August 20, 2006
Brazil Gets 40 Percent Of Gas Vehicle Fuel From Ethanol

Even adjusted for ethanol's lower energy content per gallon ethanol is cheaper than gasoline in Brazil.

Pull up to most service stations in this country of 185 million people and you will find fuel pumps offering three choices: ethanol, gasoline or premium gasoline. The labels are slightly misleading: The gasoline varieties are blends that contain at least 20 percent ethanol. The ethanol is usually significantly cheaper -- 53 cents per liter (about $2 per gallon), compared with about 99 cents per liter for gasoline ($3.74 per gallon) in Sao Paulo this past week.

Note that ethanol contains less energy per liter or gallon. So you need a third or a half more ethanol to drive the same distance as you can with pure gasoline. But the Brazilian gasoline has already been diluted with ethanol and therefore already takes you a shorter distance than pure gasoline. So the 53 cents per liter price for pure ethanol is much cheaper per mile or kilometer driven than the 99 cents per liter gasoline/ethanol blend. Mind you, in the United States, ethanol is much more expensive and costs more than gasoline per mile travelled.

Ethanol from sugar cane has replaced 40% of gasoline in Brazil. (but see the post update at the bottom for why this is less impressive than it sounds - in a word "diesel").

Ethanol is not solely responsible for Brazil's newfound energy independence -- domestic oil exploration has exploded in recent years -- but it has replaced about 40 percent of the country's gasoline consumption, according to Caio Carvalhal, an analyst with Cambridge Energy Research Associates in Rio de Janeiro.

The Brazilian sugar cane industry funded development of cheaper ways to convert sugar cane into ethanol. The government eliminated subsidies for ethanol back in the 1980s.

Through it all, the Center for Sugarcane Technology in Sao Paulo state -- a research facility created in the early 1970s and funded by the sugar industry -- continued working to improve efficiency in ethanol production by tinkering with almost everything from the genetic structure of sugar cane varieties to the industrial components of extraction. By the time oil prices began to steadily rise in the early years of this decade, ethanol producers had reduced production costs of a liter of ethanol from about 60 cents to about 20 cents.

By contrast, the US sugar cane industry funds lobbyists to keep the US government blocking Brazilian sugar from the US market. They argue this saves US jobs. But lots (all?) of the field hands are brought in from Caribbean islands. Money provides plenty of incentive for people to deceive and use government for their benefit and the expense of others. But I digress.

You might be thinking "Hey, why not just import Brazilian ethanol and lower our cost of vehicle fuel while at the same time reducing net carbon release into the atmosphere?" Hah! When large sums of money are involved lobbyists Archer Daniels Midland and the corn farmers have got you beat. A large tariff on Brazilian ethanol makes it much more expensive in the United States.

President Bush suggested earlier this year eliminating a tariff of 54 cents per gallon of Brazilian ethanol, but corn growers and their congressional allies have stymied that idea.

I feel compelled to digress again into trade politics but only because I have a really great idea for Brazil. My advice to the Brazilians: Stop letting any of your models come to the US and pose in Victoria's Secret catalogs until the US government agrees to let in Brazilian sugar and sugar cane ethanol. American citizens might tolerate having to pay more for Breyers than Dreyers in order to get sugar rather than corn syrup as ice cream flavoring. But American men will only find the backbone they need to stand up to the corn farmers and ADM when they find out that the farmers are preventing them from looking at Gisele Bündchen, Michelle Alves, Shirley Mallmann, Isabeli Fontana, Fernanda Tavares, and Ana Beatriz Barros.

Brazilian sugarcane ethanol is about 20% cheaper than US corn ethanol to manufacture. I'm surprised that the difference in price is so small.

The U.S. Department of Agriculture released a report last month that concluded sugar is not economically feasible for a domestic ethanol industry under current conditions.

The cost of producing a gallon of ethanol from sugarcane would be $2.40 while Brazil makes it for 81 cents, the report said. U.S. corn-based ethanol costs about $1.03 per gallon to make, it said.

The price quoted for Brazilian ethanol is in gallons. The 81 cents for a gallon is lower than the about $2 per gallon quoted in the first article above. But part of that difference is due to the cost of distribution and retail sales. Perhaps there's a tax on it as well.

The demand for ethanol as an oxygenator fuel additive to lower pollution emissions has driven the price of ethanol in the US far above the manufacturing cost.

Considering that in the past three months the price of ethanol has jumped 54 percent to $3.67 a gallon -- more than double what it costs to manufacture -- companies that produce ethanol are enjoying an enormous profit, at the expense of the American consumer and taxpayer. Archer Daniels Midland, the Illinois-based conglomerate that controls nearly one-fourth of the ethanol market, will earn an estimated $1.3 billion from ethanol alone during the current fiscal year. No wonder ethanol companies are hot investments on Wall Street.

Does pure ethanol really cost that much at wholesale? If so, even with a tariff Brazilian sugar cane ethanol should be cheaper than the market price for US corn ethanol. Does anyone know what is going on behind these figures?

A Chevrolet Silverado gets only three quarters as many miles per gallon when running on ethanol.

So I'll save money if I use ethanol?

Actually, no. Ethanol contains less energy than gasoline, which means mileage is lower. In city driving, for example, the base model Chevy Silverado pickup truck gets 16 miles per gallon of gasoline, but just 12 miles per gallon of ethanol.

Here is a nice web page explaining how to compare ethanol to gasoline when looking at prices. Ethanol has another cost though: You have to stop and get gasoline more often. Time is money. Ethanol costs you more in time.

US farmers face one big problem competing with Brazilian ethanol: Brazil gets more sunshine than the United States. On the other hand, Brazil's sugar cane ethanol cost advantage is not that large. The development of cellulosic technologies to extract all the energy out of the cellulose portion of plants would cut US ethanol costs substantially. However, it would do the same to sugar cane.

Question: Is the currently unusable cellulose fraction of a corn plant larger than the currently unusable fraction of a sugar cane? If so, then cellulosic technology will reduce Brazil's ethanol manufacturing cost advantage.

Another question: How much rain forest will get cut down as Brazil expands ethanol production?

In 2005 the United States used 9.125 million barrels of gasoline per day At 42 gallons per barrel that is 383.25 million gallons per day and times 365.25 days per year that is 139.982 billion gallons per year. To replace that with biomass energy would require perhaps 210 billion gallons of ethanol. Let us put that into perspective. The optimistic view is for ethanol production to expand to meet the demand for about 7% of the gasoline vehicle liquid fuel market.

If oil prices remain high, then U.S. fuel consumption of ethanol could at least double from the 2005 level of four billion gallons. But industry executives like Archer Daniels Chief Executive Patricia Woertz have set their sights even higher. Last week, in announcing the company's record earnings for its June 2006 fiscal year, Woertz said ethanol demand could triple.

"It looks like it has room to grow to 14 billion or 15 billion [gallons per year]," she said, "which is a full 10% blend in the gasoline pool in the United States." Unfortunately, before ethanol refiners can reach that goal, they might reach the limits of the country's corn supply. America's entire corn crop would satisfy just 12% of gasoline consumption, leaving no corn to feed livestock and humans. So there just won't be enough corn for corn ethanol to grow from a fuel additive into a large-scale substitute for fossil fuel. Crop years vary, too.

In 2006 5 billion gallons of ethanol will be produced in the United States.

There are 101 ethanol production plants in the nation, producing 5 billion gallons of ethanol per year, Siekman said.

Thirty-three plants to produce another 2 billion gallons are in the works in various areas throughout the country, Siekman said.

The 5 billion gallons represents about 2 and a half percent of the energy used to propel gasoline-burning vehicles (as distinct from diesel powered vehicles). We need cellulosic technologies to allow ethanol to displace most gasoline from the market. I'm guessing we'll see cellulosic technologies move into ethanol production in the next 10 years. Oil prices are so high that lots of groups have plenty of incentives to solve the problems for how to release the energy in cellulose.

High oil prices are going to continue to drive a shift to alternatives. For transportation ethanol looks to be the alternative that can ramp up most quickly. As better battery technologies come out hybrids will grow as well. Eventually when batteries improve enough pluggable hybrids will set the stage where ethanol will compete with wall socket electricity to power vehicles. Therefore the use of gasoline looks set to peak before oil production peaks.

Update: The American Petroleum Institute argues that the biggest cause of Brazil's energy independence is an increase in domestic oil production.

The U.S. is producing slightly more ethanol than Brazil, however gasoline demand in Brazil is only 4.28 billion gallons compared to the U.S. where gasoline demand is 140 billion gallons (23 times that in Brazil). The U.S. can't build an ethanol market based on sugarcane. Nor can the U.S. ramp up corn-based ethanol in the same way Brazil increased ethanol use with sugarcane. Brazil energy independence is due to increased crude oil production - not ethanol.


How did Brazil achieve energy independence? Not with ethanol but by increasing crude oil production. Brazil’s increase in crude oil production over the 2004 to 2005 period was 9 times larger than its increase in ethanol production over the same time period, and was 4 times larger over the 2000 to 2005 period. Clearly, it has been Brazil’s increased crude production, particularly over the 2004 to 2005 time frame, which has been the dominant factor in pushing Brazil towards energy independence.

