Friday, July 18, 2008

Energy independence, the stinky way

Problem: Industrial civilization faces a shortage of liquid fuels, mainly because we've based our fuels on fossil fuel hydrocarbons. These are becoming increasingly scarce, and even if they weren't we can't afford to keep burning them because of the environmental impacts. Liquid fuels have certain inherent, irreproducible advantages that make them desirable if we believe (as I do) that there is a reasonable amount of automobile use that deliver a real human benefit of increased mobility.

Possible solution #1: Biofuels derived from plants. There's a whole bunch of different candidates here -- from ethanol to biodiesel, made from corn or super-efficient algae, but the latter has proved far more difficult to recreate in practice than early theory hoped. The more fundamental problem with crop-based biofuels is that it naturally tends to displace food production, and unless we want to have the government get in to the business of telling farmers what crops they can grow on which soil (and we do some of this already) we put ourselves in to the unenviable position of trying to balance our energy and food production on the same scarce resource -- arable cropland.

An additional consideration is that any carbon we do capture via sustainable crop harvests really ought to be buried somewhere to start sequestering the carbon we've been dumping in the air for more than a century.

Possible solution #2: Electricity as a transport fuel. Yes, more, and as fast as possible please. But there are some hard limitations to electricity, namely recharging speeds and energy density. Even the as-yet-vaporware EEStor capacitors will have a range more limited than most drivers would prefer, and charging them quickly (i.e., a few minutes) is a daunting engineering challenge. To deliver 50 kwh in a matter of 10 minutes requires 3 megawatts 300 kilowatts of power. Now, you might be able to engineer the cabling and safety features, but deliver that kind of power to, say, 12 "pumps" would require building a small gas-fired plant at any charging station along the highway. (We're assuming you'd only need rapid charging for long trips, as daily commuting could be slowly charged overnight.) Meanwhile, liquid fuels can be pumped quickly and efficiently, without large infrastructure changes. [graf changed -- see below]

This is why, btw, plug-in hybrids are such a great idea -- rather than attempt to build a "perfect" electric car, let electricity do the lion's share of the work (small daily trips) but keep a small internal-combustion engine around for long-distance drives. And if fuel cells ever materialize for automotive purposes, the plug-in paradigm still has a role to play, though this time the car can also be a generator as well as a consumer.

So if it would be prudent to limit our use of biofuels, but we will still need a decent volume of liquid fuels for mobility, what's the appropriate technology? Well, ideally we want a liquid fuel (liquids are preferable for their energy density) that isn't dramatically less efficient than gasoline, can be created, stored, and moved with existing technology, and wouldn't add pollution to the atmosphere, land, or waters of the Earth. Corn ethanol passes on some counts, but overall is a failure. Cellulosic ethanol or biomass gasification will probably do better, and should be considered. Are there options that have been ignored?

Well, maybe not ignored, but I've just been reading about one of them lately -- ammonia. If you're in the mood to read a whole bunch of engineering reports on the progress various companies have made on using ammonia as a fuel, read some of the conference papers here. Suffice it to say that this is an active and interesting field for an energy nerd like me.

The properties of ammonia are initially promising: under modest pressures it's a liquid, in which form it carries almost twice as much hydrogen as a similar volume of liquid hydrogen would. It can be combusted in a regular spark-ignition engine and, if you believe the claims of the Hydrogen Engine Center, still emit less NOx pollutants than a comparable gasoline engine. It is one of the most commonly produced chemicals on Earth, meaning there's a robust and proven infrastructure for production, storage, and transportation.

The downsides: It's about half as energy-dense as gasoline, it smells foul, and it's currently produced using natural gas, a fossil fuel. If we were to continue using ammonia derived from natural gas, it would be more efficient to simply burn natural gas directly. But the other two factors are not show-stoppers: indeed, a foul smell is a safety feature, and given that a fair medium-term estimate for fuel efficiency in new cars is 60 mpg, an ammonia-fueled car could still travel about as far in a future vehicle as gasoline-powered cars do today.

