Options for Tackling Climate Change

Dear Michel,

In a brief conversation the other day you expressed the commonly held view that nuclear power is unacceptable due to the very long-lived waste that it creates. I anticipate that you are also concerned about safety and the risk of contaminating the environment as a result of a release of radioactive material. I don’t have sufficiently good French to sustain a discussion on this complex issue, which is also for many people a very emotional topic leading to stormy arguments, and so I said little or nothing, but I've finally decided that I wish to express my views on the subject. 

Climate Change - Conservation is not Enough
I’m sure that we can both agree that climate change is the single most serious challenge facing the world today. The question is: what is the best way of tackling the problem of rising energy demand, and demand for electricity in particular, at the same time as limiting the emission of man-made greenhouse gases, principally carbon dioxide? 

Whilst we can make some progress within developed economies by improvements in efficiency, conservation and demand management, the net result of these measures will be grossly insufficient to counter the legitimate requirement for more energy from developing economies. In this video “The Magic Washing Machine” by Hans Rosling (now sadly deceased) he explains why we in the developed world have no right to tell people in developing economies that they can’t use more energy because of Climate Change.

The Role of Renewables
So as well as conservation measures what is the best way forward?  Clearly renewables have a significant part to play in the future of energy production, and they are getting cheaper, but they are still being subsidized and, in common with other energy production technologies, they are not without an impact on the environment.

For example land area. Electricity generation from wind and solar energy is based on diffuse energy sources and is consequently land hungry. The graphic below illustrates the areas required for different energy generating sources for a capacity of 1000MW (1GW). The units are in square miles. To generate 1GW from wind would require about 300 square miles and from solar about 60 square miles.

In contrast nuclear power plants, and those using fossil fuels, are burning energy dense substances and so require only about 1 square mile. In terms of area, a gas fired plant would be comparable with a nuclear power station and a coal fired plant would be a little bigger due to the space required for stockpiling coal and ash.
In Europe, where most non-urban land is already farmed, I question the overall viability of covering fields with solar panels. Near us, in South West France, there is an application to do just that on an area of land that we visit each Spring to hear nightingales and look at orchids.

Variability of Wind and Solar
Depending on location, wind power installations are normally expected to produce electricity 30-50% of the time.  Their output is variable over both short and long term timescales and also by geographical situation. The latter can be an advantage since a diversity of locations can result in a degree of smoothing of output power levels.

On dull overcast days solar photo voltaic power can be largely unproductive, especially in northern latitudes. So when demand peaks on cold winter evenings when there is no sunlight, if there is also no wind, both solar and wind generators will not supply power.

Solar panels produce most efficiently in hot sunny places, like sparsely inhabited deserts.  In certain locations, if geo-political constraints involving distance and international borders can be overcome, it may be an economic proposition to generate solar power in one country and transfer it to another. Long transmission lines,with their associated costs and losses (in the USA in 2007 transmission losses were estimated at 6.5%), would then be needed to deliver the power to centres of population. 

High Voltage DC lines have been used in some long distance situations. These have the advantage that the capacitance of the line and hence its charging current is no longer a length limiting consideration as it is with alternating current systems. One of the disadvantages is that DC voltages cannot be easily reduced by transformers without first utilising inverters, with their associated costs and efficiencies, to recreate alternating current.

So both solar and wind power are, by their nature, highly variable and cannot provide continuous reliable output. In almost all electricity supply grids there is no significant storage capacity so, to meet the demand from consumers, electricity utility operators therefore have to schedule the right mix of flexible and inflexible power generation capacity from different types of generators. As part of the day to day management of the grid, this requires them to predict the weather conditions and hence the output from renewables. If there is an unpredictable drop in output from renewables, then flexible generation capacity must be quickly brought on line. Electricity from flexible sources commands a much higher price in wholesale electricity markets than that from inflexible base load plants, so getting this right is therefore important for both utility companies and consumers. This paper from the US National Renewable Energy Laboratory explains the problem in detail. 

If, as in Germany, there is a law forcing utilities to accept energy from renewables as a priority then, as the proportion of total generating capacity that wind and solar power provide increases, there will be an increasing need to maintain backup power supplies with their associated costs and energy consumption. If not the system would become unstable and power cuts would result.

In effect without grid scale energy storage, to satisfy the demand when renewables can't, there will be a limit to the proportion of electricity production that can be provided by these sources. So what options are available for large scale energy storage?

Energy Storage
There are many ways of storing and recovering energy. For electricity the one we are most familiar with is the use of batteries.