Robert Rapier also says the ethanol success of Brazil is very much exaggerated.

On the production side, in 2005 Brazil produced 627 million barrels of oil, for an annual per capita oil production of 3.4 barrels per person. The U.S. produced 2.5 billion barrels of oil in 2005, for an annual per capita oil production of 8.4 barrels per person. The annual shortfall between oil consumption and oil production in Brazil was 0.2 barrels per person in 2005. In the U.S., the shortfall between consumption and production was much larger at 16.9 barrels per person.

The question then arises: “Just how much did widespread use of ethanol in Brazil contribute toward their energy independence?” The answer is: “Not much”. In 2005, Brazil produced 4.8 billion gallons of ethanol, or 114 million barrels. However, a barrel of ethanol contains approximately 3.5 million BTUs, and a barrel of oil contains approximately 6 million BTUs. Therefore, 114 million barrels of ethanol only displaced 67 million barrels of oil, around 10% of Brazil’s oil consumption. In other words, Brazil’s energy independence miracle was 10% ethanol and 90% domestic crude oil production. Brazil did not farm their way to energy independence.

Over on his R Squared blog Rapier says that diesel is used for half of all vehicle fuels in Brazil.

According to a March 2006 presentation by the Brazilian Ministry of Mines and Energy, the actual breakdown of vehicle fuels in Brazil at the present time (by volume) is 53.9% diesel, 26.2% gasoline, 17% ethanol and 2.9% natural gas.

17% is a lot less than 40%. Also, that is 17% of a small total amount of fuel and a small per capita use of fuel as compared to the United States.

Thanks to The Ergosphere's Engineer-Poet for alerting me to the relatively large role that diesel plays in Brazil for vehicle fuel.

Update II: Engineer-Poet makes an argument on why ethanol from biomass is a bad idea. I happen to agree that biomass energy has big downsides. But my guess is that coming advances in cellulosic technology will so lower the cost of biomass ethanol that ethanol usage will increase greatly in the next 10 years. I do not treat that as a happy prospect. Large scale biomass energy production will cause humanity to compete (too successfully) with nature and take habitat away from other species. Since I'm fond of other species I'm not happy about that. I'd much prefer we use nuclear, solar, and other energy sources that use much smaller ecological footprints.

Share |      Randall Parker, 2006 August 20 01:24 PM  Energy Biomass

Engineer-Poet said at August 20, 2006 3:14 PM:

Your headline is misleading, and the situation is worse than you paint:  Brazil may get 40% of its non-diesel liquid motor fuel volume from ethanol, but 52.9% of Brazil's motor fuel is diesel and another 2.9% is natural gas (26.2% gasoline and 17% ethanol).  On a per-energy basis this works out to about 10% ethanol for transportation fuel.  On an economy-wide basis, it's going to be much smaller.

People touting ethanol as our solution are either pushing for the fossil-fuel lobby, the farm lobby or are just unable to handle quantitative information.

odograph said at August 20, 2006 4:47 PM:

I think you should update your title so that it says 10%, with 40% struck out, just so that the casual skimmer gets the message. (I was going to illustrate, but I guess comments don't get to use the strike tag.)

Wolf-Dog said at August 21, 2006 7:11 AM:

Correct me if I am wrong, but it seems that burning biomass to spin turbines that will generate electricity to charge the batteries of electric vehicles, would be 10 times more efficient.

Robert McLeod said at August 21, 2006 10:41 AM:

Yeah but why would you want to produce electricity by engaging in soil mining? There are so many better ways to produce electricity and heat than burning the products of our arable land. Biomass only makes (some) sense if you need the versatility, ease of handling, and energy density of hydrocarbon fuels. It gives you a product that is chemically relatively close to your desired product so you don't need to introduce too much energy to arrive at a fungible liquid fuel.

James Bowery said at August 21, 2006 2:41 PM:

Another question: How much rain forest will get cut down as Brazil expands ethanol production?

If the US is willing to sacrifice oil availability invading Iraq and possibly Iran, why not sacrifice availability of sugarcane from Brazil and invade to stop the destruction of the rainforest?

Engineer-Poet said at August 21, 2006 3:28 PM:

Burning biomass to spin turbines is more like 3-4 times as efficient as conversion to liquid fuels.

"Soil mining" need not be a problem.  We must grow plenty of crops, harvest wood, etc. anyway, and the byproducts from these functions can yield quite a bit of biomass without disturbing a bit of soil.  To the extent that e.g. carbonization of biomass for gas yields excess charcoal, it can be used to add carbon and nutrient-holding capabilities that natural soils never had.

Randall Parker said at August 21, 2006 4:41 PM:


If MovableType filters out the strike tag in comment posts then that is news to me.


Since a hundred million cars on US roads burn liquid fuel the market for liquid fuel is immense. Given the high price of oil and the lower cost of biomass ethanol I expect the market will continue to allocate biomass toward production of liquid fuel.

People can debate the wisdom of biomass and I certainly question it. But as long as production cost of a gallon of ethanol remains less than two thirds the production cost of gasoline I expect to see an increase in ethanol usage. Ethanol is the fuel of the future because the market says so.

Brett Bellmore said at August 21, 2006 4:42 PM:

I fully understand that there are transition costs involved in changing over to a fuel that isn't largely interchangable with gasoline or diesel, but internal combustion engines are perfectly capable of being built to run off of cellulose without chemical alteration. Diesel engines have been made to burn powdered solids and slurries. Though admittedly it's easier to use such fuels in external combustion engines such as steam or Sterling engines.

I wonder how much of a fuel cost advantage would be necessary to make a car powered by a solid fuel such as powedered biomass marketable? It would not be significantly more inconvenient to use than an electric vehicle, I would imagine.

To ease the transition, we might imagine equipping one's home to be heated by the same fuel, and the storage system for your furnace incorporating some mechanism for conveniently refueling the vehicle. You'd be set, then, for any trip that didn't require refueling away from home. An optional liquid fuel tank for those long trips, (Maybe a bladder that could fill the solid fuel hopper in emergencies?) and you'd have a system that was actually more convenient than electric cars.

Wolf-Dog said at August 21, 2006 8:03 PM:

Engineer Poet wrote:
"Burning biomass to spin turbines is more like 3-4 times as efficient as conversion to liquid fuels."

But are you taking into account the fact that gasoline powered cars are only 20 % efficient, and that
electric powered cars can be more than 80 % efficient?

1)It is said that if you burn one gallon of gasoline in a turbine to generate electricity to charge the batteries of an electric car, then you would get the equivalent of 100 miles per gallon, which is
already about 3-4 times more efficient than an an ordinary internal combustion car.
2)Also burning biomass directly to spin a turbine is a lot more efficient than converting the biomass to ethanol.

Combining 1) and 2) above, this is why perhaps an electric car might be 10 times more efficient than
an internal combustion car running on ethanol.

Any comments?

Engineer-Poet said at August 21, 2006 9:13 PM:

Yes, Wolf-Dog, I'm accounting for that.

Randall, I'm sure you're shooting low for motor-fuel consuming devices (~200 million registered light vehicles in the USA), but I'm pretty sure liquid fuels aren't the least-cost way to the future.

And I'm off until Friday night.  Bye!

Dezakin said at August 22, 2006 2:53 AM:

"But as long as production cost of a gallon of ethanol remains less than two thirds the production cost of gasoline I expect to see an increase in ethanol usage. Ethanol is the fuel of the future because the market says so."

Cite the study that states this. I saw the origian csm article where they quote some guy of dubious credentials as saying that, but the real numbers I strongly suspect arent nearly so generous and are almost definately not scalable.

Randall Parker said at August 22, 2006 4:47 PM:

Dezakin, here's from a USDA report from July 2006 entitled "THE ECONOMIC FEASIBILITY OF ETHANOL PRODUCTION

Here is the URL:

First, here are the feedstock costs:

The estimated 2003-05 average quantities and values of byproducts, gallons of ethanol produced per unit of feedstock, and net feedstock cost per gallon for converting corn and sugar feedstocks into ethanol are presented in Table 16. With an average market price for corn of $2.16 per bushel during 2003-05, the estimated net feedstock cost per gallon of ethanol produced from corn is $0.53 for a dry mill plant and about $0.40 for a wet mill plant, factoring in byproduct revenues. The net feedstock costs for sugar beets, sugarcane, raw sugar, refined sugar, and molasses account for the acquisition cost of the feedstock and do not include credits for beet pulp, bagasse, vinasse, and carbon dioxide produced during the conversion into ethanol. The value of these byproducts is included in the cost of processing sugar feedstocks into ethanol.

Of the five sugar feedstocks evaluated in this report, molasses is the most cost competitive with corn. Molasses net feedstock cost is about $0.91 per gallon. Net feedstock costs per gallon of ethanol made from sugar beets and sugarcane were $1.58 and $1.48 per gallon, respectively. These feedstock cost estimates are based on the 2003-05 average sugar recovery rates of 15.5 percent for sugar beets and 12.26 percent for sugarcane and on 2003-04 sugarcane and sugar beet market prices per ton. Valuing raw cane sugar and refined beet sugar at market prices, the estimated feedstock cost of using these materials for conversion to ethanol is estimated at $3.12 and $3.61 per gallon, respectively.

