Small Modular Reactors – “The Real Nuclear Renaissance?” Institute of Physics, Nuclear Industry Group Birchwood, 25 May 2016

Juan Matthews FInstP Dalton Nuclear Institute and School of Materials The University of Manchester

Presentation to IOP 25 May 2016 1 In the beginning there were just small reactors

GEN I GEN II GEN III Large reactors

Medium reactors

Small reactors Reactor power (MWe) power Reactor

Compact reactors

Presentation to IOP 25 May 2016 2 Small reactor markets There are are essentially two markets for small reactors: 1. Small Modular Reactors (SMRs) – Sizes from ~20 to 300 MWe – Designed for grid connection – Need to compete with other grid generation 2. Very small reactors (compact reactors) – Sizes from ~1 to 20 MWe – Designed mainly for off-grid and isolated generation but could be integrated into a grid – Need to be easily transportable and installed – Need to compete with diesel and other off-grid generation The U-Battery is an example of 2, in this presentation we look at 1.

Presentation to IOP 25 May 2016 3 USA interest in SMRs • The American interest in the 1980s for SMRs was driven by EPRI and the US Department of Energy on a strategic basis, looking at requirements for nuclear power both domestically and for export. • Around 2000 the focus had changed to the need to replace coal fired power stations in a fragmented deregulated generation industry. This momentum has, at least temporarily, been broken by the availability of cheap shale gas. • The US DoE is currently running an SMR Technical Licensing Support Program

Presentation to IOP 25 May 2016 4 USDOE view of SMR benefits

• Modularity: in two senses – factory fabrication of modules for a simpler assembly on site and reactor units as modules that can be added to match demand; • Lower Capital Investment: lower unit of investment and lower investment per unit power from factory fabrication and shorter construction times; • Siting Flexibility: smaller footprint, less demanding infrastructure and possibility of siting on existing fossil sites nearer habitation; • Gain Efficiency: use of heat for industry and other applications and coupling with other generation sources for more efficiency; • Non-proliferation: Depending on type of SMR could lead to reduced transport and handling of nuclear materials and longer refuelling times, possibility of sealed fuel units; • International Marketplace: Opens a new nuclear market.

Presentation to IOP 25 May 2016 5 Other potential advantages of SMRs

• Multiple build enables reduced costs through learning; • Enables existing licensed sites to be used where space is limited; • Small reactors can fit into limited electrical grid in remote regions, islands and in developing countries; • Small reactors are simpler and more able to use natural convection and passive safety features and to be located underground.

Presentation to IOP 25 May 2016 6 UK INTEREST IN SMRS

Presentation to IOP 25 May 2016 7 UK first interest in SMRs

• UK Interest in SMRs dates back to the late 1980s when it was realised that the privatisation of the electricity supply industry would make the investment in large reactors difficult. • Studies at that time looked at the effect of scale on cost and the industry was still firmly looking at large reactors. • The UK (Rolls Royce and AEA Technology) joined the SIR project (Safe Integral Reactor) with ABB and Stone & Webster, but in the end the dash for gas won.

Presentation to IOP 25 May 2016 8 UK current interest in SMRs

• UK Nuclear Industrial Strategy, March 2013: “To be a key partner of choice in commercialising Generation III+, IV and Small Modular Reactor (SMR) technologies worldwide”. • NIRAB (Nuclear Innovation and Research Advisory Board) initiated a feasibility study with BIS support in 2014. • The feasibility study was published in December 2014 by the National Nuclear Laboratory and focussed on the integral PWR design. • In May 2015, DECC commissioned an in-depth Techno- Economic Assessment, which has just reported. This covers: the economic case; grid integration, market assessment and future trends; safety and manufacturing.

Presentation to IOP 25 May 2016 9 UK current interest in SMRs

• A report by the Energy Technologies Institute published in October 2015 looks at SMR siting and identifies SMRs as a solution to the shortage of new nuclear sites and the lack of a low carbon solution to energy for heating. • The Government Spending Review in November 2015, specifically mentioned small modular reactors in the context of £250 million increased funding for nuclear. • The 2016 Budget also mentioned small modular reactors. • The Government is now moving to assess designs and lead companies to decide if is will support an SMR project. • A £30 million “Phase One” of a competition was announced on 17 March 2016 and bids had to be in place by 6 May. A roadmap will be published in September 2016.

