Actually *reading* the article better than reporting on the reporting of the article ... by the author!
I'm not sure how the Pimentel and Patzek study first entered into the blogosphere. Surely people don't read Cornell News Service press releases, do they? But that story got picked up verbatim by several other press outlets, and now it's "science". Do you suppose it's possible that the Cornell News Service was notified of this research breakthrough by David Pimentel, a researcher at ... Cornell?
If you look on the web, you can see that Pimentel, an entymologist, has been an activist for some time. While that's not bad in and of itself, it seems safe to say, based on comments that he has made, that he assumes his conclusions are true, and then sets out to prove them. That's no different than, say, the Department of Agriculture and the Department of Energy, who also "know" that ethanol fromADM corn is a positive energy producer. The problem is that without doing the research ourselves, we can't tell who is right or wrong, and we certainly can't rely on either group to be disinterested observers.
One thing we can determine, however, is that the reporting on the subject lacks depth (DeLong might say "Why, oh why, can't we have decent science reporting..."). Consider this "money quote" included in the Cornell press release:
The Pimentel/Patzek article contains some hard-to-justify assumptions. For example, energy associated with 11 kg of stainless steel, 21 kg of steel, and 56 kg of cement is included "per 1,000 kg biodiesel oil from soybeans". An additional 160,000 kcal of "Cleanup water" (not my units) is charged for every 1,000 kg biodiesel, but no explanation is made of why this water can't be reused (or how water is measured in kcal - is that the energy to heat it? to pump it?).
By far, however, the energy and cost required to make biodiesel from soy is dominated by the price of the beans and the lime used in their production, according to the article's assumptions. Apparently, lime is only required when rotating soybeans with corn. Under circumstances where soybeans were grown exclusive of corn (for example, after oil tops $180/bbl), it would be reasonable to neglect lime from both the energy and the cost calculation. Do that, and the energy inputs fall from 11,878 to 9,064 kcal. Further, I am going to make theadmittedly rather risky assumption that wear & tear on stainless steel, steel, and concrete is negligible, and that I can recycle my "cleanup water" (by using a filter whose cost is negligible over dozens of batches). My new energy input is 8,394 kcal, and my output is still 9,000 kcal (from the note on Table 7 in the article), so I actually have a net energy output of +7%. The cost with these new assumptions is $925.10/ 1,000 kg of biodiesel, or $0.625/kg after "taking credit for the soy meal", and therefore $0.58/l using their specific gravity of .92. That compares favorably to the $0.84 they got, and is less than twice as expensive as petrodiesel. See what a difference a few assumptions make?
I don't see any process chemicals in the paper's description of the biodiesel transesterification process, so maybe he is assuming a different process or that the catalysts are 100% reusable. That would count against renewability. I also don't see any description of capturing the heat given off by the slightly exothermic reaction (that would count in favor of renewability). They didn't subtract the portion of the energy that went into the meal from their inputs, though they did subtract that from their monetary costs (in favor). They assume coal is used in the generation of electricity, but I don't see any reason why solar couldn't be used to provide either the electricity or the steam, or why the water couldn't be captured by cistern (two counts for). That's the way Mike Reynolds would do it.
And I damn sure didn't see any analysis of rapeseed, jatropha, algae or waste cellulose as biodiesel feedstocks, since all are more promising than soybeans or sunflowers. For the cellulose, I maintain that the energy inputs to the plant mass shouldn't even count since they would be committed to the production whether or not the biodiesel is made, and they are captured in that part of the plant used for food.
Dr. Doty does a much better analyzing the assumptions in the comments at Green Car Congress.
What of the lost corn? Most of it goes to cattle feed anyhow, so it would drive up the price of beef. Dr. Pimentel doesn't seem to have a problem with that. Also, it is worth pointing out to the commenters at GCC that Pimentel is hardly "in the pocket of big oil". He's a solar and wind energy advocate. However, I wonder if he knows just how hard the case for wind power is? Or whether he has considered that PV is only net energy producing over a very long period? Suppose it's 4:1 over a 20 year period. A biomass-based fuel that gets only 1.2:1 EROIE will achieve that same output over the same period of time when used iteratively (of the 1.2, plow 1 back into the next season's crops and sell the .2).
Pimentel may well be right about corn ethanol and soy biodiesel. The fact that these two products must be subsidized to be economically viable, and that ADM and others lobby for the subsidies and try to block competitors, should say something. But that doesn't mean that all biomass is untenable, so it would be nice if scientific reporting was up to snuff on this topic.