So is it possible to produce ammonia renewably? Absolutely. Electrolysis of water gives you the hydrogen you need, and nitrogen from the atmosphere is abundant and omnipresent on Earth. The problem is that this has traditionally been pretty expensive when compared to natural gas-based ammonia. The search for cheap abundant hydrogen for fuel cells has already started to produce potentially cheaper technologies to electrolyze water. The other cost is the electricity to power all this, but this I'm less concerned about: between nanosolar, wind, and geothermal the price of all renewables is coming down very very quickly.

A number of midwestern American universities are looking at ammonia production as a way to exploit "stranded wind", windy areas that are too far from the established grid to make them profitable. (Note that the richest solar resources are similarly "stranded" because few of us live in the deserts.) Ammonia production also offers a way to make renewables less intermittent, if they're paired with ammonia-fired turbines. From an engineering standpoint, there seems to be little that would stop widespread adoption of ammonia as another solution to keep in our basket of options.

Ah, but can we make enough of the stuff to be worthwhile? First, consider that of course while we look for a replacement liquid fuel, we're trying to minimize the amount of fuel we need, through mass transit, walkable cities, and more fuel-efficient cars. So we're not looking at replacing 100% of our oil usage with ammonia -- much of it will be replaced with electricity, or walking. I think a fair guess is that we only want to replace 10-20% of our liquid fuel use. So it should be eminently possible for, say, solar power in the American Southwest (or Sahara or Arabia) to generate enough clean, cheap electricity, send it via cable to the nearest shores, and make ammonia out of seawater and air.

After we've made it, can we move the stuff around? "Easily" might be too strong a word, but with way less difficulty than, say, liquid natural gas. It could certainly be shipped over enormous distances by sea or by pipeline. Ammonia can be stored and transported in much the same way we move propane around. The US midwest already has about 3,000km of ammonia pipelines thanks to the fertilizer industry. In the long run, it might be profitable to build massive floating solar farms in the oceans to generate ammonia which they could offload on to tankers. A balloon inflated with liquid ammonia, but submerged at a depth of 200-300 feet, would keep it under pressure and cooled without further energy inputs. Depending on the engineering, you might not even need a tanker -- just drag the balloon back to shore and feed the contents in to a pipeline. A tugboat would be all you needed.

And can we use the stuff, on the scale we're talking about, without harming the Earth? Burned in a turbine or engine, ammonia will release small amounts of NOx, but probably no worse than existing gasoline engines, with no carbon dioxide emissions whatsoever. In fuel cells, it should be possible to eliminate NOx production entirely, meaning that the ammonia would be decomposed in to what we started with -- inert nitrogen gas and water.

So we've got a liquid fuel, with passable energy density, that can be produced on massive scales using renewable electricity, and used without unduly harming the Earth. Not bad. Some things this can't do:

1) Bring us back to the days of cheap driving. I would honestly be surprised if, even with optimistic assumptions, somebody could plausibly produce ammonia as a automotive fuel for less than current prices of gasoline. The days of $1/gallon are long behind us.

2) Bring back cheap flight. Ammonia will forever have too low an energy density to be used as a jet fuel, I would think. It might do fine as a fuel for trains, though.

3) Make us less stupid. Given the technologies at hand, it really would be insane to try and use ammonia to replace all motor fuels, but we may actually try to. Our love of suburbia and big cars is insatiable, and any alternative fuel faces the problem of people not wanting to change.

I'm sure there are problems I haven't covered, and issues that need to be discussed. That's why God invented comment threads.

UPDATE: Corrected a mathematical error above, and it deserves a little elaboration. I overstated the amount of electricity needed to fast-charge a capacitor by an order of magnitude, so, uh, oops. That said, even a 90% reduction in the amount of power required doesn't change things that much.

Patric -- who pointed out my math error, in comments -- says that people wouldn't want internal-combustion cars when an electric car would be cheaper to maintain and last longer, and there's a number of reasons to believe he's right: most of the things that break in modern cars don't exist (or don't need to) in an EV.

