Chapter4

llfbtlme offirstwalls
analyses
JournalofNuclear
Materials
85 & 86 (1979)135-145
0 North-Holland
Publishing
Company

A REVIEW OF LIFETIME ANALYSES FOR TOWS*

S. D. HARKNESS(1)and B. CRAMER(')

(1) Argonne National Laboratory,Argonne, Illinois

(2) McDonnell Douglas AstronauticsCompany - East, St. Louis, Missouri

System studies have shown that economic fusion power can best be achieved from the use of long
lived components. The stresses generated in a first wall module are a complex function of its
geometry, the chosen structuralmaterial and the tokamak burn cycle characteristics. A means of
applying ASME Code Case 1592 to preliminarydesign has been established. Methods of incorporating
some of the material property changes expected from irradiationare discussed. Cyclic stresses
imposed by tokamak operationare expected to cause fatigue related propertiesto govern the life
of the structure. Stress assistedbubble growth is also discussed. This may be the critical
mechanism in establishingthe creep rupture life of a fusion first wall component.

1. INTRODUCTION

To be a viable part of the nation's energy

supply system, fusion based reactorswill have
to prove the-elves both as reliable and as eco-
h Pw=NEUTAON WALL LOAD ~MWlm21
I

nomical alternativeenergy sources. A very

important factor in whether or not fusion re-
actors ever become importantproducers of elec-
tricity is the expected lifetime of key reactor
components. This paper reviews the status and
results of the lifetime analysis work performed,
to date, for tokamaks.
System study [1,2] results are useful for
putting into perspectivethe importanceof
achieving long lived designs. Figure 1 compares
the operating costs as a function of neutron
wall loading and wall lifetime assuming the
physical and mechanical propertiesof an ad-
vanced structuralmaterial such as vanadium.
The achievementof relativelyhigh 2-3 MU/m2
wall loadings?has been increasinglywidely
recognized [3] as a key means of reducing the
size and, in turn, the capital cost of a toka- 5 IO 15 20 25 30

mak-based reactor. WALL LIFE,YW~yrlm2

The main reason for the economic incentive

shown in Fig. 1 for longer lived designs is the
long time required for rebuildingof a tokamak
system. To date, no one has been able to
identify a means of in-situ repair so that the
tedious and laborious disassemblyof large,
welded, radioactivecomponentsmust be en-
visioned. A length of 80 days to rebuild a
tokamak system was used in analysis that led
to Fig. 1. Viewed in another way, the lost
revenue from one days operation of a 1000 MWe
plant where electricityis sold at five cents
RWh is 1.2 million dollars.
This work is supportedby the U.S. Department
of Energy.
All wall loadings noted in this report are
in terms of the neutron energy and in units
of MW/m2 unless otherwise noted.
Fig. 1. Effect of neutron wall loading and
structuralmaterial lifetime on the cost of
energy in tokamaks.
The effect of planned maintenance,alone, is
sufficientto justify (indeed dictate) the
developmentof a long lived design. However,
there is additional incentivefor such a de-
velopment from a considerationof the relation-
ship between expected componentlifetime and the
occurrenceof unscheduledshutdowns. Failurb
is by its very nature a statisticalphenomenon.
It is impossibleto say preciselywhen a compo-
nent will fail, as one is restrictedto a
statementof the sort that there is a 95%
probabilitythat failure will not occur before

