Chapter 4

Power Reactor Core

PWR

• About 70 - 75% of US commercial reactors

• T wo separate coolant loops

– Primary: Reactor c ooling t o s team generators (single phase)

– Secondar y : Steam g enerators to Turbines to Condenser (phase change)

• A bout 150-200 fuel assemblies

– 1000 to 1200 MWe

– R oughly 21 cm x 21 cm

PWR

Public domain image from wikipedia.

PWR

• Light water coolant and moderator: no physical separation

– Inlet: 275 C

– O u tl e t : 315 C

– P ressure: 15 MPa

– 32-33% efficient

Public domain image from wikipedia.

P W R : F u e l A s s e m b l y

• U O 2 f u e l : c e r a m i c f o r m

• E nrichment 3-4.5% (max of 5% in US)

• Z i r c a l o y cladding

• 14 x 14 to 17 x 17 fuel rods

• 4 m eters long

Public domain image from wikipedia.

PWR: Control Mechanism

• B oron in moderator ( coolant ) : Boric acid mixed in the water, B-10 has a high thermal x.s.

– C oncentration is adjusted to keep reactor critical

• C ontrol rod banks

• R oles

– C on t ro l excess reac ti v it y

– E nable startup

– Cou nt e r ac t

• t he fission product poisons

• negative temperature feedback

• fuel depletion

PWR: Important Component

• P ressurizer: Maintains p ressure inside p rimar y coolant loop

– If coolant temperature increases, water density will decrease, thus taking more space. The water expands in the pressurizer and raises the water level, comp ressing the steam at the top and increasing the loop pressure.

• C old water is sprayed on top of the steam, thus condensing it and reducing p ressure . If pressure keeps increasing , r elease v alves open.

– If coolant temperature decreases, water density increases, thus taking less space. The water level of the pressurizer drops .

• H eaters kick in to boil some of the water and increases qty of steam, thus increasing the pressure in the loop.

PWR: Emergency C ore C ooling

• H i g h p ressure in j ection

• I ntermediate pressure injection

– P rovide water to the core when water pressure remains relatively hi g h

• C old leg accumulators

– D oesn’t require electrical power

Borated w ater in a t ank with a p ressurized n itrogen bubble a t t he top

– K icks in when significant pressure drop

Low pressure injection

– R esidual heat removal

– A lso takes water from sump pump and re-circulates in long outages .

BWR

• About 25 - 30% of commercial reactors in the US

• Only one cooling loop

• U p to 800 assemblies in a core

– Roughly 15 cm by 15 cm

– A pprox 1000 MWe

BWR

Public domain image from wikipedia.

BWR

• Light water c oolant and moderator: no physical separation

– I nlet: single phase

– Outlet: 285 C , boils , two - phase

– Pressure: 7 . 5 MPa

BWR Fuel Assembl

• UO2 fuel: c eramic form

• Enrichment 3 - 4 . 5 %

(max of 5% in US)

• Zircaloy cladding

• 6 x 6 to 10 x 10 fuel rods

• 4 meters long

BWR C ontrol Mechanism

• Cruciform c ontrol blades

– C ontrol rods move to adjust core power

– Shutdown m echanism

– D riven from the bottom of the core

• Coolant feed flow

– I ncrease

• Less void, more moderation, power increases

– D ecrease

• M ore void, less moderation, power decreases

BWR: Emergency C ore C ooling

• H i g h p ressure in j ection

– P rovide water to the core when water pressure remains relatively high

• A u t oma ti c D epressur i za ti on sys t em

– R educes pressure of the core if high pressure in j ection is not workin g

• Low pressure injection

– P rovides water to the core in large breaks

• C ore spray

– S prays water on the top of the core

Advantages

• P W R

– N egative Doppler coefficient

– Negative c oolant void coefficient

– U ses light water

– No radioactive contamination of steam generators

– Less complex to operate than BWR

– S maller footprint than BWR

• B W R

– N egative Doppler coefficient

– Negative c oolant void coefficient

– U ses light water

– No boric acid in the moderator

– Less pressure than PWR

– L ower f ue l t empera t ure than PWR

New designs – AP1000

• Advanced PWR ~ 1100 MWe

• P roposed and designed by Westinghouse

• Si mp l er d es i gns

– Less piping

– F ewer valves

– F ewer pumps

– …

AP 1000

• P assive safet y s y stem

– R elies on natural convection to cool the core in the event of an accident

– “ … t he AP1000 relies o n t he natural forces of gravity, natural circulation and compressed gases to keep the core and containment from overheating.”

