Saturday, August 2, 2008

Wealth of Nations...

Megan McArdle posted the map at right (along with two other related and equally interesting maps) a couple days ago. Her post is mainly concerned with the indisputable fact that the wealthiest countries are all either part of the English-speaking world (the “Anglosphere”) or are near them.

First a word about this map (which you can click to enlarge): the darker the shade on the map, the more GDP per capita. The latter is a simple calculation: take the total value of all goods and services produced in a country (the Gross Domestic Product, or GDP) and divide it by the number of people who live in that country. The scale is not linear – look carefully at the numbers on the legend.

Something else jumped out at me from this map: many of the “high-tension” parts of the world are around places where there are large disparities in GDP per capita: U.S. and Mexico, Israel and its neighbors, Europe and the Balkans, Nepal and China, etc. Is all conflict essentially economic? Or is economic disparity a prerequisite for violent conflict? Or is it just coincidence?

Solar Power Breakthrough?

Just a couple of days ago, MIT released a rather breathlessly worded press release proclaiming a new breakthrough technology their scientists had discovered – promising cheap, efficient solar power for everyone within 10 years. It's as bad a distortion of the scientists' accomplishments as you might see from a technology company self-adoring their latest product.

The MIT press release is very misleading – the new technology actually has nothing whatsoever to do with solar power, though it could certainly be used in solar power systems. And the new technology isn't ready to go to market; it's only been shown in a lab, and the vast majority of such “lab-ware” fails somewhere along the line between the lab and production. Nonetheless, there are some good reasons to be optimistic about this particular technology – from the descriptions released it seems quite simple and inherently cheap. But I wouldn't place any bets on it quite yet – many risks loom between here and your Solapower 2G5 on the shelves at Walmart…

What does their new technology actually do? Well, at the simplest level it provides an easy and efficient way to split ordinary water into hydrogen and oxygen gas (remember H2O – two hydrogen atoms and one oxygen atom in every water molecule). Once you have separate supplies of hydrogen and oxygen gas, you can recombine them into water, generating electricity in the process. With a system including the new water-splitter, some tanks, and a fuel cell (along with a few technical details I'm ignoring), you can put electricity into it now, and take electricity out of it later. That's the same thing a battery does – but this system is far more scalable (meaning that it can store much more power) and potentially far less expensive. Not to mention that it should be much more reliable.

So what does that have to do with solar power? Well, the dirty little secret of the solar power industry is that they have no reasonable way to power your house at night. Actually, it's far worse than that: even in the land of the sun where I live (Southern California), on a summer day we'd only get 8 - 10 hours of power, and only 6 - 8 in the winter. You can add batteries to your system, and more solar panels, to get 24x7 power – but only at huge costs, both initially and ongoing, as batteries only last 3 - 5 years.

Just last summer I calculated what it would cost for me to take my home completely off the electrical grid, relying only on solar power. The answer: approximately $220,000 initial investment, and about $35,000 per year in battery replacement costs. Of that cost, only $42,000 was the solar panels themselves (and their associated cleaning and pointing systems).

MIT's new technology has the potential to reduce that storage problem to something much more manageable. It seems at least possible that a storage system based on this new technology, large enough to power my home for a few days, might be built for $10,000 to $20,000 in the not-too-distant future. In that sense, this technology is indeed a potential breakthrough for practicable solar power.

It's also useful for a myriad of other purposes, including large-scale commercial power storage to smooth out power demands – something that we have no practicable way to do today. In the area where I live, daytime demand for power (because of air conditioning) can be over 50% higher than nighttime demand. Right now, with no storage capability, we have to have power plants capable of supplying the peak power demand. If you added large scale power storage to the grid, then excess power generated at night (when demand is low) can be stored for use the next day (when demand is high). The net result is that a given set of power plants can handle higher peak power loads.

There's a decent technical description at Popular Mechanics.