135
136 S.D.Harkness and 3. Cramer i Review of lifetime analyses for tokamaks
a certain point in time. The number of failures
to be expected in the system will be the product
of the number of componentstimes the proba-
bility of failure of one componentand may have
to include an environmentalweighting factor.
Most fusion reactor system designs involve
hundreds or even thousandsof first wall modules.
For a short lived design the number of projected
component failureswould be much higher than
for a long lived componentover the lifetime
of the reactor plant.
Thus considerationsof the economicsof fusion
power indicate the incentiveto achieve an ex-
pected componentlifetime equal to that of the
reactor plant or ti30 years, To be useful, this
life must be achieved for wall loadings of from
2-3 MW/m2, thereby creatingan overall goal of
as high as 90 MW-yrfm2.
2. LIFETIME ESTIMATES
Lifetime estimatesare dependenton criteria
developed in three categories. The first of
these includes those processes that actually
lead to failure of the component,either
through leakage of the coolant or by a com-
plete failure of an element. Creep-fatigue
and flaw propagationwould fall into this
category.
The second category involves limits on di-
mensional change that would cause some planned
operation to become impossible. Swelling and
creep are the main considerationsin this cate-
gory. Setting limits for allowabledimensional
change is design dependentand can only be done
after all planned maintenanceoperationsare
well established. A more stringentrestriction
on the amount of swellingmay well result from
an analysis of the effect of the stresses
inducedby the swelling gradients.
The third category compriseslimits on de-
formationthat experiencehas shown are im-
portant to prevent the structurefrom failing
in an unspecifiedmode.
To analyze the performanceof a reactor module
in terms of these three categories,one must
first characterizethe environmentand determine
the expected stress and thermal histories.
3. THEE-HYD~ULIC CONSIDERATIONS
Any estimate of the lifetime of a tokamak
first wall module must be based on a complete
thermal-hydraulic analysis,such as is included
in references 141 and [5]. For the purposes
of this review, it is worth noting that certain
fundamentallimitationson performanceresult
directly from the selectionof a structural
material and a wall thickness. The maximum
temperatureof the first wall module can be no
lower than the coolant temperatureplus the
quantityWst/k where W, is the surface heat flux,
t is the wall thickness,and k is the therms1
conductivityof the wall. The actual wall
temperaturewill, of course, be higher since
the convectiveheat transferbetween the wall
and the coolant stream will never be perfect.
If one assumes as a basis of comparisona
coolant stream temperatureof 225*C, the re-
lationshipbetween temperatureand wall thick-
ness for three candidatematerials are as pre-
sented in Table 1. While each of these three
candidates,Type 316 stainlesssteel, a stabi-
lized 9 Cr ferritic stainless,and a titanium
allay (Ti-6Al-4V)have an upper temperature
limit of about 5OO*C in a fusion reactor environ-
ment, It can be seen that widely differentwall
loading limits result strictly from the radial
temperaturegradient. Based on this alone,
titanium alloys are restrictedto service at
wall loadingsbelow % 4 MW/m2 if a wall 3 mm or
greater in thicknessmust be used.
Table 1. Relative TemperatureLimitationsof
Different StructuralAlloys.
AT Across
1 cm wall,
First Wall
Alloy I l. Mw$i?, OC
$16 108
192
ri-6Al-4V
II?erritic,9-Cr 86
L
* Assumes coolant inlet temperatureof 225'C.
The temperaturegradientsdevelopedwithin
the wall result in secondary thermal and radia-
tion swelling stresses. The magnitude of these
stresseswill be consideredIn the next section.
4. SOURCES OF STRESS
The results of an analysis for one tokamak
wall design is presented in Fig. 2 to illustrate
the changing stress pattern that must be ex-
pected with time. Primary stresses In a tokamak
system result from the coolant pressure. Mag-
netic forces also result in primary stresses
if a magnetic structuralmaterial such as
ferritfc stainlesssteel is used. The coolant
pressures range from atmosphericfor a falling
bed concept to as high as 13.7 MFa for some
water cooled designs. Representativepressures
are shown in Table 2 for each of 5 coolant
choices as developedby recent blanket studies
161. The magnetic force that results from the
use of a ferromagneticstructuralmaterial can
 S.D. Harkness and B. Cramer/Review of lifetime analyses for tokamaks 137

wall surface facing the plasma and a corres-

ponding tensile stress to be generatedon the
coolant side of the wall. The initial level of
% these secondary stresses depends on the material
i -100 selection,the wall loading, and the wall thick-
k ness. Representativevalues for several ma-
:: terials are shown in Table 3.

1 -200
0 10 20 30 40

Table 3. Initial Radial Thermal Stresses Gener-

ated in a 3 mm Thick First Wall

Yield
Thermal Stress at
Strength
Structural at 500°C Wall Loading MWW/m2
Material MPa (psi) 1 3 1 5
Fig. 2. Longitudinalcompressionstress at
outer wall surface during burn period. t

Table 2. Nominal Coolant Pressures ;9Cr stabi- 310

lized (45,000)
ferritic

Typical iTitanium 480

Pressure (69,600)
Coolant PfPa(psi)

Helium 5.2 (750) Figure 3, as taken from the work of Mukherjee,

Lithium 0.35 (50) et al. [9], illustrate5the importanceof in-
cluding the anelastic component in the analysis
Molten Salt 1.4 (200) as it results in a recoverablereduction in
Water 13.7 (2000) stress of about 15% in the example case of an
austeniticstainless steel structure.
Falling Bed 0.14 (20)

I I I I I I I I I
*
2.2
MY/m.
TM,=503
'C,316
STAINLESS STEEL
- 30
also result in significantstresses. Recent 200 -
--- THERMOELASTIC YOOEL
calculations[7] for a 9-Cr stabilized ferritic L
s -
_--_ THEAYOELASTIC YODEL INCLUDINB ANELASTIC 2
stainless steel structure indicate pressuresof EFFECT AN0 STRESS RELAXATION
2,1 atm will be imposed if this material is i d
used.
A possible source of cyclic primary stresses
are pressure fluctuationsdue to turbulencein
the coolant. These could contributeto fatigue
failure in first wall structures. In addition,
plasma disruptionsmay cause electricalcurrents
to flow in first wall structuresresulting in
internal forces. These must be combinedwith
other primary loads in determiningstructural
thicknesses. These dynamic forces and their
resultingeffects on stress have been shown to I I I I I I I I I
have a significanteffect on structuraldesign 0 IO 20 30 40 SO 60 70 60 SO 100
and on resulting component life [8]. Because TIME. I
these disruptionsmay occur frequently (assumed
once each 10 cycles in Ref. [El), they may have
to be considereda normal loading condition.
Fig. 3. A calculationof the thermal stresses
In addition to these primary stresses, at the at the plasma side of a 3 aaathick, hemis-
onset of operation the incident electromagnetic phericallycapped first wall module during
and particle fluxes that comprise Q 25% of the a 60 second burn cycle with a 15 second dwell
total fusion energy cause large compressive period.
thermal stresses to be generated in the first
138 S.D. Harkness and B. Cramer /Review of lifetime analyses for tokamaks

