Tata motors has released the Nano for the India market, a $2000 car. They have already announced that they're looking into building a factory in Eastern Europe to build more of these, and there can be no doubt that Chinese versions will appear shortly. We can expect that in the next five to ten years, tens of millions of new drivers will appear in Asia, and that number will only increase. Meanwhile, there's a good chance that the total oil production on the earth has already peaked, so we're looking at increasing demand for a decreasing commodity. It's not hard to project that the price of oil will continue to climb under these circumstances. What will we do about this, as the price of gasoline rises to $4 / gallon and beyond? This is a look at various replacements for gasoline, and a projection for where cars are headed.
Gasoline contains a lot of energy: there is about 36 kilowatt-hours in each gallon. In 5 minutes you can easily put 500 kw-h of energy into your gas tank at any gas station for about $35, and then drive around town holding this 100 pounds or so of gasoline. 500 kw-h is roughly the total electricity you use in your house in a month. By contrast this same amount of energy stored in current batteries would take 300 hours to charge up, about 12 days, and the batteries would weigh about 20,000 pounds, ten tons, about the weight of three pickup trucks. This is why gasoline has been so popular for the last 100 years.
We need to start looking for a replacement for this energy for several good reasons: gasoline is getting very expensive; burning gasoline pollutes the air; we have far better industrial uses for oil than burning it; many of the people who profit from gasoline use the money for purposes which are questionable at best and deadly at worst. At the time of this writing gasoline is down from about $4.50 a gallon last year to perhaps $2.25 per gallon, and that price will likely hold for another year or two due to the recession, but I think no rational person projects gasoline will be under $3 per gallon in 2013.
Today we're using ethanol, alcohol made from corn or sugar, to supplement our gasoline. This solution has several serious problems: ethanol is not a very good automotive fuel. You need special engine parts so that the ethanol doesn't rot your engine. Using corn to make ethanol means we have to choose between eating and driving. It takes oil products to fertilize corn, and to plant and harvest it, enough oil that it's very unclear that corn ethanol replaces more gasoline than it uses. All told this is a poor solution.
Many companies are now working on what's called cellulosic ethanol, alcohol made from wood, trash, corn stalks. This technology has the promise of using trash to fuel our cars instead of fuel. Cellulose is familiar to everyone, it's the science name for wood. Cellulose molecules are basically sugar molecules strung together in long chains, so they contain a lot of energy - that's why we burn wood. Early estimates are that we should be able to make this ethanol for $1 / gallon. That's the good news. If gasoline went up to $5 / gallon, a 50-50 mix of gasoline and ethanol would be $3.
The bad news about cellulosic ethanol is it still rots engines, and we would still need to devote a lot of farm land to growing grasses to supply the cellulose. Roughly speaking, an acre of land can make about 1000 gallons of ethanol per year. A gallon of ethanol has a bit less energy than a gallon of gas, it takes ten gallons of ethanol to replace six gallons of gas. So to completely replace gasoline in the US we would need to plant about 400,000 square miles of grasses. This is roughly equal to the total area of the states of Nebraska, Kansas, Iowa, Missouri, Illinois, Indiana and Ohio. Thus cellulosic ethanol may become an important part of the solution, but we can easily see that we're not going to replace more than perhaps 10% to 15% of our current gasoline use with ethanol.
Growing enough ethanol to replace all the gasoline we burn uses most of our farmland.
There used to be a lot of talk about the coming "hydrogen economy," but that's all quietly disappeared lately. The reason is that hydrogen holds even less energy per gallon than ethanol. It's estimated that it will take $8 to $10 worth of hydrogen to replace a gallon of gasoline. Hydrogen in a pressure tank tends to evaporate at about 1% per day. It would cost an estimated $50B to build plants to make hydrogen and convert gas stations to hydrogen stations. This is a huge investment for a technology that's never going to work very well. Fuel cell cars are not happening. I drive past the California Fuel Cell Partnership building twice a day (space currently available for lease), and I laugh at it twice a day.
Another solution is plug-in hybrids. A normal car burns gasoline to accelerate, then uses its brakes to convert energy to heat to slow down. A hybrid has regenerative braking - when you use the brakes lightly to moderately a generator puts the energy is put into a battery, then the next time you accelerate the battery is used to power an electric motor. The Honda Insight has an 85hp gasoline engine and a 13hp electric motor, and uses the combination to achieve 50+ miles per gallon overall. However, in the end all the power comes from gasoline or ethanol. A plug in hybrid has larger batteries that you can charge overnight. You can then drive for a while on stored electricity only before using any gasoline at all. The advantage to this is that the electric motor in these cars is about 95% efficient, while the gasoline motor is only about 25% efficient, with 75% of the gasoline being wasted as heat. The Chevy Volt will travel about 40 miles before the gasoline engine ever starts, at an energy equivalent of 180 miles per gallon.
Because of the efficiency of using electricity, you only need about a quarter as much energy to drive around on battery power as on gasoline. So there are two ways to compare electricity and gasoline: by energy content, or by energy per mile. It's energy per mile that really concerns us. This is good, as the 150B gallons of gasoline we use translates into 5.4tw-h of electricity, about 50% of all the energy used in this country. However, to replace gasoline with electricity we don't need to double our electricity production. Because electric motors are so much more efficient than gas motors we only need to add about a third again to the electric generation we already have. We could consider using coal, oil, nuclear, wind, or solar power to supply this energy. We can't use more hydro-electric power, as we've already built pretty much all the useful dams we can build in the US. We'll ignore coal and oil as sources for this electricity as they are very much out of favor right now.
There are two popular ways to turn solar power into electricity: you can use solar panels, or you can use mirrors to focus the sun's energy to heat water, then use steam turbine generators just as a coal, oil, or nuclear reactor does. We have built large pilot plants for both of these technologies. The result: either way, solar plants generate about 35kW per acre, capturing about 5% of the sunlight's power. Driving cars in the US means replacing about 160 Gw of power, so solar energy would require about 10,000 square miles. Even better, the best land for solar has a lot of sun, no clouds, no rain - in short, a desert - so solar power doesn't compete with farm land. We can both drive and eat. 10,000 square miles would cover much of the west Texas desert, which, personally, I would consider an improvement over what's there now. That's the good news. The bad news is that this would cost about $300B for solar cells, or $500B for solar thermal power. However, this must be compared to buying oil, which costs us about $800B per year, so switching to solar would pay for itself in a year or two. Another problem with solar is that the power is generated during the day, but we want to use it to charge our cars at night. We need an inexpensive way to store a lot of power with good recovery, like pumping a bunch of water up a hill during the day and using it to generate power at night. This is already done in Los Angeles, where water is pumped up to Pyramid lake at night when power is cheap and lowered to Castaic lake during the day when power is expensive.
Solar power to replace all the gasoline we burn uses much of West Texas.
Another possible solution is wind power. The costs on this have come down to the point where it is price competitive with everything but coal. Wind power is currently about 5 cents per kw-h, which corresponds to gasoline at $2.10 per gallon in terms of energy content, or gasoline at about 60 cents per gallon in terms of mileage. In order for wind power to be cost competitive, there has to be wind in excess of about 12 mph most of the time, like in North Dakota. In the map below, medium blue like in most of Nebraska and Iowa is marginal, dark blue and black are excellent.
Wind power is mostly available in the mountains and on the northern prairies.
Another possibility is nuclear power. A typical nuclear reactor makes about 1G watt and costs about $2B, so it seems that nuclear power is about the same cost as solar power. We would apparently need about 150 nuclear reactors to power all our cars. Nuclear reactors must be water cooled, which means our coast lines and major rivers would have a nuclear reactor about every 75 miles. One can see that there would be a certain resistance to building all these reactors. Furthermore, we would run out of uranium in a couple hundred years. There's more power available on the earth in coal than in uranium. However, with further development of breeder reactors, we could probably extend the useful life of nuclear power by several hundred years.
All of these electrical approaches require we make some important changes in our infrastructure. We need a new transmission grid that can economically carry wind power from the mountains, solar power from the southwest, and nuclear power from the coasts and rivers to the cities where it's needed. We need energy storage facilities that can pump small lakes up and down mountains so that the power is available when it's needed, not when it's made. And we need smart meters on houses so that power consumption can be charged at different rates at different hours. I'm currently with PG&E, a loathsome monopoly, and I would be charged 42 cents per kilowatt hour to recharge a car. At this rate gasoline is cheaper than electricity. With a smart meter and a bit of appropriate legislation, that would be more like 8 cents per kilowatt hour when used between 11pm and 6am. In fact, to his credit, Obama is starting these projects.
What we can expect in the next decade is a combination of these approaches: we will continue to buy some oil, hopefully a quarter or less than what we're buying now. We will make some cellulosic ethanol, mostly made from trash, farming leftovers like corn stalks, and prairie grass. Much of our gasoline will be mixed with 15% to 50% ethanol to reduce the price. Many of our cars will be plug-in hybrids which travel 10 to 50 miles on a charge, and then burn gasoline for longer range or higher performance. We will have a bunch of new nuclear reactors, solar farms in Arizona, New Mexico, Nevada, Southern California, Texas, and wind farms in the Sierras and Rockies. We will continue to be told that cheap, safe, clean, reliable fusion power is about three years away.
We will need a new and improved electric grid to efficiently distribute all this power. Cars are normally junked after about eight to ten years on average, so a decade from now most of the cars on the road will naturally be hybrids of some sort. It's going to cost a lot to build all this new power generating capacity, but the result will be far less dependence on foreign oil and far less air pollution in our country. Perhaps more importantly, the technology we develop will be used around the world as people in Asia start driving in large numbers, which will be important to the earth's ecology as a whole.
gallon gasoline = 130MJ = 36KW-H = 4W-YR
US use: 150B gallons gasoline / year = 5400 Tw-h = 630 gw @ 8750 hours per year.
required electricity (including electric motor efficiency) = 1350 Tw-h = 160 gw @ 8750 hours per year.
current US electricity consumption: 3.8 Tw-h
Bio-ethanol = 1000 gallons / acre / year = 600 gallons gas equivalent / acre /year.
250B gallons ethanol = 250M acres = 400,000 square miles
Solar insolation = 200 w / meter average (S.US) = 800kw / acre = 500Mw / sq mi.
Solar power = 5% of insolation = 40 kw / acre = 25Mw / sq.mi.
Chevy volt travels 40 miles on 8kw-h = 200w-h/mile = 180mpg equivalent.