According to a 2003 survey, approximately 61 percent of world ethanol production is being produced from sugar crops (Berg, 2004). Ethanol production using sugar as a feedstock is economically feasible in countries like Brazil due to several factors including the relatively low price of raw sugar on the world market in most years and the use of molasses as a major feedstock. Table 17 presents U.S. and world sugar market prices for the 1991-2005 period. Estimated feedstock cost per gallon of ethanol produced has been estimated at $0.30 per gallon for ethanol made from sugarcane in Brazil, compared with $0.97 per gallon for ethanol made from sugar beets in France (Berg, 2004).

But there are production costs as well. Well, total costs are cheap! Multiplied by 1.5 per gallong to adjust for the lower energy content of corn per gallon as compared to gasoline the total costs are still cheap cheap cheap!

Processing costs of converting corn and grain sorghum to ethanol are very similar. Currently in the United States, a very small amount of grain sorghum is used in production of ethanol. Ethanol plants have been surveyed by USDA in recent years to obtain data to estimate average ethanol production costs. In late 1999 and early 2000, USDA surveyed 28 ethanol plants, both wet and dry mill, to estimate their 1998 costs of production (Shapouri, Gallagher and Graboski). These ethanol plants processed more than 400 million bushels of corn and sorghum in 1998 to produce more than 1.1 billion gallons of ethanol. The average variable cost of production of ethanol was estimated at 93.9 cents per gallon. The net feedstock cost averaged approximately 53 cents per gallon for dry mill plants and 48 cents per gallon for wet mill plants. In 2003, USDA surveyed 21 dry mill ethanol plants to estimate their 2002 production costs (Shapouri and Gallagher). These plants produced 550 million gallons of ethanol in 2002. Total production costs, including feedstock costs, averaged 95.7 cents per gallon. Net feedstock costs ranged from 39 to 68 cents per gallon for the plants surveyed.

Ethanol costs of production for the wet milling process were updated from the 1998 estimates to values for 2003-05. Prices paid indices for intermediate materials for energy and other inputs, employment cost index for manufacturing wages, and labor benefits were used to update ethanol production costs for the wet milling process from 1998 to 2003-05. Dry milling ethanol costs of production were updated from 2002 base values to 2003-05 utilizing the above indices used in updating wet mill ethanol costs of production.

Ethanol processing costs per gallon of ethanol produced by the wet milling process increased from $0.46 in 1998 to $0.62 in 2004 and to $0.70 in 2005 (Table 18). Electricity and fuel costs increased from $0.11 per gallon in 1998 to $0.21 cents per gallon in 2005. The net corn cost per gallon of ethanol declined significantly from $0.48 per gallon in 1998 to $0.30 in 2005. Lower prices for corn in 2005, higher prices for corn oil and corn gluten meal increased the value of byproduct credits and lowered the net corn feedstock cost. High costs of energy in 2004 and 2005 were offset by lower prices for corn and higher prices for byproducts.

Ethanol processing costs per gallon for dry mill plants increased from $0.41 per gallon in 2002 to $0.58 per gallon in 2005 (Table 19). Higher natural gas and electricity prices increased the energy expenses used in the production of ethanol from 17 cents per gallon in 2002 to 27 cents per gallon in 2005. Corn prices declined in 2005 due to very large ending stocks. Corn costs per gallon of ethanol declined from $0.89 per gallon in 2004 to $0.71 cents in 2005. The value of byproduct credits declined from 30 cents per gallon in 2003 and 2004 to about 22 cents per gallon in 2005.

Again, those low prices as compared to gasoline (which I think has a production cost over $2 per gallon at this point) mean that ethanol production will keep rising. Brazil can sell as much as it can make since the export market beckons. Cellulosic technologies will drive down ethanol production costs even further.

I'm not a fan of biomass energy. I've stated my reasons why. But the economics of biomass ethanol as compared to gasoline mean that biomass ethanol has a big growing future, with or without government subsidies.

I expect biotechnology to continually reduce biomass production costs. I also expect chemical tech and biotech advances to reduce the cost of cellulosic ethanol below existing production costs.

Dezakin said at August 22, 2006 5:35 PM:

These seem to be numbers that reflect price distortions from subsidies, and unstable ones at that.

All that has to happen is a ratcheting up of demand, and the corn price (as well as any other feedstock) spikes through the roof.

simon said at August 23, 2006 8:11 AM:

It seems to me that there's already quite a bit of infrastructure in place in Brazil to process sugar cane, mostly into sugar. So there's probably little need to invest in a whole new network of factories to make fuel ethanol. That's probably not the case in the US, where to process, switch grass, corn or the other flavours of the month, would probably need a load of stuff building and logistics organised. Wolf-Dog and Engineer-Poet's point about burning biomass to generate power might sound like a waste of resources, but its going to be heading towards carbon neutrality, and with the greater efficiency of electric motors looks sensible to me. Now if we can only get the range and speed of electric vehicles up to gasoline powered vehicles, you might persuade people to buy them....

Heiko Gerhauser said at August 23, 2006 9:17 AM:

The numbers are likely on an energy content basis already. IEA statistics for 2003 are:

They give 2003 biofuels production as just under 10 million tonnes (around 0.2 million barrels per day) and motor gasoline at just under 12 million tonnes. Assume similar density, but 27% lower energy content

And on an energy content basis, biofuels in 2003 would have been around 37% of combined gasoline/ethanol demand. That should be above 40% by now.

E-P hates ethanol and has a rather biased view. His solution is to go for batteries. Why those aren't making much headway, either in battery electric vehicles or plug-ins (no major manufacturer is offering them, Toyota, Honda, GM all agree that while there may be promise, plug-ins just aren't viable based on present battery technology), he chooses to ignore.

Randall's analysis is spot on, ethanol is price competitive today and that's why production is shooting up.

John Thacker said at August 23, 2006 9:28 AM:

Does pure ethanol really cost that much at wholesale?

Right now, yes, but because of a recent massive jump. Not only the recent phaseout of MTBE, but also the requirement that more gasoline be sold with up to 20% ethanol in the recent energy bill caused a massive increase in ethanol demand. This caused prices to skyrocket-- but note all those plants that are being built to refine ethanol to increase production. It'll come back down from those absurd scarcity-induced highs once those are online.

Coyote said at August 23, 2006 11:30 AM:

Brazil presumably uses fuel in the production of Sugar Cane and the conversion of the cane to ethanol. Your estimates of ethanol substitution therefore strike me as inflated, since if ethanol production and the sugar cane production that supports were to just go away, then the total fuel requirement for the country would go down as well. There would be a gap in energy consumption unment by oil, I guess, but the gap would be less than the current ethanol consumption. If you believe the studies of the amount of energy that is required to make ethanol, the gap would be a LOT less.

Dezakin said at August 23, 2006 1:27 PM:

I just cant see how corn based ethanol ever makes sense compared to algae based biofuels, oil, or coal.

consider the production costs of oil (at most 20 dollars per barrel, often around 2 dollars per barrel), and the capacity (85 million barrels per day)

The price imbalance today suggests that ethanol is a poor substitute at best, and certainly can't ramp up capacity; And when you compare production costs of algae based biodiesel, I cant see how its anything but a giant givaway to farm lobbies. Consider the source of the study as well... usda, not doe.

Randall Parker said at August 23, 2006 4:26 PM:


I was quite surprised to read the real production costs for ethanol. The costs are much lower than I expected.

People who want far better battery technology (and I have repeatedly argued for this) need to recognize that right now batteries cost too much and weigh too much. When I separate my normative judgements from the economic facts it seems clear to me that ethanol costs much less than gasoline to produce and therefore ethanol production is going to skyrocket.


You assert - without evidence - that the production costs I've cited include subsidies. The subsidies exist. But I see no evidence that the production costs I've cited from the USDA report include the subsidies. The way the report breaks down the production costs suggests this is not the case.

I showed you a credible and quite detailed source of information for biomass ethanol production costs. How about doing the same with algae based biodiesel?

Corn ethanol makes economic sense because 1.5 times $1 gallon is still well less than the production cost of gasoline. At $65 per barrel of oil gasoline costs $2.20 to produce.

The economics make sense. Middle East tensions and other factors have pushed the oil price higher: In June it averaged $65 a barrel. At that price, it cost $2.20 to produce a gallon of gasoline - about $1.56 for the oil itself and 64 cents for refining costs, according to the federal Energy Information Administration.

$2.20 is more money than $1.50. Coal diesel might be able to compete. But ethanol's production cost will fall as cellulosic technology matures and genetically engineered plants yield more per acre with less inputs.

Randall Parker said at August 23, 2006 4:46 PM:


Some of the energy used in corn and ethanol production is in the form electricty which is almost entirely generated by using energy sources other than oil. Also, the big run-up in oil costs had little effect on corn costs because other costs of corn production fell.

