Sustainable Energy – with even less hot air


I’ve read David MacKay’s Sustainable Energy – without the hot air – a few years too late, sadly. There’s a lot of very good stuff in it, but there’s a few big holes, too.

In the preface, he says, Or if they do mention numbers, they select them to sound big, to make an impression, and to score points in arguments, rather than to aid thoughtful discussion.
This is a straight-talking book about the numbers. The aim is to guide the reader around the claptrap to actions that really make a difference and to policies that add up.

Fine words – but does the rest of the book live up to them? Or is there still an element of hot air?

My impression is that MacKay is not deliberately generating hot air of his own, but in some places he’s (inadvertently?) propagating other people’s hot air for them. There’s a lot of hot air on all sides of this argument, and he’s pretty good at demolishing the hot air from the green camp, while being careful not to dismiss their arguments where they’re valid. He’s not so good when it comes to dismissing the hot air from the other side, particularly from the nuclear industry.

In just one place, one might almost get the impression that he’s really a member of the nuclear lobby, masquerading as an impartial judge. At the bottom of p.171, there is this: The rate of new build was biggest in 1984, and had a value of (drum-roll please. . . ) about 30GW per year – about 30 1-GW reactors. So there!

When it comes to the analysis, there’s a bit of nitpicking to be done, but only a few issues that I think are really important to raise, relating to onshore wind and solar and nuclear power.

Onshore wind

On p.33, he limits consideration of wind farms to covering a maximum of 10% of the UK, thereby limiting the contribution of onshore wind to about 20kWh/d/p. Wind farms don’t cover the land they stand on; over 95% of the land is still available for agriculture. The visual impact affects the whole area, but what if we had to weigh that against the loss of almost the whole of our electricity supply?

Interestingly, MacKay doesn’t compare the output of windfarms with our electricity consumption, but with the energy demand of a car running an arbitrarily chosen distance on hydrocarbon fuel.

Using MacKay’s own figures, covering just 5% of the UK with windfarms – not forgetting that 95% of the land within a windfarm site is still available for other uses – could supply ALL our electricity demand. We need more if we’re going to use electricity to replace other energy sources, but a car running on electricity will not use anywhere near as much energy as one running on hydrocarbon fuel. It cannot – we don’t have any means of carrying so much electrical energy on a car.

I agree that even the 5% level of construction would meet huge opposition – but what if the only alternative were reduction of the energy supply to a tiny fraction of what we currently use?

I for one would rather that 5% of the land within a 100km radius of my home were host to wind farms dotted with wind turbines than that there were two new nuclear power stations (each with nearly three times the capacity of the UK’s largest existing nuclear power stations) within that radius, which is roughly the equivalent.

MacKay says that there are other domestic renewables which could contribute significantly, but that their combined contribution cannot anywhere near meet demand. To generate our entire electricity supply using solar panels would require 2,400km² of panels – more than the area required by the turbines, but less visually intrusive if placed on existing roofs, the roofs of new buildings, or on roofs placed over road or car parks. Designing them into the roofs of new buildings, rather than adding them as an afterthought, would bring down the combined cost very considerably – the cost of solar installations is now largely the cost of the supporting structure, which would be entirely shared with the roof structure.

To put 2,400km² into context, it’s 37.5m² per person, comparable to the area of roofs, and considerably less than the area of roads.

Using a combination of solar and wind power reduces the need for storage or back-up fossil fuel power – and of course would reduce the areas covered by wind turbines or solar installations. Other renewables can also help somewhat in reduction in demand for storage or back-up, but as MacKay says, can only contribute a relatively small amount to the total energy supply.

Nuclear power

MacKay argues that nuclear power is sustainable, safe, and can be built fast enough to meet our needs. He basically regurgitates the nuclear industry’s claims here – in many places breaking his own rule: Or if they do mention numbers, they select them to sound big, to make an impression, and to score points in arguments, rather than to aid thoughtful discussion.

Nuclear Sustainability

In order to meet the sustainability criterion, there are four options: we can greatly increase the supply of uranium, we can use uranium more efficiently, we can use thorium as well as uranium, or we can turn to fusion.

The only way to greatly increase the supply is to extract uranium from seawater. Technology to do this is at an early experimental stage, and it’s far from clear whether it can scale to the level required to meet even our present uranium consumption never mind a greatly increased demand. My particular concern is with the supply of sufficient quantities of the special fibres that do the absorption, but there are also concerns about the huge areas of the sea where the absorbers would be deployed, and finding enough places where the flow through the absorbers would be sufficient.

If we simply scaled up the Japanese technique, which accumulated uranium passively from the sea, a power of 1GW would need cages having a collecting area of 4.8 km² and containing a weight of 350,000 tons of adsorbent material – more than the weight of the steel in the reactor itself. Steel is a relatively cheap and abundant material compared with polymers in general, never mind the kind of special polymers required for ion-exchange processes of the kind involved here.

