Surely you have heard the claim. Proponents of ethanol will claim that it takes less fossil fuels to produce a BTU of ethanol than to produce a BTU of gasoline. Here is the claim from a Minnesota Department of Agriculture site (1):
A United States Department of Agriculture (USDA), Economic Research Service Report number 814 titled “Estimating The Net Energy Balance Of Corn Ethanol: An Update” was published in July of 2002. The Conclusion states in part: “Corn ethanol is energy efficient, as indicated by an energy ratio of 1.34; that is, for every Btu dedicated to producing ethanol, there is a 34-percent energy gain.” A similar study done in 1995 indicated only a 1.24 energy ratio.
The concept of “input efficiencies for fossil energy sources” was introduced as a component of the study. This was meant to account for the fossil energy used to extract, transport and manufacture the raw material (crude oil) into the final energy product (gasoline). According to the study, gasoline has an energy ratio of 0.805. In other words, for every unit of energy dedicated to the production of gasoline there is a 19.5 percent energy loss.
In summary, the finished liquid fuel energy yield for fossil fuel dedicated to the production of ethanol is 1.34 but only 0.74 for gasoline. In other words the energy yield of ethanol is (1.34/0.74) or 81 percent greater than the comparable yield for gasoline.
I have dealt with the USDA studies in previous essays, showing the shoddy and misleading methodology they use. But let’s now examine this claim of energy efficiency. Would it surprise you to know that not only is this claim false, it is WAY FALSE?
Let’s do some quick calculations to demonstrate this. A barrel of crude oil contains 5.8 million BTUs (2) of material that will ultimately be turned into gasoline, diesel, jet fuel, etc. It is well-documented that the average energy return on energy invested (EROEI) for crude oil production is around 10/1 (3). Therefore, we will use up about 580,000 BTUs from our barrel getting it out of the ground. The other major input occurs during the refining process, and it also takes roughly 10% of the contained BTUs in the barrel of oil. The total energy input into the process is 1.16 million BTUs, and the energy output was 5.8 million BTUs. The EROEI is then 5.8 million/1.16 million, or 5/1.
For ethanol, the USDA study reference above showed that for an energy input of 77,228 BTUs, an energy output (when co-products were included) of 98,333 BTUs were generated. The EROEI is then 98,333/77,228, or 1.27/1. The efficiency of producing gasoline is then 4 times higher than for ethanol, which makes sense when you think about it.
Crude oil is a highly energy dense mixture. It is contained in underground deposits, and just needs to be pumped out of the ground. During the refining step, large amounts of water don’t need to be distilled out of the product. Contrast this to ethanol. The corn must be planted, grown, and harvested. Processing must take place to turn the corn into crude ethanol. The crude ethanol is actually mostly water, which must be removed in a highly energy intensive distillation. The final product, ethanol, contains only about 70% of the BTU value of the same volume of gasoline. So it would appear that even without doing any rigorous calculations, producing ethanol would be far less energy efficient than producing gasoline.
So, where did the claim that ethanol is more energy efficient originate? I believe it originates with researchers from Argonne National Laboratory, who developed a model (GREET) that is used to determine the energy inputs to turn crude oil into products (4). Since it will take some amount of energy to refine a barrel of crude oil, by definition the efficiency is less than 100% in the way they measured it. For example, if I have 1 BTU of energy, but it took .2 BTUs to turn it into a useable form, then the efficiency is 80%. This is the kind of calculation people use to show that the gasoline efficiency is less than 100%. However, ethanol is not measured in the same way. Look again at the example from the USDA paper, and lets do the equivalent calculation for ethanol. In that case, we got 98,333 BTUs out of the process, but we had to input 77,228 to get it out. In this case, comparing apples to apples, the efficiency of producing ethanol is just 21%. Again, gasoline is about 4 times higher.
OK, so Argonne originated the calculation. But are they really at fault here? Yes, they are. Not only did they promote the efficiency calculation for petroleum products with their GREET model, but they have proceeded to make apples and oranges comparisons in order to show ethanol in a positive light. They have themselves muddied the waters. Michael Wang, from Argonne, (and author of the GREET model) made a remarkable claim last September at The 15th Annual Symposium on Alcohol Fuels in San Diego (5). On his 4th slide , he claimed that it takes 0.74 MMBTU to make 1 MMBTU of ethanol, but 1.23 MMBTU to make 1 MMBTU of gasoline. That simply can’t be correct, as the calculations in the preceding paragraphs have shown.
Not only is his claim incorrect, but it is terribly irresponsible for someone from a government agency to make such a claim. I don’t know whether he is being intentionally misleading, but it certainly looks that way. Wang is also the co-author of the earlier USDA studies that I have critiqued and shown to be full of errors and misleading arguments. These people are publishing articles that bypass the peer review process designed to ferret out these kinds of blatant errors. I suspect a politically driven agenda in which they are putting out intentionally misleading information.
One of the reasons I haven’t written this up already, is that 2 weeks ago I sent an e-mail to Wang bringing this error to his attention. I immediately got an auto-reply saying that he was out of the office until March 31st. I have given him a week to reply and explain himself, but he has not done so. Therefore, at this time I must conclude that he knows the calculation is in error, but does not wish to address it. In the interim, ethanol proponents everywhere are pushing this false information in an effort to boost support for ethanol.
Look at the Minnesota Department of Agriculture claim again: “the energy yield of ethanol is (1.34/0.74) or 81 percent greater than the comparable yield for gasoline”. If the energy balance was really this good for ethanol and that bad for gasoline, why would anyone ever make gasoline? Where would the economics be? Why would ethanol need subsidies to compete? It should be clear that the proponents in this case are promoting false information.
1. Ethanol versus Gasoline
2. BTU Content of Common Energy Units
3. Alternative energy: evaluating our options
4. Allocation of Energy Use in Petroleum Refineries to Petroleum Products
5. Updated Energy and Greenhouse Gas Emissions Results of Fuel Ethanol