Biotechnological advances will decrease the need for ammonia for fertilizer and therefore will reduce the use of natural gas in corn production. The shift toward no-till farming also reduces energy used by reducing farm tractor use.

If energy costs played as large a role in corn production as some analysts claim then it would cost a lot more to grow corn than it does.

Randall Parker said at August 23, 2006 5:35 PM:

Better batteries would be great. But when will we get them? A New York Times article entitled "Need for Battery Power Runs Into Basic Hurdles of Science" reports battery capacity is making slow but steady progress.

Microcells intrigue companies that make laptops, cellphones and other portable devices because they can store far more energy than comparably sized batteries. Methanol-based microcells, for instance, have roughly 10 times the energy density, creating the prospect of wireless laptops that could run all day without recharging, according to Rick Cooper, vice president for business development of PolyFuel Inc. The company, based in Mountain View, Calif., supplies components to several Asian manufacturers that have been working on such devices.

“The energy capacity of batteries is increasing 5 percent to 8 percent annually, but demand is increasing exponentially,” Mr. Cooper said.

The tweaking of materials and chemicals in the lithium-ion battery will extend its usefulness for at least another decade or more, said Gao Liu, a scientist at Lawrence Berkeley National Laboratory. He expects innovation to come slowly.

I see liquid fuel stretching many years into the future. Though the lower energy density of ethanol will narrow the advantage of liquid fuel over batteries.

Dezakin said at August 24, 2006 12:41 AM:

You assert - without evidence - that the production costs I've cited include subsidies. The subsidies exist. But I see no evidence that the production costs I've cited from the USDA report include the subsidies.

Subsidies drive down the cost of corn but artificially inflating the supply.

It doesnt matter if farmer X is getting no subsidies at all if farmer Y is and Y is still unloading into the market. Given the US farm bill is in the tens of billions, one can assume there are some price distortions in the cost of corn on the market.

"Corn ethanol makes economic sense because 1.5 times $1 gallon is still well less than the production cost of gasoline. At $65 per barrel of oil gasoline costs $2.20 to produce."

Gasoline and ethanol are not substitutes, and this is a gross distortion of the economics. Generally, you compare the production cost of oil. Its not that much extra to send it to the refinery. Your ethanol has to compete against new oil exploration, coal liquefaction, and tar sands, all of which have lower production costs. Comparing it against the crude price that the refinery pays simply isnt appropriate.

The production costs today for coal liquefaction are cheaper, and that produces gasoline and diesel fuel, a far higher value fuel.

I showed you a credible and quite detailed source of information for biomass ethanol production costs. How about doing the same with algae based biodiesel?

There still is a place for ethanol; cellulistic ethanol can be very useful for waste reclamation and certainly has a market in making biodiesel. But corn or sugar based ethanol simply cant expand to take up large shares of the fuel market, because there isnt the production capacity on this planet.

Randall Parker said at August 26, 2006 1:32 PM:


You offer no price data on the algae biodiesel.

Yes, of course the farm subsidies distort the corn market. But production costs are low.

The US 2006 corn crop will be 11 billion bushels.

USDA on Friday projected farmers this season will average 152.2 bushels per acre nationwide for a total production of 10.98 billion bushels of corn.

Corn sells for about $2 per bushel for about a $22 billion crop. (the price varies but these numbers close enough for this discussion)

Corn subsidies in 2004 were about $4 billion per bushel. That works out to about 36 cents per bushel. But some of that money goes to administration and other ways that do not lower the cost of corn. So remove the subsidies and the effect on prices would be minimal - and certainly not enough to make corn ethanol uncompetitive.

See above where corn feedstock costs as an input to corn ethanol manufacture ranges from 40 cents to 53 cents a gallon. Even if an end to subsidies increased per bushel costs by 36 cents that'd only be 18%. 18% more of 53 cents feedstock costs would only push the price up for feedstock costs by about 9 cents a gallon. $1.09 a gallon would become $1.63 for a gasoline equivalent amount of energy. That's still cheaper than the $2.20 it costs to produce a gallon of gasoline.

Yes, 1.5 gallons of ethanol is an effective substitute for a gallon of gasoline. The ethanol will move a car down the road. So why claim it is not a substitute?

Refining is not much extra? So far you've made a string of assertions and haven't provided the real numbers of back up any of them. I've provided sources for my numbers.

Biodiesel would certainly be preferable to me over ethanol because it reduces the need to stop at gas stations. But soy oil and canola/rapeseed oils cost more as energy sources.

As for the ethanol subsidy, Biodiesel is getting used due to a 99 cent a gallon subsidy on it

Biodiesel gets the same mileage per gallon as regular diesel, O'Rourke said, and a government rebate of 99 cents per gallon to those who use alternative fuels keeps the cost in line with the price of regular diesel, which he says averages $2.99 per gallon nationwide. That comes in handy for the 30 to 40 tours the company works per year.
Randall Parker said at August 26, 2006 2:10 PM:

A 2004 analysis from the US Department of Energy web site shows biodiesel gets lots of subsidies:

For the past several years, the USDA has offered grants for biodiesel production through the Commodity Credit Corporation (CCC). The CCC payments for expansion of biodiesel production in the fiscal years37 2004-06 are $1.45-$1.47 (2002 dollars) per gallon for soybean oil biodiesel (Table 4) and 89-91 cents per gallon for yellow grease biodiesel (Table 5). Base production payments apply to production up to the level of the prior fiscal year, and additional production payments are for production above the level of the prior fiscal year. CCC payments for producers with output levels of 65 million gallons per year or less are shown in Tables 4 and 5. Payments for output levels above 65 million gallons per year are approximately 30 percent lower than the values shown in Tables 4 and 5.

The CCC payments effectively reduce the variable cost of additional soybean oil and yellow grease biodiesel to $1.10 and 53 cents per gallon, respectively, in fiscal year 2004. Additional units produced in fiscal year 2004, however, become base units in fiscal year 2005 and are eligible only for much smaller, and declining, base production payments. The variable cost of soybean oil and yellow grease biodiesel added in fiscal year 2004 jumps to $2.32 and $1.27 per gallon, respectively, in fiscal year 2005.

Biodiesel has energy costs for creating it that are close to ethanol's cost per gallon. Though adjusted for energy content per gallon the biodiesel energy costs at the processing stage seem lower (ignoring farm energy costs):

NREL provided estimates of other components of biodiesel production costs, based on transesterification of oil with methyl alcohol catalyzed by sodium hydroxide, yielding methyl esters (biodiesel) and glycerol. Operating expenses were estimated at 31 cents per gallon (2002 cents), excluding the cost of the oil or grease and energy, and the sale of the glycerol was estimated to reduce the cost by 15 cents per gallon of biodiesel.29

The biodiesel production process uses, for each gallon, 0.083 kilowatthours of electricity30 and 38,300 British thermal units (Btu) of natural gas.31,32 EIA estimates energy costs (in 2002 cents) of 18 cents per gallon in 2004 and 16 cents per gallon in 2005 and 2006.33 A new biodiesel plant is estimated to cost $1.04 per annual gallon of capacity. EIA assumes that the plant is financed by equity with an annualized return of 10 percent over 15 years. Treating the hypothetical income stream as an annuity over the 15 years, the estimated capital cost is $1.36 million per year, or 13.6 cents per gallon (2002 cents) at full output.

The article shows the production cost of soy oil biodiesel as at about $2.50 a gallon. That's higher than the production cost of gasoline at $65 per barrel oil. The yellow grease biodiesel costs less. But the supply of waste grease is small enough that the potential for scaling up yellow grease biodiesel is very limited.

Also, biodiesel suffers to a lesser extent from the same problem that ethanol suffers from, biodiesel gets lower miles per gallon:

Another disadvantage of biodiesel is that it tends to reduce fuel economy. Energy efficiency is the percentage of the fuel’s thermal energy that is delivered as engine output, and biodiesel has shown no significant effect on the energy efficiency of any test engine. Volumetric efficiency, a measure that is more familiar to most vehicle users, usually is expressed as miles traveled per gallon of fuel (or kilometers per liter of fuel). The energy content per gallon of biodiesel is approximately 11 percent lower than that of petroleum diesel.14 Vehicles running on B20 are therefore expected to achieve 2.2 percent (20 percent x 11 percent) fewer miles per gallon of fuel.

Biodiesel also worsens the biggest general diesel emissions problem: oxides of nitrogen. Though blending it with some other stuff can cut back on the emissions. See the previous link for the details.

Engineer-Poet said at August 26, 2006 3:26 PM:

The potential for scaling up corn ethanol production is rather small, too.  All the economics of corn ethanol depend on the use of the spent grains (DDG) as animal feed; without that, the fuel product has to make up the full cost of all the inputs and the price skyrockets.  Another factor is the price of fertilizer and distillation heat; the shrinking supply of N. American natural gas has driven both of them up steeply.  There are ways around this (using coal instead of natural gas to make nitrates, distilling with spent steam from powerplants instead of dedicated boilers) but they're not in wide use.

