Cellulosic Ethanol Reality Check

I firmly believe we should be aggressively researching the potential of cellulosic ethanol. This was after all the topic of my graduate school work at Texas A&M. But I think the hype has gotten way out of touch with reality at this point in time. There is a reason that nobody today is making money with cellulosic ethanol. It is quite possible that they never will, and in this essay I will discuss the reasons for that.

I did an interview with a major publication last week on the topic of cellulosic ethanol. I won’t divulge any details, because I don’t want to leak out anything before it is published. But during the course of the interview, I made the point that one of the challenges is in securing a large and steady local supply of biomass to run through the plant. This was one of the points I made in my essay Cellulosic Ethanol vs. Biomass Gasification. Then I was asked about just how much biomass it would take to support a cellulosic ethanol plant. So I decided to do a little calculation.

Iogen, probably the closest to commercializing cellulosic ethanol, has reported that the theoretical yield of biomass to ethanol is 114 gallons of ethanol per ton of biomass. However, what they actually achieve in practice is 70 gallons per ton. Let’s consider a typical mid-sized 50 million gallon per year ethanol plant. Using Iogen’s demonstrated yields, the biomass requirement would be 50 million/70 = 714,286 tons of biomass per year. According to Dr. Bruce Marcot, an ecologist at the USDA Forest Service the average Douglas fir yields about 1660 lbs of pulp (90% of the tree’s weight). So, to run a mid-sized cellulosic ethanol facility would require the equivalent of 714,286 tons * 2000 lbs/ton /(1660) or 860,585 Douglas firs PER YEAR. That’s a lot of biomass, and it puts into perspective the issue of a declining EROEI as biomass must be secured from farther afield.

What does this mean? Even if one achieved the theoretical limit of 114 gal/ton, it is still going to be very difficult to grow enough biomass to keep the plant going. Furthermore, consider the conversion penalty that is being paid when compared to corn ethanol. Let’s presume for a moment that the conversion for corn is the same as for cellulosic ethanol. There are 56 lbs in a bushel of corn. In our example above, it took 714,286 tons to run the 50 million gallon per year facility. This much biomass is equivalent to 714,286 tons * 2000 lbs/ton * 1 bushel/56 lbs = 25.51 million bushels of corn. In a conventional corn ethanol plant, that much corn would produce about 25.51 * 2.8 = 71.43 million gallons, or 43% more than we would get from the same amount of biomass in a cellulosic ethanol plant.

On an energy equivalent basis, converting 860,585 Douglas firs to ethanol will displace 0.02% of our annual gasoline supply on a gross basis (not counting the fossil fuel inputs to produce and process the ethanol). If you look at the USDA reports on corn ethanol, they say that to produce 75,000 BTUs of ethanol the fermentation/distillation requirement is 50,000 BTUs on average. That is from actual plant surveys as reported in their 2004 report. However, that is for solutions that are 15-20% ethanol. Cellulosic ethanol produces a crude product that is only 4% alcohol, meaning it will take quite a bit more than 50,000 BTUs to separate it out. I can tell you from experience that once you get down to 3% alcohol, it is classified as a waste stream and sent to wastewater treatment.

Future cellulosic ethanol plants are envisioned as being supplied by something like switchgrass or miscanthus. Will they yield more or less biomass per acre than corn? According to Questions & Answers about Miscanthus:

Over large areas, under typical agricultural practices, an average of about 8t/ha (3t/acre dry weight) may be expected at harvest-time.

That means our 50 million gallon ethanol plant, displacing 0.02% of our annual gasoline demand, would require 714,286/3 = 238,000 acres. To displace 50% of our current gasoline consumption of 140 billion gallons per year would take 70 billion/0.65 (this is for the lower energy content of ethanol) * 238,000/50 million, for a total acreage requirement of 513 million acres. This is about 13% of the land area of the United States; land which is presumably being currently used. This is also about 7 times the land area currently utilized for corn production.

A similar story yesterday came out that echoed this theme:

Study: Up to 100 million acres needed for renewable energy crops

Some excerpts from this story:

As many as 100 million acres of cropland and pastures would have to be dedicated to cultivating biomass fuels like switchgrass to support a national goal of 25 percent renewable energy use by 2025, a University of Tennessee study says.

Of course the reason cellulosic ethanol is so attractive is that the payoff would be huge, as the story explains:

But the rewards could be great. The study projects \$700 billion in new economic activity including: a \$180 billion growth in net farm income over the next 20 years; creation of 5.1 million jobs to support renewable energy enterprises; and government savings of more than \$15 billion in crop subsidies.

