The Ergosphere
Sunday, July 30, 2006

Open letter to Vinod Khosla

Dear Mr. Khosla,

I hope this letter finds you well.  I have no doubt it finds you wealthy.

You appear to have taken a position in re:  ethanol, claiming that it will provide "the solution" to our petroleum woes.  This has been challenged.  Robert Rapier has already done a nice job of debunking the claims you've made.  I'm just a pseudonymous analyst (and an amateur at that), but I hope that you'd explain the details behind what you've chosen to acknowledge versus what you've chosen to disregard.

The implications of what you've chosen to disregard are considerable.  Your position (please correct me if I'm wrong, I cannot view your video presentation) appears to be this:

I've done the math, and I can't see how you justify this position.

The facts about ethanol vs. oil

For background, the USA uses approximately 20 million barrels of crude oil per day, 7.3 billion barrels per year.  At ~6.1 gigajoules of energy per barrel, that's roughly 44 exajoules (42 quadrillion BTU, or quads) per year.

That's a whale of a lot of energy.  Even the fraction used for motor fuel is enormous.  Gasoline consumption amounts to roughly 140 billion gallons (17.6 quads @ 126,000 BTU/gallon) per year.  Diesel fuel (about 2.7 mmbbl/day out of 4 mmbbl/day of total "distillate" consumption) comes to roughly another 5.8 quads at the pump.  If we want to replace just our imported petroleum, we have to replace all of that and more.

I'm wondering how you expect ethanol to do this.  You appear to acknowledge that grain ethanol (from corn, wheat, etc.) is a dead end.  That's good.  What isn't good is your claim that ethanol from cellulose is going to be The Solution.  That's only good if it's true, and as far as I can tell it's a long way from that.

Let's forget a few inconvenient details for the moment:

For the moment, let's just look at what we could make from what we've got here in the USA.

The estimated biomass available from sources not currently used is 1.3 billion tons per year.  Biomass does not have as much energy per ton as hydrocarbons (around 16 GJ/ton), so the total energy available from this is on the order of 21 EJ or 20 quads.  This is only about 45% of the energy the USA obtains from oil each year.  But it gets worse.

Conversion from biomass to liquid fuel is lossy.  Iogen claims a yield of 330 liters (87 gallons) of ethanol per ton from their enzymatic process.  This is slightly under 50% energy efficiency; the 1.3 billion tons would produce 113 billion gallons (~10 quads) of ethanol, with the energy of about 70 billion gallons of gasoline.  Conclusion:  enzymatic ethanol from unused or waste biomass not only cannot replace petroleum, it can only replace about half of gasoline.

Other processes promise somewhat higher efficiency.  Syntec's gasification process is claimed to yield 114 gallons of ethanol per ton (presumably of biomass; the highest theoretical yield would require more energetic inputs, like scrap rubber and plastic... or coal).  This is about 60% efficiency, roughly 30% better than Iogen.  But another 30% only gets up to replacing 65% of gasoline (and no diesel, no jet fuel, no heating oil, no LPG, and nothing for petrochemicals, coal or natural gas).

The end of the line

And once you've used all your 1.3 billion tons, that's it.  You're out of feedstock; you can't get any more out of the inputs without:

  1. Building a new system which is able to use the inputs with higher efficiency, and
  2. Scrapping part or all of the previous system (which may be brand-new), losing the investment in it.

It may be even worse than that.  If the feedstocks are locked up by long-term contracts, it may be legally impossible to convert even if the consequences of failure are disastrous.  A set of entities which have cornered the feedstock market may even scare off the venture capital necessary to create the more-efficient alternatives.  I'm sure you understand this very well.

But while that's the end of the line, it's far from the end of the issues.

Carbon dioxide and the climate

Climate change is the elephant in the room of all these discussions.  As atmospheric CO2 increases, the climate warms.  This has knock-on effects all down the line, including a reduction of plant productivity due to heat and water stress.

Recall that 1.3 billion ton number?  That's only a historical projection based on what the climate used to be.  The steep rise in forest fires in the Western USA may very well indicate that we're not going to be able to harvest that much in the future; a lot of that biomass is going to burn, and hotter, drier conditions mean that it will grow back more slowly if at all.