Even if the market for DDG couldn't be saturated, the projected 11 billion bushels of corn would produce a mere 27.5 billion gallons of ethanol.  That's about 18.3 billion GGE, compared to US consumption of ~140 billion gallons/year.  It's no silver bullet; at best, it's a silver BB.  Cellulose from new crops might provide enough feedstock, but it's a very close thing.  That's a problem, because we need room for expansion.

Regarding batteries, claims that they "aren't ready" are somewhere between nonsense born of ignorance and outright lies.  The Tesla roadster and DeWalt tools with A123Systems cells are proof that the batteries we need are here, and they're already cheaper and more available than hydrogen fuel cells.  Several different technologies have eliminated the thermal runaway issue and greatly improved the specific power, charging rate and both the cycle and calendar lives.

This isn't just talk.  Hybrid manufacturers are already planning to switch from NiMH to Li-ion as a consequence.

The problems with ethanol are that it requires some kind of biomass as a feedstock, the carbon output in the fuel proper cannot be sequestered, and the end-to-end efficiency is quite low (under 10%).  Batteries can use electricity from any source, carbon used in stationary powerplants is far simpler to sequester (an issue I know Randall doesn't take seriously... yet) and the end-to-end efficiency can be far higher.

(psst... if you aren't going to allow the <acronym> tag, at least stop stripping it from the user's text so they can put the text in parentheses or something.)

Randall Parker said at August 26, 2006 5:14 PM:


If the corn feedstock cost of ethanol doubled then it would increase the per gallon cost of corn ethanol from the current $1 (see above) by 20 to 26.5 cents. Translated into gasoline terms it would increase the cost by 30 to 40 cents per gasoline gallon. That would raise the gasoline equivalent cost of corn to $1.80 to $1.90 per gallon of gasoline. That is still way less than the current production cost of gasoline.

A tripling of the cost of corn feedstock for ethanol would still leave it competitive with gasoline. That potential will fund a large expansion in corn production. Marginal increases in production will cost more per gallon. But the cost gap between corn ethanol and gasoline is large enough to fund expansion of corn ethanol production at higher marginal costs.

What can stop a big expansion in corn and corn ethanol production at this point? I see a few possibilities:

1) A deep lasting recession that cuts total energy demand. But the recession would just delay eventual corn and corn ethanol production expansion.

2) A lowering of tariffs against sugar cane ethanol imports. But the corn farmers and ADM ought to be able to block that.

3) A scaling up of coal to liquid production. We'd shift toward diesel for vehicles. The recently announced GM diesel engine that can meet all tough diesel emissions standards will help make that possible. Coal diesel can probably compete with ethanol on price. But we'd get a lot more pollution. The first coal-diesel plant has been announced.

4) Cheap cellulosic technology. This might shift biomass ethanol production in the direction of other plant sources. Or maybe full corn stocks would be used as feedstock. It is not clear to me whether corn or switch grass would get used for cellulosic technology ethanol production. Anyone have insights on this?

Current gasoline production costs make corn ethanol and coal diesel cheaper than gasoline. Why won't these both expand greatly in response? The cost gap is large enough to pay for big expansion in either direction.

Randall Parker said at August 26, 2006 5:29 PM:

As for the problems with fossil fuels inputs to corn and corn ethanol production: You need to separate your normative judgement from your economic analysis. Ethanol is cheaper to make than gasoline. Ethanol production could be expanded greatly before the marginal cost could exceed the marginal cost of gasoline production.

One company has already announced plans to start producing ammonia fertilizer from coal.

Farmers will respond to higher energy prices by finding ways to produce a bushel of corn with less energy inputs. Biotech advances will help them do that.

Cheaper and higher energy density batteries are an eventual source of competition for ethanol. But right now batteries are not doing much to dampen demand for liquid fuels. When will that picture change? The price and physical shape of the future Tesla suggests that day is not real soon.

Engineer-Poet said at August 26, 2006 9:27 PM:

FWIW, both Robert Rapier and Robert McCleod are ripping Heiko's arguments apart too.

At perhaps 2.75 gallons of ethanol per bushel, doubling the price of corn from $2.50 to $5.00 would increase the ethanol price by nearly a dollar a gallon.  Fertilizer prices have a knock-on effect:  farmers are already shifting production away from corn to crops which require little or no nitrogen, like soybeans.  That's the problem with corn; its inputs are largely fossil-derived, so its price (and the price of the ethanol produced from it) will track the price of fossil fuels.  IIRC, the price of ethanol is currently over $3.00/gallon; it is not competitive with gasoline, and never has been.

The only things which keep ethanol in use at all are the oxygenate mandates and the blender's tax credit.  If those were dropped, ethanol use would disappear very rapidly.

I believe it likely that the market price of corn is about to get as uncertain as the market price of oil is now.  Coupling the price of corn to the price of oil is bound to do that.  And this is going to have political implications:  when the price of food shoots up in response to the price of oil because demand-pull inflation is added to cost-push inflation, the public will rebel.  Even people who don't drive have to eat.  They'll demand relief, and the ethanol mandates (and maybe even the credits) will be repealed.  That will be the end of ethanol from corn.

This appears likely because of global warming.  Each 1° increase in average temperature decreases grain yield by about 10%, so the surpluses which are the sole political justification for corn ethanol are going to disappear.

Ethanol from cellulose (available both as byproducts and as waste) would only be affected by the productivity issues.  On the other hand, cellulose to ethanol is between 45% and 60% efficient, after which you have the average 14.9% efficiency of vehicles; the end-to-end efficiency is less than 9% and perhaps under 7%.  If you burned the biomass in combined-cycle gas turbines to make electricity, you'd get closer to 50%; with the electrical grid, batteries and vehicles probably being 60% or better, your throughput would be at least 30%.  If you used direct-carbon fuel cells, my BOTE calculation says that you could reasonably 70% electric efficiency[1] and 42% or better throughput.

There's a reason Vinod Khosla & Co. aren't pushing this, though.  Ethanol would be up against expensive oil; even if it was expensive to make, the scarcity would guarantee profits for the makers.  Making electricity from biomass could eliminate coal and natural gas, but the availability of other sources eliminates the scarcity and drives down the price and the profits.  This is great for consumers and the economy, but bad for venture capitalists like Khosla.

I don't think public policy should be aimed at creating or maintaining inefficiencies so that certain investors can see bigger profits.  Do you?

[1] If the biomass is carbonized in an integrated system, roughly half the chemical energy is yielded as hot gas and the other half as charcoal.  DCFC's can convert the charcoal to electricity at 80% or better efficiency, for a yield of 40%.  The waste heat from the DCFC's, plus the heat and chemical energy in the gas, comes to 60%; if this can be converted to electricity at 50% efficiency in a combined-cycle turbine system, the yield is 30% for a total of 70%.  CC turbines can get as much as 60% efficiency, so the throughput might someday hit 76%.

Randall Parker said at August 26, 2006 10:23 PM:


The price of ethanol is far above its production cost. The ethanol plant operators are making a mint right now. That's due to regulatory requirements that have greatly pushed up demand for ethanol. But the rapid construction of ethanol plants will drive down the cost of ethanol eventually.

A bushel can produce 2.5 to 2.6 gallons of ethanol (I found a range but the US government says 2.5):

One bushel of corn can produce at least 2.5 gallons of ethanol. It can also make 1.6 lbs. of corn oil, 10 lbs. of high protein feed, 2.6 lb of corn meal, or 31.5 lbs. of starch to make beverages or sweeteners.

A web page at the Chicago Board of Trade points out that distillers dry grain (DDG) is a by-product with value:

The crush refers to the process of refining a feedstock, such as corn, into fuel ethanol. While costs vary, corn is the principal input cost, representing approximately 70 percent of the per gallon cost of creating ethanol. The CBOT corn crush measures the difference between the sales value of finished fuel ethanol and the price of corn. Because ethanol is traded in dollars per gallon and corn in cents per bushel, a conversion of prices into equal units is necessary. One bushel of corn yields approximately 2.6 gallons of ethanol and two major bi-products, distillers dried grains (DDGs) and carbon dioxide. Therefore, the price of ethanol must be multiplied by 2.6 in order to convert it into a price per bushel to corn. The price of corn is then subtracted from the converted price of ethanol (in dollars per bushel) to obtain the corn crush.

A huge increase in demand for corn to use to make ethanol will yield more DDG. A huge amount of cheap DDG might turn out to be a boon for cattle farmers.

Beef Cow Rations: There is very little research on feeding distillers grains to beef cows. Only a few trials with limited feed combinations have been tried. In a trial at the University of Illinois with lactating beef cows on limited intake rations, the cows were fed diets containing from 50 to 75 percent distillers grains with solubles on a dry matter basis. Based on this limited experience, it may be possible to feed rations containing large quantities of distillers grains.

In that case ethanol buyers will drive down the price of beef and other meats. The corn DDG could substitute for soy for some uses.

The price of corn for this harvest is expected to be at around $2 at harvest time. So a doubling would bring it to $4, not $5. Though you could argue that you are referring to spring prices.