The bottom line is that it is going to take enormous swaths of land to supply these cellulosic ethanol plants, and it is questionable whether a farmed source of biomass can be counted on to run the facilities. Better to locate cellulosic ethanol facilities close to a massive source of waste biomass – say a very large municipal dump in which paper is sorted out, a paper mill, or some other consistent source of large volume biomass. If you then use the unconverted waste biomass for process heat, you could end up with a workable process.

I certainly don’t advocate giving up on cellulosic ethanol, but we do need to approach this with a realistic and sober outlook. Men once desired to turn lead into gold. That was ultimately a futile quest (unless you want to try something like a nuclear reaction), but with cellulosic ethanol there is much more at stake. My impression is that many people in our government are basing energy policy decisions on the presumption that cellulosic ethanol is a done deal. My advice would be to have several backup plans.

40 thoughts on “Cellulosic Ethanol Reality Check”

1. Great post!

This is also about 7 times the land area currently utilized for corn production

Just curious, where did you get this stat?

2. The USDA publishes figures for land planted to various crops.  FYI, corn planted for grain and corn planted for silage appear to be listed separately.

One thing this analysis does not do is consider that cellulose is often a byproduct of other crops, particularly corn.  Production of excess corn stover (above what’s needed for erosion control) is about 2.5 tons/acre.  Also, the figures for Miscanthus productivity are well below what UIUC is reporting.

Nevertheless, meeting US transportation energy needs with ethanol from any renewable source looks to be between extraordinarily difficult and impossible.  This observation, and a proposal for a scheme which could actually work, is the subject of my on-going writing effort.  Robert has already seen my work in progress, and The Oil Drum will probably run it when I’m done.

3. Just curious, where did you get this stat?

I found a news article that stated:

“In terms of acreage, I’ve been suggesting that we may have to push acreage up to 88 million to 89 million acres of corn,” Hurt said Monday in a statement. “That would be a 10 million-acre increase from 2006 and would put us at the highest acreage planted to corn in the United States since 1946.

So the current acreage planted in corn is somewhat less than 80 million acres.

Cheers, Robert

4. Production of excess corn stover (above what’s needed for erosion control) is about 2.5 tons/acre.

I thought about that when I was doing the calculations, but couldn’t find a reference as to whether harvesting stover over a long period of time would lead to more soil erosion. In fact, I almost left the corn analysis completely out of the article, because I thought it might be confusing. But I wanted to show the kind of efficiency hit that would be taken for corn if you presumed cellulosic conversion numbers.

Also, the figures for Miscanthus productivity are well below what UIUC is reporting.

Actually, the article that I referenced said the same:

Speculating from European data on small plots in agricultural experimental stations, the crop may attain as much as 25 t/ha (10 t/acre) dry weight by Fall.

But it said that under typical conditions, you could expect an average yield of 3 tons/acre. I figure on a very large scale, you are more likely to get typical yields as opposed to the highest yields ever observed at an ag experimental station.

Nevertheless, meeting US transportation energy needs with ethanol from any renewable source looks to be between extraordinarily difficult and impossible.

And that is the take home message. I think this cellulosic hype just needed a nudge back toward reality. I have read a lot lately that indicates that too many people are taking success for granted here.

Cheers, Robert

5. I am a big believer that RD&D in cellulosic ethanol will reap big rewards for society. But while enzymatic hydrolysis may help squeeze out some return from agricultural waste, I am not confident that it justifies the full frontal RD&D that the D.O.E. seems to be endorsing.

The best feedstock for cellulosic ethanol is waste, which can be converted into biofuels using gasification and syngas fermentation. We are not talking about growing anything here or using any extra acreage. We are talking about converting agricultural, industrial, forest, and urban blight into cleaner air, cleaner lands, and cleaner fuels. In short, we are talking about extending recycling on a mass scale.

In L.A. we (the utilities including the LA/Department of Public Works and the LA Co. Department of Sanitation) are working on diverting landfill garbage to biorefineries – not just because of the electricity and biofuels we can generate but because we have a “Peak Landfill” problem. It’s a much cleaner solution that waste-to-rail – shipping unrecyclables on 3-mile long trains each day 200 miles to the desert.

I just wrote an article about the forestry products industry call for using similar biorefineries to convert paper and pulp mill waste into electricity and biofuels using gasification. No extra trees needed – just blight to reprocess and fire-prone, diseased trees to convert. And the industry recognizes that the time is ripe because all their combustion boilers are hitting the end of their lifecycles.

We need to have a new industrial revolution focused on replacing combustion with gasification so that we can control carbon emissions. Syngas fermentation will help sequester the carbon in biofuels, green chemicals, and other new products.

6. But while enzymatic hydrolysis may help squeeze out some return from agricultural waste, I am not confident that it justifies the full frontal RD&D that the D.O.E. seems to be endorsing.

