Thinking about hydrogen

It seems some German scientists have found a new way to produce hydrogen straight from sunlight and water. This is excellent. But it took me a while to see why it is excellent, hence this post; it’s not obvious at first glance.

Brief recap of basic physics: To make hydrogen (and oxygen as a worthless byproduct) from water, you put in energy. In particular, the binding energy. The binding energy is the heat released when two fancy-free hydrogen atoms commit to a stable relationship with an oxygen atom, forming water. This energy is fixed. There is no trick, no technology, no scientific advance that will permit you to produce hydrogen from water for anything less. Hence my immediate reaction on seeing the article: “So what?”, quoth I. “The binding energy is fixed; unless they’re claiming to have a perpetual-motion machine, it’s no better than any other production method.”

On reflection, however, I realised that there can be a gain; because what you can do is find more efficient ways of supplying that energy. Usually we supply it in the form of electricity, that is, a voltage difference, which has to be generated somehow. That generation is not usually very efficient. Other than nuclear and geothermal power, all power generation ultimately comes from the Sun, generally through many steps of not-very-high efficiency. Fossil fuels, for example, are fantastically inefficient; the deposits we’ve pretty much burned through in a century were laid down over millions of years of sunlight. That’s why they’re so great as fuels: Your average lump of coal concentrates the sun-energy of a large tract of land over hundreds of years. More accurately, it concentrates a tiny fraction of that sun-energy. The rest was re-radiated into space as waste heat, the eventual fate of all energy that doesn’t get bound into chemical or other storage. Coal is a very effective source of energy: You get a lot of bang. But it’s not very efficient, relative to the sun-energy that went into producing it: Per unit solar constant, you get practically nothing. Generally we don’t care about that, because we don’t have to supply the input.

So, what this does is remove several low-efficiency steps from hydrogen production. The old order would go sunlight->(fossil fuel / hydro / wind)->electricity->hydrogen. Three steps, each with some inefficiency. In the case of fossil fuels, the inefficiency is so large that once you run out of coal deposits, you can’t produce hydrogen this way at all; it’s only the accumulation of several million years of an inefficient process that lets us use coal as an energy source. The new process is sunlight->hydrogen, one step! Whatever the efficiency, it’s sure to be higher than the chain above.

A little counter-intuitively, this should not be thought of as contributing to the hydrogen economy. What it does is to make large-scale solar power viable. (Assuming, of course, that the process scales up, which may not be true.) Consider: At the moment, to get electricity from sunlight, you have to have a solar panel, made from semiconductors and whatnot. On any sort of industrial scale, this is prohibitively expensive. What’s more, there is no way to store the energy: At any given moment, what you see of the Sun is what you get. If it’s the middle of the night and you need extra power because all of California is staying up late to watch a comet, too bad – rev up them coal-burning power plants. Or if it’s a nice sunny day and you produce more than you need, well, you can pump water into towers or other sorts of mechanical energy storage, but that adds two steps and you lose most of the energy. With hydrogen, that problem is solved: If you’re producing more hydrogen than you need, put it in your storage tank for the later comet-watching party. And what’s more, you don’t have to do a transformation up and down in voltage, and build huge electric lines from expensive copper, to move the energy from the production site in the middle of the desert to where people live: Instead, you load it on the back of a tanker and put your power plant in the middle of the city. (Modulo safety concerns, that is; hydrogen is explosive, after all. But you can get a lot closer than the middle of the Mojave.)

Our entire energy problem is, in effect, a problem of storing energy, and transporting it. Energy, as such, is practically free, in the form of sunlight. The problem is that, while it’s raining lots and lots of soup, the rain is a light drizzle at any given spot – more of a mist, really – and we don’t have big enough buckets, so we can’t collect as much as we’d like to eat. (That is, the total solar power coming in is huge, but so is the Earth, so it’s very thinly distributed.) Solar and wind power are the equivalent of opening your mouth in such a rain: You can get enough for that particular moment and that spot, but you can’t store up a lot of soup for when you need it later, or need a lot of it at one point. Oil is the equivalent of a field full of buckets, which you don’t know how to make: It has been filling up for hundreds of years from the rain, and you can easily carry it elsewhere, but a hundred thirsty people are going to empty it out a lot faster than it refills. Hydro power is a little like a stream: It concentrates the rain in one spot, so you can drink your fill, but it only has so much flow, and only occurs in certain spots.

Hydrogen, in this analogy, is a bucket that we know how to make: It lets you move the soup around easily. Sunlight-to-hydrogen is an efficient way of filling that bucket – an artificial concentrator of the soup, as it were.

Advertisements

Leave a comment

Filed under Economics, Physics

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s