Some farms will fare better than others, but Hurt said the region will likely have a strong corn harvest this fall. Farmers may want to consider waiting to sell the crops, though. Farmers could reap an additional 35 to 75 cents a bushel by storing the corn and selling it in the spring, Hurt said. He predicts spring prices for corn could hit $2.60 a bushel, while harvest prices will likely hover closer to $2.
Engineer-Poet said at August 27, 2006 9:47 AM:

Your DDG link mentions that high rates of DDG in feed have not been tested ("it may be possible..." emphasis mine).  It also mentions that high fractions of fat and protein (concentrated in DDG) can either be wasted or even slow growth.

Which still doesn't change the fact that corn ethanol isn't anything close to a solution to the USA's petroleum woes.  Since it's so profitable, it's time to eliminate the subsidies and mandates and let it compete on its merits.

Randall Parker said at August 27, 2006 10:28 AM:


I'm all for eliminating the subsidies.

But if the USDA study is correct then the unsubsidized production cost of ethanol from corn is pretty low. Elimination of the subsidy to corn farmers won't change that fact (said subsidy is only $4 billion per year and some of that goes to admin costs). Neither will elimination of the sudsidy to alternative fuel producers.

The points I'm making:

1) The production cost of gasoline is now expensive enough to make biomass ethanol competitive.

2) Technological advances will cut biomass ethanol production costs and make it even more competitive.

3) Yes, higher biomass ethanol production will increase the cost of feedstock inputs. But before the feedstock inputs get too expensive we are going to see a big increase in biomass ethanol production.

Heiko Gerhauser said at August 27, 2006 10:44 AM:

Presently batteries provide virtually no petroleum displacement, even if we were to count hybrids. There are about 500 million cars on the road, and half a million hybrids. Very generously assume that the hybrids have doubled fuel economy and their total saving amounts to 1 in 2000 or 0.05% of gasoline demand.

E-P, we'll see. If batteries are that good, we'll get plug-ins, which though would still leave a market for liquid fuels (gasoline, ethanol or synthetic fuels from coal, gas, biomass). Major automakers don't seem to believe the battery hype though. Nor do venture capitalists. Guess what, those could make lots of money from those improved batteries, if they were really as good as claimed.

Randall Parker said at August 27, 2006 11:28 AM:


It turns out a far larger body of research exists on feeding DDG to livestock and poultry. It appears that an extensive research program at the University of Minnesota on DDG use for feed has produced a lot of results. If you click on the technical information section items on the left of that page you can see lots of research papers for each livestock type.

Here's a paper on DDG feeding to swine:

Cook et al. (2004) conducted a farrow-to-finish study that used diets containing 0 to 30% DDGS. No overall effects of DDGS on ADG, ADFI and feed efficiency were observed. Mortality rate decreased linearly (P < .05) from 6.0% (0% DDGS) to 1.6% (30% DDGS). However, a linear decrease (77.3 to 75.6% P < .05) in carcass yield was reported. Clearly, these observations heighten the importance that factors of economic importance must be considered in addition to standard growth and carcass measurements when assessing the value of DDGS. Not all studies have shown reduced performance using as much as 30% dietary DDGS. DeDecker et al. (2005) showed that ADG, feed efficiency, and carcass composition were not different among DDGS treatments (0, 10, 20, 30% DDGS).

Here's a paper on DDG feed for poultry:

Noll and Brannon (2005) conducted a study in turkeys testing the acceptable inclusion level of DDGS in turkey rations as influenced by dietary protein level.They reported that up to 20% DDGS could be fed in turkey tom grower/finisher diets but that when high levels of protein were fed to turkey toms, improved performance could be managed with a 15% inclusion rate.

Most poultry rations are formulated on a least cost basis with ingredient maximums set by a nutritionist. Current practice by poultry nutritionists is to set a maximum limit of inclusion rate for DDGS in layer rations at 8-10%. More research is needed to justify going higher than these levels at this time.

One of the problems with DDG as a high protein feed is an imbalance of amino acids. But that's a solvable problem. Genetic engineering to increase lysine and methionine could lead to corn that makes a better high protein feed. Some additional genetic engineering could improve other nutrient balances in the DDG.

Dezakin said at August 27, 2006 1:07 PM:

Ethanol is cheaper to make than gasoline.

Randall, this is more than just a bit distorted. This is only true when your production costs are from oil company to refinery, when the real production costs are from well to pump, and the average production cost in industry for new oil is at most 15 to 25 dollars per barrel, and refining costs add maybe 5-10 cents on the gallon. A barrel of oil yields about 20 gallons of gasoline, and maybe 10-15 gallons of fuel product with similar market value and that gives productions costs of maybe 60 cents a gallon.

Randall Parker said at August 27, 2006 1:18 PM:


No, it is not distorted to use the costs that refineries have to pay for oil. We can't buy oil at the cost Saudi Arabia has for pumping it out of the ground.

I'm comparing the cost for a refinery to produce a gallon of gasoline with the cost of an ethanol plant to produce a gallon of ethanol. That's the pair of prices that will determine in the market how many corn ethanol production plants get built.

I do not know what the average production cost is for additional oil. But whatever it is has not been enough to prevent oil from going north of $70 a barrel and staying there for months.

Randall Parker said at August 27, 2006 1:28 PM:

The other big competitor to gasoline is coal-to-liquid which, if Rentech is to be believed, should cost $25 per barrel. But is that just the conversion cost or also the cost of the coal input?

Note that is the cost of a barrel of diesel which is worth a lot more than a barrel of unprocessed oil. At 42 gallons per barrel that's only 60 cents a gallon. I've read other estimates for coal-to-liquid. See here for a claim that coal to liquid competes with $45 per barrel oil.

Heiko Gerhauser said at August 27, 2006 2:03 PM:

Hi Randall,

capital costs are a much larger component of overall cost for CTL, that makes it much more risky than conventional ethanol plants, the capital costs of which are so low that in the current price environment payback is between 1 and 2 years, I believe. In the case of CTL, there's the risk that oil prices might drop to less than $45 maybe even before the plant is built, and it'll take years to amortise the large capital initial capital outlay.

I've been trying to get an idea of what the limitations for ethanol are likely to be.

The National Corn Growers Association project 5.5 billion bushels for ethanol by 2015/16.

That would be a bit over a third of that year's corn production, it would be about half of last year's corn production.

They also think that'll be achieved with minimal increases in land use.

And they point out that in 1931 109 million acres were planted with corn, compared to 81.6 million last year, while simultaneously production went from 2.2 billion to 11 billion bushels.

Getting land use up to the 1931 level, together with expected yield increases would about double corn production. And as most is for feed use and exports and ethanol already and HFCS and other uses only make up 13% of current usage, getting to 90% of doubled production by 2015/16 looks quite feasible in principle.

Compared to last year's 4 billion gallons, the NCGA suggests 18 billion gallons in 2015/16 from 5.5 billion bushels of corn.

Dedicating 20 billion bushels would yield some 60 billion gallons of ethanol.

Total US land is 2.3 billion acres. 109 million acres is less than a twentieth of that.

Current cropland is 442 million acres, 109 million acres is less than a quarter of that.

To replace all gasoline with corn ethanol at likely future yields would require something like 10-15% of US land.

And now for the big BUT. Firstly, that's a lot of land (and the US has a lot of land compared to its population). Secondly, is there enough land of the right quality to achieve the needed yields. And thirdly, would water be a limitation (I don't think this would be insurmountable, but I haven't found much on the subject that would answer the question)?

And finally, is this acceptable from a land use / environmental effects point of view (monocultures, biodiversity and related concerns)?

I suspect it's too aggressive to use 10-15% of US land (or more than half of current cropland) to replace all gasoline use. On the other hand, it seems a lot more feasible on the face of it than you might think at first. A favourite anti-ethanol line is, after all, that even using all current corn production would replace only 10% of gasoline or so. It kind of gets neglected that corn currently only takes up something like 2-3% of US land, and that yields are very likely to go up further.

Randall Parker said at August 27, 2006 2:36 PM:


Dupont is working on the cellulosic technology problem and their researchers are tinkering with microorganisms to solve various aspects of the problem. The ability to use the corn stalk would boost ethanol from corn:

One acre's worth of harvested corn kernels can be converted into 432 gallons of ethanol, DuPont said. But if just half of the stalks and leaves left behind after the corn is harvested could also be turned into ethanol, the yield could be boosted by 36 percent to 586 gallons, according to DuPont researchers.

Even with the additional yield, however, it would take 239 million acres of corn - three times the amount planted this year - to meet annual U.S. gasoline demand of 140 billion gallons.

Could all of the stalk be used? That'd boost yield by 72%. That article also implies that the corn stalk contains less total energy than the corn ears. I've always wondered about that question.

But there's an additional problem with the stalk: It is more moist - about 40% to 50%. The article says that they need to get down to 15% for storage. Getting the moisture out takes energy. Seems to me this is a job for solar energy. Design a passive drying building which could remove the moisture using the autumn sun's rays.

But once we get the ability to use cellulose as an energy source then we move into the realm of being able to use many plant types. Would wheat, flax, and other crop stalks be useful? Would switch grass produce a few times more ethanol per acre as some claim?

