Context

1. The Path to Fusion PowerChris Llewellyn SmithDirector UKAEA CulhamChairman Consultative Committee for Euratom on FusionChair ITER Council

2. Context • Huge increase in global energy use expected + needed to lift billions out of poverty • Meeting demand in an environmentally responsible manner will be an enormous challenge • A ‘portfolio’ approach is needed (no silver bullet) (NBElectricity only = 1/3 of total primary energy demand) • improved efficiency (encouraged by fiscal measures) • renewables*when appropriate, *none can meet a large % of the world’s needs, except solar which could provide 100% in principle - butbig breakthroughs in cost and storage needed • must including large scale sources of base-load power, for which only options are: hydro (but potential limited), continue burning fossil fuels (so carbon capture and storage important), fission, and potentially fusion

3. FUSION • powers the sun and stars • and a controlled ‘magnetic confinement’ fusion experiment at the Joint European Torus (JET) • (in the UK) has produced 16 MWof fusion power • so it works The big question is - when will it work reliably and economically, on the scale of a power station? First: What is it? Why bother? Why is it taking so long? s

4. A “magnetic bottle” called a tokamakkeeps the hot gas away from the wall Challenges: make an effective “magnetic bottle” (now done ?) a robust container, and a reliable system * ten million times more than in chemical reactions, e.g. in burning fossil fuels while a 1 GW coal power station would use 10,000 tonnes of coal a day, a fusion power station would only use 1 Kg of D + T Most effective fusion process involves deuterium (heavy hydrogen) and tritium (super heavy hydrogen) heated to above 100 million °C: Helium Deuterium + energy (17.6MeV)* Tritium Neutron WHAT IS FUSION ?

7. A “magnetic bottle” called a tokamakkeeps the hot gas away from the wall Challenges: make an effective “magnetic bottle” (now done ?) a robust container, and a reliable system * ten million times more than in chemical reactions, e.g. in burning fossil fuels while a 1 GW coal power station would use 10,000 tonnes of coal a day, a fusion power station would only use 1 Kg of D + T Most effective fusion process involves deuterium (heavy hydrogen) and tritium (super heavy hydrogen) heated to above 100 million °C: Helium Deuterium + energy (17.6MeV)* Tritium Neutron WHAT IS FUSION ?

8. A Fusion Power plantwould be like a conventional one, but with a different fuel and furnace The blanket captures energetic neutrons produced in the fusion process, which: - react with lithium in the blanket to produce Tritium ( fuel the reactor) - deposit their energy  heat which is extracted through a cooling circuit and used to boil water and produce steam to drive a generator

9. Why bother? • Lithium in one laptop battery( tritium from the reaction: • neutron (from fusion) + lithium  tritium + helium) • + 40 litres of water(from which ‘heavy water’/deuterium can easily be extracted),used to fuel a fusion power station, would provide 200,000 kW-hours = • (EU electricity production for 30 years)/(population) in an intrinsically safe manner with no CO2 • Unless/until we find a barrier, this is sufficient reason to develop fusion power

10. FUSION ADVANTAGES • unlimited fuel • no CO2 or air pollution • intrinsic safety • no radioactive “ash” and no long-lived radioactive waste • competitive* electricity generation cost, if reasonable availability (e.g 75%) can be achieved *compared to most other carbon free electricity sources

11. FUSION DISADVANTAGES • The blankets will become radioactive • but can choose materials so that half lives ~ 10 years, and all components could be recycled  new fusion power plant within 100 years(no waste for permanent repository disposal: no long-term burden on future generations) • More research and development needed • Fusion power stations will need plasma volumes of at least 1000 m3 (ten times JET), so small scale demonstration impossible (hence - relatively slow - step by step progress)