Scott,

That was exactly why I decided to write about this issue. I don’t think it wise to throw too many of our eggs in this basket. Fund it? Yes. Presume it is going to work, and allow that to affect our energy policy decisions? Very foolish.

Cheers, Robert

7. ….the biomass requirement would be 35 million/70 = 714,286 tons of biomass per year.

I think you meant 50 million/70 = 714,286

Your corn analysis is confusing. Your numbers imply someone would feed corn kernels into a cellulosic process and not use the stover. In reality the exact opposite would occur. If you want to show conversion efficiency just do a simple mass calculation. The side trip into acreage requirements is completely misleading.

Finally, there is a huge disconnect between your 210 gallon/acre switchgrass yields and Khosla’s 2000+ numbers. He may be dreaming, but surely SOME improvement in Miscanthus yield and cellulosic conversion efficiency is possible.

8. I think you meant 50 million/70 = 714,286

Yes, I did an analysis for a 35 million gallon plant and a 50 million gallon plant. I need to adjust for that.

Your numbers imply someone would feed corn kernels into a cellulosic process and not use the stover.

The point was simply to demonstrate that lower yields from cellulose versus corn starch. I agree that they acreage analysis is confusing, and I will probably modify it later.

Finally, there is a huge disconnect between your 210 gallon/acre switchgrass yields and Khosla’s 2000+ numbers. He may be dreaming, but surely SOME improvement in Miscanthus yield and cellulosic conversion efficiency is possible.

My numbers merely show the status quo, to demonstrate that their is an enormous gulf to be crossed. Khosla is certainly daydreaming; in fact his yield numbers are more than is theoretically possible. That’s what happens when you compound improvements at 7% per year for 30 years. You get totally unrealistic results.

Thanks for the comments. They were helpful.

Robert

9. It looks to me like cellulosic ethanol has roughly 5% efficiency: 2000 lbs of biomass to about 500 lbs of ethanol (25%), which would be burned at perhaps 20% efficiency (with luck).

OTOH, burning it for electricity for use in electric vehicles or rail could be 50% efficient. You could easily supply the perhaps 20% of swing fuel needed to make a renewable power supply work.

10. I went ahead and edited to clear up the confusion on the corn example. The primary purpose was to show the conversion penalty of a corn ethanol process versus a biomass process. Therefore, I took the language about the necessary acres of corn out of the essay.

Cheers, RR

11. I’m all for rational analysis cutting through the hype and I appreciate your tempered analysis of cellulosic ethanol, Robert. I’m a bit more optimistic than you are that the technology can be commercialized and will prove to be an important component of a transition towards a sustainable energy future. I don’t think it will displace 100% of our transport fuel use – that is probably impossible – but I think it could certainly provide a 1/4 or 1/3 of our current transport fuel. The rest can probably come from savings from increased fuel efficiency and a transition to plug-ins/EVs/electrification of transport.

Also, I find it a bit hard to reconcile your pessimistic outlook with the slough of announcements about cellulosic ethanol plants opening in the next year or two. These ones came across my RSS reader just today:

-> Xethanol Acquires Cellulosic Ethanol Plant in NC (RenewableEnergyAccess.com)

It seems like there are more like this every day as a variety of companies move closer to opening pilot projects, or even commercial-scale plants. It looks like most are targeting 2007-2008 to begin production, with commercial-scale plants targeting 2009-2010.

I’m sure that many of these projects won’t prove profitable and won’t pan out. But there seems like there are a dozen or more companies commercializing cellulosic ethanol, each with different technologies and processes, and it seems like one or more should be able to hit on a succesful process.

Maybe that’s just my youthful optimism, and I certainly defer to your experience in the field, but I’m crossing my fingers that cellulosic ethanol can and will prove to be an important component a sustainable energy future.

12. I’m a bit more optimistic than you are that the technology can be commercialized and will prove to be an important component of a transition towards a sustainable energy future.

The point is that the hype has completely lost touch with reality. It’s like we have a bunch of companies selling tickets to Mars, and yet nobody has actually been to Mars. It is just too incredibly costly.

I think it could certainly provide a 1/4 or 1/3 of our current transport fuel.

This is where a sober analysis is required. Calculate the biomass requirement to replace 1/3rd of our transport fuel. Let’s just focus on the gasoline. To replace a third would take 47 billion gallons. If we assume that the yield increases all the way to the theoretical yield, then that is going to take 412 million tons of biomass. Doing our Douglas fir calculation, this will take the biomass equivalent of almost half a billion Douglas firs per year.