Randall Parker said at August 27, 2006 3:03 PM:

BTW, Dupont's 432 gallons of ethanol per acre seems a tad high. If we divide 432 gallons by 2.6 gallons per bushel then we get 166 bushels acre. I typically see 150 to 160 bushels per acre for most states and most years. Though in a good year Minnesota can get over 170 bushels per acre:

In Cottonwood County, the scouts made four stops and found corn yields ranging from 139-196 bushels per acre. This compares to 2005, when the average corn yield was 184 bushels/acre.


In 2005, the average yield in Minnesota was 174 bushels/acre. The U.S. Department of Agriculture estimated the 2006 corn yield for the state, as of Aug. 1, at 160 bushels.

How rapidly will corn yields increase?

As for the fertilizer need: Monsanto is working on genetically engineered corn that fixes nitrogen from the atmosphere. This will reduce the need for fossil fuels to produce ammonia-based fertilizers and therefore increase the net energy gained by producing ethanol.

Nearby are 121 more growth chambers, each housing different bioengineered seed projects, ranging from nitrogen-absorbing corn to soybeans modified to produce heart-healthy omega-3 fatty acids. Each room has its own air circulation system and can replicate the temperature and humidity of almost any place on earth. St. Louis-based Monsanto's researchers may grow 10,000 plants to find one with the right genes in the right place. From the chambers, plants go to one of the 26 greenhouses on the roof. The electricity bill for the building runs $4 million a year. "We operate this process as a single machine, from lab to field," says Tim Conner, one of Monsanto's food technologists.

Biodiesel is more convenient. What I'd like to know about biodiesel: What are the prospects for developing a crop that produces much more oil? Currently soy oil is the cheapest but still yields biodiesel that costs over $2.50 per gallon - and biodiesel is 10% less energy dense than petroleum diesel. Are there any prospects for coming up with a crop that could double the oil produced per acre? Biodiesel would be a lot more convenient than ethanol. Could a plant even be genetically engineered to produce an oil that would be easier to convert to biodiesel?

Randall Parker said at August 27, 2006 3:31 PM:


Since you are interested in the economic angles of corn used as an energy source also see my post Corn Ethanol Production Expands In United States and note the calculations I do there comparing the cost of heat from burning natural gas versus corn. The use of corn for ethanol production might drive corn heat cost above natural gas heat cost. On the other hand, natural gas prices could rise again and keep it more expensive as a heat source.

Heiko Gerhauser said at August 27, 2006 3:40 PM:

I think they are likely using a slightly higher conversion efficiency than 2.6 gallons per bushel (have a look at the NCGA link I posted). I was aware of the idea of getting corn to fix N2, we'll see how successful that pans out to be.

Soybeans have a pretty low oil yield, they are mostly grown as high protein cattle feed (and to fix N2) I believe, not so much for their oil.

Palm oil is cheaper and has a much higher yield (4 tonnes per hectare currently, 10 tonnes is feasible), but it only grows in tropical weather.

BTL is another option for a biomass derived liquid fuel. Look at the CHOren site for a company trying to commercialise it (BTL's much less economic than ethanol currently and more expensive than gasoline/diesel, palm oil is just about cost competitive with crude oil at the moment, but less cost competitive than ethanol)

Heiko Gerhauser said at August 27, 2006 4:08 PM:

Yes, an interesting post, and I see that you also did a calculation comparing with total US land area (assuming present yields, no conversion losses, and displacing all petroleum and nat gas).

At the moment nat gas is a lot cheaper than gasoline ($7 per MMBTU for nat gas compared to something like $14 per MMBTU for gasoline), so from that perspective it's harder to compete against it.

From years back I was reminded of the following link regarding heating with corn:

I believe that the efficiency of burning both the corn and the nat gas (or propane in the above link) heavily depend on how much gets invested into the boiler. For the same efficiency, I also believe nat gas boilers are invariably substantially cheaper than corn fired ones.

For consumers there is also the issue of corn storage and logistics. A farmer who already stores loads of corn anyway and who isn't connected to the gas grid and therefore has to use more expensive propane trucked into the rural Mid West will have much better economics for burning corn than a home owner in California connected to the gas grid.

Engineer-Poet said at August 27, 2006 4:09 PM:

One or the other source linked above claimed 2.75 gal/bu for dry-milled maize, so 432 gallons would come out of 157 bushels - in the ballpark.

Again, though, this comes down to efficiency.  The extra 154 gallons/acre comes down to about 3100 vehicle-miles; unimpressive.  But you would get roughly 2.5 bone-dry tons of excess stover from that acre, with about 40 GJ of energy in it.  You could carbonize it to about .75 tons of charcoal with about 20 GJ of energy, and another 20 GJ of heat and chemical energy in the off-gas.  Burn the off-gas in something like a Capstone microturbine @ 26% efficiency, you get 5.2 GJ or 1444 kWH; at 300 Wh/mile, you've provided ~4800 vehicle-miles from the stover's off-gas alone.  Put the charcoal through a DCFC @ 80% and you get 16 GJ (plus heat, which is also hot enough to drive a gas turbine), which is 4444 kWH or almost 15,000 mile's worth at 300 Wh/mile.

You could get maybe 8600 miles out of the corn kernels (11,500 using the whole plant) via ethanol, or almost 20,000 miles out of the stover alone via electric conversion.  Go with ethanol and you've guaranteed scarcity due to low efficiency and the single-sourcing of energy (grain today, and biomass in general... someday); go with electricity and you can not only get more out of the waste product than the ethanol path gets from the whole plant, you have multiple sources of energy in case biomass gets expensive or has more important uses (chemical feedstock, carbon sequestration).

FWIW, an acre producing 150 bu of maize and 2.5 short tons of stover is making 4.6 tons of biomass all told.  Switchgrass can yield 10 tons/acre and Miscanthus has yielded as much as 27 tons/acre!  10 dry tons of grass @ 16 GJ/ton through the above carbonization/electric process would yield ~85 GJ of electricity, sufficient for about 78,000 vehicle-miles @ 300 Wh/mile.  On top of this, grass (not grain or DDG) is what cattle are evolved to eat.

If we want to move our economy off of oil and be able to respond to shortages anywhere in the supply network, ethanol is not the best thing to make from biomass.  It's not even the third-best.

Randall Parker said at August 27, 2006 4:42 PM:


A wild thought: Would farm tractors be more easily converted to run on electricity than cars? Imagine a system whereby no batteries were needed.

In theory one could run a cable out to a moving tractor and have an apparatus stretching across the field to keep the cable at the right length and tension as the tractor moved back and forth across the field.

How to do this?

Run a cable down from a very long pole to the tractor? I think the cable would have to be too heavy to stretch across many acres and the pole would have to be too high.

How about a sort of long cart train that stretches across one dimension of the rectangle of a field? The cable would run from the long cart to the tractor. As the tractor moved along its length the cable's end anchored to the cart would slide along a groove on top of the cart. Or something like a trolley car's metal roof apparatus for reaching overhead electric could be used. The long cart would gradually move across the field to keep it the same distance from the tractor.

I fear this apparatus would cost too much to build and move. But maybe with modern materials it could be done?

Another thought: A hydrogen tank on the cart filled by electric power to hydrolyze water. The tractor could have some way to get filled up automatically and quickly once every n passes across the field. The hydrogen tank would not need the sort of capacity that a vehicle would need for much longer trips. Might be workable just using pressure.

You see some way to use the limited area of a farm field to advantage to allow alternate fuels to be used? How about burning corn to run a steam engine on a tractor? Corn is a much cheaper heat source than oil.

Engineer-Poet said at August 27, 2006 4:48 PM:

I've pondered the possibilities of tractors powered by zinc-air fuel cells, but I have no figures handy.

I did calculate that the typical production of corn stover would be more than sufficient to power the required farm machinery from charcoal via a gasifier.  This is far less efficient than fuel cells, but it would be more than good enough for severe fuel shortages and other emergencies.

Heiko Gerhauser said at August 28, 2006 2:37 AM:

Hi Randall,

while I don't know about tractors being operated through trailing cables, I do know about large scale open cast mining excavators that are:

Niederzier is the place where I grew up, the open cast mine is located about 2 km away from my parents' home.

I've even seen these excavators being moved from one mine to another, very slowly (10 metres/min), but using electric power and trailing cables. These excavators are massive:
weight 13500 metric tons
excavation throughput 240000 metric tons per day
height 96 m, length 225 m
and they are worth about 100 million Euros.

Garson Poole said at August 28, 2006 4:47 AM:

Blog readers might be interested in another proposed source of ethanol that is discussed in an article entitled Fast-Growing">">Fast-Growing Trees Could Take Root as Future Energy Source. Researchers at Purdue are investigating the possibility of using hybrid poplar trees as a bioenergy crop. The trees would be planted "in rows just like any field crop."