This stress is further relaxed by thermal and

radiationenhanced processes that result in
permanent deformationof the structureso that
eventuallythe thermallyinduced secondary 5~s.
2.33 MW/m2 MAX TEHP.BOO% I-MN 8URN
._1
do ,
stresses are low during the reactor burn and 5
approach the thermoelasticstress less the an- RELAXATION DUE TO THERMAL CREEP
elastic recovery during the dwell portion of the RELAXATION DUE TO RADIATION -
burn cycle. Since it is this stress that has
been found to govern the fatigue and flow propa-
gation estimates,the importanceof the inclusion
of the anelastic recovery is evident since crack
growth rates are a function of the stress level
raised to an exponent ranging from 2 to 4. Thus
the reductionof the stress by 15 percent could
result in a factor of 2 reduction in the crack / I ‘c
0
growth rates. lo-’ 100 10’ I02 IO3 IO’ Id IO4
CYCLES
It should be noted that radiationenhanced
stress relaxationis particularlyeffectivein
reducing the level of stress nresent during the Fig. 4. The expected relaxationof thermal
burn due to both its amplificationof the pro- stresses.
cess and its linear dependenceon stress. This
latter characteristicallows the relaxationpro-
cess to proceed to much lower stresses than
would be reached by the much more highly stress
dependent thermal stress relaxationprocess
(Fig. 4). Radiation enhanced stress relaxation
eventuallycauses the plasma side of the wall
to be under rather high tensile stress during
the dwell period which may have importantim-
plicationsfor the fatigue propertiesas surface
damage due to blisteringand sputteringmay aid
both crack initiationand propagationunder
tensile loading. A very importantaspect of
the stress relaxationis that it tends to mini-
mize the importanceof creep rupture as a life
limiting property since the high thermal
stresses are only imposed for the relatively
short periods between burns and at times when
wall temperaturesare lower and the neutron flux
is absent. This has the effect of translating
the tokamak first wall problem into one where
fatigue and flaw propagationconsiderationsare
of increased importance.
In addition to the thermallyinduced secondary
stresses,another importantsource of secondary
stress is differentialswelling. As shown in
Fig. 2, for a swellingmaterial such as Type
316 stainlesssteel, gradientsin void volume
within the material.result in rather high
stresseswithin the material at long times. Fig. 5. A calculationof the stresses generated
The gradientsin swelling can result in in a D-bend mDdule by differentialswelling in
rather complex stress distributionsin a struc- Type 316 stainless steel.
ture constructedof a material such as Type
316 stainlesssteel. An elastic analysis is histories indicate the relaxationof com-
presentedas Fig. 5 to dramatize this point. pressive stresseswith time due to creep and
subsequentincrease in compressivestresses due
Design dependentstress distributionswill to swelling. This relaxationof compressive
also result from coolant pressuresand tempera- stresses results in residual tensile stresses
ture distributions. Stress histories in the during the "non burn" portion of the cycle.
first wall structures,includingeffects of These stress historieswere calculatedfor a
irradiationswelling and creep, have been calcu- uniaxial element in the first wall. They are
lated on an inelasticbasis in references [9] shown for various levels of neutron wall loading
and [lo]. Results of the analyses from refer- and include a temperatureeffect (temperature
ence [9] are shown in Fig. 2. These stress
 S.D. Harkness and 3. Cramer / Review of fifetimeanalyses for tokamaks 139

increaseswith increasingwall loading) as well

as a time dependent irradiationeffect. The
results shown in Figs. 2 and 5 provide an indi-
cation of the complexityof first wall stresses
due to their time dependent nature, cyclic na-
ture, and the redistributionof stresseswhich
will occur.
In-depth analysis of a near term doublet
shaped plasma chamber has been reported [S, 11,
12 and 131. Because of the rather short opera-
ting life ("70 days) of this machine, unirra-
diated material propertieswere used. However,
a detailed analysis of thermal stresses and
plasma disruption stresseswas included in pre-
dicting wall life. A resulting cyclic stress
history for an Inconel 625 wall is shown in
Fig. 6. Results included effects of.plasticity
to determine thermal stress distributionsand
optimum wall section geometriesas well as a dy-
namic analysis to determine response of the Fig. 7. Distributionof stresses in an unre-
wall to plasma disruptions. Cyclic stresses strained hemisphericalstainlesssteel wall as
due to both effects have a significanteffect a functionof the semiangle,$. The wall thick-
on first wall life. ness is 3 mm.