But that’s on a gross basis. On a net basis, we will have only replaced a fraction of the 47 billion gallons. If we get to the theoretical limit of conversion, the crude ethanol solution will be about 4%*(114/70), or 6.5% ethanol. The rest will be water, which takes a lot of energy to remove. Current corn ethanol plants yield about 15-20% ethanol in their fermented product, and distillation is the biggest energy sink in the process. It will be an even bigger sink in a cellulosic ethanol plant, but you may be able to partially offset this by burning the biomass for process heat. But then we still have all of the inputs for getting the biomass to the plant, for disposing of the waste biomass, for harvesting the biomass, etc. How much of our 47 billion gallons are we going to burn up harvesting half a billion mature trees and getting them to the ethanol plant? It’s a tall challenge. A very tall challenge.

Xethanol Acquires Cellulosic Ethanol Plant in NC (

You mean this Xethanol? Read that story and you will understand some of the reasons for my skepticism.

I’m crossing my fingers that cellulosic ethanol can and will prove to be an important component a sustainable energy future.

No doubt we need it to be successful. But that doesn’t mean it will be. My bet is still on some variation of a biomass gasification process.

Cheers, Robert

13. “If we assume that the yield increases all the way to the theoretical yield, then that is going to take 412 million tons of biomass. Doing our Douglas fir calculation, this will take the biomass equivalent of almost half a billion Douglas firs per year.”

But we’re not going to be using Douglas firs for cellulosic ethanol conversion, nor are we going to be using corn kernels, which is why – although I like what you are going for – I find your analysis a bit misleading.

Let’s take a look at where we could get 412 million dry tons. Here’s what the DOE and USDA estimates is currently available (current unexploited resources – there’s more available in a growth scenario):

a) corn stover, rice hulls and other agricultural wastes 144 million tons;
b) forestry wastes and wood from fire treatment thinning operations – 101 million dry tons;
c) forestry products industry wood residues – 8 million dry tons;
d) urban wood residue – 28 million dry tons;

Total: 281 million dry tons, and that’s just what’s already out there going to waste (note, that doesn’t include 15 million dry tons of grains – corn mostly – assumed to go to ethanol and biodiesel, nor does it include non-wood organic waste in our urban waste stream which could be used in gasification processes).

If we were to develop a strong cellulosic biomass industry, the estimated total annual avaiable biomass yield is over 1 billion dry tons (1.3 billion dry tons per year, by mid century).

So, yes, 412 million dry tons is a lot, but there’s a lot of avaible dry tons out there, and nobody’s talking about cutting down whole swathes of doug fir forests or devoting all of our agricultural land to get it (well, at least I’m not).

Anyway, the logistics of harvesting all of that biomass and getting it to ethanol plants is not an insignificant problem, but the availability of biomass isn’t really the problem, it seems.

As for the ethanol-water solution being very dilute and needing lots of energy to seperate, every cellulosic ethanol process I have read any details about generates plenty of extra energy, usually either steam for export or excess electricity, after providing all of the process heat, steam and electricity needed for the ethanol plant (i.e., Iogen, BRI, etc.). Gasification-based processes obviously produce plenty of hot syngas (and then steam), while fermentation based processes leave plenty of lignin to burn for process energy, so I don’t think the issue of seperating out the ethanol is really an issue at all. Of course, I again defer to your more detailed experience; maybe I’m missing something there.

Anyway, I agree that ethanol is overhyped (Xethanol could be a particularly bad culprit, but again it’s only one of many developing cellulosic ethanol processes), and that it isn’t the silver bullet (there probably is no silver bullet anyway). I also appreciate that you are delving deeper with some analysis, but I think your use of doug firs or corn is a bit misleading. Maybe looking at available waste biomass resources would be a better focus.

All the best,

Jesse

14. But we’re not going to be using Douglas firs for cellulosic ethanol conversion, nor are we going to be using corn kernels, which is why – although I like what you are going for – I find your analysis a bit misleading.

The use of Douglas firs is because people can appreciate the mental image. They know what a tree is, and if I say it is going to take a million trees, they can appreciate that. If I say a million tons of biomass, they can’t appreciate that because they don’t really know how much that is. The point is not that we are going to use Douglas firs, but that we are going to have to use that much biomass.

Let’s take a look at where we could get 412 million dry tons.

The problem is not one of availability. It is one of concentration. You need a very large, concentrated supply of biomass. You can’t afford to transport biomass too far because then you will be seriously cutting into the EROEI. That is the problem, and that is the whole point of the Douglas fir analysis. In order to run that 50 million gallon plant, you have to have that much biomass in close proximity. That waste crop biomass is not going to fall into that category.

As for the ethanol-water solution being very dilute and needing lots of energy to seperate, every cellulosic ethanol process I have read any details about generates plenty of extra energy, usually either steam for export or excess electricity,..

