters at a time to commercial operations running fermenters processing thousands to many millions of liters per year. Thus, once in possession of the relevant strain of microbe, increasing production of a biofuel may well be feasible at many scales, thereby potentially matched closely to changes in demand. Because of this flexibility, there is no obvious lower bound on the scale at which bio-production is economically and technically viable.

The scalability of microbial production of biofuels depends in part on which materials are used as feedstocks, where those materials come from, and how they are delivered to the site of production. Petroleum products are a primary feedstock of today's economy, both as a raw material for fabrication and for the energy they contain. Bio-production could provide fuel and materials from a very broad range of feedstocks. There is no obvious fundamental barrier to connecting the metabolic pathways that Amyris and other companies have built to produce fuels to the metabolic pathways constructed to digest cellulose for ethanol production, or to the pathways from organisms that digest sewage. Eventually, these biological components will inevitably be enhanced by the addition of photosynthetic pathways. Conversion of municipal waste to liquid biofuels would provide a valuable and important commodity in areas of dense human population, exactly where it is needed most. Thus microbial production of biofuels could very well be the first recognizable implementation of distributed biological manufacturing.    

The NYT reports that a company called E-Fuel has developed a refrigerator-sized box that turns yeast and sugar into ethanol.  This home fermentation and distillation unit is described as having a variety of technological improvements, such as semi-permeable membrane filters, that reduce the cost of separating ethanol from water.  The price point for the E-Fuel 100 Microfueler is suggested to be $9995, though few other details are given.

Regular readers will recall that I am not particularly enthusiastic about ethanol, but -- assuming it is real -- the Mircrofueler might be an interesting step forward because it ought to work for higher chain alcohols such as butanol.  The physics is fairly straightforward: there is an increase in enthalpy from mixing alcohol and water, which is in principle the only energy you have to add back to the system to separate them.  In practice, however, the only way to achieve this separation is to heat up the mixture, which requires considerably more energy because water has such a large specific heat.  Any technology that helps reduce the energy cost of separating alcohol from water could substantially lower production costs.

E-Fuel might therefore have a way to help Amyris, or LS9, or even BP lower the costs of separating fuels from aqueous production mixtures, and to do so with a box that could sit in consumers' garages.  This raises all sorts of questions about where the bug comes from, whether for the purposes of cost those bugs are consumables, and where the revenue stream comes from in the long term.  I suspect the answer, long term, is that the feedstock and the hardware are the only way to make money.

For example, let's say the University of Alberta 2007 iGEM team (the "Butanerds"), who continue to work on their project, are successful in building a bug that can crank out butanol from sugar.  That bug will be full of Biobrick parts, which at present sit in the public domain.  Acquiring a working circuit made of Biobrick parts will always be substantially less expensive than building a proprietary circuit.  In other words, if a bunch of (talented) undergraduates manage to get their "open source" biofuel production bug working, then it isn't clear that anybody else will be able to charge for a bug that does the same thing -- unless, of course, a proprietary bug is much more efficient or has other advantages.  But how long would the relevant genetic circuits even stay proprietary?  DNA sequencing is cheap, and DNA synthesis is cheap, so reproducing those circuits is going to be easy.  Nobody is going to get anywhere with "biosecurity through obscurity".

Either way, you then still need some sort of box that houses the bugs during fermentation or synthesis of fuels, and also serves to separate the fuel from the soup in which the bugs grow; enter the Microfueler.

So on the one hand we have a new piece of hardware that supposedly will allow the user to produce fuel at home from sugar (or, perhaps, starch, cellulose, and waste), and on the other hand we are starting to see efforts to build organisms that produce a variety of fuels that might be processed by that hardware.

Here is what I really want to know: How long will it be before we see a partnership between E-Fuel and a company (or an iGEM team) to put butanol (or other fuel molecule) "biorefinery" in your garage?  It could even be a company other than E-Fuel, because they are unlikely to have a corner on the technology necessary to build the relevant hardware.  Or perhaps there will even be an open-source "microbiofueler/biomicrofueler" emerging from a garage or university project?

Distributed biological manufacturing, here we come.

I just received word that Amyris has now officially announced a partnership to use Brazilian sugar cane to produce biofuels.  Amyris and Crystalsev are aiming to hit the market with cane-derived biodiesel in 2010.  This deal has been in the works for a while; the press release is here.

Amyris now has access to large amounts of sugar, as well as substantial fermentation capacity, and they need to get their bugs up to snuff for producing large volumes of fuel.  Based on what I have seen in my travels, I expect they will get there.  In principle, the market will be the ultimate judge of the accomplishment, with consumers in the role of jury.  But the market, at present, is also skewed by subsidies and tariffs that place Brazilian products at a disadvantage compared to less efficient domestic production.  It is yet unclear whether the new biofuel will be subject to the same uneven playing field facing Brazilian ethanol.

(Update:  Since this post seems to be getting lots of traffic, I will point those interested to a couple of my earlier posts on the role of Amyris and synthetic biology in producing biofuels: "The Intersection of Biofuels and Synthetic Biology", and "The Need for Fuels Produced Using Synthetic Biology".)