That said, energy density is still a key problem: it's cheaper and easier to pump energy on large scales in the form of a liquid than it is to move it quickly over wires. For long-term trips, people aren't going to want to have to wait even 1 hour for their car to recharge before getting back on the road again. If the tank can be filled in 5 minutes and take them on their way again for another 300-400 miles, that's preferable and even arguably more efficient. (Time costs money, too!)

That's why I think that, aside from unreconstructed urbanites who almost never drive long distances on highways, a pure EV is not actually an ideal solution.


patric said...

Very interesting blog, I will be sure to book mark it.

However, you stated that the problem of "fast charging" an electric car as follows: "To deliver 50 kwh in a matter of 10 minutes requires 3 megawatts of power." In fact, it only requires one tenth that much, or 300 kilowatts (you made a mathematical error I believe). This is hardly "high power" in distribution terms. Regular pole mounted transformers (the size of a small garbage can) as you would find in the streets of Toronto deal with those power levels and much more.

In fact, there is no need to build "distribution" stations all over the place to deliver that power. Each user or EV owner could have a smallish unit that is recharged using normal house hold current during the day, and then used to charge up the car in 10 minute bursts. Household current of 40 amps and 220v yields a power rate of 8.8 kw per hour, allowing for a complete recharge of a 50 kwh battery in less than 6 hours. This is the approximate maximum current drawn by cloths dryer or electric stove/oven.

However, you're point is well taken. Wide spread EV use will drive up the demand for electric power. If that extra electricity demand is provided by coal, then there is still a net CO2 savings b/c electric cars are much more energy efficient than gasoline powered cars. However, the net CO2 savings would be far greater if the electricity was provided by something other than coal (insert greener technology here).

The main advantage of electric cars; however, is simple. If the electric storage factor is solved (which is SLOWLY happening), then nobody will want a gas powered car. Electric motors are cheap, super reliable and require no maintenance. EV's, if produced in numbers (and if they were equipped with some "dream battery" like an ultra-capacitor), would cost less than half of a gas powered car, would require zero maintenance and would last for decades. Since electric motors are so robust and reliable, you could drive an electric car for 30 years or more, changing only the tires every few years. With that kind of competition, no one would want a gas powered car.

Catelli said...

What are your thoughts on pure Hydrogen as a fuel? BMW already has a flex-fuel car in limited production that can burn either hydrogen or conventional gasoline in an IC engine.

Something that allows us to use our existing fuel infrastructure while building up a replacement makes a hell of a lot of sense to me. Going this route allows the future development of (purportedly more energy efficient) fuel-cell technology while actively promoting (and driving investment) into green hydrogen extraction technologies.

Win-Win. We're not relying on massive social change, but yet we still lower our dependence on oil as fuel and green up as well.

It appears so simple and elegant a solution that I'm obviously missing something.

john said...

Pure hydrogen is possible, and in some cases it might be preferable to ammonia or other liquid fuels. But for automotive purposes H2 isn't great -- basically it can't be stored in sufficient quantities. You need super-compressed hydrogen (10,000 psi) and it still doesn't get you as far. Liquid ammonia holds 1.7x as much Hydrogen as liquid H2 would, and we have no idea how to build a liquid H2 tank for a car -- it needs to be chilled to just above 0 kelvins.

But yes, I agree that it would be sensible to retain the benefits of internal-combustion engines until (or if) fuel cells or something better comes along. Ammonia is also a potential fuel-cell fuel, if you have an on-board reformer (something to strip the hydrogen from the nitrogen before it reaches the cell.) This is necessary because in most cases ammonia poisons fuel cells.

Gar Lipow said...

I don't think pure electrics are as bad a choice for most people as you think. The average car trip is around 33 miles, and very seldom do most people drive more than 50 miles in a day. The TESLA has a 200+ mile range albeit at an absurdly high price. The Triac has a 100 mile range for $20,000. Mass production could bring these costs down. Essentially they are being produced today buying parts at close to retail. Most people travel more than 200 miles in day seldom enough they can rent a plugin hydrid for rare occasions when they need something to travel longer.