If you are going to burn biomass to produce energy, does it make sense to separate ethanol from a 96% water solution? The vast majority of your BTUs are just wasted. You would be far better off burning that biomass to make electricity, and then using that electricity to support an electrified transportation system. Or gasifying the biomass and then using the syngas to make methanol or diesel.

Cheers, Robert

15. Three cheers for you Robert! Well done blog. Glad I found it.

Cheers,
G-Man

16. I believe that the solution to the biofuels issue, and renewable energy in general, is going to be very gradual and decentralized.

You are right, Robert – ethanol will not solve every problem. But it is a near universal fuel extender for internal combustion engines – and gradual infrastructural change is what we need right now (unless we are prepared for nuclear reactors and all-electric cars).

Ag is great for producing energy feedstock but it requires alot of input energy in the form of fertilizer, harvesting, and transport. Ag, forestry, and urban waste “harvests” are much more predictable, much more decentralized, and requires no excess cultivation.

I wish people would look more closely at waste as a cheap feedstock (some even say it is negative cost feedstock). It is the existing residual to our very wasteful industrial and waste management infrastructures. Not even the D.O.E. looks at it seriously enough and it is a WIN-WIN-WIN solution resource. Even if the net energy was zero, at least we would mitigate the waste disposal problem.

Carting wood to centralized biorefineries is not the low hanging fruit of biomass conversion. Instead, pulp and paper mills will very probably replace their boilers with gasification units and process their existing pre- and post-processing waste on site. This will extend their existing practice of producing much of their own energy while providing them with additional income opportunity. See Renaissance of the Forest Products Industry.

Same with municipal solid waste (MSW). Biorefineries will process the waste at the sorting centers where, at least in the L.A. plan, there is plenty of extra facilities space. This will save 2/3 of the diesel used to send the residual to landfills – and save 3/4 of the need for the landfill. See Expanded Recycling.

Since gasification allows for blended feedstock, new decentralized biorefineries will have great flexibility to respond to resource changes. For instance, wood-based or bagasse-based biorefineries in the Southeast could, in my scenario, respond to Katrina-type catastrophes by processing the excess C&D waste of the destruction. This would save the region the multiple blights of the destruction while creating jobs, reducing disease, and resupplying some of the lost energy.

17. Anonymous says:

Here is a reality check of a different sort, not looking at acreage required for biomass, the type of biomass, the source of biomass or even energy balance for ethanol. Let’s just think of the “material handling” and scale issues of keeping a large cellulosic ethanol industry going. Let’s base the reality check upon the U.S. pulp and paper industry where there is significant experience in moving large quantities of “biomass” into facilities (pulp and paper mills) and producing a finished product. The feedstock to a mill may be log wood, chips, and/or wastepaper collected from a variety of sources – we can call it “biomass”.
According to the American Forest & Paper Association (AF&PA), total U.S. paper production is approximately 100 million tons per year. (Actually slightly less) Even at a crude 50% yield for a pulp and paper mill, approximately 200 million tons per year of feed stock are required to keep the ENTIRE U.S. pulp and paper industry running.
So let’s use some “quick and dirty” numbers to look at the reality of cellulosic ethanol.
If one uses the USDA/DOE “Billion Ton Annual Vision” to replace only 30% of our current petroleum consumption, it requires ONE BILLION tons of biomass per year. Therefore, we would have to move enough biomass (even if it is free or readily available as municipal waste, forest waste or new agricultural production) in such quantities that it would supply the entire U.S. annual paper production five times over! Most people don’t think about entire industries, much less the pulp and paper industry, but the U.S. pulp and paper industry is the largest in the world.
I’m not saying that 30% replacement over twenty four years is impossible, but in light of the current dismal yields for cellulosic ethanol, most proponents are not thinking out the scale of their dreams.

18. As another data point, the US uses roughly 1 billion tons of coal per year and ships quite a bit of it from e.g. the Wyoming Powder River basin to far-flung plants all over the nation.

This problem may be difficult, but it is in no way unmanageable.

19. Good point by E-P, coal does get transported huge distances in the US, and it’s not actually all that much more energy dense than biomass. Also, US railways could be electrified.

Yes, biomass transportation is an issue, but how much of an issue? We need numbers there and not just of the status quo, but also of what could easily be.

Similarly, I think you need to say a lot more about separation of water and ethanol, than just the bland assertion that it’s energy intensive.

20. Good point by E-P, coal does get transported huge distances in the US, and it’s not actually all that much more energy dense than biomass.

Two things on the coal issue. First, it is being burned, not fermented. If the cellulose was being gasified instead of fermented, then you could afford to transport it long distances.