Presentation to IOP 25 May 2016 10 What is the UK looking for

• SMRs present an opportunity for the UK to have a stake in the development of a system that may lead eventually to involvement in GEN IV reactor construction and to be global nuclear player again. • The hope is to find a project partner that will make space for UK participation and IP generation – perhaps on the model of how aircraft are built. • There are currently too many SMRs designs. Inevitably most will fail but the winners could eventually pick up 100s of orders. The trick is to pick the winner. • The number of sites suitable for large reactors is limited and SMRs are an opportunity to use smaller sites, often nearer centres of population, to expand nuclear capacity.

Presentation to IOP 25 May 2016 11 Potential market for SMRs

Source: NNL SMR feasibility study 2014

The more recent ETI study suggests this may be an underestimate as the estimate up to 21 GWe for the UK by 2050 worth £60 billion. Presentation to IOP 25 May 2016 12 Energy Technologies Institute Study

Small Modular Reactors – UK Energy System Requirements For Cogeneration (October 2015) • Need for nuclear power to contribute to heat supply as well as electricity generation • Need for more flexible generation - load following • Need to use sites too small for GWe stations • Need to remove dependence on cooling water • Need to site smaller stations closer to conurbations to allow the use of heat for industry and homes • Potential to use over 20 GWe of SMR capacity by 2040!

Presentation to IOP 25 May 2016 13 Current and historic thermal power station locations in England and Wales From “Power plant siting study” August 2015, prepared for the ETI by Atkins.

Presentation to IOP 25 May 2016 14 WHAT SMR DESIGNS ARE WE CONSIDERING?

Presentation to IOP 25 May 2016 15 Range of SMR concepts

• In addition to small versions of current reactors, there are over 60 SMR concepts that have been studied in the last 10 years but few are likely to be built. • The main thrust of recent SMR development is with Integral PWRs – PWRs where the steam generator is integrated into the reactor pressure vessel (RPV). We will focus on this direction as several designs are close to market • High temperature gas cooled reactors are inherently small reactors to enable heat removal in loss of coolant accidents. We have the UK U-Battery very small reactor project and China is building a twin 105 MWe pebble bed reactor station. • A number of small sodium and lead alloy cooled fast reactor designs – at one time looked like they might be built but currently are seen as a step too far for SMR development. • There is a recent interest in molten salt reactors, which would be in the SMR range.

Presentation to IOP 25 May 2016 16 The basic components of a PWR

Primary circuit

Secondary circuit

Core

To coastal, river or cooling tower heat sink

Primary pump

Presentation to IOP 25 May 2016 Source USNRC17 Integral PWR (origin marine reactors - Otto Hahn, submarines)

Pressurizer

Steam Steam Control outlet generator rods

Primary circuit

Core

Water Primary inlet Secondary Secondary pump circuit pump

Apologies to USNRC for merging their Presentation to IOP 25 May 2016 PWR and BWR animated GIFs 18 Integral PWR concepts

Power RPV size (m) Core size (m) Primary MWth MWe Diameter Height Diameter Height circuit SIR (UK, USA) 1000 320 5.8 19.2 2.8 3.6 Forced IRIS (international) 1000 335 6.2 21.3 2.3 4.26 Forced CAREM (Argentina) 100 25 3.2 11 1.1 1.4 Natural ACP-100* (China) 310 100 3.19 10 1.85 2.15 Forced SMART (S Korea) 330 100 6.5 18.5 1.85 2.0 Forced mPower (USA) 530 180 4 27 2.0 2.4 Forced NuScale (USA) 160 50 2.9 17.4 1.5 2.0 Natural Westinghouse SMR 800 225 3.7 28 2.3 2.4 Forced SMR 160** (USA) 525 160 2.7 15 1.64 3.7 Natural IMR (MHI) (Japan 1000 350 6.0 16.8 2.95 2.4 Natural RITM 200* (Russia) 175 50 3.3 8.5 2.2 1.65 Forced

*External pressuriser **Steam generators and pressurizer section of RPV offset with a dog-leg Presentation to IOP 25 May 2016 19 Advantages of the integral PWR concept

Careful design will give a considerable reduction in the NSSS cost but the main advantages are related to the safety of the system • The primary circuit is kept entirely within the RPV, so no primary pipework and no active circuit outside RPV • No penetrations of the RPV larger than 50 mm diameter so no large break loss of coolant accidents • Increased shielding of RPV from fast neutrons • Lower core power density and larger volume of water in RPV • Larger surface area to power ratio, making passive decay heat removal easier • Uses standard PWR fuel, but with shorter fuel rods.

Presentation to IOP 25 May 2016 20 Safe Integral Reactor (SIR)

Power – 300 MWe Power density – 55 MW/m3 (similar to BWR) Special design features – 12 modular steam generators, integral with RPV There is an extremely low fast neutron dose to the reactor pressure vessel because of the large water gap between the reactor core and the vessel wall. Vessel diameter – 5.8 m Vessel height – 19.2 m

Source: Matzie et al, Nucl. Eng. and Design 136 (1992) 73 Presentation to IOP 25 May 2016 21 • IRIS (International Secure and Innovative) was a Primary collaboration led by Westinghouse that included Pressurizer pumps BNFL in the UK and organisations in Italy, Spain, Russia, Brazil and Japan. • The main design work was done in the early 2000s, but the influence of the project can be CRDM seen in all current integral PWR designs. • The design is just outside the normal definition of Steam an SMR at 335 MWe. generators 8 helical • IRIS has the features that can be seen in designs once like the Westinghouse SMR, mPower, SMART, through CAREM and some Russian designs modules Core • IRIS is clearly influenced by the SIR design (Westinghouse acquired ABB Combustion Engineering) but the main innovation is the sealed CRDM inside the RPV.

Presentation to IOP 25 May 2016 22 Source: IAEA • Westinghouse has moved on from its role in IRIS to create its own 225 MWe SMR design Claimed to be derived from the AP1000 design but clearly built on the IRIS experience • The design is much longer and thinner than IRIS and a second flange is added below the steam generators to allow easier access to the core for fuel reloading. The coolant pumps are also located lower. • The Westinghouse design is well developed and has advanced along the USNRC approval but there are no specific plans to build a first plant. Westinghouse have been lobbying the UK Government since the 2015 Spending Review. © Westinghouse Electric Co Presentation to IOP 25 May 2016 23 Source: Westinghouse • The mPower design comes from B&W in the USA and was being developed as a JV with Bechtel. It is said to be based on its experience with US naval reactors, but the influence of IRIS is obvious • The design has been rated at various powers but currently aims to deliver 180 MWe. In many 530 respects the design is very similar to that of 180 Westinghouse but the coolant pumps are still located at the top of the RPV. • The containment concept is less ambitious than Westinghouse, NuScale or ACPDiscussion-1000, with a on construction on larger dry containment.the TVA Clinch River Site. Plan to submit • After a very bullish start, several US DoE grants Design Certification application to the NRC and substantial investment by B&W, the projects by late-2014 for approval by 2018 is now essentially mothballed. B&W are looking for a partner to take a majority stake, but still hopes to be the main manufacturer. © Generation mPower LLC Presentation to IOP 25 May 2016 24 Source: Generation mPower LLC • NuScale started as a concept at of Oregon State University and Idaho National Laboratory but is now being developed by Fluor. From the point of view of containment and heat removal it is the most innovative of the new designs. • As the only design the UK is looking at, that uses natural convection during normal operation, the reactor power is just 50 MWe. It is intended to be used in blocks of up to 12 modules (a total of only 600 MWe). Despite this an overnight cost of only $5000/kWe is claimed • The CRDM system is external using conventional technology, which means the linkages are very long. • Rolls Royce has given support to NuScale in its US DoE 2016 funding applications. Currently the plans are to build the first plant at Idaho and it is expected that the construction and operation licence application to USNRC will be made in 2017, NuScale are lobbying

© NuScale Power LLC the UK Government since the 2015 Spending Review. Presentation to IOP 25 May 2016 25 Source: NuScale ACP-100 and ACP-100+ (CNNC, China)

• The Integral PWR project closest to Pressurizer implementation is the the Chinese ACP-100 design which is currently being assessed for safety by the IAEA CRDM and construction is waiting Central Committee approval. Steam Primary generators pumps • The design for a 100 MWe reactor is being replaced with a 120 MWe development with several important design improvements, notably placing Core the CRDM and pressurizer inside the RPV. • The new reactor is designated for the time being as ACP 100+ and this design is the one that the UK is looking at a ACP-100+ possible development partnership.