If we consider moving away from passive collection to active pumping, it’s easy to calculate that the energy required for the pumps is an order of magnitude greater than the energy available in the uranium (unless we use breeder reactors).

To use the uranium most efficiently, we would have to use fast breeder reactors, to be able to make use of the 238U as well as the 235U. Fast breeder reactors exist, but they’re much less well-established than ordinary reactors, and their safety and reliability are questionable. The system also requires reprocessing of fuel, which is also an area where there are major problems.

Even to use the 235U to the greatest extent possible requires reprocessing. You can only use a few percent of the 235U before you need to reprocess the uranium to re-enrich it and remove fission products.

Using thorium, to any extent that makes any significant difference, also requires breeder reactors.

Thorium can be completely burned up in simple reactors (in contrast to standard uranium reactors which use only about 1% of natural uranium). Thorium is used in nuclear reactors in India.

Thorium isn’t a nuclear fuel at all – it has to be converted into 233U in a breeder reactor first. That’s certainly not a ‘simple’ reactor. India’s use of thorium is still very limited and experimental.

See The Thorium Myth for more detail.

Fusion? To quote MacKay: Fusion power is speculative and experimental. I think it is reckless to assume that the fusion problem will be cracked.

I’d say much the same thing about all four nuclear sustainability options, to be frank. Fusion is certainly more speculative than the other three, but none of them are well-established processes – unlike wind turbines and solar panels.

I’d add another caveat about fusion. Again, to quote MacKay: tritium, a heavier isotope of hydrogen, isn’t found in large quantities naturally (because it has a half-life of only 12 years) but it can be manufactured from lithium.

Let’s look at how tritium is manufactured from lithium. It’s done using neutrons, so we need a supply of neutrons. How very handy that a fusion reactor produces copious neutrons! Unfortunately we also need copious neutrons to manufacture the tritium, and the fusion reaction doesn’t produce enough. At first sight, it produces just exactly enough – one neutron from each tritium in the fusion reaction, and one tritium from each neutron in the lithium – but you can’t herd neutrons like sheep, and a lot of them will go off and do their own thing instead of converting lithium into tritium. There are in fact a couple of ways of getting two tritium nuclei for one neutron, but a careful study of reaction rates reveals that this helps only a little, and cannot hope to make up for all the many ways we can lose neutrons.

The neutrons that are lost are themselves quite a major problem, damaging reactor components and inducing radioactivity.

To make up the supply of tritium we need some more neutrons from somewhere – and the only possible source of sufficient numbers of neutrons is fission reactors. Unfortunately, fission reactors produce about seven times fewer neutrons for a given power output than fusion reactors. They need over a third of them simply to sustain their own reactions, and inevitably lose quite a few because they can’t be herded.

Overall, to keep the fusion reactor supplied with tritium we’re going to generate more power in the fission reactors that supply it than we do in the fusion reactor. And those fission reactors aren’t breeder reactors, which need every available neutron to breed plutonium (or 233U if they’re starting from thorium).

See Nuclear Fusion? for more detail.

Nuclear Safety

Even if we have no guarantee against nuclear accidents in the future, I think the right way to assess nuclear is to compare it objectively with other sources of power.

This would be wonderful. Unfortunately it’s not easy, and reassuring noises from the nuclear industry are as much hot air as anyone else’s. Their numbers are pure fiction. We simply don’t know what the real figures are, because we don’t know how effective the containment of nuclear waste will be in the long term, nor do we know the real probabilities of unlikely but catastrophic events. We don’t even know the real casualty figures from events that have already occurred.

Coal power stations, for example, expose the public to nuclear radiation, because coal ash typically contains uranium.

This is an old canard, and pure hot air. The quantity of uranium in coal ash is comparable to that in ordinary garden soil. The quantity of uranium in the ash produced in a coal fired power station is greater than the quantity of uranium in the fuel for an equivalent nuclear power station – but the total radioactivity of the fission products is far greater than the radioactivity of the uranium. This doesn’t matter as long as the fission products don’t escape, but when they do, it’s quite another matter.

There are already plenty of places that are off-limits to humans. I may not trespass in your garden. Nor should you in mine. We are neither of us welcome in Balmoral. “Keep out” signs are everywhere. Downing Street, Heathrow airport, military facilities, disused mines – they’re all off limits. Is it impossible to imagine making another one-square-kilometre spot – perhaps deep underground – off limits for 1000 years?

Of course not – if we could guarantee that the material would stay put. Who knows what’s going to happen in the next thousand years? For one thing, it’s beyond reasonable doubt that the climate will change, probably with considerably higher sea levels and altered patterns of precipitation, with some currently-arid areas becoming wet and some wet areas turning to wind-blown dust.

Well – I say “of course not” – but “Keep out” signs don’t generally last a thousand years, and nor do the prohibitions they refer to. Who knows what developments might happen at Balmoral, Downing Street, Heathrow etc. in the next thousand years? Who even knows where mines only 150 years old are? We know some of them, but certainly not all – we don’t even know everything about much more recent nuclear activities. The nuclear industry itself doesn’t.

After 1000 years, the radioactivity of the high-level waste is about the same as that of uranium ore. Thus waste storage engineers need to make a plan to secure high-level waste for about 1000 years.

Uranium ore is not hazard-free – it’s far more hazardous than that coal ash.

Compare this 25ml per year per person of high-level nuclear waste with the other traditional forms of waste we currently dump: municipal waste – 517 kg per year per person; hazardous waste – 83 kg per year per person.

But that’s not comparing like with like. Even the most hazardous chemical wastes – a very small proportion of those 83 kg – are very much less hazardous, kg for kg, than radioactive fission products. And we shouldn’t be happy about some of those hazardous chemical wastes, anyway.

Perhaps the biggest accident-related issue isn’t really to do with deaths and injuries anyway – it’s to do with land use, see below.

Rate of Nuclear Construction

...the required rate (3000 new reactors over 60 years) is 50 new reactors per year.

The planned life expectancy of nuclear reactors is 40 years, so if we build them at a rate of 50 new reactors a year, we’ll reach a steady state of 2,000 operating reactors in 40 years time. That’s 2,000 GW – that is, roughly current global electricity consumption, just 13% of current global energy consumption, and it’s not achieved until about 2055.

It doesn’t look quite so bleak for the UK, if we ignore the rest of the world, and assume that the UK, with 1% of the world’s population and 2% of the world’s current electricity consumption, could probably build as many new nuclear power stations per year as France did for a short while, or maybe as many as five a year.

Land use

Land use for nuclear power is really not a big issue as long as we don’t have accidents. In the event of accidents – at uranium mines, reactors, reprocessing facilities or waste disposal sites – land ‘use’ can become a very big issue indeed.

The exclusion zone around Chernobyl, following the accident, is 2,600 km², and that around Fukushima 650 km². These areas are lost to agriculture for a very long time. Eventually some of the peripheral areas will become fit for use, but large areas will remain out of bounds for many generations.

It’s no use saying such things couldn’t happen here. The more reactors we build, the more accidents will happen. Compare these areas with the area occupied by wind farms – not forgetting that over 95% of the area of a wind farm is still available for agricultural use. No, you can’t put exact figures on it, because you can’t predict how many accidents there will be, or how serious they will be.

Global food security will become even more of an issue if too many such exclusion zones become necessary – or we’ll have to accept a deterioration in our health and expectation of life, due to radioactivity in our food, air and water – or both.

Another issue

p. 209: In the NIMBY plan, we reduce the contribution of nuclear power to 10 kWh/d/p (25GW) – a reduction by 15GW compared to plan D, but still a substantial increase over today’s levels. 25GW of nuclear power could, I think, be squeezed onto the existing nuclear sites, so as to avoid imposing on any new back yards.

One of the big problems with existing nuclear power station sites is that they’re almost all on the coast. There’s a good reason for this – availability of cooling water – and MacKay recognizes this elsewhere, when he calculates the proportion of the coast that would be occupied by nuclear power stations if we build large numbers of them. However, with sea levels rising, the coast has its own problems. The destructive power of the sea increases disproportionately as sea level rises, because there is, almost everywhere around the coast, a wide belt of relatively shallow water just off shore – in some places a wave-cut rock platform, and in others, a depositional feature. This exists because sea level has been stable at approximately its present level for a few millennia. The rapid sea level rise we expect over the next few decades means that this platform will be increasingly deeply submerged, reducing its effectiveness in protecting the coast from storms. (See Wave Cut Platforms & Coastal Erosion.) In addition, storms are themselves likely to become more powerful as a consequence of global warming.


MacKay never mentions two potentially important grid storage technologies: submarine compressed air, and thermal. See Storing Energy. Neither renewables nor nuclear energy can supply more than a small fraction of our energy demand without storage on a far larger scale than Dinorwig and the like can provide.

Is this nonetheless a useful reference work?

In his review, Tony Juniper, Former Executive Director of Friends of the Earth says, It will be a core reference on my shelf for many years to come. As I said, there’s a lot of very good stuff in here. I’d like to use it as a reference, too – it looks like a good source for values of all sorts of quantities that are hard to find elsewhere – but are they correct?

I’d assumed they were, until one jumped out at me that I could see was obviously wrong: on p.334, under the heading Calorific values of fuels, he says Methane has a density of 1.819 kg/m³.

No it doesn’t. The density of methane is 0.7 kg/m³. This depends on temperature and pressure, but it’s very close to this value at either STP or NTP (Standard or Normal temperature and pressure – 0°C or 20°C and one atmosphere). I’m sure this isn’t part of any plot to deceive, it’s merely a slip – but how many of the numbers that I don’t know already are subject to such slips? I don’t know where that 1.819 comes from – it’s close to the density of carbon dioxide or propane, but not exactly either, at either STP or NTP.

How many more errors are there? I’ve no idea.