Is Nuclear Power Globally Scalable? by Derek Abbott

From ScienceForSustainability
Jump to navigation Jump to search

This is a review of "Is Nuclear Power Globally Scalable?" which is a "Point of view" article by Derek Abbott, published in in the Proceedings of the IEEE in October 2011.

Derek Abbott starts by asserting that "robust debate" exists "over climate science and remaining oil reserves". This statement is at odds with the scientific consensus on climate change, and it is characteristic of the so-called "merchants of doubt" who seek to cast doubt on the strength of the consensus.

The gist of Abbot's opinion piece is conveyed by his statement:

Given the awesome power density delivered by nuclear stations, it is a valid question to ask if nuclear power can be massively scaled in order to meet our global energy needs. We shall explore the consequences of a future where nuclear power is the main energy source. Currently, the total global power consumption by mankind is about 15 terawatts (TW) - so the question we address is: Can nuclear power feasibly supply at least 15 TW?

Abbot proposes that any technology which cannot supply 100% of the planet's needs is of little use and "major investment" should be reserved for a technology which could supply 100%:

If we can show that nuclear power can viably provide massive power at this level, for millennia to come, then the investment in improving and scaling-up nuclear technology is justified. However, if we find it does not scale up, then major investment must be redirected to a different solution that is truly scalable.

This view is not supported by the findings of IPCC WG3 who find that we need multiple clean energy sources - renewables, nuclear, and CCS - for effective climate change mitigation. Abbott also gives no rationale for his criterion that nuclear should be capable of providing energy "for millennia to come". Even over the last millennium our energy sources, and most other technologies, have changed radically and in ways which nobody at the time could have foreseen.

Abbot continues:

It has been argued that the one renewable energy solution that is scalable well beyond 15 TW is solar thermal technology[5] - this is where large mirrors are used to focus sunlight to heat water thereby creating superheated steam, which can then generate electricity via a conventional steam turbine. The potential is enormous, as the amount of solar power that reaches ground level is 5000 times our present world power consumption. Therefore, the pertinent question is to ask how nuclear power compares to solar thermal power as an energy resource on a massive global scale.

The reference, [5], which Abbot gives for his assertion that concentrating solar thermal electricity generation (CSP) is a scalable alternative, is to his own paper, "Keeping the Energy Debate Clean: How Do We Supply the World's Energy Needs?", in which he finds that CSP alone, with a variety of storage and energy transfer mechanisms, can supply all the world's needs. This finding is not supported by the IPCC's assessments.

Abbot proceeds to discuss various issues which he identifies as


This review covers some of the above points in Abbot's article:

The Nuclear Site Location Problem

Abbot asserts that "One has to find locations away from dense population zones, natural disaster zones, and near to a massive body of coolant water."

In his first criterion Abbot doesn't specify how far away he thinks nuclear power stations should be from population zones, or what density of zones he thinks should be avoided.

Abbot's second criterion is that locations must be away from natural disaster zones. He doesn't give examples of such zones. Areas at risk of inundation by lava flows from active volcanoes are indeed probably best avoided. However Japan experiences some of the most severe earthquakes in the world and has 54 nuclear reactors, all of which survived, undamaged, the Great East Japan (Tōhoku) earthquake of 2011 which was the 4th most violent earthquake ever recorded anywhere. The reactors included designs dating from the 1960s and 1970s. (The only reason some of the reactors at Fukushima Daiichi were damaged was because the owners, TEPCO, had negligently failed to design for the magnitude of the tsunami encountered.) Nuclear power plants have also not only survived, but continued providing power through hurricanes such as Harvey and Florence in the USA.

The requirement for proximity to "a massive body of coolant water" is puzzling. As Abbot himself points out nuclear power plants can, and frequently do, employ air cooling, as do other thermal power plants sporting distinctive cooling towers. Abbot's preferred solar thermal generators are subject to the same laws of thermodynamics as nuclear and other thermal generators, and he proposes locating them in deserts where they will have to be air cooled, so the increase in cost Abbot assigns to air-cooled nuclear applies equally to them. As for the requirement for cooling water in emergencies, many nuclear power reactors are now designed to safely shut down and dissipate afterglow heat without external cooling.

The Land Area Problem

Abbot cites the work of Mark Z. Jacobson who claims that nuclear power plants require as much as 20.5 km2 of land area including not just the footprint area of the power station itself, but also its exclusion zone, associated enrichment plant, ore processing, and supporting infrastructure. He concedes that this is less than the area of a desert-based solar system, but claims that CSP takes up only "unused desert area, whereas nuclear stations tend to take up prime area adjacent to sources of coolant water". However deserts can be ecologically fragile and biodiverse environments which vast arrays of mirrors and generating equipment, and corresponding human activity, can disrupt or destroy.

Also, as noted above, nuclear power plants do not have to be sited adjacent to sources of cooling water.