Presentation to IOP 25 May 2016 26 Source: CNNC SIZE MATTERS

Presentation to IOP 25 May 2016 27 Effect of economies of scale on reactor size

Source: Locatelli et al, Prog. in Nucl. Energy, 73 (2014) 75. Presentation to IOP 25 May 2016 28 Is reactor construction time related to reactor power? 12

Japanese construction times 10

PWR 8

BWR 6 (years)

4

2 ieomsato osrcin to ccommercial operation Timeform start of construction

0 0 200 400 600 800 1000 1200 1400 Reactor net power (MWe)

Presentation to IOP 25 May 2016 29 Claimed construction times (from first concrete to grid connection)

Main vendor offerings

VVER 1200 EPR

RITM 200 ATMEA APR1400 IMR (MHI) AP 1000 ABWR NuScale Smart mPower

Westinghouse SMR SMR 160 DMR (Hitachi)

Current SMRs concepts

Presentation to IOP 25 May 2016 30 Variation of reactor vessel diameter with power

PWR Integral PWR

SMRs

10

SMART IRIS IMR SIR

mPower ACP100 Westinghouse CAREM SMR160 NuScale RITM 200

Reactor vessel inner diameter (m) ABV 6M

P ∝µr3 P ∝µr2 1 10 100 1000 10000 Thermal Power (MW)

Presentation to IOP 20 April 2016 31 Variation of reactor vessel diameter with power

PWR BWR Integral PWR HTGR LMFBR SMRs

10 Reactor vessel inner diameter (m)

P r3 P r2 1 10 100 1000 10000 Thermal Power (MW)

Presentation to IOP 25 May 2016 32 Comparison of size of NSSS of an SMR to a regular PWR AP 1000 NSSS net power 1117 MWe mPower NSSS SMART Net power 180 MWe NSSS Net power 100 MWe

27m

12m 18.5m

6.5m OD 4m OD 4.4m OD Presentation to IOP 25 May 2016 33 Source: IAEA ARIS data and images Comparison of size of primary containment for integral and regular PWRs.

AP1000 Westinghouse unit 1117 MWe 225 MWe

mPower unit 180 MWe NuScale unit 45 MWe

Presentation to IOP 25 May 2016 34 Source: IAEA ARIS data Primary containment of Westinghouse SMR

© Westinghouse Electric Co Presentation to IOP 25 May 2016 35 Source: Westinghouse Containment SMR Containment

NuScale • One of the opportunities for SMR safety design is to choose a small-volume pressure Reactor vessel suppression containment more usually found with BWR systems. • The NuScale, Westinghouse and ACP-100+ designs have below ground tight steel high pressure containments. Modules with • Westinghouse and NuScale can operate with own containment the containment under vacuum, so that the in water pond in reactor has no insulation and is easier to concrete cells cool by flooding the containment in a loss of cooling accident. © NuScale Power LLC • The Westinghouse and ACP-100+ containments have internal water reservoirs for pressure suppression. • The NuScale design has up to 12 reactors sitting in a large fuel handling pool.

Presentation to IOP 25 May 2016 36 Source: NuScale Footprint of SMRs – are they smaller?

Reactor Nuclear island area Nuclear and non- Total area nuclear Island area m2 m2/MWe m2 m2/MWe Hectares m2/MWe

Current designs relevant to UK EPR ~10000 6 ~25000 15 Twin 170 ~500 AP1000 ~5000 4.5 ~10000 9 Twin 125 ~560 ABWR ~3250 2.5 ~7500 5.5 Twin 50 ~185 SMR designs mPower ~1225 7 ~3675 20 Twin 16 ~450 Westinghouse SMR ~1124 5 ~4500 20 Single 6.5 ~290 NuScale ~1458 32 ~6561 146 x12 18 ~333 SMART ~3600 36 ~7200 72 Single ~9 ~900 HTR-PM ~1300 12.5 ~3000 29 Twin ~2 ~100

SMRs occupy less land in absolute terms but are comparable relatively

Presentation to IOP 25 May 2016 37 WHAT ARE THE SMR CHALLENGES FOR THE UK?

Presentation to IOP 25 May 2016 38 Overnight costs of US nuclear plant

Range of US SMR claims

AP 1000 US EIS data

Presentation to IOP 25 May 2016 39 SMR overnight capital cost estimates £/kWe

Data from “SMR feasibility study”, National Nuclear Laboratory, December 2014 The study had access to 4 of the best developed SMR design, the “adjusted” data is where the original vendor data was reassessed by the study team.

Presentation to IOP 25 May 2016 40 SMR levelised electricity cost estimates

Levelised cost estimates are shown for discount rates of 5%, 8% and 10% for FY 2012 £s Data from “SMR feasibility study”, National Nuclear Laboratory, December 2014

Presentation to IOP 25 May 2016 41 Importance of reducing manufacturing costs

• An Energy Technologies Institute study has show that overturning the economies of scale for SMR is vital to their viability. • The use of factory production to reduce costs might be a way of getting substantial cost reduction with production of many units. • The use of residual heat Cost model projections and breakeven capex levels for for district heat makes electricity only generic SMRs the viability of SMR From “The role of nuclear in a low carbon energy system” and much more attainable. Energy Technologies Institute Insight Report October 2015.

Presentation to IOP 25 May 2016 42 ETI study on cost reduction

250

200

150

100

50

0

(Weighted average capital cost) From: “System requirements for alternative nuclear technologies” Mott MacDonald Report the Energy Technologies Institute, August 2015. Presentation to IOP 25 May 2016 43 What can we conclude about SMR costs? • SMRs should be built in substantially shorter times than larger reactors • SMRs will occupy much less land than larger reactors but the area occupied/kWe is not very different. • Overnight costs/kWe of SMRs are similar to those of larger reactors. • Levelised costs of electricity from SMRs may be smaller than from larger stations, but there is a large uncertainty which can only be resolved when an SMR building program is started. • The unit cost of capital for a twin SMR station is still large for a deregulated electricity industry at around £2 billion.

Presentation to IOP 25 May 2016 44 Manufacturing issues

• The only way the overnight costs for SMRs can compete with larger reactors is to have simple completely modular designs that can be made efficiently in factory conditions and quickly assembled on site. • If SMRs are successful then the number of systems built could be hundreds and experience from the aircraft and automotive industry might be useful. • It would then be worth making the investment in tooling for automated fabrication

Presentation to IOP 25 May 2016 45 Modular construction • Modular construction should extend to the whole plant. • Experience from the aircraft industry shows that complex modules, weighing >100 tonnes containing multiple services can be made by different factories and assembled on site, but it needs precision.

A380 Module Source: Airbus Presentation to IOP 25 May 2016 46 Modular construction for whole plant

• Replacement of shuttering and rebar setting with structural modules for shield walls and main plant structures. Modules are filled with concrete once in place. • Structural modules with built-in piping , cable trays, ductwork, HI-VAC and other services. • Functional modules that contain a plant systems, eg water treatment. • Modules need to be small enough to be transportable and where possible they should be interchangeable.

Presentation to IOP 25 May 2016 47 SMR R&D needs • Integral PWRs mostly use the same technology as other LWRs and the same fuel, reduced in length. So the materials, fuel and structural integrity issues are the same ones as being faced in BWRs and PWRs, eg Zr alloy oxidation and radiation damage of the RPV. • The really distinct issues for SMRs are related to: – specific design features; – safety; – and manufacture. • For lower power SMRs there is the possibility of cassette fuel replacement.

Presentation to IOP 25 May 2016 48 NNUMAN and Nuclear AMRC • There are two major activities in the UK at the heart of advanced nuclear manufacturing: – The EPSRC New Nuclear Manufacturing program (NNUMAN) which is does research at TRL 2-4 with £8 million investment. – An 8000m2 research factory, with £30 million investment, at the Nuclear Advanced Manufacturing Research Centre (AMRC)

Manufacturing Technology Research Laboratory The University of Manchester

Nuclear AMRC Research Factory University of Sheffield

Presentation to IOP 25 May 2016 49 Use of advanced manufacturing technology • It has proved very difficult to introduce the latest technologies into current new build, because of barriers of licensing of established designs. • SMRs offer an opportunity to avoid the barrier, decreasing costs and manufacturing times, while improving reliability. • Technology being developed by NNUMAN and the Nuclear AMRC – Electron and laser welding – narrow gap, thick section – Automatic welding of complex components – steam generators – Assisted machining – cryogenic machining – Robotic machining and process to part manufacture – Diode laser cladding of primary circuit facing structures – Near net shape and additive manufacture – eg HIP – State-of the-art metrology with virtual and augmented reality Presentation to IOP 25 May 2016 50 Modular structure of the RPV in the mPower and Westinghouse designs.

Pressurizer Steam generator Upper flange Steam outlet and feedwater Mid-flange inlet Pumps

CRDM

Core

Westinghouse SMR mPower

Presentation to IOP 25 May 2016 51 Main components for the mPower NSSS

• Pressure 14 MPa Max T 320°C • Main structural components low alloy SA-508 Gr. 3 steel with primary facing surfaces clad in 308 stainless steel or alloy 57. • Currently large structures are likely to be forged and machined. Cladding of surfaces is done by overlay welding. Studies are being done on the use of powder metallurgical methods • Steam generator tubes are Alloy 609 or a similar nimonic. • The secondary circuit is largely based on low alloy steel so there are no transition welds

Source: Sandusky and Lunceford, PNNL-22290 2013 Presentation to IOP 25 May 2016 52 Main components for the mPower core structures

• Core structures are mainly made of 304L or 316 nuclear grade stainless steels, but these have poor radiation resistance, largely through the high nickel content. No alternatives are available yet and this could be a life limiting feature on components. • Components like support grids are made by machining large forging with high fractions of material removal, as welded structures may not be acceptable. There are opportunities for new advanced manufacturing methods Source: Sandusky and Lunceford, PNNL-22290 2013

Presentation to IOP 25 May 2016 53 Main components for the NuScale NSSS and primary containment Differences from the mPower design include: • No recirculating pumps as the system uses natural convection; • Control rod drive mechanism (CRDM) is external so the top dome has many penetrations and the control rod extensions are very long • The steam generator configuration is the opposite of mPower – the secondary circuit passes through the boiler tubes. • The primary containment is a steel vessel fitted tightly around the reactor vessel. Source: Sandusky and Lunceford, PNNL-22290 2013

Presentation to IOP 25 May 2016 54 Design issues • Primary coolant pump – The primary pumps are usually fixed externally in a “canned design” with a penetration through the RPV for the drive to the impellors. Some designs dispense with pumps and operate wholly on natural convection. • Control rod drive mechanism (CRDM) – The long RPV and the location of steam generators and pressurizer makes design of the CRDM difficult. Some designs use a sealed internal CRDM. • Steam generator – The steam generators have to be compact and efficient. Many systems use modular steam generators that can be isolated and separately removed if there is a tube leak. • Pressurizer – The location and small volume available for the pressurizer is a challenge.

Presentation to IOP 25 May 2016 55 Westinghouse SMR CRDM

“The CRDMs are a high-temperature and pressure version of the recently developed AP1000 CRDM, which has already been tested to eight million steps, to be used within the pressure boundary of the SMR integral reactor. Elimination of control rod penetrations through the reactor pressure boundary has resulted in reduced cost of manufacture and the elimination of the normally-postulated rod ejection event.”

Source: Nuclear Engineering International

Presentation to IOP 25 May 2016 56 Safety issues • Smaller reactors are easier to cool passively • Some designs are very conventional in their approach to safety systems, but others, notably NuScale, take the opportunity to completely rethink heat transfer in accidents. • The use of tight low volume containments. • Sitting the containment in a large volume fuel- handling water pool. • The use of multiple cooling routes and with low power cores the possibility of dry cooling by air convection without the melting of the reactor.

Presentation to IOP 25 May 2016 57 NuScale safety features

In operation there is a vacuum Decay heat removal from Emergency core cooling by between the containment external feed water passive convective cooling by vessel and the RPV. There is no accumulators and from main flooding containment vessel. need for insulation on the RPV water pool. Steam generators Any primary circuit steam is making inspection easier. are used to cool primary circuit. vented through reactor vent valves Presentation to IOP 25 May 2016 58 © NuScale Power LLC Source: NuScale Power LLC Long term cooling in NuScale

© NuScale Power LLC Presentation to IOP 25 May 2016 59 THAT ALL FOLKS!

[email protected]

Presentation to IOP 25 May 2016 60