use of restraint in system design must be

approachedcarefullywhen dealing with struc-
tural materials that are dimensionallyunstable
under irradiation. For example, Wolfer and
Watson [14] found it was necessary to avoid
all restraintwhen designing a graphite first
wall structuredue to the stresses that would
otherwisebe generated in this dimensionally
unstable material.
A number of ways to minimize first-wall
temperaturegradientsdue to the electromagnetic
and particle flux from the plasma have been de-
vised. These generally take two forms. One is
by reducing the fraction of energy that is de-
posited on the wall through the use of a divertor
to interceptthe particle transport. The other
is to spread out the time that energy is radiated
to the wall so that some heating continues
during the dwell period by means of a liner
or limiters.
The effects of a magnetic divertor on the
Fig. 6. Stress levels expected during the opera- temperatureand temperaturegradients generated
tion of a tokamak device. Note the possible in the first wall of a tokamak have been
"ringing"effect that could occur for certain analyzed by B. Misra and V. Maroni (41. Their
designs as the result of a plasma disruption. results reflect the fact that over half of the
heat flux incident on the first wall surface
is due to parti&@ transport. Therefore,the
5. DESIGN APPROACHESTO REDUCE THE LEVEL OF use of a divertor effectivelyallows a factor
INDUCED STRESSES of two higher wall loadings to be imposed on
the system for a given allowable thermal stress
The designer has at his disposal several level.
methods of reducing the level of thermal The applicationof a liner as a protective
stresses imposed on the first wall of a tokamak. first wall device has been analyzed by Kearney
One of the most importantconsiderat,ions et al. [51. A liner is usually included in
deals with the use of restraintas a means of tokamak designs to protect the first wall from
limiting bending stresses. For example, If plasma disruptionsthat result in a sudden de-
the hemisphericallycapped first wall module position of the plasma's energy onto the first
were not rigidly clamped, then the thermal wall. This device, which is usually conceived
stresseswould be as shown in Fig. 7. Clamping to be fabricatedfrom a coated graphite or
the module eliminatesthe bending stresses. The silicon carbide, has the additionalbenefit of
140 S.D. Darkness and 3. Cramer /Review of lifetime anaiyses for tokamaks
flatteningthe rate of energy depositiononto
the first wall during normal operation. Figure
8 compares the expected first wall surface
temperaturefor the case when a liner completely
covers the inside of the tokamak,when re-
fractory limiters comprise 10% of the surface
area, with that for a bare wall as calculated
by Smith and Krazinski [15]. The use of a
liner was calculated [5] to increase the crack
growth limited lifetime from 2.6 to 89 years
in a separate study of a tokamak based reactor.
460
I I I I
440 -
Pw= I MWd
Y Type 316 stainlesssteel are shown in Fig. 9
6420 - BARE WALL
0 IO 20 30 40
CYCLE TIME,
Fig. 8. *The first wall surface temperatureas
a functionof time for a 60 second burn pulse
and a 15 second dwell period,
It should be noted that after stress relaxa-
tion has worked to limit the presence of the
thermallyinduced stresses to the downtime
between burns, ingeniousmethods to generate
energy in the blanket such as developedfor the
Numak [3] reactor might serve to aggrevate the
problem. The Numak approach uses the solidifi-
cation of a Pb-Li eutectic during the dwell
period to even the temperatureswing seen by
the turbine. This could serve to also raise
the temperatureof the coolant side of the
first wall, thereby causing an increase in the
stress levels already present since stress re-
laxation during the burn has caused the coolant
side to be under compressionduring the dwell
part of the cycle.
6. LIFRTIME CRITERIA
In developinga method for predictingthe
lifetimeof a fusion reactor first wall module,
the logical foundationto build upon is the
ASME Code Case 1592 for Class 1 Componentsin
ElevatedTemperatureService [16]. C. K.
Youngdahl and D. L. Smith [17] have developed
this approach for a fusion reactor first wall
in such a way as to relate the maximum allow-
able surface heat flux as a functionof the
average wall temperaturefor a number of candl-
date structuralalloys.
These authors suggest a two step process.
In the first step the structure is analyzed
for those processesthat result in actual
failure of the component such as fatigue, stress-
to-ruptureand excessiveprimary loading. Based
on the use of Code Case propertiesfor the un-
irradiatedmaterialswhere they existed, and
estimatesbased on availableexperimentaldata
when they didn't, the authors were able to pro-
ject good performancein this first step for all
materials from annealed Type 316 stainlesssteel
(the worst performer)to vanadium alloys (the
I I I best) and still comply with the Code Case re-
quirementson tensile, stress-to-rupture and
fatigue properties. These authors' results for
along with an additionalplot of the maximum
wall loading possible at a given wall tempera-
ture based on the simplifiedanalysis presented
in the thermal hydraulicssection of this paper.
It can be seen that these temperaturelimits re-
duce the attractivenessof low temperature sys-
tems.
50 60 70 80
s
2.0
t- \ / -+.o
0-0
400 450 500 550 600 650
1 'C
O"9'
Fig. 9. The maximum allowable thermalwall
loading as a functionof the temperatureof a
Type 316 stainlesssteel first wall as defined
by fatigue, creep rupture and primary loading
considerations.
Figure 9 was developedby first establishing
the allowablemodule wall thicknessfrom Sm t
the primary membrane stress limit through the
relation 8 = PR/S, t. S, t is the lesser of
the time independe& tensile limits and the time
dependentstress-to-rupture related properties.
P is the coolant pressure and R is module radius
 SD. Harkness and B. Cramer / Review of lifetime analyses for Tokamaks
in the above relationship. This value of wall
thicknesswas then used in a fatigue based
equation
u&W
Et (0 =- 2k(l-w)
where Q(L) is the code given allowable strain
to achieve the desired lifetime,W is the ther-
mal wall loading (equals* 25% of the neutron
wall loading),a is the thermal expansionco-
efficient,k is the thermal conductivity,v is
Poisson's ratio, and 6 is the wall thickness.
In the second step of the analysis, deform&ion
limits on thermal ratchetingand thermal creep
were examined based on sn approach for analyzing
the effects of ratchetingin a thermallycycled
system developedby Bree [Ml. When Youngdahl
and Smith extend their analysis to cover
ratcheting,the allowable design point drops
from 1.3 to 0.9 HWfm2 (i.e. from point A to
point B in Fig. 10) where the fatigue curve
plotted in Fig. 10 is taken from Fig. 9. Again
the results of includinga temperaturelimi-
tation reduces the attractivenessof using a
low structuraltemperaturedesign.
Wall Thickness, 6, mm
Fig. 10. An adaptationof the Bree Diagram
for a stainlesssteel module operated at 482'C.
This promising approach to design stops short
of including the propagationof existent flaws
in the analysis. In one study [191 flaw propa-
gation was found to be the life limiting proper-
ty. This analysis is dependenton both the
assumptionsmade for initial flaw sizes and the
applicabilityof the available data which is
for unirradiatedmaterial tested in air.
Within these constraintsthe crack growth
propertiesnecessary to achieve a given life-
time can be calculated. For Type 316 stainless
steel the assumptionof the presence of an
0.124 cm flaw depth, a flaw aspect ratio of 0.5,
141
in a wall of 0.25 cm thicknessresulted in the
informationpresentedas Fig. 11 [19]. For
these conditions,a significantimprovementin
the crack propagationpropertiesare necessary
just to achieve a 10 MW-yr/m2 life at a wall
loading of 2 HW/m2. Based on this calculation,
only low stress, low temperaturecombinations
would allow the achievementof the minimal 10
MW-yr/m2.
In a study of an Inconel 718 first wall
module, Kearney, et al. [5] found that wall
loadingswere limited to less than Q 2.2
MW/m2 from their impositionof a 6OO'C maxi-
mum temperature. For a neutron wall loading
of 2 HWfm2, it was found essential to detect
all flaws larger than 0.75 mm in depth in
order to achieve acceptablelifetimes (> 10
MW-yr/m2) for their 5 rm~thick structure.
u
7
@
I.1
-#
lo- .s -
w- .4_
-5 -
-6-
A&
rtr
AK--psIJIR
Fig. 11, Required flaw growth rate properties
dependingon stress levels to give a 10 MW-yr/m2
lifetimebased on the assumptionof a 90 minute
plasma burn.
The question of what is the appropriate
initial flaw size to choose is dependent upon
flaw detection capability. Fusion devices will,
in all probability,be field erected and consist
of joining massive components. While it is
true that techniquesare currently in practice
to detect scratchesof 0.025 mm in thin walled
tubing, such schemes may not be applicableto
fusion first wall modules. The benefits to
be gained in going to better nondestructive
 142 S.D. Harkness and B. Cramer /Review of lifetime analyses for Tokamaks
testing systems are shown in Fig. 12 for a wall
thickness of 0.25 cm. If a scratch depth of
only 0.025 cm (10% of the wall thickness) is
assumed, a substantial improvement is still re-
quired in crack propagation properties if eco-
nomically interesting wall lifetimes are to be
achieved.
a
&Gq t
u material whose ductility {as dropped below 5%
AK - psi'_Ji;;
Fig. 12. Effect of initial flaw size on re-
quired flaw rowth rate properties to achieve
a 10 MW-yr/m !! life.
From this discussion, it is evident that
flaw propagation must be included in any life-
time analysis of a fusion device. This should
be done after the structure has been found
acceptable in an analysis such as outlined by
Youngdahl and Smith.
With this analysis framework in mind, the next
step is to extrapolate the available data to
conditions presented by the fusion environment.
7. EXTRAPOLATION OF DATA
The high energy neutron bombardment character-
istic of a fusion device will lead to major
microstructural changes [20,21]. These effects
are best documented for the austenitic stainless
steels where swelling and ductility loss are
perhaps the most important property changes
with fluence. An important paper 1221 has
recently been presented that deals with the
difficult problem of how to incorporate the
ductility loss into a design approach. In
the referenced approach, each of the ASME 1592
Code Case requirements were addressed from the
viewpoint of what additional conservatisms
should be included when designing with a low
(where the definition of low changes with the
requirement) ductility material. A major
assumption made in this work is that swelling
and irradiation creep do not inherently damage
the material.
For example, in setting the criterion for
the primary membrane stress intensity for un-
irradiated materials, the level must be kept
below the yield stress, S , while for irradiated
(or as can be shown to be equivalent, whose
yield strength to ultimate strength ratio has
exceeded 0.6), the allowable level becomes 0.6
of the ultimate strength.
When including the effect of ductility loss on
the use of the Bree diagram to protect against
ratcheting, the authors conclude that it must
be amended when the uniform elongation is re-
duced to approximately 1% for an austenitic
stainless steel. At this point the work of
Leckie and Ponter [23] is used to modify the
boundaries in the Bree diagram. These authors
have shown that ratcheting strains will be
small when creep is significant if loads are
kept less than n/n+1 of the elastic-plastic
shakedown loads where n is the exponent in
the creep rate relation E, = AU". If the thermal
creep expression is used, then n's of 4-6 are
appropriate and relatively small movements of
the Bree diagram boundaries result. If the ex-
pression for radiation enhanced creep is used,
then an n value near 1 should be used and much
larger changes in the Bree boundaries are re-
quired. The use of any thermal creep related
deformation laws for the prediction of per-
formance of a component during neutron radiation
may at first seem questionable since, in
the temperature range of interest, the time de-
pendent deformation mechanisms for the matrix
of the grains of the structural alloy are not
those that operate during thermal creep. In
other words, cFeep during irradiation is not a
superposition of a radiation enhanced process
and a thermal one. However, the importance of
including a limit on rhtcheting has more to do
with avoiding conditions where cracks are ini-
tiated within the material than on the absolute
amount of deformation that occurs. Therefore,
what one would really like to know is the amount
of grain boundary sliding that occurs for some
given amount of total strain. If one assumes
that the grain boundary sliding process is
little affected by radiation, then the calcu-
lation of a nonexistent amount of thermal creep
strain may serve as an adequate estimate of the
level of the grain boundary sliding (some small
fraction of the actual calculated creep strain)
that can be incurred without crack initiation.
For this reason, the suggestion by Nelson,
et al., to use a total strain limit that is
based on a summation of the ratio of the plastic
strain, 6~p, to the plastic strain limit, E:,
 S.D. Harkness and B. Cramer i Review of lifetime analyses for Tokamaks 143

