Yesterday I went to a one day conference on energy efficiency at the Department of Physics here in Cambridge. They have a new programme on the physics of sustainability and this is their first major event.
The meeting was mainly aimed at scientists, and as a non-specialist, there was often rather more detail than I needed, but it is encouraging to see some great minds applying themselves to the practicalities of sustainability technology. I do now have a much better idea of how solar PV cells work though!
A few more general points of interest to share with you though. One aspect I haven’t seen in these kinds of talks before is to ask what the theoretical physical limits are. Normally, it’s just a presentation of some advance in what can be done, but it’s helpful to know what else might be possible. For instance, the maximum theoretical efficiency of a single-material solar PV cell turns out to be 31%, and the best ones we have now are nearing 20%, so we can expect some further improvements but not substantial ones.
A second strand in the conference looked at the efficiency of computer processing and data storage. This currently expends in the order of 2% of world energy use, and trends are upward. Here too traditional forms of computing are beginning to hit some physical limits in terms of how far energy use can be reduced. However, the human brain is a factor of a million more efficient than this, and not too far above the absolute physical limits for transferring and storing data. One of the more interesting talks, by Simon Laughlin, explained just how it was so efficient: a combination of dense wiring, only sending on information when it is needed, and mixing chemical and electrical processing so that the more efficient is generally used. Certainly strong support for the claim that it was designed carefully.
It might be possible to try and design a computer which works more like a brain in these respects but it would be very different from our current ones, and certainly rather slower (and perhaps not completely accurate). Programming it could be interesting, but this is the kind of thing we’ll need to consider doing if we want to keep exponentially increasing processing and storage power.
The first talk of all, by Malcolm Keay, was by a social scientist, and he challenged the audience that energy efficiency wouldn’t automatically lead to sustainability, mainly due to a variety of rebound effects. For instance, although lighting technology is now several hundred times more efficient than in 1800, the amount of lighting we do has increased by over ten thousand, so our per capita consumption has risen by a factor of twenty or so. Things vary enormously from field to field, but in general people don’t see limits in what they can consume, and the amount spent on a given type of consumption stays roughly the same as a percentage of earnings.
He therefore advocates trying to make energy efficiency savings more in areas away from the front-line of consumption to prevent this kind of effect, but it goes to show that behavioural change is fundamental (if difficult to obtain).
There’s hence a lot of hope for good things from the scientists, engineers and technologists, but we can’t rely on them alone to get us to a sustainable future.