Furthermore not all of the land area assigned to nuclear power plants has to be, or even can be, co-located: Uranium mines may be on the other side of the world from the stations they supply.

Jacobson's estimate is larger than the largest figure calculated by Vaclav Smil for 1 GW of nuclear reactor power. Smil's figures range from 0.61 to 13.6 km2.

The Embrittlement Problem

Abbot claims that neutron embrittlement of metal surfaces will require reactors to be limited to 40-60 years lifetime, leading to a requirement that one nuclear power station will need to be built and another decommissioned every day. However alternative technologies also have limited lifetimes. The biggest CSP plant to date is Noor in Morocco which has just over half the output of Abbot's notional 1 GW nuclear power stations, but at a far lower capacity factor - possibly 1/3 for ideal locations, or less. Assuming a Noor-type CSP plant has a similar lifetime to a nuclear plant we would have to be building and decommissioning 3 of these every day. Phase I of Noor has half a million mirrors.

The Entropy Problem

Abbot states that "In the same way that any electrical device or machine heats up and eventually fails, the same is inexorably true for a nuclear station." Why this is supposed to be a limitation for nuclear power stations and not other machines which heat up (and/or are subject to other stresses) and fail, such as concentrating solar power stations with millions of independently steered mirrors operating under hot and dusty desert conditions, Abbot does not make clear.

The Nuclear Waste Problem

Abbot claims that "there is still no universally agreed mode of disposal" of nuclear waste and that "nuclear waste still raises heated controversy". The first claim is untrue; there is a scientific consensus that disposal in Deep Geological Repositories is a safe and effective method of dealing with spent nuclear fuel. On the second claim it is certainly true that the anti-nuclear movement makes much of the lack of facilities for disposing of nuclear waste, whilst simultaneously opposes plans for building such facilities.

The Accident Rate Problem

Abbot extrapolates from historical accident rate data (claiming that there have been "about 11 accidents of the magnitude of a full or partial core melt") to a rate of "major accident somewhere in the world every month". He claims that we are not justified in taking into consideration that "the engineering of safety features for nuclear plants has surely improved" because "There are many unforeseen pathways to an accident and there are large rare events that can knock out redundant backup systems in parallel". However for large rare events (such as the enormous Tsunami which followed the Great East Japan Earthquake in 2011) to cause repeated accidents presupposes that the nuclear power industry does not learn anything from the first occurrence, which is clearly not true since the industry has re-assessed the safety of existing and new plants in the light of experience from the Fukushima, Chernobyl, and Three Mile Island accidents. A better model might be air travel, where numbers of deaths per passenger mile have declined dramatically over the decades.

The Proliferation Problem

Abbot cites a study on fast breeder reactors to claim that "The presence of nuclear power creates an infrastructure where materials and expertise for weapon making can proliferate". Russia is the only country currently operating fast breeder reactors, although China is building a Russian-designed one and the USA developed a mature design. Britain and France have also operated FBRs in the past. All these countries already have nuclear weapons. Many countries have nuclear power reactors but have shown no interest in developing weapons, and all states which have developed weapons did so before, or without, developing civil nuclear power, with the possible exception of India. International regulations covering the supply and use of nuclear material militate against any state employing nuclear energy from developing weapons (with the obvious exception of the nuclear "club" - the USA, Russia, Britain, France, and China which already had weapons when the non proliferation treaty was drawn up).

Abbot suggests that "all nuclear fuels and all nuclear products can be utilized in a dirty bomb, if not a nuclear bomb". The nuclear products which can be used in a dirty bomb include Americium in domestic smoke alarms and various medical isotopes. Indeed nuclear material is probably not even necessary: if a modest explosive device were simply claimed to contain radioactive material radiophobia would probably produce the effect the perpetrators desired.

The Energy of Extraction Problem

Abbot's claims regarding the energy required to extract, enrich, and process Uranium to make fuel for nuclear reactors relies on the discredited, non-peer-reviewed "StormSmith" study, whose estimate of the energy consumed by Uranium mining and milling in Namibia was higher than the energy consumption of the entire country.

Other "Problems"

Abbot's remaining claims could be similarly examined. One in particular is worth noting for its bizarre logic. In his section "Nuclear Power and Climate Change" Abbot says:

The fervor with which the number of nuclear advocates have taken up the cause of climate change appears some what opportunistic. They propose a rapid upscaled nuclear power program to avert a global warming crisis. This is as deeply suspicious as an undertaker who sponsors a keep-fit program to promote longevity.

Presumably in this analogy nuclear power is the undertaker and averting the global warming crisis is the keep-fit program to promote longevity. Does Derek Abbot not think that longevity, and averting global warming, are desirable outcomes, or does he not accept the scientific evidence that keeping fit, and nuclear energy, are found by experts to be part of the means to these ends?