Researchers believe that using the hybrid poplar in its present form could produce about 70 gallons of fuel per ton of wood. Approximately 10 tons of poplar could be grown per acre annually, representing 700 gallons of ethanol. Corn currently produces about 4.5 tons per acre per year with a yield of about 400 gallons of ethanol. Changing the lignin composition could increase the annual yield to 1,000 gallons of ethanol per acre, according to experts. Planted on 110 million acres of unused farmland, this could replace 80 percent of the transportation fossil fuel consumed in the United States each year.
Engineer-Poet voiced a concern that future climate changes might reduce the availability of ethanol from some plants. He said that "Each 1° increase in average temperature decreases grain yield by about 10%, so the surpluses which are the sole political justification for corn ethanol are going to disappear." The poplar article suggests that the trees currently grow in diverse climates so this potential problem might be reduced.
"We need a bioenergy crop that can grow many places year-round," Meilan said. "The genus Populus includes about 30 species that grow across a wide climatic range from the subtropics in Florida to sub-alpine areas in Alaska, northern Canada and Europe.”

Bob Badour said at August 28, 2006 7:15 AM:


Your idea for farm equipment has a serious flaw: Modern farmers do not farm a single field. My neighbours farm over 2000 acres with some fields adjacent to the Gulf of St. Lawrence and other fields on the Northumberland Strait. Compared to farms in the midwest or Idaho, my neighbours are a small farm.

Also, some people specialize in just one aspect of farming. For example, I heard a conversation last week regarding combines and how in the prairies combine owners will buy a $300k combine and move it across the continent with the progression of the ripening crops. The combine gets used every day for six months of the year and is sold as used equipment at the other end of the trip.

Heiko Gerhauser said at August 28, 2006 8:03 AM:

Hi Garson,

it sounds like they are talking about cellulosic ethanol. While a lot of people believe there is great promise, things such as cost and net energy and yield are projections.

Interestingly, if the lignin content is reduced to near zero, with the biomass being nearly only cellulose/hemic-cellulose, there'll be no lignin left over to burn for process energy, and it is the burning of the lignin that yields the high projected net energy (ok they could burn cellulose, corn ethanol plants could burn corn stover or corn itself).

Dezakin said at August 28, 2006 2:35 PM:

I just dont see how crop based ethanol can compete with biodiesel derived from algae on a large scale. First, you can get about 20 times the yield per acre of algae than the nearest competitor, oil palms, yielding 15000 gallons of biodiesel per acre.

Then theres all the logistics... flexibility of fertilizers (almost anything) irrigation is ideally saltwater and theres no shortage of saltwater aquifers even in deserts, and harvesting is a pump from the pond to the processing plant.

Now I'm less sure that it can compete with coal or fossil oil, but the logistics of algae versus any crop based ethanol convince me.

Chris Smith said at August 28, 2006 2:59 PM:

Using ethanol as fuel for cars is a very preliminary step towards building a renewable energy economy. Even if all of the corn produced in the U.S. were processed into ethanol, this would still provide less than 15 percent of the nation's annual transportation fuel needs. Brazil will probably never export much of its ethanol to the U.S., even if trade restrictions are lifted. The real issue is limited resources. Ethanol produced from sugar cane has a higher energy content than its counterpart produced from corn, but Brazil does not have the capacity to supply both domestic and foreign markets. Of course, there are environmental considerations, such as overtaxing the soil and clearing more of the Amazon rainforest, as well as displacing food crops.

The public has enthusiastically accepted ethnanol as a fuel; this could serve as a test case for rolling out future technologies, such as fuel cells. More research is needed into other methods of ethanol production that would supposedly yield far more energy content, such as from cellulosic materials.

What about producing hydrogen from ethanol? Here is an article from the Phyorg website, posted on Feb 9, entitle "Scientists develop new inexpensive technology to produce hydrogen":

By mimicking a protein found in nature and putting it to work, a group of scientists in Montana and New York is looking at producing alternative fuel using inexpensive sources and a unique chemical reaction. The invention is aimed at producing hydrogen as a fuel using inexpensive ingredients, although the inventors say more development is needed.

"Currently the energy industry produces hydrogen by using fossil fuels and re-forming them into hydrogen," said MSU chemistry professor and co-inventor Trevor Douglas. "That's a zero-sum game."

This invention--a hydrogen production reactor--would use organic acids or ethanol and water along with either the naturally occurring protein or a synthetic equivalent to create hydrogen.

"In principle, this is an incredibly efficient, renewable, environmentally friendly source of hydrogen," Douglas said.

The scientists face hurdles, however, before the invention can supply fuel. One involves getting enough of the protein, which the scientists have learned how to make in the laboratory, to drive large-scale energy production.

But the invention is far enough along, Douglas said, to interest potential fuel manufacturers.

Commercial challenges exist as well, including the lack of a hydrogen-fuel infrastructure to support large-scale distribution and usage similar to that for petroleum fuels.

But Douglas said once hydrogen reactors are commonplace, a hydrogen distribution and usage system is likely to follow.

In addition to Douglas, the other inventors are MSU chemistry professor John Peters, MSU plant pathology professor Mark Young and Hamilton College (New York) scientist Tim Elgren. A patent on the invention is pending.

I don't know how much ethanol the above method would consume, but I'm certain that it would far less than the amount needed to power an internal combustion engine. Also keep in mind that current fuel cell technology is two to three times more energy efficient that an internal combustion engine; in the future, fuel cells will probably be even more efficient. Cellulosic ethanol could be an essential energy source, as part of a chemical reaction to produce hydrogen from water. I would think that the amount of energy derived from a given amount of organic material would be greatly enhanced, transcending current limitations.

Bob Badour said at August 28, 2006 4:03 PM:


What you posted does not give enough information to draw your conclusion. What are the efficiencies involved?

The problem with breaking up water is one generally has to add a lot of energy to do it. While it is probably possible to lower that hurdle somewhat, I doubt it is entirely possible to eliminate it.

Randall Parker said at August 28, 2006 5:30 PM:


Let us be generous and assume 1000 gallons of ethanol per year from 110 million acres of poplar trees. Okay, that's 110 billion gallons of ethanol. But we have to divide by 1.5 to adjust to gasoline equivalent gallons. That's 73 billion gallons of gasoline equivalent energy.

But the US uses about 140 billion gallons of gasoline per year. Also, from that page we use 4110 times 42 times 365.24 = 63 billion gallons of "distillate fuel oil" (which I think includes diesel and heating oil - which is diesel) per year.

So we use over 200 billion gallons of liquid hydrocarbons - the vast bulk of that for ground transportation. We also use another 25 billion gallons or or so flying around.

So we'd need about (225/73)*110 = 339 million acres of poplar to fuel all that.

We have about 9.1 million square kilometers of land. That 9.1 million square kilometers converts to 2248 million square kilometers of land.

I do not want to use (339/2248)*100 = 15% of the land area of the United States (and a much higher percentage of total area that can grow trees) for transportation fuel production.

Also, the population of the US is growing (we should stop immigration to stop that but I digress). We also have rising per capita consumption of energy. While the rising per capita energy usage has stalled it will eventually resume. So the land area usage would get progressively worse.

I've stated repeatedly that biomass energy is a bad idea due to the land usage problem. At the same time, I'm expecting biomass energy production to soar. The ecomomics for it are too favorable as far as I can tell. You can't always get what you want.

Garson Poole said at August 28, 2006 10:51 PM:

Thanks for your excellent posts examining the numbers behind the feasibility of obtaining bioenergy from corn, poplars and other plants (and thanks to other commenters). Unsurprisingly, proponents of specific schemes often seem to report figures selectively and choose optimistic projections. I share your concerns about bioenergy from plants. On the other hand, every significant form of energy extraction and/or collection that I know of has a large impact footprint on the environment. No single energy source is adequate today and that will likely be true in the future.

As you have stated in the past, advances in solar energy are highly-desirable. Solar collection devices might be placeable on existing man-made structures such as roofs and perhaps even roadways as you discuss in this 2004 article entitled Structures In United States Cover Area Equal To Ohio. That would reduce the environmental footprint of energy collection, but considerable progress is needed.

Engineer-Poet said at August 29, 2006 4:17 PM:

Biomass energy works if efficiency gets high enough; with sufficiently low losses, you can supply the nation's current transport energy needs from waste.

Higher efficiency would tend to drain the feedstock from the less-efficient pathways; compared to the ethanol maker, the manufacturer of charcoal for DCFC's could pay more for biomass and still sell his product at a lot less per vehicle-mile or kilowatt-hour.  But that assumes that the alternative pathway gets a chance.

That alternative pathway is mostly electric.  This breakthrough has been waiting in the wings since the 1970's, never given a real chance through a combination of industry intransigence and regulatory blunders.  Or were they blunders?  We've seen the mistakes enough times that we should stop making them, but they happen again and again.  It's looking more like a conspiracy.

The recent GCC item about Richard Lugar talks about ethanol, ethanol, ethanol... but Lugar says nothing about batteries.  Is this because a steep increase in efficiency and REAL alternative energy (wind, solar) might threaten both petroleum and ethanol profits and even the subsidy system?  Your guess is as good as mine, but this tip-toeing around electricity has gone on too long and too consistently to be mere oversight.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©