The following graphic, which plots energy density in MJoules/litre against MJoules/kg, puts into context the energy density of Lithium Ion and Zinc-Air batteries relative to other substances. The significance of energy density is that gasoline, for example, effectively stores about 50 times more energy per kg than lithium ion batteries and so is much more portable. 

Battery storage technology is developing slowly; but current batteries are still expensive, have low energy densities and a limited life. In my view both energy density and longevity would need to improve by at least one or possibly two orders of magnitude before the use of batteries to smooth out variable production from renewables would become economic on a grid scale.

Other energy storage systems are being proposed and investigated, and some may be promising in the long term, but most are not yet proven to be economically viable or available for widespread deployment. 

Generation of Hydrogen
An interesting example is the proposal to use surplus energy generated by wind and solar, that is not required by the grid, to create hydrogen which can be stored until it is used to generate electricity during periods of low output from renewables. The difficulty with this concept is its overall efficiency. The efficiency of electrolysis is 60-70% and the ongoing conversion back into electricity is 40-60% (depending on the method used). Thus the overall efficiency of the round trip is 24-56%, which is much lower than batteries, and there is also the energy cost of compressing or liquefying the gas but, providing there are periods of overcapacity of installed renewable generation, the electricity used is effectively free. There are many possible means of implementing such a scheme and there are also other uses for hydrogen such as in powering vehicles. Work is ongoing to establish its overall economic viability.

Liquid Air Energy Storage
A further method which is getting serious consideration and development is the use of liquid air as an energy storage medium. Similarly to the production of hydrogen, electricity production from wind and solar is used to liquefy air at atmospheric pressure and when required it can be evaporated under pressure to drive generators. A 5MW/15MWh demonstration plant, constructed by Highview Power, has been in operation since 2018 at Pilsworth, Bury, UK.  Construction has started on an £85m project at Trafford Energy Park, Manchester, UK, which is due for completion in 2022 and will store 250 MWh of energy. The round trip efficiency of such systems is low at around 25% but this can be enhanced with the use of a cold store to 50%. If the facility is located near a source of waste heat like an industrial process or thermal power generation system, the efficiency can be further enhanced, and Highview Power claims 70%, without the need to augment the capital cost with heat recovery systems.

Electric Vehicles
Another example is the idea that, when required, electricity could be fed back into the grid from the batteries in electric vehicles: and therefore at some future date a large battery storage capacity would be available to smooth out shortfalls between generating capacity and demand. Although this concept is now being marketed by Nissan and Ovo, it's unclear how it would be managed on a large scale, but it would probably need an infrastructure that had connection points at the majority of parking places; as well as some form of smart metering that credited the vehicle owner if stored electricity, that had already been paid for, was drawn out of the vehicle's batteries. These connection points, or the vehicles themselves, would also need to be equipped with inverters capable of converting DC battery current into AC power synchronised with the grid.

The transition to electric vehicles, which is happening at a surprisingly fast rate, will make a contribution to the reduction in greenhouse gases from burning hydrocarbon fuels. Even if the electricity to charge vehicle batteries comes from gas-fired power stations there will be some efficiency savings and hence less carbon dioxide and nitrogen oxides will be discharged to the atmosphere.

Pumped Hydro
One system that has been proved to be viable for energy storage is pumped hydroelectricity, which has had plants in operation for decades.  They require two large lakes, one several hundreds of metres above the other. Water is pumped up when electricity costs are low and released through turbines to generate electricity when required. Due to pump/turbine efficiencies and the two way conversion, the electricity recovered is only about 70-80% of the electricity input. The geographical opportunities for such installations are few, and so far they have only been used for providing flexible power for relatively short durations at peak times when the feed-in tariff for electricity is high. 
There are ten such schemes under construction in Europe totalling 1,339MW of capacity. To put this into context, this adds less than 0.1% of the total energy consumption of the 28 EU countries in 2015 as storage capacity. 
(This was 12,609.25TWh, equivalent to a continuous consumption of 1.44TW. A Terawatt TW is 1 million Megawatts).

Other Renewables

Except in particular locations bio-fuels are heavy consumers of agricultural land and are unlikely to provide more than marginal amounts of electricity on a global scale but they have a role to play. In Brazil, for example, they are successfully replacing fossil fuels for vehicles with blends of gasoline and ethanol derived from sugar cane. Biodiesel, in various blends, is established in North America and Europe. Biofuels are being encouraged by the EU, but there are concerns that by displacing food production from agricultural land and clearing forests, which are net absorbers of CO2, the production of biofuels could actually release more carbon dioxide than they save.

Tidal power could also contribute more to renewable energy production, and some sites have been in operation since the 1960's. Tidal power generation is very predictable and could remove much of the need for back up fossil fuel  generation for other less predictable renewables, but tidal barrages and lagoons have an impact on ecosystems and the energy generated is on a lunar cycle, which does not coincide with diurnal demand. There are few suitable sites in North America, China or India and its contribution is likely to be marginal.

Submerged tidal turbines placed in locations where currents are created by tidal flows are another possible source of tidal power. Studies are ongoing but environmental impacts are likely to be much less than tidal barrages.

Wave power is another source of renewable energy which could be further developed and wave farms have been installed in a number of coastal locations. Currently the largest operational wave farm has a capacity of 2.4MW and most are in the fraction of a MW range. The graphic below shows the potential in terms of kW/m for the world's oceans.

Although there are opportunities for improving output by replacing old machinery, and also by installing small scale plants, hydro-electricity is largely fully developed in Europe, the USA and some parts of Asia.

My conclusion is that, in the absence of economic grid scale energy storage technologies, over any specific 24 hour period, renewables can only satisfy a proportion of the total energy demand on any supply grid.

Backup Power
Backup power production capacity is therefore necessary to supply the demand for energy when wind and solar can’t and, because it takes time to bring generating capacity on line, some backup systems have to be kept running even when the demand is being satisfied by renewables.  

Under the German EnergieWende, in which it is intended to move away from fossil fuels and nuclear power to a low-carbon energy economy, they have been running lignite and hard coal fired power plants (red and dark green) to replace the energy generated by nuclear plants (lime green) which have been taken out of service, and to provide backup to wind and solar (blue). During the last few years the consumption of gas (yellow) has also been reduced: so the increase in energy from renewable sources has been offset by the reduction in the least carbon dioxide producing sources i.e. nuclear and gas.

Coal, and particularly lignite, is the most polluting of all the fuel options since, due to lower efficiencies, legacy plants produce more carbon dioxide per MWh than other fossil fuels and release, into the environment pollutants: including particulates; sulphur and nitrogen oxides; and ash. The latter contains uranium and is 100 times more radioactive than nuclear waste. It also contains heavy metals which will never decay and become less polluting. This ash is typically dumped or stockpiled with minimal control and has caused serious ash-slides necessitating dangerous clean up work. Air pollution from burning coal is even more serious because it travels further. You may remember the acid rain arising from UK coal burning power stations that acidified Scandinavian lakes in the 1980's. The horrific air pollution in China also dramatically illustrates the results of burning coal and is responsible for many premature deaths. 

There is a debate in Germany about the phasing out of coal and the government have published a draft bill which is expected to be passed by the end of this year (2019). This will commit Germany to closing all of its coal fired power stations by 2038 but it will clearly be very difficult to phase out both nuclear and coal fired generation, which represent more than 50% of current capacity, and replace it with intermittent renewables.

Another effect of relying more and more on renewables has occurred in Germany on some sunny and windy days. Because German law forces the Grid to accept renewable energy in preference to that from fossil fuels, and electricity production from fossil fuels cannot easily be ramped down, on occasions the price of electricity has become negative in response to an over-supply, meaning that commercial consumers are being paid to burn more electricity!

In effect the German EnergieWende amounts to an experiment on a national scale.

More recently gas and oil have become cheaper internationally, as a result of fracking and the exploitation of shale oil in the USA, but gas and oil, while more efficient and less polluting than coal, still produce carbon dioxide.

Environmentalists for Nuclear and Renewables

So what would it really take to limit or even reverse climate change if today's renewable technologies by themselves aren't enough? Ross Koningstein and David Fork are engineers at Google, who worked together on the bold renewable energy initiative known as "RE>C".   Their research led them to think that a new technology is required, which will disrupt the existing status quo, but they don't specify which, or deal with the need to act now and not wait in the hope that some new technology will emerge.

Taking all this into account I've come to the conclusion that, alongside renewables, the only energy production technology that is available to be deployed on a global scale over the next twenty to thirty years, which will not contribute to climate change, is nuclear power.

In this video James Hansen explains why he has reached the same conclusion.

Other environmentalists also expand on their reasons for changing their view of nuclear power in this video.



So, together with a growing community of engineers, scientists and environmental campaigners, I am proposing nuclear power alongside renewables as the future for reducing carbon emissions.

But Not Pressurized Water Reactors!
I am, however, not happy that solid fuel, second generation pressurized light water reactors (PWRs), the most common type of reactor currently in service, are the best option for the future other than as a short term stop-gap solution. They were originally intended to be used in submarines for which, being surrounded by emergency coolant they are ideally suited, and represent a design that dates from the late 1960's. There has been little improvement since then. 

I don’t like the fact that, in order to generate high temperature steam at around 300 deg C, PWRs run at a pressure of over 300 bars and therefore carry an inherent risk of explosion. To guard against a reactor vessel failure they need thick heavily reinforced concrete containment vessels. They burn enriched uranium fuel which generates waste with very long lived radioactive transuranics and actinides. Also they can only burn a small proportion of this fuel before the fuel rods deteriorate and must be replaced. 

Another major concern with the designs currently in service is that, when a reactor shuts down unexpectedly, and there is no power available from the grid, they need backup power supplies in order to run pumps and therefore maintain the cooling needed to remove the heat from the radioactive decay of fission products. This was what failed at Fukushima Daiichi, where the backup generators were in a location which was only designed to resist a 3 metre tsunami. The generation III+ designs currently awaiting approval or under construction do, however, incorporate passive cooling meaning that, once shut down, they don't need power to dissipate decay heat.

Other types of reactor, operating in the fast neutron spectrum, like the liquid metal fast breeder LMFBR, which uses molten sodium as the coolant medium have other disadvantages. Sodium metal in liquid form is a good coolant, which operates at higher temperatures and lower pressures than water and is a very good heat transfer material but it reacts with air, and violently reacts with water. So the possibility of a leak between the liquid sodium and the steam generating sides of the primary heat exchangers is a serious safety concern. These types of reactors can, depending on their size and design, also have an overall positive  temperature coefficient of reactivity, meaning that as the temperature rises nuclear reactions increase and create more heat. Such matters can be overcome by good design. The Experimental Breeder Reactor at Argonne (EBR II) included passive safety features and was successfully tested in 1985 under conditions simulating power failure. 

Nuclear Safety
In spite of all these objections, and the age of most of the plants currently in service, on a global scale, there have been very few incidents involving nuclear reactors which have had major consequences, and two of those were at plants with fundamental design flaws. 

At Chernobyl, where 64 people died of acute radiation sickness after emergency work, the reactors were never provided with containment vessels and they had other design flaws which would never have been permitted in Western countries.

At Fukushima Daiichi the backup power plant was flooded by the very large tsunami that we tend to forget resulted in over 18,000 people dead or missing. The several dozens of direct casualties at Fukushima are attributed to the effects of the evacuation of the surrounding area on the sick and elderly, but some workers have received radiation doses which exceed lifetime safety limits.

At Three Mile Island operators did not react appropriately when a pilot valve failed and, by sticking open, allowed coolant to escape. This fault had occurred on 11 previous occasions. 

Estimates of future deaths related to radiation exposure from these accidental releases of radioactive material vary widely depending on their source, the methods used and the assumptions made. For example for Chernobyl in 2006 the WHO estimate of radiation related premature deaths was 4,000 whereas the Greenpeace estimate was 200,000! In 2008 another WHO report urged caution in the development and use of projections (paragraph 110). 

These estimates, are controversial and the methods for calculating them are disputed by respected scientists. A full discussion would occupy too much space in this context, as would an examination of the number of fatalities in other industries associated with energy production, like coal mining. All energy production carries risks and is responsible for premature deaths.

The specific point I wish to make is that even safer nuclear options are available than those which are currently in operation and these will be discussed below. 

Molten Salt Reactors for Safer Nuclear Electricity

In view of these potential safety issues a shift away from PWR’s and fast breeders designed in the 60's and 70's towards inherently safe plants would be very welcome. Ideally these should: 

• operate at atmospheric pressure and so couldn’t explode; 
• be unable to overheat or meltdown; 
• be designed to be walk-away-safe in the event of power failure; 
• have a negative temperature coefficient of reactivity; 
• allow load following and rapid output changes;
• produce much less waste with a much shorter radio-active half-life.  

Of the various proposed Generation IV nuclear reactor designs there is one which fulfills all of these criteria, the liquid fuelled thorium reactor (LFTR).

On 14^th July 2011, thanks to Ken Pottinger of French News Online (now also sadly deceased) I became aware of an alternative to current PWR technologies, which is based on liquid fuelled thorium reactors LFTRs and the Molten Salt Reactor Experiment. At the time I had no idea that nuclear reactors could have many design variants and even work in the liquid phase.  I spent several days researching the topic and wrote this piece on my blog summarizing my findings. A prototype molten salt reactor, operating at atmospheric pressure, ran at Oak Ridge National Nuclear Laboratory (ORNL) in the 60’s and 70's. The film below, made at the time, shows how this pilot scale plant was designed, built and operated.

The Molten Salt Reactor Experiment ran for more than 13,000 hours at full power, without significant materials problems, and successfully demonstrated the viability of the concept. To turn it into a commercial product there is more development work needed on the optimization of the waste processing stages, and also on the materials necessary to resist the high temperature and intense radiation environment over the long term. At the time that the project was shut down, ORNL had already started to work on these areas. We are fortunate that the work on the MSRE at Oak Ridge was very fully documented and these documents are in the public domain.

Kirk Sorensen has been actively promoting this technology and he explains in this video how he came to rediscover molten salt reactors and Thorium as an alternative fuel.

What About Nuclear Waste?
But your specific concern was with the management of nuclear waste from PWR’s running on the Uranium 235/238 fuel cycle. 

The composition of nuclear waste depends on the fuel used in the reactor and the degree to which the fuel is burned. Solid fuel reactors can only burn about 1% of the fuel contained in their fuel rods which have to be replaced every 18 months. During this operation the reactor is taken out of service.

Liquid fuelled reactors can burn a much larger percentage of their fuel than solid fuelled reactors because it is so much easier to remove gaseous fission products like Xenon 137, which poison nuclear reactions by absorbing neutrons. Also by incorporating on-line processing to remove waste from a side stream of the main molten salt fuel, there is no need to shut down the reactor to change fuel rods. Therefore, for the same amount of energy generated, molten salt reactors can produce 35 times less waste. 

LFTR versions running on Thorium also produce waste with a much shorter half-life of 300 years as opposed to tens of thousands of years. This very clear video, again from Kirk Sorensen with others, explains how fission products are created and what they can be used for.

In this rather more detailed video Kirk Sorensen projects forward the results of radioactive decay over time on waste from the Uranium 235/238 fuel cycle. He expands on the idea of recycling nuclear waste and asks, is it really all waste? 

But as he says you can also dispose of waste from PWR’s using liquid fuelled waste burning reactors. Leslie Dewan explains how in this video.

Energy Cheaper Than Coal
But without internationally agreed and binding carbon taxes, which would make fossil fuels more expensive, safer nuclear power just won't happen unless it is cheaper than other options. 
Robert Hargraves develops the argument that in order to replace fossil fuel burning power plants you must be able to generate electricity by low carbon methods at an overall cost less than that of fossil fuels. 
In this detailed talk he examines the costs of generating electricity from different sources including wind and solar. He also proposes using liquid fuelled reactors and points out that the higher operating temperatures of such reactors offer higher power generation efficiencies.

If you’ve got this far you are probably suffering from information overload but congratulations on your persistence! I hope that you are now beginning to question the dogmatic opinions of anti-nuclear campaigners because there are plenty of numbers and facts which support the statements above.

Why Hasn't it Been Developed Before?
When I first researched liquid fuelled reactors burning thorium I found it hard not to think that there was something that was being hidden from me. Such as some reasons that explained why such obviously better technology hadn’t been developed! Finally I was convinced that there isn’t anything of the sort and the reasons why it has languished for 60 years are almost entirely political in origin. This google tech talk, again by Kirk Sorensen, explains the background. 

Where Will Safer Nuclear Power Happen First?
This brings me finally to the current political environment. 

With an American President in place who doesn’t believe in man-made climate change and the need to reduce carbon emissions, and in spite of his recent public statements supporting the US nuclear industry, I feel that there is little or zero chance of any US government funding for this technology, at least for the next few years while he remains in office. It’s a complex subject difficult to explain to non-specialists, or the general public, and can’t be fitted into a few tweets for people with short attention spans or other priorities. Contrast this with the ease of invoking fear among the public of accidental releases of radioactive materials.

There are numerous privately funded companies which have announced programmes to develop liquid fuelled reactors. It remains to be seen whether this multi-pronged effort can surmount the burdensome costs and difficulties that will arise when they submit their designs for approval by regulatory agencies which are entirely unfamiliar with this technology. In the case of the USA, and at the risk of being called negative, it seems to me unlikely that these companies can persuade regulatory agencies to reduce the estimated one to two billion dollar cost and ten year timescale that the US Government Audit Office currently estimates it would take to certify and license a fundamentally new design. Faced with this private investors just won't bother or, if they are really keen, they will migrate to jurisdictions which are more welcoming.

There are research programmes in several countries, including France, and the International Atomic Energy Agency technical meeting on the status of molten salt reactors in 2016 represents a good summary of recent progress. It's also clear from these proceedings that the only government which is investing serious effort and resources in molten salt reactors is China. They will be the first to recreate the Molten Salt Reactor Experiment and develop it further. They will then patent their designs and sell them internationally. I wish them every success!

A bientôt

John