This applies across the board.  Tallgrass crops can burn too (burning is how prairie prevents the incursion of trees).  Hay fields burn all too well (see Texas and Oklahoma this year alone).  Ripe grain fields can burn, and the productivity of all of these is reduced by hotter weather.  We may not have 1.3 billion tons to work with; if we can't control warming, it may be half that.

What's the relevance to your ethanol schemes?  It's the fact that we'd still need to get the following from fossil fuels:

  1. At least 35% of current gasoline consumption.
  2. 100% of current diesel and jet fuel consumption.
  3. 100% of current heating oil and LPG consumption.
  4. 100% of current petrochemical consumption.
  5. 100% of demand currently served by coal and natural gas.

All of those would still be pumping CO2 into the atmosphere.  Your proposal does involve capturing some carbon from the atmosphere, but it all goes right back; you have nothing to sequester it.  If this problem is not to get worse, all atmospheric greenhouse gas (GHG) inputs have to be at least offset by removals.

Suspicion of motives

It's very hard to believe that you don't know all of this already.  You've got net worth of over a billion dollars; you have access to the best expert advice in the world, and you have certainly used it before to make your current fortune.  Yet the reasons you've stated for supporting ethanol don't match up with the most cursory stab at due dilligence.

It's easy to come up with darker reasons for your position, each worse than the last:

All of these are reasons not to listen to you.  The last two are excellent reasons to actively fight you.  Maybe you aren't familiar with being on the wrong side of a grassroots battle, but enough people writing letters, volunteering to work for candidates who oppose you, outing your support as astroturf campaigns and voting to stop your plants can throw a monkey wrench into the smoothest business plan.

My advice, offered for nothing:  don't go there.  Either there is something very important that you don't know, or something equally important that you aren't telling.  Straighten this out, and the rest will follow.

Where this all goes

I'm sure all of my numbers are good to within 10% or so (corrections accepted).  Unless I have screwed something up by at least a factor of three, ethanol can't get us off imported oil (let alone all oil) as long as it depends upon biomass.  (Making it from coal is another matter... but mind the environment and don't go there either.)  If this problem has a solution, it's not present inside the current model; to see it, you have to step outside the box.

This is my forté.

All variants of your scheme fail because of their common element:  they propose, and even demand, reliance on the internal combustion engine.  This is essential if we aren't going to replace the vehicle fleet.  But why worry about this?  The fleet wears out and is replaced regularly as a matter of course; we didn't ban cars without catalytic converters, but they're almost completely gone nonetheless.

Where we are now

The typical light vehicle has a gasoline engine coupled to an automatic transmission using a hydraulic torque converter.  The average thermal efficiency of these is abysmal; out of the energy pumped into the tank, perhaps 15% gets to the wheels.  The vast majority of these losses are in the engine itself.  Heavy-duty diesels are better, getting up to about 45% for medium-speed truck engines.

What this means is that our 17.6 quads/year of gasoline only put about 2.6 quads of energy out to wheels.  Our ~40 billion gallons/year of diesel comes to about 5.6 quads, of which perhaps 2.5 gets to wheels.  The total actually going to useful work is about 5.1 quads.  That's only about 25% of our prospective supply of energy from biomass.  This means that the problem IS solvable, in principle.  It just can't be solved with ethanol in combustion engines.

Let me make this clear:  ethanol is not a solution; it is a niche player at best.  It cannot be a mainstay, and any program relying on it to perpetuate the status quo ante is doomed.

Where we should go

Right now you're probably asking, "If ethanol won't work, then what?"  While it's logically possible for there to be no solution (e.g. the Halting Problem), you can see that I'm not going to tell you that.  Nope, there are better solutions out there.  People are putting some of these solutions into practice themselves, because they are tired of waiting for business and government to do it.  So I'll list a few for you.

Plug-in hybrid electric vehicles (PHEV's)

This is the #1 best option for petroleum independence, and eventual fossil independence.  The PHEV is a hybrid with a bigger battery pack, a charger and a wall cord.  As little as 20 miles of all-electric range would handle the average daily commute and cover 30-40% of all driving; a PHEV60 with 60 miles of all-electric range would cover as much as 74% of all driving.  Combined with roughly 1/3 less inherent fuel consumption from hybrid economy, these cars could cut fuel demand by 50-60% with a PHEV20 to a whopping 80+% with a PHEV60.

Note that it's possible to convert a PHEV20 to a PHEV60 just by adding more or updated batteries.  Note further that the daily electric demand of these cars, perhaps 15 kWh per day for a PHEV60 driven its full range, is still small enough to carry over standard house wiring at 110 volts for an overnight charge (10 hours).  A 220 volt outlet would speed this drastically.

This would convert light-vehicle energy consumption from 17.6 quads of fuel to about 3.5 quads of fuel and perhaps another 2.3 quads of electricity (about 76 GW average).  The US consumes about 13.7 quads of electricity (around 4000 billion kWH, ~450 GW average) per year; we'd notice a 17% increase, but if it was scheduled at night, during windy periods, and when generation is otherwise in surplus the grid managers would love it.


The other half of the ground transport problem (that you don't touch) is heavy trucks.  These are almost exclusively diesel-powered, accounting for that 5.6 quads of fuel.  They're also a tougher nut to crack, because they run longer distances than light vehicles and could not get as much benefit out of conventional batteries.

There are ways to address this, too:

The last three could pretty much eliminate petroleum requirements for long-haul trucking.  The DCFC by itself could potentially replace 5.6 quads of diesel with 3.6 quads of carbon.  If we harvested 20 quads of biomass and converted it to charcoal at 50% energy efficiency (easily done with low-tech equipment), we'd be able to run the trucks and still have 6.4 quads left over.  Even zinc-air fuel cells could manage the job, either chemically (carbon reduction) or with about 6.5 quads of electricity to regenerate the zinc metal from zinc oxide.

This also ignores potential for more-efficient trucks.  Wal-Mart is aiming to double the economy of its fleet from 6.5 MPG to 13 MPG.  Cutting the problem in half makes it even easier.

Dealing with the remainder

Getting the USA completely off fossil fuels, or just imported petroleum, is a huge task.  Slashing some of the sub-tasks, like ground transport, down to size is an essential part but not the whole job.  For instance, even a full conversion to PHEV60's would leave perhaps 20% of demand for liquid fuel.  That's about 28 billion gallons of gasoline equivalent (GGE) per year, or about 135 million metric tons of ethanol.  As luck would have it, the process of producing charcoal for DCFC's also produces an off-gas which can be burned, fermented or otherwise processed into product.  Roughly 10% of the total biomass (130 million out of 1.3 billion tons) would be carbon in the off-gas.  Two companies are pushing processes for production of ethanol from syngas:

Ethanol is 24/46ths carbon by weight, or a hair more than half.  If half of the carbon in the off-gas could be converted to ethanol, that would be 130 million tons.  That might be a little short of the mark (especially given the difference between short and metric tons), but it's within striking distance of 100%.  A plant which uses algae to capture and recycle the unused carbon could boost this figure considerably, creating a surplus for many other uses.

These possibilities are far less developed than ethanol from grain and cellulose, but they actually have the potential to eliminate petroleum dependence.  They also have great possibilities for cutting atmospheric GHG's (for instance, the carbon from stationary DCFC's could be sequestered relatively easily... or recycled via algae).  Taken together, these possibilities make the three-sided problem of fossil-fuel dependence, oil imports and climate change look manageable.

What you could be doing

You're both wealthy and influential.  The first thing you should do is dump your positions in ethanol ventures and then state loudly why you did so.  Maybe you can help get the "flex-fuel" gas-guzzler loophole closed, just by making it obvious that you changed your mind and why.

You might also do one or several of the following:

In my less than humble opinion, the powers-that-be are promoting ethanol because it serves up subsidies to various interests while not threatening the status quo (oil companies).  If you can make an end-run around those interests, you could improve the environment, the economy and the prospects of the average American while making a huge pile of money.  Isn't that better than just being a shill for GM, the corn farmers and ADM?

Here's to your change of heart.

Update:  "Petajoules" corrected to "exajoules".

Labels: , , ,

"Fan mail from some flounder?"
"No, this is what I really call a message."

Have you looked over the MixAlco process developed by Holtzapple at Texas A&M? Just google MixAlco. No fancy enzymes and I believe he claims 15 to 1 EROEI. There's also a quicktime lecture posted on the A&M website.

He's also pushing a much more efficient engine to burn mixed alchohol fuels. This drops the land requirements. It's called StarRotor.
Have you looked over the MixAlco process developed by Holtzapple at Texas A&M? Just google MixAlco.

Holtzapple was my research advisor at Texas A&M. I did my thesis on the MixAlco process. No enzymes, true, but if he is getting 15 to 1 EROEI, he has made some quantum leaps in the past 10 years.
Sounds like a great way to eliminate landfills, Dovish, but the cellulose fraction of MSW is part of that 1.3 billion ton figure.  Given that, I can't see how such alcohols can provide enough energy to be more than niche players (and the low efficiency of the combustion engines they'd be used in argues against them).

Our real problem is the internal combustion engine and its lousy efficiency.  Going from 15% to 80% would solve most of our problems right there.
Our real problem is the internal combustion engine and its lousy efficiency. Going from 15% to 80% would solve most of our problems right there.

Yes, that the motivator for the star rotor concept. See if you haven't already.

Robert Rapier: yes, my memory could be way off on that EROEI.
Starrotor is just another combustion engine.  It has entropy losses in combustion, heat losses to conduction, and pressure losses to leaks.  Even if it doubles efficiency, it's still 2.5 times short.
Re: MixAlco and EROEI -

Holtzapple claims "10 to 15" to one, that is one unit of input yeilding 10 to 15 units of energy.

He says it right here about 2 minutes into the tape.
EROEI is only half the issue; the other half is quantity available.  There isn't even a ghost of a chance garbage will provide enough to continue the status quo, even if it's converted to fuel at 100% efficiency.

It's like the people saying that peak oil is bogus because reserves keep increasing.  But the rate at which those reserves can be pumped keeps going down....
I trust you understand that in addition to garbage he likes sweet sorghum and energy cane - anything from the plant kingdom that will rot in the sun.

He's got the land requirements and the economics all worked out. Yes, it's not a complete substitution but it's something like 80 percent of the way there.

Robert Rapier: please chime in - you worked with Holtzapple.

It does sound like the MixAlco process is interesting. However, there are several issues I see with it:

1. It requires addition of hydrogen at the end of the process to convert the organic acids into mixed alcohols. This is an energy intensive step, and to be affordable, that H2 would have to come (at this time) from natural gas, with CO2 generated as a waste product (he does suggest sequestration, but that problem remains to be definatively solved). If another cheap, carbon neutral way to generate H2 comes along, then we may not need mixed alcohols at all.

2. This is likely to have a lower EROEI than cellulosic ethanol techniques, because the materials are converted from CHO to ketones to acids and then back to alcohols (through the hydrogenation process), although this is not clear. The 10-15 to 1 number he quotes is impressive, but that is an awefully large spread. There's also a lot of ways to define EROEI, and I'd like to see an independent analysis along the lines of the ethanol analyses.

3. The starroter concept is again interesting, but there is not even a working prototype as of yet, and that engine will have tremendous problems with heat management.

If it works out great... a plug in hybrid with a starroter getting 50% efficiency is certainly better than a PHEV with a 15% efficient engine, though you can almost get to 40% even with today's diesel engines.

Is the mixalco process better than cellulosic ethanol (or butanol)? Is it better than the gassification process? Is it better than the solzinc process? Is it better than the carbonization/DCFC process?

It is pretty hard to say at this point. The guy is a good presenter, though.
As far as hydrogen goes, look here.

It's a good question: what is best? The old fashioned answer is: the market will sort it out.

Right now the market seems to be trending towards flex-fuel vehicles and hybrids. Can MixAlco play a part?

If the Amazon dies (a very scary thought) then look for governments to pick winners.
Interesting to see the CO2 sequestration H2 plants. With CO2 capture, though, will the process still be economic?

Will the EROEI still be so favorable?

When the process is scaled up resulting in the collection and mass transportation of heavy biomass to processing centers, and the shipping of the finished product (probably in trucks rather than pipelines) will the EROEI still be so favorable?

We'll have to see. If so, it will be very helpful (in conjunction with PHEVs lowering the amount of liquid fuels required), and Holtzapple will be very rich.

As well he should be.
Dovish, you didn't look too closely at your hydrogen link; it talks about using fossil fuels as the source (this isn't part of a hydrogen-fuel strategy, it's part of a carbon-sequestration strategy).

Hydrogen could be made by reforming some of the ketones; I don't see a big problem with that part, and it might make a great system to produce chemical feedstock.  But the quantities needed to feed internal combustion vehicles at our current scale are just too great.
E-P: My point is some very big kids are serious about producing hydrogen from FF with sequestration which can tie into MixAlco

Point taken about the scale and the alternatives to FF.
Robert Rapier: please chime in - you worked with Holtzapple.

Let me just say that Holtzapple is an eternal optimist. The glass is never, ever half empty. He is a great guy, and I loved working with him. But his weakness is that he will tend to overlook some of the negatives in the belief that problems will be easily solved.

I could go on and on with various issues that I dealt with when I was working on the process, but those things never make it into a thesis or paper.
E-P What's the approximate efficiency of a pure electric vehicle at the moment? (I'm just thinking of the pure-electric version of the well to wheels efficiency chart).
It's highly dependent upon the particular vehicle and where the energy is measured (between wall and charger, between battery and motor) but the figures I've seen for the tzero and Tango are 170-210 watt-hours/mile (out of the battery), and 250-odd Wh/mi at the charger for the Prius+.

If you take the energy equivalent of 250 Wh of gasoline (no conversion losses), it's about 0.0068 gallons; the effective "fuel" economy is almost 150 MPG.
This comment has been removed by a blog administrator.
That's 15-22% of the energy needs of my ICE at the moment.

Petroleum at roughly 8.76kWh/l works out to $0.20 per kWh at present. If you take into account the 15% efficiency, that's around $1.33 per kWh. Checking my last power bill, it's costing me around $0.06 per kWh.

What I was hoping to do was apply the losses associated with electric motors to that figure so it shows a better idea of the ongoing costs.

From where I'm sitting ($1.33 / kWh versus $0.06 / kWh) it seems like a no-brainer as to which is a better way to move ourselves around.

Obviously, this doesn't answer where the extra energy is meant to come from, but it certainly goes some distance toward justifying the cost of switching.

Of course, electric motors don't sound like V8s, so they'll always be met with opposition. Good on you humans.
On this ethanol question, you are clearly opposed to Vinod's methods for fuel...but I'm curious about the chances of all this cellulostic ethanol research developing useful alternatives for plastic feedstocks.
How likely is it that folks can find replacement (or even improved) bioplastics?
That depends on your process details and how much you want to spend.  If you gasify your feedstock and build monomers from syngas, you can make anything which can currently be made from syngas; however, losses (and costs) will be higher.  Starting with bio-molecules means that the choices of monomers and the relative costs are going to be substantially different from what petroleum chemists have to work with, and it's probably going to take a while for the industry to get to the same level of expertise it currently has with petroleum derivatives.
Post a Comment

Links to this post:

Create a Link

<< Home
Talk largely about energy and work, but also politics and other random thoughts

Mail Engineer-Poet

(If you're mailing a question, is it already in the FAQ?)

Important links

The Reference Library

Blogchild of

Armed and Dangerous

Blogparent of


The best prospect for our energy future:
Flibe Energy

January 1990 / February 2004 / March 2004 / June 2004 / July 2004 / August 2004 / September 2004 / October 2004 / November 2004 / December 2004 / January 2005 / February 2005 / March 2005 / April 2005 / May 2005 / June 2005 / July 2005 / August 2005 / September 2005 / October 2005 / November 2005 / December 2005 / January 2006 / February 2006 / March 2006 / April 2006 / May 2006 / June 2006 / July 2006 / August 2006 / September 2006 / October 2006 / November 2006 / December 2006 / January 2007 / February 2007 / March 2007 / April 2007 / December 2007 / January 2008 / May 2008 / June 2008 / August 2008 / October 2008 / November 2008 / December 2008 / February 2009 / March 2009 / April 2009 / May 2009 / June 2009 / July 2009 / August 2009 / September 2009 / October 2009 / November 2009 / December 2009 / January 2010 / April 2010 / May 2010 / June 2010 / July 2010 / August 2010 / September 2010 / October 2010 / November 2010 / December 2010 / January 2011 / February 2011 / March 2011 / April 2011 / May 2011 / July 2011 / August 2011 / September 2011 / October 2011 / April 2013 / November 2013 / December 2013 / January 2014 / February 2014 / March 2014 / April 2014 / July 2014 / August 2014 / September 2014 / October 2014 / November 2014 / February 2015 / April 2015 / October 2015 / March 2016 / April 2016 / May 2016 / June 2016 / July 2016 /

Powered by Blogger

RSS feed

Visits since 2006/05/11: