A taste for irony is a useful habit to cultivate if you happen to write about energy issues in the declining years of a civilization defined by its extravagant use of energy, on the one hand, and the dubious logic it uses to justify that extravagance on the other. One of the things you can count on, if that description fits you, is that any time you discuss one of the fallacies that has helped back that civilization into a corner, plenty of readers will respond with comments that demonstrate the fallacy in question more clearly than any of your examples could have done.
Last week’s Archdruid Report post was no exception to that rule. Regular readers will recall that it focused on the difference between the quantity of energy in an energy source and the concentration of energy in that energy source, and pointed out that the latter, not the former, determines the exergy in the source – that is, the amount of work that the energy source is able to perform. True to form, I fielded a flurry of comments that took issue with this, or with the conclusions I drew from it, on the grounds that I wasn’t paying enough attention to the quantity of energy in some favorite energy source.
The example I’d like to highlight here is far from the worst I received. Quite the contrary; it’s precisely because it’s a thoughtful response from an equally thoughtful reader that it makes a good starting point for this week’s discussion. The reader in question pointed out that the photons that reach the Earth from the Sun each contain exactly as much energy as they did when they left the solar atmosphere, and argued on that basis that a point I made about the exergy of solar power was at least open to question.
He’s quite right about the photons, of course. The energy contained in a photon is defined by its frequency, and that remains pretty much the same (barring a bit of gravitational redshifting) from the moment it spins out of the thermonuclear maelstrom of the Sun until the moment eight minutes later when it arrives on earth and gets absorbed by a green leaf, let’s say, or the absorbent surface in a solar water heater. Once again, though, that’s a matter of the quantity of energy, not the concentration. The concentration, in this case, is determined by the rate at which photons impact the leaf or the solar panel; that depends on how widely spread the photons are, and that depends, in turn, on how far the leaf and the panel are from the Sun.
Think of it this way. The individual photons that heat the planet Mercury each contain, on average, the same quantity of energy as the individual photons that heat the planet Neptune. Is Neptune as warm as Mercury? Not hardly, and the reason is that by the time they get out to the orbit of Neptune, the Sun’s rays are spread out over a much vaster area, so each square foot of Neptune gets a lot fewer photons than a corresponding square foot of Mercury. The photons are less concentrated in space, and that, not the quantity of energy they contain, determines how much of the hard work of heating a planet they are able to do. There are stars in the night sky that produce photons far more energetic, on average, than those released by the Sun, but you’re not going to get a star tan from their light!
This may seem like an obvious point. Still, it deserves restatement, because so many contemporary plans for using solar energy ignore it, fixating on the raw quantity of solar energy that reaches the Earth rather than the very modest concentration of that energy. A habit of comforting abstraction feeds that sort of thinking. It’s easy to insist, for example, that the quantity of solar energy falling annually on some fairly small fraction of the state of Nevada, let’s say, is equal to the quantity of energy that the US uses as electricity each year, and to jump from there to insist that if we just cover a hundred square miles of Nevada with mirrors, so all that sunlight can be used to generate steam, we’ll be fine.
What gets misplaced in appealing fantasies of this sort? Broadly speaking, three things.
The first is that familiar nemesis of renewable energy schemes, the problem of net energy. It would take a pretty substantial amount of highly concentrated energy to build that hundred square mile array of mirrors, counting the energy needed to manufacture the mirrors, the tracking assemblies, the pipes, the steam turbines, and all the other hardware, as well as the energy needed to produce the raw materials that go into them – no small amount, that latter. It would take another very substantial amount of concentrated energy, regularly supplied, to keep it in good working order amid the dust, sandstorms, and extreme temperatures of the Nevada desert; and if the amount of energy produced by the scheme comes anywhere close to what’s theoretically possible, that would probably be the only time in history this has ever occurred with a very new, very large, and very experimental technological project. Subtract the energy cost to build and run the plant from the energy you could reasonably (as opposed to theoretically) expect to get out of it, and the results will inevitably be a good deal less impressive than they look on paper.
The second is another equally common nemesis of renewable energy schemes, the economic dimension. Plenty of renewables advocates say, in effect, that people want electricity, and a hundred square miles of mirrors in Nevada will provide it, so what are we waiting for? This sort of thinking is extremely common, of course; mention that any popular technology you care to name might not be economically viable in a future of energy and resource constraints, and you’re sure to hear plenty of arguments that it has to be economically feasible because, basically, it’s so nifty. There’s a reason for that – it’s the sort of thinking that works in an age of abundance, the kind of age that’s coming to an end around us right now.
The end of that age, though, makes such thinking a hopeless anachronism. In an age of energy and resources constraints, any proposed use of energy and resources must compete against all other existing and potential uses for a supply that isn’t adequate to meet them all. Market forces and political decisions both play a part in the resulting process of triage. If investing billions of dollars (and, more importantly, the equivalent amounts of energy and resources) in mirrors in the Nevada desert doesn’t produce as high an economic return as other uses of the same money, energy, and resources, the mirrors are going to draw the short end of the stick. Political decisions can override that calculus to some extent, but impose an equivalent requirement: if investing that money, energy, and resources in mirrors doesn’t produce as high a political payoff as other uses of the same things, once again, the fact that the mirrors might theoretically allow America’s middle classes to maintain some semblance of their current lifestyle is not going to matter two photons in a Nevada sandstorm.
Still, the problems with net energy and economic triage both ultimately rest on thermodynamic issues, because the exergy available from solar energy simply isn’t that high. It takes a lot of hardware to concentrate the relatively mild heat the Earth gets from the Sun to the point that you can do more than a few things with it, and that hardware entails costs in terms of net energy as well as economics. It’s not often remembered that big solar power schemes, of the sort now being proposed, were repeatedly tried from the late 19th century on, and just as repeatedly turned out to be economic duds.
Consider the solar engine devised and marketed by American engineer Frank Shuman in the first decades of the 20th century. The best solar engine of the time, and still the basis of a good many standard designs, it was an extremely efficient device that focused sunlight via parabolic troughs onto water-filled pipes that drove an innovative low-pressure steam engine. Shuman’s trial project in Meadi, Egypt, used five parabolic troughs 204 feet long and 13 feet wide. The energy produced by this very sizable and expensive array? All of 55 horsepower. Modern technology could do better, doubtless, but not much better, given the law of diminishing returns that affects all movements in the direction of efficiency, and most likely not enough better to matter.
Does this mean that solar energy is useless? Not at all. What it means is that a relatively low-exergy source of energy, such as sunlight, can’t simply be used to replace a relatively high-exergy source such as coal. That’s what Shuman was trying to do; like most of the solar pioneers of his time, he’d done the math, realized that fossil fuels would run out in the not infinitely distant future, and argued that they would have to be replaced by solar energy: “One thing I feel sure of,” he wrote, “and that is that the human race must finally utilize direct sun power or revert to barbarism.”
He may well have been right, but trying to make lukewarm sunlight do the same things as the blazing heat of burning coal was not the way to solve that problem. The difficulty – another of those awkward implications of the laws of thermodynamics – is that whenever you turn energy from one form into another, you inevitably lose a lot of energy to waste heat in the process, and your energy concentration – and thus the exergy of your source – goes down accordingly. If you have abundant supplies of a high-exergy fuel such as coal or petroleum, that doesn’t matter enough to worry about; you can afford to have a great deal of the energy in a gallon of gasoline converted into waste heat and pumped out into the atmosphere by way of your car’s radiator, for example, because there’s so much exergy to spare in gasoline that you have more than enough left over to send your car zooming down the road. With a low-exergy source such as sunlight, you don’t have that luxury, which is why Shuman’s solar plant, which covered well over 13,000 square feet, produced less power than a very modest diesel engine that cost a small fraction of the price and took up an even smaller fraction of the footprint.
This is also why those solar energy technologies that have proven to be economical and efficient are those that minimize conversion losses by using solar energy in the form of heat. That’s the secret to using low-exergy sources: heat is where exergy goes to die, and so if you let it follow that trend, you can turn a relatively diffuse energy source to heat at very high efficiencies. The heat you get is fairly mild compared to (say) burning gasoline, but that’s fine for practical purposes. It doesn’t take intense heat to raise a bathtub’s worth water to 120º, warm a chilly room, or cook a meal, and it’s precisely tasks like these that solar energy and other low-exergy sources do reliably and well.
It’s interesting to note that Augustin Mouchot, the great 19th century pioneer of solar energy, kept running up against this issue in his work. Mouchot began working with solar energy out of a concern that France, handicapped by its limited reserves of coal, needed some other energy source to compete in the industrial world of the late 19th century. He built the first successful solar steam engines, but they faced the same problems of concentration that made Shuman’s more sophisticated project an economic flop; a representative Mouchot engine, his 1874 Tours demonstration model, used 56 square feet of conical reflector to focus sunlight on a cylindrical boiler, and generated all of 1/2 horsepower.
Yet some of his other solar projects were quite a bit more successful. For many years, the French Foreign Legion relied on one of his inventions in their North African campaigns: a collapsible solar oven that could be packed into a box 20 inches square. It had the same general design as the engine, a conical reflector focusing sunlight onto a cylinder that pointed toward the sun, but it worked, and worked well; the Mouchot oven could cook a large pot roast from raw to well done in under half an hour. Another project, a solar still, proved equally successful, converting wine into brandy at a rate of five gallons a minute – rather good brandy at that, “bold and agreeable to the taste,” Mouchot wrote proudly, “and with...the savor and bouquet of an aged eau-de-vie.” Again, notice the difference: low-exergy sunlight doesn’t convert well to mechanical motion via a steam engine, due to the inevitable conversion losses, but it’s very efficient as a source of heat.
The implications of this difference circle back to a point made by E.F. Schumacher many years ago, and discussed several times already in these essays: the technology that’s useful and appropriate in a setting of energy and resource constraints – for example, the Third World nations of his time, or the soon-to-be-deindustrializing nations of ours – is not the same as the technology that’s useful and appropriate in a setting of abundance – for example, the industrial nations of the age that is ending around us. Centralized power generation is a good example. If you’ve got ample supplies of highly concentrated energy, it makes all the sense in the world to build big centralized power plants and send the power thus produced across hundreds or thousands of miles to consumers; you’ll lose plenty of energy to waste heat at every point along the way, especially in the conversion of one form of energy to another, but if your sources are concentrated and abundant, that doesn’t matter much.
If concentrated energy sources are scarce and rapidly depleting, on the other hand, this sort of extravagance can no longer be justified, and after a certain point, it can no longer be afforded. Since much of the energy that people actually use in their daily lives takes the form of relatively mild heat – the sort that will heat water, warm a house, cook a meal, and so on – it makes more sense in an energy-poor society for people to gather relatively diffuse energy right where they are, and put that to work instead. The same point can be made with equal force for a great many industrial processes; when what you need is heat – and for plenty of economically important activities, such as distilling brandy, that’s exactly what you need – sunlight, concentrated to a modest degree by way of reflectors or fluid-heating panels, will do the job quite effectively.
This is another reason why Schumacher’s concept of intermediate technology, and a great many of the specific technologies he and his associates and successors created, provide a resource base of no little importance as the world’s industrial societies stumble down the far slopes of Hubbert’s peak. When concentrated energy is scarce, local production of relatively diffuse energy for local use is a far more viable approach for a great many uses. This will allow the highly concentrated energies that are left to be directed to those applications that actually need them, while also shielding local communities from the consequences of the failure or complete collapse of centralized systems. The resulting economy may not have much resembance to today’s fantasies of a high-tech future, but the barbarism Frank Shuman feared is not the only alternative to that future; there’s something to be said for a society, even a relatively impoverished and resource-scarce one, that can still reliably provide its inhabitants with hot baths, warm rooms in winter, and well-done pot roasts – and, of course, good brandy.