Second, according to this the energy density of anthracite coal is 31.4 MJ/kg. The corresponding numbers for dry wood, wheat straw, and bagasse are 15.5, 18.9, and 19.0. That’s a pretty big difference in energy density.

Similarly, I think you need to say a lot more about separation of water and ethanol, than just the bland assertion that it’s energy intensive.

I have covered this topic in some previous essays. If you look at the USDA reports on corn ethanol, they say that to produce 75,000 BTUs of ethanol the fermentation/distillation requirement is 50,000 BTUs. That is from actual plant surveys as reported in their 2004 report. However, that is for solutions that are 15-20% ethanol. Cellulosic ethanol is only going to come in at around 4% ethanol, meaning it will take quite a bit more than 50,000 BTUs to separate it out. I can tell you from experience that once you get down to 3% alcohol, it is classified as a waste stream and sent to wastewater treatment.

Cheers, Robert

21. Similarly, I think you need to say a lot more about separation of water and ethanol, than just the bland assertion that it’s energy intensive.

After rereading the essay, I agree. So, I put some information in on this point. Thanks for the comments.

Cheers, Robert

22. Robert, you wrote “First, it is being burned, not fermented. If the cellulose was being gasified instead of fermented, then you could afford to transport it long distances.”

Why do you say this? If the yield is similar for a fermentation-based process as it is for a gasification-based process, and right now, they are at least in the same ball-park, why can you afford to transport feedstocks farther for gasification-based plants? Just curious, as I don’t immediately see the rationale.

It seems to me that a gasification-based approach is better, because of the feedstock tolerance as well as the variety of different outputs that can be made from syngas – ethanol, DME, FT diesel or gasoline, pipeline gas, or electricity – but I don’t immediately see why it would allow for longer distance feedstock transport.

And thanks to the anonymous poster who mentioned the scale of the U.S. pulp and paper industry. That’s provides a good sense of scale I wasn’t aware of. Cheers,

Jesse

23. If the yield is similar for a fermentation-based process as it is for a gasification-based process, and right now, they are at least in the same ball-park, why can you afford to transport feedstocks farther for gasification-based plants?

The yields are not close to the same. In a gasification process, all of the carbon-based compounds are converted into syngas, or combusted all the way. This includes lignin, which ends up as a wet waste in a fermentation process. Lots of carbon ends up this way in a fermentation process.

However, I would not gasify and then turn the gas into ethanol. A far more efficient approach would be to gasify and make electricity. If a liquid fuel is a must-do, then it would be much more efficient to produce methanol or Fischer-Tropsch diesel. Gasification is much more efficient, and gives you a lot more options.

Cheers, Robert

24. If biomass was 3 kJ/kg and coal 300 kJ/kg, I’d understand why biomass could at best be moved 10 km, while coal could be moved 2000 km, but, when it’s 15 kJ/kg compared to 30?
(which is what I meant with “not all that much”)

On ethanol/water separation I suppose I ought to do a little bit of researching the issue myself, rather than asking others to do all the hard work 😉

25. Anonymous says:

What about Iogens use of wheat straw. You live in MT, am I the only one seeing unused straw bales wasting away. I see ethanol potential. Just a thought.

26. If biomass was 3 kJ/kg and coal 300 kJ/kg, I’d understand why biomass could at best be moved 10 km, while coal could be moved 2000 km, but, when it’s 15 kJ/kg compared to 30?

Again, only if you are going to burn the biomass as you are burning the coal. You can afford to move coal across the country because 1). It is energy dense; and 2). We are burning it, not fermenting it. If the coal was destined for an inefficient fermentation, you couldn’t justify moving it across the country.

What about Iogens use of wheat straw. You live in MT, am I the only one seeing unused straw bales wasting away. I see ethanol potential.

Why do you think Iogen has been announcing for 4 years that they are going to build a plant, but they never do? Again, back to the essay. To run a mid-sized ethanol plant will take the wheat straw equivalent of 850,000 Douglas firs a year. Actually, it would be a bit less than this because of the higher energy density of wheat straw. A quick calculation says we need the wheat straw equivalent of 700,000 Douglas firs a year. There aren’t that many straw bales laying around. Furthermore, once you ferment it, you are going to end up with a 4% ethanol solution. You are going to consume probably around 90% of the BTU value of the final product in just the distillation step. That’s why it would make more sense to gasify that straw, or co-feed it into a coal-fired power plant. Fermentation is inefficient. Combustion is efficient.

I think that’s the kind of trajectory we need to be on. Electrify the transport system. Start supplementing coal with biomass, until eventually you are deriving all electricity from renewable sources (biomass, wind, solar). I believe this strategy has long-term staying power.

