The third of this series of posts assesses whether it is feasible for the UK to obtain the electricity required to deliver its net zero plans.

The government’s long-term plan for energy, largely set out in its White Paper Powering Our Net Zero Future (2020), has several main elements.  Firstly, demand for energy will be restrained by measures to improve the energy-efficiency of buildings and industrial processes (1).  Secondly, many uses of energy which currently rely heavily on fossil fuels, notably domestic heating, road transport and industry, will in future be powered largely by electricity, which will become by far the most important form in which energy is consumed.  There will also be contributions from hydrogen and biofuels, and from fossil fuels with carbon capture and storage (CCS) (2).  Provided the electricity is generated from low-carbon sources, and the hydrogen is produced using low-carbon electricity, this implies a substantial reduction in carbon emissions as well as in other air pollutants.  Thirdly, electricity will be generated mainly from four zero- or low-carbon sources: wind, solar, fossil fuels with CCS, and nuclear (3).  Generation from fossil fuels without CCS will be phased out.  Fourthly, there will be new approaches to the problem of keeping electricity supply and demand in balance at all times.   In the past, this has been achieved almost entirely by adjusting the amount of generation from fossil fuels, but increasing reliance on intermittent generation from wind and solar will require new sources of flexibility.  In future there will be much greater storage of electricity by batteries and in other ways, and much more emphasis on encouraging users to manage the timing of their demand for electricity (4). Finally, suitable financial arrangements and incentives will be put in place in support of these plans.  In particular, the government has recognised a need for fundamental reform of electricity markets to ensure adequate flexibility at reasonable cost, although in this case it appears to have no specific plan for reform, having issued a consultation document which sets out options without coming to a clear conclusion (5).

In response to events of 2022, the government has made some marginal adjustments to the above plan.  Its Policy Paper British Energy Security Strategy (2022) envisages that the UK will “reverse decades of underinvestment” in nuclear power so that by 2050 nuclear will contribute “up to 25%” of projected electricity demand (6).  The Paper is quite concise and does not seem to specify which other sources of electricity would make a reduced contribution.  It also contains various plans to improve energy security in the short and medium term.  Deployment of offshore wind is to be accelerated via a package of measures including a relaxation of planning and environmental controls, and development of low-carbon hydrogen production will be supported (7).  These changes, though driven by energy security considerations, will contribute to reducing carbon emissions.  However, the Paper also envisages accelerated exploitation of the remaining reserves of North Sea oil and gas, and appears to re-open the possibility of exploiting onshore shale gas (fracking) (8).  The implication is that UK production of fossil fuels will be larger than it would otherwise have been, although it is possible that the extra production will simply substitute for imports and not increase territorial emissions.  

This post will focus on the technical feasibility of generating the huge amount of electricity required by the above plans.  Energy policy will be considered in a later post.  I make no apology for using a lot of numbers; it is what the topic needs.  But the analysis uses no high-powered maths, just basic arithmetic applied to figures obtained from reputable sources, and care with units.  Note especially the sequence tera-, giga-, mega-, kilo-, each a thousand times the next.  Just as the familiar kilowatt-hours (kWh) are sensible units for the electricity used by a single household, so terawatt-hours (TWh) are appropriate units for the electricity used by the whole of the UK. 

According to Powering our Net Zero Future, annual UK electricity demand in 2050 will be of the order of 680 TWh (9), about twice as much as it was in 2020 due to electrification of heating, transport and industry together with economic growth.  Its scenarios envisage that about 70% of this will be from renewables, 20% from nuclear and 10% from gas with CCS.  The 2022 paper suggests a slight adjustment to 65-70% from renewables, 20-25% from nuclear and 10% from gas with CCS. While small contributions from other sources are also envisaged, these can be ignored in the context of the broad numbers I will consider. 

It may be asked how hydrogen features in all this.  There is, quite rightly, much interest in the potential of hydrogen as a clean, storable fuel.  However, hydrogen is not a primary energy source: it has to be produced first and that requires energy from some other source.  It can properly be described as an energy carrier.

To assess the feasibility of generating 65-70% of 680 TWh, or 440-480 TWh, from renewables, I will make use of some of the analysis in David Mackay’s Sustainable Energy Without the Hot Air.  Although this book was published in 2009, and in some respects is now out of date, its particular merit is that it offered a carefully reasoned attempt – based on physical principles and practical constraints – to calculate permanent limits to the energy the UK could obtain from various sources.