If you look on the web, you can see that Pimentel, an entymologist, has been an activist for some time. While that's not bad in and of itself, it seems safe to say, based on comments that he has made, that he assumes his conclusions are true, and then sets out to prove them. That's no different than, say, the Department of Agriculture and the Department of Energy, who also "know" that ethanol from
One thing we can determine, however, is that the reporting on the subject lacks depth (DeLong might say "Why, oh why, can't we have decent science reporting..."). Consider this "money quote" included in the Cornell press release:
"The United State desperately needs a liquid fuel replacement for oil in the near future," says Pimentel, "but producing ethanol or biodiesel from plant biomass is going down the wrong road, because you use more energy to produce these fuels than you get out from the combustion of these products."That's a fairly broad statement, but what does the actual article in question, "Ethanol Production Using Corn, Switchgrass, and Wood; Biodiesel Production Using Soybean and Sunflower", have to say about the subject?
(3) For biodiesel production, there are two problems: the relatively low yields of oil crops ranging from 1,500 kg/ha for sunflower to about 2,700 kg/ha for soybeans; sunflower averages 25.5% oil, whereas soybeans average 18% oil. In addition, the oil extraction processes for all oil crops is highly energy intensive as reported in this manuscript. Therefore, these crops are poor producers of biomass energy.So, which is it? Is the problem "biodiesel from [all] plant biomass", or "biodiesel from soybean and sunflower"? Since the set "all plants" is not limited to soybeans and sunflowers, it makes a difference, but the meme is passed through sloppy reporting, with titles such as "Corn Ethanol and Biodiesel Net Energy Losers".
The Pimentel/Patzek article contains some hard-to-justify assumptions. For example, energy associated with 11 kg of stainless steel, 21 kg of steel, and 56 kg of cement is included "per 1,000 kg biodiesel oil from soybeans". An additional 160,000 kcal of "Cleanup water" (not my units) is charged for every 1,000 kg biodiesel, but no explanation is made of why this water can't be reused (or how water is measured in kcal - is that the energy to heat it? to pump it?).
By far, however, the energy and cost required to make biodiesel from soy is dominated by the price of the beans and the lime used in their production, according to the article's assumptions. Apparently, lime is only required when rotating soybeans with corn. Under circumstances where soybeans were grown exclusive of corn (for example, after oil tops $180/bbl), it would be reasonable to neglect lime from both the energy and the cost calculation. Do that, and the energy inputs fall from 11,878 to 9,064 kcal. Further, I am going to make the
I don't see any process chemicals in the paper's description of the biodiesel transesterification process, so maybe he is assuming a different process or that the catalysts are 100% reusable. That would count against renewability. I also don't see any description of capturing the heat given off by the slightly exothermic reaction (that would count in favor of renewability). They didn't subtract the portion of the energy that went into the meal from their inputs, though they did subtract that from their monetary costs (in favor). They assume coal is used in the generation of electricity, but I don't see any reason why solar couldn't be used to provide either the electricity or the steam, or why the water couldn't be captured by cistern (two counts for). That's the way Mike Reynolds would do it.
And I damn sure didn't see any analysis of rapeseed, jatropha, algae or waste cellulose as biodiesel feedstocks, since all are more promising than soybeans or sunflowers. For the cellulose, I maintain that the energy inputs to the plant mass shouldn't even count since they would be committed to the production whether or not the biodiesel is made, and they are captured in that part of the plant used for food.
Dr. Doty does a much better analyzing the assumptions in the comments at Green Car Congress.
What of the lost corn? Most of it goes to cattle feed anyhow, so it would drive up the price of beef. Dr. Pimentel doesn't seem to have a problem with that. Also, it is worth pointing out to the commenters at GCC that Pimentel is hardly "in the pocket of big oil". He's a solar and wind energy advocate. However, I wonder if he knows just how hard the case for wind power is? Or whether he has considered that PV is only net energy producing over a very long period? Suppose it's 4:1 over a 20 year period. A biomass-based fuel that gets only 1.2:1 EROIE will achieve that same output over the same period of time when used iteratively (of the 1.2, plow 1 back into the next season's crops and sell the .2).
Pimentel may well be right about corn ethanol and soy biodiesel. The fact that these two products must be subsidized to be economically viable, and that ADM and others lobby for the subsidies and try to block competitors, should say something. But that doesn't mean that all biomass is untenable, so it would be nice if scientific reporting was up to snuff on this topic.
Labels: energy
<< Home