Cheers, RR

27. Actually, there’s a fair amount of straw around.  One recent test found 4 tons/acre of available rice straw, and a test in Washington state found up to 1.3 tons/acre of available wheat straw.

If we assume 1 ton/acre of wheat straw and 48.8 million acres planted, that’s 48.8 million tons.  Not a huge amount, but not trivial either.  The 3.3 million acres planted to rice doesn’t amount to much, unfortunately.  Corn stover is the winner; at 80.7 million acres planted in 2004 and the 2.5 tons/acre collected by the corn stover collection project, it scores 202 million tons.

28. Robert,

I am not sure whether I’ve given you any info on my background yet. I am a research fellow at the University of Aston in Birmingham, and have been working on fast pyrolysis of biomass for more than seven years now (which was also the topic of my PhD). My full name is Heiko Gerhauser, gerhaush is the first bit of my email address (@aston.ac.uk), and Heiko my first name.

Let me expand Jesse’s point a bit. There aren’t any commercial scale BTL (gasification + Fischer-Tropsch) or cellulosic ethanol plants anywhere in the world yet.

So, estimates of liquid yield and cost are just that, and there’s a fair bit of uncertainty. But within the envelope of that uncertainty, it seems that the two processes come out about the same, ie 40% yield, of the order of \$100 per barrel cost.

So, in both processes a majority of the biomass needs to be burnt for process energy. The BTL route doesn’t have distillation as a big energy drain, but overall process energy consumption seems to be about the same.

The reason I am so keen on knowing more about the scope for improvement for distillation, is because it’s such a key factor in the overall performance of the cellulosic ethanol route.

If the promise of membrane processes pans out, and energy use for water/ethanol separation gets cut by 90%, we’d have a big improvement in the energy efficiency of cellulosic ethanol. We wouldn’t need to burn half the biomass anymore, and in the case of wood, do something else with the lignin by-product (eg BTL), or we could use a feedstock (like incidentally corn stover) that hasn’t got very much lignin and get a much better liquids yield.

I appreciate your argument about electrification, but electricity and heat can come from many renewable sources (wind, heat pumps, solar), while biomass is particularly suited as a source of fixed carbon, which will be required at least for some applications, and be it only chemicals and kerosene for air transportation.

29. I am not sure whether I’ve given you any info on my background yet.

Of course I know your background. I looked you up through your profile a long time ago. 🙂
But within the envelope of that uncertainty, it seems that the two processes come out about the same, ie 40% yield, of the order of \$100 per barrel cost.

The estimates I have seen, linked to in this essay, were a theoretical yield of 114 gallons per ton for fermentation and 230 gallons per ton for gasification. However, the fermentation/distillation process is going to take a lot of energy inputs because the ethanol is going to be very dilute. The gasification process is going to be self-sustaining, and you are going to convert all of the lignin to syngas. Based on what I have seen (first-hand) from natural gas partial oxidation, I would expect >90% of the carbon atoms in the biomass to get converted to syngas. The disadvantage I see for the gasification process is in capital costs.

The reason I am so keen on knowing more about the scope for improvement for distillation, is because it’s such a key factor in the overall performance of the cellulosic ethanol route.

But distillation has been around forever. I don’t expect any revolutionary advances there. Extraction is tough because water and ethanol have such high affinity. If you had something like butanol, extraction would be much easier. But separating ethanol from water is going to be tough to do with low energy inputs.

Cheers, Robert

30. The efficiency I was talking about is based on 100 kWh of wood input into a plant, and on the assumption there are no other inputs and all process energy (heat/electricity) comes from the wood itself.

In the case of cellulosic ethanol, my understanding is that the cellulose, once converted to sugar, will convert to ethanol at near 100% efficiency (as you said in an earlier discussion, the microbes consume very little energy indeed), but pre-treatment of the cellulose and hemi-cellulose, and particularly distillation will require loads of process energy, essentially consuming all the lignin (and some dregs left over by the bacteria).

In the case of BTL, oxygen blown gasification will be followed by gas cleaning, and before we get to clean syngas for Fischer-Tropsch we may lose some 20-30% of the energy in the wood, and Fischer-Tropsch then loses us another 20-30% to give us a product that is heavy in waxes, or heavy in methane. For transportation fuels, it seems, we’d want to maximise the proportion of waxes, because those can be more easily converted to diesel (using conventional hydro-cracking) than methane.

The theoretical yields you give don’t seem to be energy yields, I think the gasification yield appears to assume an external input of hydrogen or some other process energy input.

At least a quick back of the envelope seems to tell me that 230 gallons per dry ton (~900 kg) would contain more energy than the original biomass. 230 gallons should be some 700-800 kg and ethanol has nearly twice the heating value of sugar.