Although I shall use a number of MacKay’s figures, I shall reverse his reasoning in this respect: instead of asking how much electricity we could generate given the extent of availability of key inputs, I shall ask how much of the key inputs we need to generate the electricity required by the government’s net zero plan.  Noting that the UK currently obtains much more electricity from onshore and offshore wind than from solar, but also that there is considerable opposition to the spread of onshore wind turbines, I shall take the necessary 480 TWh to be delivered by the following mix: solar 100 TWh; onshore wind 100 TWh; offshore wind 280 TWh.

Considering solar first, MacKay stated that the average power of sunshine falling on a south-facing roof in the UK is 110 W (watts) per square metre (10 p 38).  Without going into details, I am satisfied that this figure makes due allowance for cloudy weather, day and night, and seasonal variation. It isn’t clear, however, exactly how MacKay defines a south-facing roof (does he include roofs facing east or west which still receive some sun?).  He assumes that solar panels can convert sunlight to electrical energy with an efficiency of 20% (11 p 39).  Today, most panels are between 15% and 20% efficient, but efficiencies of up to about 23% are available (12).  Allowing a little for further efficiency improvements, I will assume average efficiency in 2050 of 25%.  On this basis, the average electricity generated by one square metre of solar panel in one hour is 110 x 25% = 27.5 Wh (watt-hour).  Multiplying by 24 and then by 365, the electricity generated over a whole year would be 241,000 Wh or 241 kWh. 

To find how many square metres of panels would be needed to generate 100 TWh, we need to divide 100 TWh by 241 kWh, noting that 1 TWh equals 1 billion kWh.  So the calculation is 100 billion divided by 241 which is 415 million square metres. The number of houses (excluding flats) in the UK is about 23 million (13).  If we only consider houses, therefore, the average requirement of solar panels per house is therefore 415 million divided by 23 million or about 18 square metres.  That may be feasible, but statistics on average roof areas, let alone those which are south-facing, do not seem to be available, and in any case many roofs cannot be completely covered by solar panels due to the standard sizes of panels and to obstacles such as chimneys and loft windows.  My judgment is that the average south-facing roof area per house that could be covered in solar panels may be somewhat less than 18 square metres.  On the other hand solar panels can also be sited on the roofs of non-residential buildings or, though they compete with other land uses such as growing food, as arrays in countryside.  On balance, 100 TWh from solar appears technically feasible.

Turning to wind, MacKay used physical principles to show that the electricity generated by an array of wind turbines depends mainly on the wind speed and the area of the array.  The size of individual turbines makes relatively little difference because larger turbines must be spaced further apart to work well (14).  For onshore wind speed, MacKay based his calculations on an average speed of 6 m/s but later cast doubt on this figure and suggested that 4 m/s might be more realistic (15).  That makes a big difference since the power a turbine generates at any time depends on the cube of the wind speed.  Cutting through various complications, I am therefore going to reduce by half MacKay’s estimate that onshore wind can generate 2 W per square metre of land (16).  Therefore my estimate of the average electricity that can be generated by wind in one hour is 1 Wh per square metre of land.

Scaling up, the corresponding figure for one year is 1 x 24 x 365 = 8,760 Wh or about 9 kWh.  To generate 100 TWh annually from onshore wind, therefore, the land area required is 100 billion divided by 9 which is about 11 billion square metres or 11,000 square kilometres.  Since the land area of the UK is 244,000 square kilometres, that’s about 5%, perfectly feasible, though (both directly and via associated infrastructure such as access roads) occupying land that could be put to other uses, adversely affecting scenery and wildlife, and potentially creating a health hazard via low-frequency noise if turbines are sited near to homes.

For offshore wind, MacKay assumed a power of 3 W per square metre of sea (17), the wind generally being stronger at sea than on land.  As for onshore wind, however, I will reduce this by half, to 1.5 W per square metre.  The corresponding figure for one year is 1.5 x 24 x 365 = 13,140 Wh or about 13 kWh.  To generate 280 TWh annually from offshore wind, the area that must be covered in arrays of wind turbines is 280 billion divided by 13 which is about 22 billion square metres or 22,000 square kilometres.  To allow for shipping corridors, MacKay applies a factor of 3 (18) which would increase the required area to 66,000 square kilometres.  That is quite feasible since the area of UK territorial waters to a depth of 50 metres is about 120,000 square kilometres (19).  What’s more, the scope for offshore wind has been considerably extended by the development of floating offshore wind, a technology not considered by MacKay.  In a recent auction by Crown Estate Scotland, over half of the capacity of the successful bids were for floating offshore wind (20), with many of the sites being in waters deeper than 50 m. 