Given the price of grain, and a dislike of genetically modified crops, Europe might soon be eating meat grown in a vat.  Stay with me:

The press is full of noise about the price of food.  Whatever the real impact of biofuels production on food prices -- which is probably very hard to pin down quantitatively -- the grain surplus we have enjoyed for decades is now over and demand exceeds supply.  This condition is probably permanent, and in order to keep the economy running we need to figure out how to get more production out of limited arable land.  This in turn raises the issue of improving yields and overall harvest through the use of genetically modified crops.  GM crops are widely grown and consumed in the Americas, but have met with governmental and consumer resistance elsewhere.

The general embrace by U.S. farmers of GM crops, and contemporaneous rejection of those crops by European consumers, produces interesting complexities within markets.  While the European region is presently a net food exporter, much of the feed for European livestock and poultry comes from the Americas.  Yet the strict safety testing and labeling requirements for food or feed containing GM plants amounts to a European zero-tolerance policy for importation of GM products.  While GM sugar beets and some varieties of GM corn may be officially approved for sale in Europe, consumers appear to avoid products with the GM label.

This policy has fascinating secondary consequences, namely that it is on track to force dramatic reductions in European livestock production due to increasing fractions of GM feed grains.  In an article in the October 2007 issue of Nature Biotechnology, "Europe's anti-GM stance to presage animal feed shortage?" (the full text of which you can find online here, PDF here), Peter Mitchell writes:

...If a solution isn't found, European farmers will be forced into wholesale slaughter of their livestock rather than have the animals starve. Europe will then have to import huge quantities of animal products from elsewhere—ironically, most of it from animals raised on the very same GM feeds that Europe has not approved.

Mitchell cites a report from the European commission that production of meat could fall by between 1 and 44 percent over the next two years, depending on actual supplies of non-GM feed.  Changes in attitude that produce a marketing environment friendlier to GM products may alleviate this problem.  Yet consumer resistance to GM products in Europe is both deep and broad.  Even in the face of economic hardship, brought on by reduced food exports and increased domestic prices, consumers and interest groups may take many years to change their minds.

The New York Times reports that pressure is growing in Europe to change policies on GM crops.  According to an article by Andrew Pollack in the 21 April, NYT:

In Britain, the National Beef Association, which represents cattle farmers, issued a statement this month demanding that “all resistance” to such crops “be abandoned immediately in response to shifts in world demand for food, the growing danger of global food shortages and the prospect of declining domestic animal production.”

Despite these pressures, Pollack writes that, "Since the beginning of the year France has banned the planting of genetically modified corn while Germany has enacted a law allowing for foods to be labeled as “G.M. free.”"

So, in a world with declining GM-free feedstocks, where is Europe going to get GM-free meat?  The science fiction vision of meat grown in vats could be economically relevant sooner than one might think.

Earlier this month at Wired News, Alexis Madrigal wrote about the recent In Vitro Meat Symposium in Norway.  A report was presented that claimed, "Meat grown in giant tanks known as bioreactors would cost between $5,200-$5,500 a ton (3,300 to 3,500 euros)" -- more or less competitive with current European beef prices."  Madrigal reports that according to Jason Matheny at Johns Hopkins University, "The general consensus is that minced meat or ground meat products -- sausage, chicken nuggets, hamburgers -- those are within technical reach.  We have the technology to make those things at scale with existing technology."  Matheny is the founder of New Harvest, a non-profit working on producing meat substitutes.

Madrigal's story carried a skeptical tone, and suggested that commercialized in vitro meat was probably many years away.  I have been wondering whether the market would, um, serve up cultured meat sooner than that, and this week brought an interesting surprise.

People for the Ethical Treatment of Animals (PETA) just announced a US$ 1 million prize for, "The first person to come up with a method to produce commercially viable quantities of in vitro meat at competitive prices."  It may not be long before PETA writes that check.

If it is already possible to produce "meat-like products" at prices competitive with those in Europe today, then continued increases in the price of products grown on the hoof or claw should make in vitro meat even more attractive economically.  The feedstocks for meat cells grown in culture would be fairly basic, just sugars and amino acids, and possibly some lipids.  These in turn can be produced by plants, yeast, and bacteria.  In principle, a co-culture of non-GM animal cells and hacked/engineered microbes that serve as feeder cells could provide a fairly high-efficiency conversion of sunlight to meat.  That might not pass muster in Europe, but it would probably sell like hotcakes Big Macs in other countries.  And how could you tell the difference?

Pushing further down this road, how long will it be before an iGEM team produces a circuit that facilitates the differentiation and culture of stem cells from fowl, fish, and mammals to produce a better burger?  I suppose an intermediary step is to hack filamentous E. coli so that it grows to have the texture of muscle tissue for minced meat.  Those clever undergrads have already made coli smell like bananas and mint, so why not add a few more metabolic products: "Yum! Tastes just like chicken!"  Or lamb.  Or yak.  Yuck. You could even enjoy a nice "coliburger" ("bactoburger"?) that intentionally contained bacteria and not have to worry about kidney and liver damage.  (Oh, yes, this is waaaay more fun than finishing the last chapter of my book.)

Just as long as our in vitro meat isn't actually made of people (see 1:34:29).  Can't wait for the t-shirt.