and the ratio of the thermal creep strain 6~’

to the creep strain limit cc where the limits,
eL and E; are in turn allowkd to vary with
fluence and temperature appears valid. That is
URIRRADIATED

y -C 8 (where S < 1).

E.

This approach replaces the fixed 1% limit on

strain imposed by the ASME Code Case in a
6@?6ii& ZONE IN
manner that is more in keeping with the intent IRRADIATED SPECIMEN
of minimizing the chances of crack formation
in a material whose ductility changes as a
function of time. In a similar way, the
authors recommend a linear summation of the
fraction of rupture life and fatigue life
where each of these is again a function of
fluence as well as temperature and stress.
While the linear summation aspect [24] of ENLARGED VIEW OF
the approach is open to debate, it does serve MICROCRACK FORMED

to emphasize two properties of special im- FROM LINK-UP OF

portance; the fatigue and creep lifetimes ex- SHEARED VOIDS

pected under tokamak reactor conditions.

Some experimental information has been obtained

on the effect of neutron irradiation in the void Fig. 13. Possible changes in flaw growth be-
forming temperature range on fatigue lifetimes. havior in irradiated materials.
Michel and Korth (25lhave suggested an adaptatiol
of the Universal Slopes Equation to effectively mechanism dependent on the formation of voids
use the increases in yield strength and reduction These would be
ahead of the fatigue crack tip.
in area parameters to adjust for the effect8 of expected to form at much different rates in the
radiation on the fatigue properties. This presence of a highly mobile self-interstitial
approach results in a prediction of an improve-
population whose migration will be a sensitive
ment in properties under high cycle conditions
function of the stress state present around the
with a reduction at low cycles to failure. The
crack tip.
Universal Slopes Approach has the inherent prob-
lem of choosing the proper ductility parameter Besides affecting the propagation rate of a
to use. For example, if the reduction in area fatigue crack, irradiation will also be ex-
of an irradiated sample is measured at low mag- pected to reduce the fracture toughness proper-
nifications, very low values are found while at ties. Several studies have evaluated the growth
higher magnification examination of the same of existing flaws to a coolant leak condition.
sample will reveal relatively high local re- Failure by coolant leakage has generally been
ductions due to its tendency to deform by dis- considered to be more critical than fracture.
location channeling, It is not evident which This is due primarily to the relatively thin
of the values of reduction in area is appro- gauge8 of the first wall concepts evaluated so
priate to use. that the critical crack size for fracture is
larger than the material thickness. These
Furthermore, this difficulty relate8 to the
analyses have generally been based on linear
larger issue that it may be unreasonable to ex-
elastic fracture mechanics which requires
pect an empirically derived relationship to be
appropriate fracture toughness data. Because
valid when the basic deformation mechanism8
first wall gauges are relatively thin, stable
change. A8 discussed by Grossbeck, [20] the
flaw growth with significant plastic deformation
clustering of point defects into voids and
at the crack tip might be expected to occur
dislocation loops leads to a microstructure in
prior to the onset of crack instability. How-
which dislocation channeling can occur. In this
event the nature of the plastic zone advancing ever, the embrittlement which occurs due to
in front of a fatigue crack would be expected to the irradiation environments is expected to
change dramatically as shown in Fig. 13. Addi- make brittle fracture from small crack8 more
t ionally ) the shearing of voids might be ex- probable. Therefore, the linear elastic frac-
ture mechanics analysis and associated plane
pected to cause branching cracks to form in
front of the main fatigue crack. strain fracture toughness properties are con-
Other
mechanisms proposed for fatigue crack propa- sidered relevant to the first wall structures.
gation would also be expected to change under Fracture-toughness Is expected to be severely
irradiation. For example, as reviewed by degraded by irradiation. Hawthorne and Watson
Weereman. [26] Yokubari has suggested a [27] have studied fracture toughness trenda in
144 S.D. Ilarkness and 6. Cramer /Review
AISI 300 series stainless steels welds. They ob-
served large decreases in postirradiation frac-
ture toughness in 308 stainless steel welds,
based on fatigue precracked Charpy-V specimens
and J-Integral assessment procedures. For
EBR-II radiation exposures at temperatures
above 370°C, fracture toughness values were
reduced to as little as 20% of unirradiated
values. Reductions of this magnitude will have
an extremely detrimental effect on fusion reac-
tor first wall design and life.
Since at present it is impossible to predict
on theoretical grounds how these different
effects will combine, it is imperative that
in-pile fatigue, crack growth, and fracture
toughness data be obtained at once on candidate
structural materials. Until that time, the
designer has little option but to use a Uni-
versal Slopes type approach to the prediction
of fatigue life.
The creep rupture lifetime of a metal ex-
posed to a fusion neutron spectrum may well be
determined by the stress assisted growth of
helium filled gas bubbles at the grain bound-
aries. The high rate of helium generation
caused by the energetic fusion spectra will
make this mechanism much more important than
in fission applications. Experimental veri-
fication of the potential importance of this
mechanism has been demonstrated by Grossbeck,
et al. [21] who were able to show good corre-
lation between a calculated lifetime based
on a Hull-Rimmer [28] type calculation of
bubble growth rate and the experimentally
measured rupture time.
An analysis of this mechanism shows that it is
linear in stress and reduced by a decrease in
grain size as well as an increase in the frac-
tion of helium that can be trapped in the
matrices of grains. Much more attention needs
to be given to this gas bubble growth mechanism
which heretofore has been mainly of academic
Interest.
0. CONCLUSIONS
The major points emerging from this review
include:
1. Long lived first wall structures are
imperative for the achievement of economic
fusion power.
2. The stress state in a tokamak first wall
component is a complex function of the
module geometry, the choice of structural
material and the irradiation conditions.
3. Stresses can be reduced by the use of
selected materials, such as the ferritic
steels, and the incorporation of design
features such as divertors and liners.
4. A framework for design of fusion reactor
first walls has been developed based
of liferime analyses for Tokamah
on the ASME Code Case 1592 due primarily
to the work of Youngdahl and Smith [17]
and to Nelson and others [22].
5. Fatigue and stress-to-rupture properties
may be critical to the achievement of long
lives by tokamak first wall modules.
6. When the propagation of undetected flaws
is included, much shorter lives result
than predicted by the ASME based
approach.
7. Effects of radiation on crack propagation
and fracture toughness need to be much
better understood through the gathering
of in-reactor experimental information.
a. Stress-assisted bubble growth may have a
significant effect on the creep rupture
life in a fusion environment. In-reactor
studies are needed under high helium
generation conditions to verify the lm-
portance of this mechanism.
ACKNOWLEDGEMENTS
The authors particularly benefited from
discussions with C. Youngdahl and D. L. Smith
of ANL and with C. Y. Li and S. Muhkerjee of
Cornell University. The typing and editing
of the report by Ms. Carolyn Poore is grate-
fully acknowledged.
REFERENCES
1. M. A. Abdou, et al., "Impact of Technology
Considerations on Economic Aspects of
Tokamak Power Plants," Presented at the
IAEA International Conference on Plasma
Physics and Controlled Nuclear Fusion
Research, Innsbruck, Austria (August 23-30,
1978).
2. G. M. Fuller, "Fusion Reactor First Wall
Systems Analysis," McDonnell Douglas Astro-
nautics Company, F.R-582 (November, 1977).
3. R. W. Corm. G. L. Kulcinski, and C. W.
Maynard, "NUMAK: An Attractive Medium Field,
Medium Size. Conceptual Tokamak Reactor,"
University of Wisconsin. UWFDM-249 (May,
1978).
4. B. Misra, H. Stevens and V. Haroni, "Thermal
Hydraulic and Power Cycle Analysis of Liquid
Lithium Blanket Designs," ASMB Heat Transfer
Conference, Salt Lake City, Utah, (August
15-17, 1977).
5. D. W. Kearney. et al., "Mechanical and Ther-
mal Design of a Gas-Cooled Fusion Blanket
Module," General Atomic Co., GA-Al4671
(September, 1977).
6. D. L. Smith, et al., "Fusion Reactor Blanket/
Shield Design Study," McDonnell Douglas Astro-
nautics Co., and Argonne National Laboratory,
ANL/FPP-79-l (1979).
7. L. R. Turner, "Calculation of Magnetic Forces
on a Ferritic Stainless Steel First Wall,
Argonne National Laboratory (1979).
 S.D. Harknessand B. Cramer / Review of lifetime analyses for Tokamaks 145

a. PCFR Plasma Chamber Design Report to 21. M. J. Grossbeck, J. 0. Stiegler, and J.

General Atomic, McDonnell Douglas Astro- Holmes, "Effects of Irradiation on the
nautics Co., (September, 1978). Fracture Behavior of Austenitic Stainless
Steels," Radiation Effects in Breeder Reactor
9. S. Mukherjee, S. D. Harkness and C. Y. Li,
Structural Materials. edited by M. L. Bleiburg,
"Mechanical Analysis of a First Wall Model
Bleiburg, p. 95. (19771.
by Using a State Variable Approach,w this
conference. 22. D. V. Nelson, J. D. Stephen, M. M. Abo-el-ata,
and R. G. Sim, 'Development of Structural
10. "Fusion Reactor First Wall/Blanket Systems
Design Criteria for Highly Irradiated Core
Analysis," Research Report 472-1, Final
Components," ASME publication 78-PVP-78,
Report, McDonnell Douglas Astronautics Co.,
(June, 1978).
(October, 1976).
23. F. A. Leckie and A. R. S. Ponter, "Deformation
11. Plasma Chamber, Volume 6 of GA-ANL TNS
Bounds for Bodies which Creep in the Plastic
Scoping Studies, McDonnell Douglas Astro-
Range," Journal of Applied Mechanics, Trans.
nautics Company for General Atomic Company,
ASME, June 1970, pp. 426-430.
GA Report No. GA-Al4614 (October, 1977).
24. S. Majumdar and P. S. Maiya, "A Unified
12. M. J. Delaney and B. A. Cramer, "Structural
and Mechanistic Approach to Creep-Fatigue
Stability Analysis Considerations in Fusion
Damage," Proceedings of the Second Inter-
Reactor Plasma Chamber Design," Proceedings
national Conference on the Mechanical Be-
of the Third Topical Meeting on the Tech-
havior of Materials, p. 924 (1976).
nology of Controlled Nuclear Fusion, Santa
Fe, New Mexico (May, 1978) (to be published). 25. D. J. Michel and G. E. Korth, "Effects of
Irradiation on Fatigue and Crack Propaga-
13. M. J. Delaney, B. A. Cramer, C. A. Trachsel,
tion in Austenitic Stainless Steels,"
D. A. Bowers and J. W. Davis, "Structural
Radiation Effects in Breeder Reactor Struc-
Performance of Plasma Chamber Materials for
tural Materials, edited by M. L. Bleiberg
a Power Generating Tokamak," this conference.
and J. W. Bennett, p. 117 (1977)
14. W. G. Wolfer and R. D. Watson, "Structural
Performance of a Graphite Blanket in Fusion 26. J. Weertman, "Fatigue Crack Propagation
Theories." to be published as part of the
Reactors," University of Wisconsin, UWFDM-
ASM Symposium on Fatigue and Microstructure,
258 (June, 1978).
(October, 1978).
15. D. L. Smith and J. A. Krazinski. "First
Wall/Limiter Design for Tokamak Experimental 27. J. R. Hawthorne and H. E. Watson, %xplora-
Power Reactor," ANS Winter Meeting, Washing- tion of the Influence of Welding Variables
on Notch Ductility of Irradiated Austenitic
ton, D.C., ANS Trans. 30, 65 (1978).
Stainless Steel Welds," Radiation Effects in
16. "Nuclear Components in Elevated Temperature Breeder Reactor Structural Materials,
Service," Case 1592. Interpretations of edited by M. L. Bleiberg and J. W. Bennett,
ASME Boiler and Pressure Vessel Code, p. 327 (1977).
(April, 1974).
28. D. Hull and D. E. Rimer, "The Growth of
17. C. K. Youngdahl and D. L. Smith, "Stress Grain-Boundary Voids under Stress,"
and Lifetime Limitations of First Wall Phil. Msg., 4, p. 673 (1959).
Structural Materia1s.u this conference.
18. J. Bree, "Elastic-Plastic Behavior of Thin
Tubes Subjected to Internal Pressure and
Intermittent High-Heat Fluxes with Appli-
cation to Fast-Nuclear-Reactor Fuel Ele-
ments," J. of Strain Analysis, 1, (19671,
p. 226.
19. S. D. liarkness, B. Cramer, et al., nAn
Analysis of the Relationship Between
Available Materials Properties and Allow-
able Fusion Reactor Design Considerations,"
Proceedings of the Third ANS Topical Meeting
on the Technology of Controlled Nuclear
Fusion, Santa Fe, New Mexico (May, 1978)
(to be published); also, Argonne National
Laboratory, ANL/FPP/TM-112 (July, 1978).
20. E. E. Bloom, "Mechanical Properties of
Materials in Fusion Reactor First-Wall and
Blanket Systems," this conference.