However, it also needs to be considered whether we can obtain enough steel for the huge number of turbines needed to generate 380 TWh annually.  Dividing by 365 and then by 24, that’s equivalent to average power of 0.043 TW or 43,000 MW.  Given that the useful life of a wind turbine is often taken to be 20 years, we can infer that turbines delivering an average power of about 43,000 / 20 = 2,150 MW must be built each year.  It has been estimated that each MW of wind power requires about 150 tons of steel (21), but that is presumably maximum power, before allowing for wind intermittency which reduces average power by a factor of about 3.  So the annual steel required would be 2,150 x 150 x 3 tons which is about 1 million tons.   Given the many uses of steel, that’s quite a big proportion of the UK’s total steel production, which in 2019 was about 7 million tons (22).  Steel might also be imported, but other countries may also need large quantities for their wind turbines.  We may conclude that it is feasible to obtain enough steel to obtain 380 TWh from wind, but that demand for steel for use in wind turbines will be a very significant economic factor affecting both the cost of wind turbines and the availability of steel for its many other uses.

For fossil fuels with CCS, our target is to generate 10% of 680 TWh or about 70 TWh annually.  Key inputs are the fuels themselves and sites to store the carbon dioxide.  Since both fossil fuels and suitable storage sites are non-renewable resources, the feasibility of generating electricity at that rate depends on our time horizon.  MacKay, focusing on coal reserves, assumed a time horizon of 1,000 years (23), and as a consequence inferred that the amount of electricity that could be generated annually was rather small.  However, in respect of a period as long as that it seems quite reasonable to point out that we cannot know what further reserves might be discovered, what new energy technologies might be developed or how energy-efficient the economy might become.  Given also that the UK’s main fossil fuel now used to generate electricity is gas, let’s consider whether it is feasible for the UK, from gas with CCS, to generate 70 TWh annually for 100 years, or 7,000 TWh.  1 cubic metre of natural gas typically contains about 10 kWh energy (24).  A gas power station can be over 50% efficient (25), but CCS itself requires energy resulting in an ‘energy penalty’ of about 15% (26), so the net electricity generated per cubic metre of natural gas is about 10 x 50% x 85% or 4.25 kWh.  To obtain 7,000 TWh would therefore require 7,000 billion divided by 4.25 which is about 1,600 billion or 1.6 trillion cubic metres.  That’s less than 1% of proven world reserves of natural gas which are about 190 trillion cubic metres (27). For comparison, it’s somewhat less than the UK’s share of world population, which is about 1%.  What’s more,  proven reserves have tended to rise over time, and that is before any consideration of coal, of which there are also very large reserves.  I conclude that availability of fossil fuels is not a technical constraint on generation of 70 TWh annually except perhaps in the very long term.

What about storage sites?  We first need to consider how much carbon dioxide needs to be stored.  A cubic metre of natural gas weighs 0.76 kg, so the gas needed to generate 70 TWh annually for 100 years would weigh 0.76 x 1.6 trillion kg or about 1.2 gigatonnes (Gt). Natural gas is mainly methane, and simple chemistry shows that combustion of 1 tonne of methane produces 2.75 tonnes of carbon dioxide (28).  So the weight of carbon dioxide to be stored is about 2.75 x 1.2 or 3.3 Gt.  According to the British Geological Survey, there is a “geological storage potential” of over 70 Gt within over 500 sites under the UK seabed, including saline aquifers and oil and gas fields (29).  The word ‘potential’ is important here.  Firstly, the weight of carbon dioxide that can be contained in a given volume depends on its state: this figure appears to assume conversion into a “high pressure, liquid-like form known as ‘supercritical CO₂’” (30).  While that is feasible, it requires energy, involves costs, and raises the question of how leakage of the high-pressure substance is to be prevented.  Secondly, much more work is needed to verify the suitability of sites for carbon dioxide storage. A project commissioned in 2015 by the Energy Technologies Institute focused on just 5 sites selected as among the most promising but also technically and geographically diverse.  It concluded that the sites were suitable, albeit subject to some “specific development risks” (31).  However, comparing the storage requirement of 3.3 Gt with the storage potential of over 70 Gt, it only needs about 5% of the latter to prove suitable.  I conclude that storage capacity is unlikely to be a technical constraint on generation of 70 TWh annually.

It remains to consider the feasibility of generating 25% of 680 TWh or 170 TWh annually from nuclear.  In 2021 the UK obtained 46 TWh from nuclear, but in the past it has obtained much more, as much as 99 TWh in 1998 (32).  170 TWh is less than twice that.  One may also point to the example of France, which has obtained far more – 379 TWh in 2019 – from nuclear (33).  It may reasonably be inferred that 170 TWh is feasible provided the necessary inputs are available in sufficient quantity.

One essential input is an adequate number of suitable sites on which to locate nuclear power stations.  Whether a site is suitable depends on the type of reactor (34).  Some types, especially those which use water as a coolant, require larger quantities of water nearby than others. But that should not be a problem given the UK’s long coastline.  For safety reasons, sites should not be too close to residential areas.  The simplest approach, and the one which the government appears to be following, is to locate new reactors at existing sites (35).  In some cases more than one reactor can be located at the same site, such as the two reactors currently under construction at Hinkley Point (36).

Another vital input is uranium for use as nuclear fuel.  Uranium is present in ores and rocks at various concentrations, and in seawater at a very low concentration.  Broadly, the lower the concentration, the higher the cost of extraction.  According to the World Nuclear Association, global reserves of about 6 million tonnes of uranium are available at a cost of no more than £112 per Kg (37).  That may seem expensive, but one kilogram can generate about 45,000 kWh of electricity (38), so the cost contribution of the uranium per kWh is only about £112 / 45,000 or 0.25p.  To generate 170 TWh for 100 years would require 170 billion x 100 / 45,000 Kg, which is about 400 million Kg or 400,000 tonnes, about 7% of the above reserves.  Given the likelihood that many countries around the world will be expanding their reliance on nuclear power as a zero-carbon source of electricity, it seems rather naive to assume that the UK, with about 1% of the world’s population, either could or should secure as much as 7% of the world’s low-cost uranium reserves.  Admittedly, larger reserves are available at higher cost, and additional reserves may be discovered.  Nevertheless, it cannot be asserted with confidence that the UK will be able to obtain sufficient uranium to generate 170 TWh for 100 years.  Nuclear power on that scale is certainly feasible by 2050 and for some years thereafter, but it is possible that a scarcity of uranium will limit its longer term role in providing zero-carbon electricity. 

Because of this doubt as to the continuing availability of uranium, and also because of safety risks regarding nuclear power, it is of considerable interest whether it would be feasible to obtain 680 TWh annually without nuclear, that is, from a combination of solar, wind and fossil fuels with CCS.  Withough going into detail, a reasonable inference from the above analysis is that it would be feasible subject to the following: for solar, rather more reliance on solar arrays occupying large areas of land; for wind, even greater demand for steel for wind turbines; and for fossil fuels with CCS, greater likelihood that coal as well as gas would be required. 

However, the feasibility of managing without nuclear – not just generating sufficient electricity without nuclear but supplying it when and where it is needed regardless of whether the sun is shining or the wind is blowing – is subject to an even more important qualification.  If 70% of the electricity is from renewables, the problem of intermittency looms large; if, as might be necessary without nuclear, 90% is from renewables, the problem looms much larger still.  The issue of intermittency will be considered in a subsequent post, but for the time being we must conclude that providing the electricity we will require to achieve net zero by 2050 without nuclear is on and perhaps just beyond the edge of feasibility.

Provided on the other hand we include nuclear in our portfolio, then our analysis suggests that it will be quite feasible for the UK, by 2050 and for many years thereafter, to generate the electricity it will need from low-carbon sources.

Notes and References

1. HM Government (2020)  Powering our Net Zero Future  https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/945899/201216_BEIS_EWP_Command_Paper_Accessible.pdf  pp 101 & 122
2. HM Government, as (1) above, pp 92, 110-112, 125-128
3. HM Government, as (1) above, p 44
4. HM Government, as (1) above, p 72
5. BEIS (2022) Review of Electricity Market Arrangements https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1098100/review-electricity-market-arrangements.pdf
6. HM Government (2022) British Energy Security Strategy  https://www.gov.uk/government/publications/british-energy-security-strategy/british-energy-security-strategy#nuclear  pp 20-21
7. HM Government, as (6) above, pp 16 & 22-23
8. HM Government, as (6) above, pp 14-15
9. HM Government (2020)  Powering our Net Zero Future  p 44 Fig 3.4
10. MacKay D (2009) Sustainable Energy Without the Hot Air  UIT Cambridge Ltd  p 38
11. MacKay, as (10) above, p 39
12. Project Solar UK (2021)  How Efficient are Solar Panels?  https://www.projectsolaruk.com/blog/how-efficient-are-solar-panels/#:~:text=The%20average%20solar%20panels%20are,of%20energy%20from%20the%20grid.
13. BRE Trust (2020)  The Housing Stock of the United Kingdom  https://files.bregroup.com/bretrust/The-Housing-Stock-of-the-United-Kingdom_Report_BRE-Trust.pdf  p 16.  23 million is the sum of the numbers of all dwelling types except flats.
14. MacKay, as (10) above, p 265
15. MacKay, as (10) above, pp 265-6
16. MacKay, as (10) above, p 32
17. MacKay, as (10) above, p 60
18. MacKay, as (10) above, p 60
19. MacKay, as (10) above, p 60-1.  120,000 sq km is the sum of 40,000 sq km “shallow offshore” and 80,000 sq km “deep offshore”.
20. Crown Estate Scotland (2022)  Scotwind Briefing p 1 https://www.crownestatescotland.com/resources/documents/scotwind-briefing-june-2022 Comparison of the map at the bottom of https://www.crownestatescotland.com/our-projects/scotwind with that on p 61 of MacKay shows that many of the sites are in waters deeper than 50 m.
21. Arcelor Mittal  Steel is the Power behind Renewable Energy  https://corporate.arcelormittal.com/media/case-studies/steel-is-the-power-behind-renewable-energy
22. House of Commons Library (2021)  UK Steel Industry: Statistics and Policy  https://commonslibrary.parliament.uk/research-briefings/cbp-7317/
23. MacKay, as (10) above, p 157
24. The energy content of 1 cubic metre of natural gas is about 37 megajoules (The Physics Factbook  https://hypertextbook.com/facts/2002/JanyTran.shtml) and 1 kWh equals 3.6 megajoules (https://www.google.com/search?q=joules+per+kWh&rlz=1C1ONGR_en-GBGB927GB927&oq=joules+per+kWh&aqs=chrome..69i57j6.6159j0j7&sourceid=chrome&ie=UTF-8).  37 divided by 3.6 is about 10.
25. Ipieca Combined cycle gas turbines  https://www.ipieca.org/resources/energy-efficiency-solutions/power-and-heat-generation/combined-cycle-gas-turbines/#:~:text=The%20overall%20electrical%20efficiency%20of,cycle%20application%20of%20around%2033%25.
26. Vasudevan S et al (2016) Energy penalty estimates for CO₂ capture  Energy 103 pp 709-14  https://precaution.org/lib/ccs_energy_penalty_for_coal_vs_natural_gas.2016.pdf  Fig 3 p 714
27. BP Statistical Review of World Energy 2021  https://www.bp.com/content/dam/bp/business-sites/en/global/corporate/pdfs/energy-economics/statistical-review/bp-stats-review-2021-natural-gas.pdf  p 34
28. The atomic weights of carbon, hydrogen and oxygen are respectively 12, 1, 16.  So combustion of 1 molecule of methane (CH₄) with molecular weight 12 + (4 x 1) = 16 yields 1 molecule of carbon dioxide (CO₂) with molecular weight 12 + (2 x 16) = 44 (and 2 molecules of water).  44 / 16 = 2.75.
29. British Geological Survey  CO₂ storage capacity estimation  https://www.bgs.ac.uk/geology-projects/carbon-capture-and-storage/co2-storage-capacity-estimation/
30. British Geological Survey Understanding carbon capture and storage https://www.bgs.ac.uk/discovering-geology/climate-change/carbon-capture-and-storage/  See drop-down labelled Carbon dioxide storage
31. Energy Technologies Institute  Strategic UK CCS Storage Appraisal  https://www.eti.co.uk/programmes/carbon-capture-storage/strategic-uk-ccs-storage-appraisal See especially  pp 6 & 8 of the Summary Report
32. HM Government  Digest of UK Energy Statistics Table 5.6  https://www.gov.uk/government/statistics/electricity-chapter-5-digest-of-united-kingdom-energy-statistics-dukes  Row 47
33. Wikipedia – Nuclear power in France https://en.wikipedia.org/wiki/Nuclear_power_in_France
34. For an explanation of different types of reactors see World Nuclear Association (2022)  Nuclear Power Reactors  https://world-nuclear.org/information-library/nuclear-fuel-cycle/nuclear-power-reactors/nuclear-power-reactors.aspx 
35. HM Government  Sites of existing and proposed nuclear power stations in the UK  https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/48352/1138-map-nuclear-power-stations-uk.pdf
36. Wikipedia – Hinkley Point C Nuclear Power Station  https://en.wikipedia.org/wiki/Hinkley_Point_C_nuclear_power_station
37. World Nuclear Association (2022)  Supply of Uranium  https://world-nuclear.org/information-library/nuclear-fuel-cycle/uranium-resources/supply-of-uranium.aspx  The figure given is US$130, which I have converted at a current exchange rate of 1.16 US$/GB£.
38. European Nuclear Society – Fuel Comparison https://www.euronuclear.org/glossary/fuel-comparison/