• Results of the P robabilistic Risk Assessment (PRA) show a very low core damage frequency (CDF) that is 1/100 of the CDF of currently operating p lants and 1/20 of the m aximum CDF deemed acceptable for new, advanced reactor designs.

A P 1 0 0 0

• 6 units under construction in China

• C ombined License (COL) applications filed for 14 units in the US

New design - EPR

• European Pressurized R eactor

~1600MWe

• D esign features

– Core catcher in the e vent of a c ore m eltdown

– F our independent emergency cooling systems ( 300% redundanc y)

– D ouble containment to resist airplane crash

EPR

• P r o p o s e d a n d design

(and under construction ) b y AREVA

– O ne unit in Finland

– O ne unit in France

• 4 COL applications filed in the US

Images removed due to copyright restrictions.

New design – ABWR

• Advanced BWR

– P assive safety

– Shorter construction time (39 months)

– D esigned by GE

• NRC approve d ABWR i n 1997

– 4 units already operating in Japan

– COL applications filed for 2 ABWRs in Texas

New design – ESBWR

• Economic Simplified BWR

– D esigned by GE- H itachi

– ~1600MWe

– N atural circulation only, no pumps

– Very low risk

• CDF less than 1/10 of the AP1000’s

– Much lower costs

– COL applications filed for 6 units in the US

Summary from this class

• Neutron M ultiplication Factor

• Essential features of PWR and BWR

• U n d ers t an d h ow an d w h en th e pressur i zer operates

• A dvantages and Disadvantages of PWR and BWR

• C ontrol Mechanisms of PWR and BWR

k : six factor formula

P NL = P FNL P TNL

P FNL probability that fast neutron will not leak out

P TNL

probability that thermal neutron will not leak out

k   f  pP FNL P TNL

k - infinity: four factor formula

k    f  p

Summary from this c lass

• Neutron life cycle i n a thermal r eactor

• F our factor formula

– D e fi n iti ons o f th e t erms

– R ange of the terms

• S ix factor formula

CANDU - History

• UK war scientists moved t o C anada; research laboratory created at the Université de Montréal in 1942. Also came Lew Kowarski, Russian émigré physicist who had worked in France and then had fled to England.

• K owarski came with very valuable cargo: almost entire world’s supply of heavy water, spirited out of Norway and then out of France .

• 1943: Meeting between Roosevelt, Churchill, and Mackenzie King. Canada enters into wartime collaboration on research into nuclear fission with UK and USA.

• T he importance of heavy water as a neutron moderator was understood, and since Canada now had an inventory of it, Canada was given the responsibility for developing a heavy - water reactor to eventually produce plutonium for an atomic bomb f or the w ar effort .

• T he Montréal Laborator y was moved to Chalk River in 1944.

• W ork began on designing NRX, which was to be t h e pro d uct i on reactor f or p l uton i um f or t h e war effort.

However , Lew Kowarski was able to get authorization, as a first step, to build a research reactor: ZEEP (Zero Energy Experimental Pile).

• Following end of war, in the early 1950s, several visionaries, among them Bennett Lewis, head of Chalk River Nuclear Laboratories (which eventually became AECL in 1952), lobbied hard to apply Canada’s nuclear knowledge to peace f u l en d s: th e pro d uc ti on o f e l ec t r i c it y.

• B ennett, a man of purpose and eloquence, convinced the Government to give AECL that mandate.

• E xcellence and success of ZEEP develo p ment made it natural to continue in the heavy-water “path” for the moderator.

• Thi s was i n contrast to t h e US d ec i s i on to develop light-water reactors for power, which followed f rom t he successful American nuclear - submarine program.

• A distinctive, world-class Canadian reactor design was born – a great technological success and a proud feat for a country with a small population .

Basic C haracteristic

• Heavy w ater moderator

– T he neutron economy of heavy water is such that natural uranium can be used as fuel.

– With light water as moderator, this is not the case: the rate of neutron absorption is sufficiently high that the reactor cannot go critical with natural uranium fuel; the uranium must first be enriched in the 235U isoto p e to increase the probability of fission relative to that of absorption.

• Natural U ranium Fuel

– I mportant for Canada: self-sufficient in its very

large u ranium resources , i t d id not have t o

develop the complex and costly enrichment capabilit y or rel y on external sources of enriched fuel.

– R emains important factor for other small countries not willing to depend on foreign sources for reactor fuel.

• CANDU fuel is uranium dioxide.

• E ach element consists of UO2 pellets encased in a zircalo y sheath.

• A number of fuel elements are assembled to g ether to form a bundle of len g th ~ 50 cm.