On distillation:

The most promising alternative seems to be zeolite membranes. The principle is that they allow water through, but not ethanol.

31. At least a quick back of the envelope seems to tell me that 230 gallons per dry ton (~900 kg) would contain more energy than the original biomass.

I have just double-checked, and they are almost identical. I came up with an energy content for dry wood of 17.2 MMBTU/ton, and 230 gallons of ethanol is about 17.5 MMBTU. So, clearly the yield is not on a net basis for either cellulosic or gasification. It looks to me like the presumption for the gasification calculation is that all of the carbon gets converted into ethanol with no regard for energy inputs.

However, we are doing something with cellulosic that we aren’t doing with gasification: We are adding a whole bunch of water to the product. That water has to be removed, and will be the single biggest difference in the net energy yields. However, the ability to convert the lignin to syngas in the gasification case is also pretty significant.

The most promising alternative seems to be zeolite membranes. The principle is that they allow water through, but not ethanol.

My understanding is that this only works well once most of the water has already been removed.

Cheers, Robert

32. Anonymous says:

Great posts on Cellulosic Conversion from wood but Doug fir is not the only available species. Take a look at the currrent ethonal production on the map and it is all in midwest with crops that will soon compete with food costs. Cellulosic wood waste is everywhere but locations that are not at current time prodcuing ethanol. Cellulosic may not completely solve the need for other liquid energy components but can make significant contribution. For instance the Paper production of wood currently flowing into the system does not consider the significant volume of wood left at harvest sites after the forest operation which can range from a 1 to 1 ton equivenlent in plantation thinnings to 4 to 1 in final harvest. Not to mention precomercial thinning and improvement cuttings to allow for higher value product timber management. What Cellulosic ethanol offers is more tools to manage existing forest resoruces and to increase value to forest management and operations. Not to mention to restore short lines of distrbution for local energy resources to be developed. In communites of the great northwest where we have decided to let our resource burn in wildfires there would be new oportunity to restore rural economic value to communities. I appreciate the discussion but there are some real significant opportunites for cellosic conversion, Also the Gentlemen from Birmingham is right!we don’t know the potential for yeilds from fast pyrolosis and gasification, but the potential is for higher production.

What we know is that wood residues are abundent outside the corn belt. Forest Fires are consuming more that 4 million acres/year, Real value can be returned to rural communties where energy production can be converted from existing underutilzed forest resources, Companies are already making investments in conversion facilites in New York, North Carolina, and Georgia. All states must soon evaluate where they will be on energy policy, What are we waiting for?

33. Anonymous says:

Actually he wrote that Doug firs were merely used to illustrate the amount of biomass that would be required. I doubt if there is a million trees worth of waste wood laying around to run a small plant.

34. Look for Heaton et al. 2007 to get a range of Miscanthus yields at different locations and soil types around Illinois over three years. It should be submitted within the next week or so. UIUC is continuing to expand field trials to get more accurate regional productivity numbers.

Great blog. I see the future holding a diverse array of feedstocks including grasses, wastes, corn grain and stover, forestry inputs. This will allow for decentralized ethanol conversion in many different climate systems around the U.S.

35. D Flater says:

Vast areas of BC and the Pacific NW pine forests are being devastated by the Mountain Pine Beetle. In spite of rapid harvesting, soon much of this timber will be unsuited for lumber production. Perhaps a better use would be to liquidate the already dead and dying forests into cellulosic ethanol. This would allow reforestation and remove a potential forest fire hazard.

36. Tom Harrington says:

At conservative yields miscanthus would require about 47,000 acres. That’s at 15tpa, a level that has been exceeded already in the US and which our fields are showing can be higher. We are far south of the EU and UK where smaller yields are common. Sugar cane also has high yields.

37. Obviously some of these bloggs are old news. Ceetol is winning the arguement. Cellulosic Ethanol is being created and cost effectively

38. Cellulosic Ethanol is being created

While that is certainly true – in fact I have created it myself, this:

and cost effectively

That is a howler. Not close to being true. Even the cellulosic guys will admit that their costs are >\$3/gallon. That is, those who actually have experience making it. Those who have plants only on paper can assert anything they like.

Those who don't have a vested interest will tell you that unsubsidized costs are more like \$5 or \$6/gallon.

RR

39. isn't it better to bury carbon in landfills rather than release it to the atmosphere?

isn't the goal to reduce atmospheric CO2?

nuclear, wind, solar, and their ilk, supplying electic cars is the only way.

conservation right now is the best bang for the current buck. led lighting, use mass transit, etc..

trash plastic bags, trash cellulose, trash carbon, bury it deep!

and use electrons from inorganic sources to power the future.