12. Why so long? • Cannot demonstrate on a small scale: (power out)/(power to operate) grows faster than (size of fusion device)2 – need GW scale to be viable • Not funded with any urgency – otherwise from agreement on basic geometry in 1969, could have reached today’s position 15 years ago (note that energy R&D boosted by oil crisis but then collapsed) • It is very challenging • need to heat ~ 2000 m3 of gas to over 100 M 0C, without it touching the walls • find robust materials with which to make the walls (able to withstand intense neutron bombardment and heat loads) • ensure reliability of very complex system Nevertheless huge progress: from T3 to JET and from JET to ITER (later)

13. Progress in Fusion has been enormous, but even JET(currently the world’s leading fusion research facility) is not large enough to be a (net) source of power T3: Volume ~1 m3 Temperature ~ 3 M 0C Established tokamak as best configuration (1969) • JET: Volume ~100 m3 • Temperature ~ 150 M 0C • World record (16 MW) for fusion power (1997)

14. JET

15. MAST Progress • Huge strides in physics, engineering, technology • JET: 16 MW of fusion power ~ equal to heating power. • Ready to build a Giga Watt-scale tokamak:ITER – expected to produce 10 x power needed to heat the plasma • [Pi =pressure in plasma; • τE = (energy in plasma)/(power supplied to keep it hot)]

16. NEXT STEPS FOR FUSION • Construct ITER (International Tokamak Experimental Reactor) • energy out = 10 energy in • “burning” plasma During construction, further improve tokamak performance in experiments at JET, DIII-D, ASDEX-U, JT- 60…further develop technology, and continue work on alternative configurations [Spherical Tokamaks (pioneered in UK), Stellarators] • Intensified R&D oni) materials for plasma facing and structural components and test of materials at the proposed International Fusion Materials Irradiation Facility (IFMIF), and ii) fusion technologies

17. ITER JET (to scale)

18. ITER • Aim - demonstrate integrated physics and engineering on the scale of a power station • Key ITER technologies fabricated and tested by industry • 5 Billion Euro construction cost (will be at Cadarache in southern France) • Partners house over half the world’s population

19. Plasma Physics Issues • Major positive developments (1980s and 90s) • ‘Bootstrap’ plasma current (predicted at Culham)  much less external power needed than previously thought • High confinement mode (serendipitous discovery at Garching)  higher pressure + more fusion power with given magnetic field • Potential Problems • New instabilities in burning plasmas? • Steady state operation in power station conditions (looks possible with help of bootstrap current: if not, could  pulsed machine, or stellarator) • Potential improvement • Better control of potential instabilites to allow higher pressure

20. After:The edge of the plasma is very sharp and energy containment improves so the plasma pressure you can maintain is bigger Transition to H mode at MAST (at Culham, UK) Before: the edge of the plasma is fuzzy and energy containment is poor

21. Spherical Tokamaks • Based on promising, more compact but less developed, configuration than JET and ITER - use magnetic field much more efficiently (but face other challenges): START (Culham, UK 1991-1998, first substantial Spherical Tokamak) raised world record for key figure of merit  ( = ratio of plasma to magnetic pressure) from 13% to 40% ! • Many STs built subsequently: world’s leading STs are • NSTX (Princeton) and MAST (Culham):

22. Spherical Tokamaks • Making important contributions to conventional tokamak physics • different shape → new perspective • Could play vital role as a “Component Test Facility” in the medium-term • A CTF, which would test whole components (blankets, welds, joints,…), is a highly desirable (perhaps essential) step between ITER and a prototype power station • Could, in long-run, be basis for (smaller and simpler) power stations • No superconducting magnets → cheaper and simpler

23. STELLARATORS(Originally pioneered at Princeton) • Helical field, needed to confine plasma, provided externally • Avoid need to drive the Mega Amp currents that provide (part of the) helical fields in Tokamaks, and are a source of instabilities • Intrinsically steady state devices. The price is greater complexity. • LHD in Japan: W7-X under construction in Germany:

24. MATERIALS • Structural materials –subjected to bombardment of 2 MW/m2 from 14 MeV neutrons • Plasma facing materials subjected to an additional 500 kW/m2from hot particles and electromagnetic radiation (much more on ‘divertor’) • Various materials have been considered, and there are good candidates that may survive in these conditions, BUT: • Further modelling +experiments essential: • Only a dedicated (€800M) accelerator-based test facility - the International Fusion Materials Irradiation Facility (IFMIF) - can reproduce reactor conditions: results from IFMIF will be needed before a prototype commercial reactor can be licensed and built

25. Materials Issues • Major positive development (1990s) • Body-centred cubic low activation steels seem able to withstand neutron damage • Potential problems • Effect ofhelium generation in the materials • Heat on ‘divertor’ (can be reduced by compromising design) • Potential improvement • Development of advanced materials (SiC ceramics,…) for much higher temperature operation

26. European Power Plant Conceptual Study • Results • Confirm good safety and environmental features of fusion • Give encouraging range for the expected cost of fusion generated electricity(9 €-cents/kW-hour for early near-term [water cooled steel] model; 5 €-cents/kW-hour for early advanced [Li-Pb cooled Si-C composites] model) • Note • Economics favours large fusion power plants  major centres of population (complementary to renewables) • Capital intensive; very low operating cost – lots of cheap off peak power  hydrogen? • Results of this study used as input to Culham ‘Fast Track’ study

27. European Power Plant Conceptual Study • Results • Confirm good safety and environmental features of fusion • Give encouraging range for the expected cost of fusion generated electricity(9 €-cents/kW-hour for early near-term [water cooled steel] model; 5 €-cents/kW-hour for early advanced [Li-Pb cooled Si-C composites] model) • Note • Economics favours large fusion power plants  major centres of population (complementary to renewables) • Capital intensive; very low operating cost – lots of cheap off peak power  hydrogen? • Results of this study used as input to Culham ‘Fast Track’ study

28. FUSION ‘FAST TRACK’ • During ITER construction • operate JET, DIII-D, JT60…  speed up/improve ITER operation • In parallelintensify materials work (approve and build IFMIF as soon as possible)and development of fusion technologies (magnets, remote handling, heating systems, fuel cycle, safety,…) • Then, having assimilated results from ITER and IFMIF, build a Prototype Power Plant (‘DEMO’) •  Fusion a reality in our lifetimes

29. Fast Track - Pillars Only

30. Role of Fusion in 2100? • A 1998 study (using MARKAL) by the Netherlands Energy Research Foundation(ECN) looked at potential role of fusion in the European Energy market* * a ‘world study’ is currently being made in the framework of the European Fusion Development Agreement • Some of the assumptions (e.g. 2100 cost of oil = $30/barrel!) no longer look reasonable, others still valid (e.g. expected cost of fusion energy) • – all such modelling is of course subject to enormous uncertainties (especially on discount rate and environmental targets) • modelling = exploration of what might happen, not prediction of what will happen

31. Outcome of ECN modelling • With no constraint on carbon, coal is dominant • Fusion plays an important role with atmospheric CO2 limited to ~ 600 ppm or less, or a carbon tax of €30/tonne or more This conclusion is relatively insensitive to other assumptions – it is very hard to meet expected demand with carbon constrained e.g. changing assumptions to allow more fission reduces gas, not fusion* * unless unlimited fission allowed at ~ current uranium price/without fast breeders – which seems unlikely

32. Conclusions on Fusion • DEMO could be putting fusion power into the grid in under 30 years, given • Funding* to begin IFMIF in parallel with ITER, plus technology development and start of design of DEMO • No major adverse surprises *world fusion funding ~ $1.5 billion pa [c/f electricity(energy) market ~ $1.5 trillion ($4.5 trillion) p.a.] • The cocktail of energy sources that are needed (plus improved efficiency) to meet the energy challenge must includelarge-scale sources of base load electricity – fusion is one of very few options