• T he elements are held together by bundle end plates.

Public domain image from wikipedia.

• T he CANDU fuel bundle is short and eas y to handle.

– N o need for special borated casks

• It has few (7) different components.

• C ANDU fuel is much cheaper than light water reactor fuel

• C ANDU fuel-manufacturing capability can readily be developed by even small c ountries which purchase CANDU reactors.

– N o need for enrichment

• N ote: althou g h natural uranium has been the fuel for CANDU since the beginning, the heavy-water moderator does not demand natural uranium.

• In fact, C ANDU is extremely f lexible - can burn enriched uranium, mixed-oxide (U/Pu) fuels, or even irradiated fuel from light-water reactors.

– DUPIC cycle

– T h cycle

– M A burning

• Latest CANDU design , the Advanced CANDU Reactor (ACR), will use slightly-enriched uranium.

Pressure Tube Design

• Canada did not have a heavy i ndustry capable of manufacturing a pressure vessel of the required size, so a contract was signed to purchase the vessel from the UK.

• H owever, the fathers of CANDU then started to be concerned about the size of the pressure vessel, not only for NPD, but even more so for the larger reactors that would follow . The pressure vessels would really have to become enormous .

• A s a result of these mis g ivin g s , the p ressure - vessel design for NPD was scrapped (with penalty to tear up contract for vessel).

• NPD was c h ange d to a pressure-tu b e d es i gn – the tubes would be the pressure boundary for

the hot coolant , the r eactor v essel (renamed a

calandria) would not be at pressure, and would be much simpler to manufacture.

• I n fact, it could be manufactured domestically, another important plus for Canada.

• N PD desi g ners , and those of all currentl y operating CANDUs, opted for horizontal pressure tubes.

• Thi s was i n t h e i nterest o f symmetry – t h ere would be no “preferred” direction for the coolant

flow , a s t here w ould b e i f t he pressure tubes

were vertical.

• With horizontal pressure tubes, the coolant could be made to flow in opposite directions in alternate channels, which would further enhance axial symmetry .

• V e r y im p ortant to note that what made the pressure tube concept viable was zirconium.

• T he large mass of metal in the pressure-tube design c ould absorb t oo many neutrons - definitely the case with steel pressure tubes: the fission chain reaction could not be made self- sustaining .

• Z irconium, “magic” nuclide with a very low neutron absorption cross section, came on the scene i n t i me.

• T his as the result of materials research in Chalk River f or the U S nuclear program .

• N ote: while the p ressure tubes are the p ressure boundary, they would tend to conduct heat from the fuel out into the moderator.

• In order to provide i nsulation f or the m oderator and prevent it from boiling in contact with the hot pressure tube, each pressure tube is surrounded by a concentric calandria tube of larger diameter.

• T he gap between pressure tube and calandria tu b e i s fill e d w i t h i nsu l at i ng gas (CO2) , a ll ow i ng it to operate at relatively low temperature (~ 70 oC ) .

• I n the p ressure-tube desi g n , the moderator and coolant are separated, in contrast to the situation in the pressure-vessel design. In principle, this allows the moderator and coolant to be different .

– A CR uses light water as a coolant

• In spite of this , all operating C ANDUs have

heavy water as the coolant. The idea for retaining heavy water as the coolant too is to max i m i ze th e neu t ron economy.

• CANDU - 6

– 380 channels

– ~ 7 m diameter

– ~ 6m long active core

– ~ 2 2 0 0 M W t h

– ~ 650 MWe

– ~ 30-32 % e ffi c i ency

– 10 MPa

– Outlet = 310 C

Online refuelling

• With pressure tubes, o n - power refuelling becomes possible - f uel channels can be “opened” individually and at full power to replace some of the fuel. On power refuelling was therefore adopted for CANDU.

• O n-power refuelling also means that “old” fuel is replaced by fresh fuel nearly continuously. Thus, very little excess reactivity is required . Batch refuelling would require a large excess reactivity at the s tart of each cycle ( as in LWR) .

• T he short CANDU fuel bundle facilitates on- p ower refuelling - c an then replace part of the fuel in a channel at each refuelling operation (

– 8-bundle shift refuellin g scheme in CAND U 6

– 2 -bundle shift refuelling scheme in ACR

– R efuel about 1-2 channels per day

• Also , horizontal c hannels s implify r efuelling – the

bundles need not be “tied” together. In Gentilly-1, with vertical channels, a central tie-rod was needed to hold the entire f uel - string together .

• H orizontal channels allow axial symmetry (no difference in coolant density between the 2 ends).

