An old friend and reader of this blog sent me a link to an article on biodiesel at an engineering magazine's website. Despite the technical qualifications of the author, he chose to quote uncritically from a widely-cited, but nevertheless misleading energy comparison between ethanol and gasoline. Since this specific piece of data has come up many times in the comments posted to this blog, it's worth examining its central fallacy. The study in question, from Argonne Labs, indicates that ethanol returns over 1.3 BTUs for every BTU of input, while gasoline returns only 0.8. On this basis, it concludes that ethanol is a more efficient fuel than gasoline. In fact, the same data can easily be used to show that gasoline is more efficient, merely by choosing the correct system for comparison. Getting this right is more than just a detail; it has implications for the future quantity of primary energy available for the global economy.
The first real Chemical Engineering class I took--as distinct from such scintillating pre-requisites as differential equations and organic chemistry--focused on accounting for all of the inputs and outputs of mass and energy in a system, and doing so consistently. It turns out that where you draw the "envelope" around a system largely determines the result you get. The researchers in the Argonne study chose a perfectly reasonable place to draw the envelope around ethanol, evaluating all its fossil energy inputs, including fertilizer and the fuel used in cultivation, harvesting, transportation and distillation. I don't doubt the validity of the positive balance they obtained. If you add up the fossil fuel that goes into making it, ethanol looks like a decent extender, especially if you have coal and natural gas to spare. Hamburger Helper for gasoline, in effect.
The problem comes from the choice of an arbitrary and--at least from an industry, rather than academic perspective--inappropriate way to draw the envelope for assessing gasoline's energy efficiency. Dr. Wang counts all the BTUs that went into producing the oil, getting it to a refinery, and processing it into gasoline. So far, so good, but then he adds the BTUs that were in the oil in the ground. This is analogous to tallying up the BTUs of sunshine falling on a desert. They're there, but what else are you going to do with them? This approach might be more defensible if there were important uses of petroleum that didn't involve refining it into fuels and other products. But we don't burn much unrefined oil, or use it for paving or anything else, since the days when it was sold in little bottles as "patent medicine." We nearly always refine it into fuel, lubricants, and other byproducts, and in this country we turn many of those byproducts into gasoline, too.
What happens when you draw the envelope around the gasoline system to reflect the inputs that went into producing the oil and running the refinery, but not the energy content of the oil in the ground, and weigh them against the gasoline coming out? Well, from the same presentation giving the oft-cited 1.36:1 ratio for ethanol, we see that producing 100 net BTUs of gasoline only requires about 25 BTUs of energy input, yielding an energy return ratio of roughly 4:1. That doesn't mean we should always prefer gasoline over ethanol, because in a world of expensive oil and unstable suppliers, we may need both. However, it does clearly show that gasoline is higher up the energy food chain than ethanol, which might explain why it beat ethanol hands down in the competition to fuel those new-fangled internal combustion engines a century ago.
Ardent supporters of corn ethanol will argue vehemently that I'm looking at this incorrectly. Instead of turning arabesques refuting this fairly simple calculation, though, perhaps they should focus on the need to close the substantial gap it reveals. Improvements in areas such as ethanol plant integration and bio-engineering of seed corn should steadily push the corn-ethanol balance toward a 2:1 energy return, even without the major improvements that practical cellulosic ethanol could bring. At the same time, the gasoline energy balance will deteriorate, as the world's oil supply shifts toward highly energy-intensive production sources such as oil sands, ultra-heavy oil, and ultra-deep water fields. As this energy gap narrows, the magnitude of the subsidy required to make biofuels competitive will shrink, and that'll be good for consumers and for taxpayers.
The bad news is that if that gap shrink faster from the top than the bottom, it will mean that our society's energy surplus is shrinking with it. In the same way that it took the creation of an agricultural surplus to free our ancestors from slavery to crop cycles and give them the time and spare workers to create a civilization, our civilization requires an energy surplus, in the form of sources of primary energy that return much more net energy than we put into their production. Biofuels have a long way to go to catch up to oil and gas in this regard.