Advantages of online r efuelling

• Constant global power shape , with

localized “ripples” as channels are refuelled and go through their burnup cycle

• Constant in - core burnup

• C onstant shutdown-system effectiveness

• P ossibility of on-power removal of failed fuel, and therefore clean HTS

Safety Advantage

• Unpressurized c alandria - no risk of catastrophic vessel “break-up”

• Reactivity devices in unpressurized environment – no “rod ejection”

• L ow excess reac ti v it y – po t en ti a l f or reactivity addition small

• V ery long prompt-neutron lifetime

• R edundant , inde p endent safet y s y stems

• S e p a r a t i o n b e t w e e n c o n t r o l a n d s a f e t y systems

• L a r g e v o l u m e o f c o o l m o d e r a t o r “ water ” – excellent heat sink in hypothetical severe accidents

• Low fissile content in fuel – no criticality concern ou t s i d e t h e reac to r

Differences in Reactor-Core Design

CANDU

 Natural - uranium fuel

 Heavy-water moderator & coolant

 Pressure tubes; calandria not a pressure vessel

 Coolant p h y sicall y se p arate d

 Small/Simple fuel bundle

 On - power refuelling

 No boron/chemical reactor control in coolant system

PWR

 Enri ched - uranium f uel

 Light-water moderator/coolant

 Pressure vessel

 N o separat i on o f coo l ant f rom moderator

 Lar g e , more com p lex fuel assembly

 Batch (of f-power) refuelling

 Boron/chemical reactor control in coolant system

Safety Systems

• Shutdown System 1

– S pring loaded shutdown rod

• Shutdown System 2

– R apid injection of poison (Gd nitrate solution)

• E mergency Core Cooling System

• C ontainment

• Liquid z one controllers

– 21 water rods that are filled with light water

• 3 zones with 7 controllers

– L ight water acts as a poison in CANDU reactors

• P ermanent control rods are inserted

– 21 rods in total

• U sually stainless steel

• At Gentilly - 2 , source

t hey irradiated Co - 59 to produce t he Co - 60

– R adioactive tracer

– Cancer treatment

Disadvantages

• Positive void coefficient

• Low fuel burnup

– L arge amoun t o f was t e

• E asy to produce Pu239

– P roliferation issue

• H eav y wate r

– E xpensive to make

CANDU ACR

• ACR: Advanced CANDU Reactor

• H eavy water moderated

• Light water c ooled

– 12.5 MPa, Outlet T = 319 C

• Tighter lattice

– F rom 28.5 to 24

• Same core size as the C ANDU - 6

– E xcept higher power density

– 1000-1100 MWe vs 600-650 MWe

• CANFLEX bundle

– 43 fuel pins

– Slightly enriched uranium

– D yprosium in fuel

• Higher fuel burnup

• F latter power profile

• Reduction in void coefficient

– A lmost negative, recent studies indicate s li g h t l y pos i t iv e

Fast Reactors

• Initially developed for breeding

– U ncertainty on how much U-235 was present

– Fear that nuclear mi g ht not be sustainable

• B reeding: Convert fertile materials into fissile

– T h-232 to U-233

– U -238 to Pu-239

• F irst fast breeder reactor

– Clementine, built at LANL in 1946

– EBR-I, built in Idaho, first to produce electricity

Design considerations

• N o moderato r

– A void low mass materials

• H igh enrichments are required

– 10 - 30% either U - 235 or Pu - 239

• H exagonal lattices

– Reduces coolant-to-fuel ratio

• F uel can be either metal of ceramic

• C oolant is usually a liquid metal (some designs use gas)

– N a, NaK, Lead-Bismuth

– Low pressure, ~ 1MPa

Public domain image from wikipedia.

Designs

• Loop design

• Pool d e s i g n

• T hree loops

– P rimary

– Intermediate

– Steam Generator

• Breeding usually o c c u r s in a b l a n k e t o f U-238 that surrounds the core

Public domain image from wikipedia.

Fuels

• M etal

– F aster (Harder) neutron spectrum

• N o moderation from oxygen

– High t hermal c onductivity, w hich compensates f or low meting point

– Lots of swelling, which prevents high burnup

• Oxide

– Lots of experience at high burnup

– High melting point (2750 C)

– Poor Thermal Conductivity

• N itride and Carbide

– High t hermal c onductivity

Fuel lattice

• W ire wrap

• C ladding

– Z irconium is not suitable , does not behave well at high

temperatures Images removed due to copyright restrictions.

– Steel is usuall y preferred

Control Mechanisms

• Control r ods

– A bsorbing material such as boron carbide

– Much less effective than in thermal reactors

• R emoving fuel

– Method has been used in EBR-II, but it is not common

Need for fast r eactors

• Transmutation of MA

– Most MAs are fissile to fast neutrons

– Higher ratio of fission/absorption

• B reeding

• P ower Production

Problems w ith SFR

• C o s t

– M uch more expensive than LWRs

• P ositive void coefficient

• H igh Pu content

– P roliferation issue

• Hi g h MA con t en t

• C ore meltdown can lead to reactivity excursion

• R eprocess i ng

– C o s t

– Proliferation i ssue

Important Ratios

• Fissile Conversion Ratio ( CR)

Fissile material produced / Fissile material destroyed

• Breeding Ratio

– S ame as conversion ratio, but only when CR

> 1

• T RU Conversion Ratio

TRU P ro d uce d / TRU d es t roye d

• E quivalent to the absorption rate in U-238 divided by the t otal fission rate in TRU

Hi g h T empera t ure G as Reactors

Basic G a s Reactor D e s i g n

• Moderator – Graphite

• C oolant – C O 2 ; Heliu m; Molten Salt

• F uel – ( U,Th)O 2 ,C,or CO

• Vessel – steel or

prestressed concrete

• Cyc l e – In d ir ect Stea m o r Direct Brayton

Public domain image from Wikipedia.

New D esign P remise?

• C ommercial

– Magnox reactors in Britain (60s-90s); Gen I

– UNGG reactors in France (60s-90s); Gen I

– F ort St. Vrain in US (70’s-80’s)

– A GR in Britain (80s-Present); Gen II

• R esearch

– P ebb l e B ed in Ge rm a n y ()

– H TTR in Japan (1999-Present)

– HTR - 10 in China ( 2003 - Present)

F u e l P a r t i c l e s

• T R(istructural) ISO(tropic) coating

– P orous Carbon, IPyC, SiC, OPyC

– Crackin g Resistant be y ond 1600 o C

• Thermal stress

• Fission Product Buildup

– O .D. of .5-1.0 mm.

– F uel Kernel Encased

• E nrichment 7-15%

Public domain image from wikipedia.

F u e l Pebble T y p e

Fuel Element Design for PBMR

5mm Graphite Layer

Diameter 60 mm

Fuel Spher e

Half Section

Coated Particles Imbedded in Graphite Matrix

Pyr olytic Carbon

Silicon Carbite Barrier Coating Inner Pyr olytic Carbon

Por ous Carbon Buffer

Diameter 0,92mm

Coated Particle

Diameter 0,5mm Uranium Dioxide

Fuel

Image by MIT OpenCourseWare.

Comparison

• P rismatic

– P ros

• O perating/Fabrication Experience (Ft. St. Vrain)

• Coolant flow and fuel positions well known - > more accurate modelling

• C ontrol rod p lacement easie r

– C ons

• High excess reactivity

• H ot spots not mobile

• N eed periodic shut down for refueling

• W ater ingress -> strong reactivity increase

Comparison

• P ebble Bed

– Pros

• C an keep excess reactivity to minimum

• More ef fective f uel utilization

• F ew S.D.’ s required

• E nrichment lowe r

• P eak fuel temperature lower

– C ons

• Dif ficult to calculate flow and temperature

• C arbon dust production (3 kg/year)

• C omplications due to uncertainty in fuel position

Factors i n N eutronics

• Block and/or pebbled bed packing fraction

• R eactivity effects of working fluid (He)

• D epen d ence o f k on t empera t ure, enrichment, core geometry

• W ater ingress effects on reactivity

• T e m p erature coefficients of reactivit y

• F uel burnup effect on k over time

Safety Considerations

• Events leading to Reactivity Insertion

– Water Ingress, Control Rod ejection, Repacking of fuel pebbles

• A ir Ingress coupled with Carbon dust formation

• D ecay Heat Removal (due to FPs)

Air Ingress Experiments

• O pen Air Chimney T e st Results; He to Air at 850 o C

Images removed due to copyright restrictions.

HTGR vs. Other Reactors

• G raphite moderator well suited for transient scenarios (high thermal inertia)

• P rimary coolant less radioactive

• Very high burnup possible (to 200 GWd/t)

• A pplications for high outlet temperatures

– Oil E x t rac ti on f rom Sh a l e an d Oil S an d s

– H ydrogen Production, Coal Gasification

– D esa li n i zat i on

• Lower Power Density (~1/30 of PWR)

• P roliferation Resistant

MIT OpenCourseWare http://ocw.mit.edu

22.05 Neutron Science and Reactor Physics

Fall 20 09

For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms .