OorOP-^/^-2<Z PREPRINT UCRL- ?

B759

Lawrence Uvermore Laboratory

Safety and Environmental Aspects of Fusion Reactor;

John P. Holdren

October 15, 1976

This paper was prepared for submission to

the ANS/ENS international Conference,
Hoshinaton, D.C., November 14-19, 1976.

This Is a preprint of a paper Intended lot publication in a |ouina! or proceedings Since changes muy be made
~3
Dotoift publication, this preprint is made available with the understanding inai <i wilt not be cited or reproduced
without the permission o! (he aulhor.

^f BOIICB ^

ability'
 5afety and Environmental Aspects of Tusion Reactors
John P. Holdren
Lawrence Livermore Laboratory
and
University of California, Berkeley
Abstract

Fusion is examined against the yardstick of fission technology

with respect to inventories of radioactivity {and associated Biological
Hazard Potentials), routine emissions, accident pathways and consequences,
radioactive-waste management, and misuse of nuclear materials. 3asd on
conceptual designs of Tokamak Fusion reactors with stainless steel
structure and t r i t i u m inventories of 10 ko per thtrmal gigawatt, the
apparent advantage of fusion is 1 to 2 orders of magnitude in most indices
of radiological hazards. Fusion's advantage is 2 to 5 orders of magnitude
in comparing damage potential of intentional '.irborne dispersnl of t r i t i u m
and plutonlum, but nonexistent in comparing medium-term radwaste hazard
potential (1000 to 100,000 years) and intentional waterbome dispersal of
t r i t i u m and plutonium. Fusion appears to have some qualitative advantages
with respect to accident pathways and safeguards considerations. Fusion
has the theoretical potential for improvements of 1 to 2 additional orders
of magnitude in short-term BHPs and 3 orders of magnitude and more in
radwaste BHPs after 10 years i f vanadium-titanium alloy can be used in
place of stainless steel i n the reactor structure. Other important
unresolved questions are how much the inventory of t r i t i u m can be reduced
by ingenious design, and what fraction of a fusion reactor's activation
products could be volatilized and released in a severe accident such as
a lithium f i r e . Overall, fusion's potential advantages are appreciable
but not automaticit w i l l require early and sustained attention to
environmental characteristics to avoid losing some of the potential
advantages in pursuit of other goals.

*Work performed in part under the auspices of the L>.5. Energy Research and
uevelopment Administration under Contract No. W-7405-Eng-'").
 Safety and Environmental Aspects of Fusion Reactors

Johr P. Holdren

Lawrence Livermore Laboratory

and
Energy and Resources Group, University of California. Berkeley

1ntroduction

The traditional rationale for developing fusion as an energy source has

been that i t w i l l be inexhaustible, cheap, clean, and safe. On any interest

ing '.ime scale, however, sunlight and fission breeder reactors are also " i n

exhaustible" energy sources. furthermore, the high construction costs of

fusion reactors probably w i l l more than offset the cheapness of the f u e l : i t

now seems unlikely that fusion w ' l l be cheaper than fission and i t may not be

cheaper thn solar. Thus the case for fusion rests, more stronqly than is

commonly supposed, upon whether i t enjoys real advantages over competing

energy sources with respect to environment and safety. This paper examines

fusion's environmental and Sofety characteristics against the more familiar

yardstick of those of fission (1). Fusion's early stage of technological

development makes such an evaluation d i f f i c u l t , but the attempt is worthwhile

because there is s t i l l time to steer fusion-reactor design in the direction

of minimizing potential hazards that ^re uncovered.

Environmental and safety characteristics of conceptual fusion-reactor

designs have been surveyed in a number of earlier reviews (for example, 2-7).

Both fusion technology and environmental assessment are rapidi q fields,

however, which makes frequent reexamination of U'. .o of nnurnjtion on

environmental aspects c' ' ,.mle. This review has benefited from

hi'tter information on neutron activation in fusiim-reactor blankets than was

available e a r l i e r , from estimates of t r i t i u m inventories that are probably

more r e a l i s t i c tnan earlier values, from the general continuing increase i n

realism and sophistication evident in conceptual fusion-reactor designs under

development around the world, and from recent work In fission-reactor safety

analysis (such as the Reactor Safety Study of the U.S. Nucleii Regulatory

Commission).

Environmental effects, broadly defined, originate in may stages of the

fuel cycles of modern energy supply (for example, explorjtion, extraction,

processing, transportation, conversion, and end use); they take many forms

(for example, routine effluents, accidents, other environmental transformations,

resource consumption, and social disruptions); and the damages are f e l t by a

variety of kinds of 'Ictlms in a variety of ways ( f o r xar>l*, prc"itinn.il

iteath. ind disease, direct health damage to members of the public now alive

and in future generations, damage to economic goods and services such as build

ings and tourism, damage to environmental goods and services such as climate

regulation and nutrient cycling, and generation of aggravations and tensions

within and between nations) (8). I w i l l confine my attention here, however, to

a narrower subset of environmental issues - namely, those issues that are generic

to nuclear energy sources and that are the subject of close if r u t i n / anrl often

controversy today In the case of nuclear fission power. These are: (a) occupa

tional and public radiation exposure as a conseauence of the routine operation

of nuclear fuel cycles (excluding storage or disposal of long-lived radioactive

wastes); (b) nonroutine releases of radioactivity from reactors or other fuel-

cycle f a c i l i t i e s owing to accidents or sabotage (but again excluding long-term

waste management); (c) the problem of the long-lived radioactive wastes; and

(d) the use of nuclear materials produced In commercial f a c i l i t i e s for weapons,

by nations or subnational groups. (The numbering is not meant to ii'i^ly an order

of importance). Information on some of the environmental issues I w i l l not

consider here - such as land disturbance 1n fuel e x t r a * t i o n , thermal discharges,

and demands on nonfuel minerals - is available i n several e a r l i e r reviews ( e . g . ,

2, 6, 9).

The present review is Sased mainly on the D-T fusion fu.7l cycle using

magnetic confinement. Sonc but not a l l of the results are specific to loUmaks,

on which thH most extensive conceptual-design l i t e r a t u r e happens to be available.

A few eomnents about advanced fusion fuel cycles are made at the end.

Radioactive Inventories and Hazard__Potentia1s

The f i r s t step in evaluating hazards from either routine or nonroutine

releases of radioactivity i\, to determine how much and what kinds of radioactivity

the System in question contains. Table 1 compares the inventories in megacuries

per gigawatt-electric (GWe) of capacity in Liquid Metal Fast Breeder fteictors

tLMFbR) and in two conceptual Controlled Thermonuclear Reactors {CTKJ All the

reactors are assumed to hive a thermal-to-electric conversion efficier.cy of

40 percent, so 1 GWe corresponds to 2.5 gigawatts-thermal jGWt). The CTRs are

l i q u i d - l i t h i u m cooled Tokamaks Identical except for the structural materials

exposed to neutrons: one uses 31> stainless steel 131bbb); the other uses an

alloy of 841 (atomic) vanadium and l $ l titanium (V-Ti). The specific CTR design

is the University of Wisconsin Tokamek UWMAK-1 (10). The LMFBfl is a sca^-uu of

the r,erman-Uelqian-L)utcn Sflfi-JOO {III.

lhe major components of the CTR t r i t i u m inventory are (a) the t r i t i u m

injection, collection, and purification systems (3 kq T uer CWt); !b) the breed

ing medium and separation unit (1 kg/GUt); and (c.i storage and e.nor'jeiicy inventory

!(> ky/(iWl), <or a total of 10 kg or 100 meg.uuries per Gut. This is u somewhat
1
Uruirr fi'jurr than tnese u t i - d in nsi earlier work; U di-ise , from assuming

5. b rn-up o< T oer pass tnrnuqh the plasma, a 'iold-uo time of ' day in tfie
u
 h
tritium c i r c u i t , and a reserve of 2 days' Supply of T to keep. t ** feat tor
running during repairs on the fuel handling system. About 'i pert on t of the
?
i n i t i a l activation in the 316 SS CTR Blanket (2 year operation at 1.?* Mw/m
neutron wall loading) is in f i v e Isotopes: ?7,8 day Cr-M (,"j0 HClf. 103 da/
Hn-64 (|ja C i j , 2.58 (tour Mn-^6 {880 HC i J, 2,6 year !>-**> (4H8 MU), .mi
71.3 day Co-iS (?2S MO), J he I n t t ' a l a c t u a t i o n in the v - t l b Unlet (sane
loading a* for 316 SS) l i mostly 3./^ minute V-W (?!. MCl), *> K minute
Ti-*1 (19B MCI) and 1.83 day it-48 (% MCi). The coolant actuation figures
r
are f o l i q u i d l i t M u m a n l arise fro*n i^&urHfes {principal*^ '^, Ci . **a. *mJ
if.) ad corrosion products.

The fission product a c t i v i t y in an l.WftR is distributed j--cr.g a tarne

array of Isotopes; among tht v o l a t i l e ones (corarable in so.-*' sense to
t r i t i u m ) , the iodines dominate {*> MCi in 1-111 through i - l J ^ j . Hearty
AH yertent of the i . . i t u l s,,-vttijrs$ a c t u a t i o n *s Mn-M (!H * X i i . Mi*-Ms
{?2 WCI) ant; 'Jo-58 (38 MCi). The coolant a c t u a t o n is IS hgyr n,,.;-' 110H
HCO and ?.6 year Ha-?4 (1 MCi). The i n i t i a l jcttntde a c t i v i t y Is donlnated
by 23 minute U-?39 and 2.4 day Np-, 19 11 T'.-l W. i .-,,r *hr "1-if.t *. !imrti

LOfitributOr ..t 30 d*ys H ?-?! H * X . O .

the (.omp^r.sons in I^hle I j r r ,or*-*h,u 'M-.tttrtfu i*s *j*of tf f i s s i o n ,
o/ virtue o the neg:-ct of inventrrtes of r j O ' r [ t i f M fuels in parts of the
fuel cycle other than the reactor i t s e l f . The plutuntipa j i t u n , in JI\ i HHW
fuel cytie outside I t * reactor is dhoijt 1'* j j r r i r n t of whit U In the <-p^, t r
U ? i , t f i U a * . by contrast, is retyt l*d u h m the fusion F r i t t e r pUtit 4fid
the inventory shown m T,tt>le I Includes the contents of this internal c y l e ,

A better irasui-c ' hajanj than radinact Wi{y in curies t-. the Fllolngml
>
"i/jr.l Potential (8>lJ '. oMarned by duidlnt} thr mvi-riln.y i f jn i-.otop*- by
i!-, K-t.^ttv- P^roiss.bi*- CQcentrjtiOf- iNCf. j r,r tiei.tKXMT.ilerf r . m i i t M H o nlr
(.y'lH r * ' t c i g* J " ' o r M d ' r r , tfir (>., I t I n.j flHP rt(jrf\<m*A (dc v o l . ,- .-* Air

or w t o r l h , * i o u l t f t><- <f.JcJ : a d i l u t e *.*< : n * r n l o r , to ;hr ;><- Mb'i> !rv.-l

T.ttilc- .' s h r v * the B'tP^, tor 'M^t'of ir -Sir i n j h r , . \ l r ( !<<! 01 >,,>. .. tiTTr\s>t>hainq

lu thr mvrntoripi ftf i t i f o a t t Sy i t ^ i n tn<- iHSftft jnO Hi * * r Of.*,*- .r>.i , ' . T i LTB\.

Tup HPCs '.t"l t o <!.-. i*<- t * < " . e n j - - * v r ' . i ' t t^Oi.- in 'c.rip ; - .HJUMJ

SUtCV 4t t m - co o ' l V i '.lli. fit* v a l u r \ 'or '.Olut'lr on<J m w l u M c furTM o'

w
JO I ' . o l Q l r J*<f', the lo-vr 'f. - <;i*in<i t h r Mijhrr SWl' - *>a-, t>( i ^ r i l here

Ttm o K l r l / u\>'<J p r i x i - d o r v f . i j .ji-.tort ( c r t . i ) " 1 vw<iar l - . i m i lor r u'Clc, thr

Iftjn for In-.olubtf farvr., tjt tiv\c rl<-=-,rnH. c i t \ t l f i r r , i i ! i i f ' , t\ MqMy in-

\0luE>)v c t H I r - . . if t*w, rmun in Ihjt ' u r t * wndpr j u l d r f t t mfvJ 11 Inns . tht-n v r

i.f t t w *?<.: 'i>r !fic -.cl^l, * r - . i n [IMP 1. j h u i a U p " " . i i v t ' r " , t * t r \
r j th.- .nn'rihu'Tn

* ttw . t t t i f i M c . ,;* ihv ,1 t i n i i j , . - . i n t * i r l * * f l . .il< 1.11 <-.>n ' TM- . 't'f j11

.t'.\.jrw<l tn-.olu!ilr t -.Keif t ^ t n w l l t P 1:* t*w F.'fP o,*J'I * * U * ' C P U/.tX'i rttltgtt

it* to 1,'IXI m i l l i o n m ! )n ; i , r -.thvr 1*il. !> ^ K 'or U*- M i l ^ i . l r 'or is

1 ; ' i p H I . - to INjt Mrt of thr r-lfrr-'url .,('". (JuTo'i'vC innrMn'j t hJt

V M O T f i l * ' . t " y 1-llNtf

In -.nth J C41V, I t i l l U r l y t f u l aXAh Ol tftp ( r l t ' u a r r l c j i e d t>ulj tw In

Ihc OL10C *ora.

With '.ftc foreijolfig j m - J l . . on*- tan lutKludr t r ' H 4 t w * l y 'rem Tjtilr .*

trial Ihc r e U l t v c ha:tm rpprrn-f'tptl by Ihr iMlturs i n i m t o r y in (h<- totxepw*!

LlRi i i JbOut t i r r e urJcri of fagnltuJr ^ m l l r f l-"un i f m r t jrvientrj by the

f" n i t o n produs.lt in iHrQBs uf i n * \ j ^ r 17rnrr.1l I*-*; u w ' H ; . ir-i th*l t*r

dCtlvdtton product! in the ClK'. r<-(irrl*M 4 r r b l i i r hj.'ird one to IMP Of-lft;

f
Of rj.;nludc i m j l l r r .hjn th.*l rri'i'ricfilcd tv ".hr 4<. 11 r>'>tiri In ;> tK.S

i ^ l . v a t t o n In th* V-Tl LIS i i i p f f r ' l a t e l y *p or.;-. ' rj.jmV,.!^ I.--.-, ' M r

that in j l 6 $ i IB jnd corpjr:'. .i> to ^ ! t ( i o h i f '.* L) ft?'

WoiltJttpn tipoiuic frar. ' j t j t t n c C?-r4l ion*.

 The second pathway 'or (r)t>us * M r i H ' ' J V I O I t'.rc^jh :f- jMoy-,

tonialrCTpnt-iyitiT boundaries, c v l g | l l / '.v r i u p * info !fw ) i r ^'n-jfi-i '."

plant * i HT or gasrout rt!9 THI* pathway ^ i c r * ! ) / h*\ Exerr. ri-^jr.;**: *\ rtiuf

to control. and the resulting elusions presided In I T - . ^ U r r A$pro.tN<v

used Include separating not i m ^ M t o r i f . fir/- coW o^r'. *f<J r ^ t t n ^ r.<-r,

design e f f o r t 10 n i n l n l i e the ftirrvr, \ur'0-ifni1r-<; "3* T j r r * ' , nl-,* , oH T ' . I I

wdll',. and employing copper, alurtiimr, or 'cr*r.l< fiw". lfj*. *\ ; i " v . ' - . ; * f - i r n r r *

l3, 17. Ifi) The Bolt proal^liij WJ< '.o r>1nl"l.-t ',ie hoi trilik."- ih cr,',or/ f

soens to be breeding i f ',H>! l i t i i . j - . nTO'jrd\ w J I < * I i - - ' c<,li-". tM-intrl

ril trt this approach. inventor (-. tr. Mankrt ami I O H I J ' . I (t '*;fc " " . ^ j , *. Uw 4-.
r
10 C'/GWc Mve been satd to be possible ( l i ) .

K
the t r t l t u n iftvcf.tory r^jsl d l f f t t j l t In rwiurr Srrrr. no! tc * t r ir.vcr.torr

H. t w breedtm nrdtuff:, however, tm! rather the i r . v r i l o r , j v . i " u'.-!! ! t*- f- d I -

Ifi'j triimn- that hai bteh injeclrd *: fuel byt p\. j j i r ; froti the r r j ( t ">n r r j l o n

without burninq. The l o w r the ' r a i t l p m l 1grii {i |>rr pass, !'rr i r r

U '.nrrf

this t rtl'lcn. (Stmoly the reserve to penal 1 1 (in: 1 nurd operation i f I ho </ilin<]

S/Stc" ' ^ r unburned t r i t i u m nai fur* * ions rust I * aixjwl 1 4 < l i ) ' / f -nrles pf

'J'*t-<Jay, where 1 is the fract'onal Lurnup r*r juss j A ;r)".tur- haixilimj -.tilf~
f 1
for 4 mirror fullon reactor has hern HcM.ritir'1 t : i a .]f. i-^ m y m i n r / ,jf ,' I-'

Ci/GWe for the entire plant. I>ul no msrrvr for n i l f unctions 1'. lm !!e<J in that

figure [20).

in addition to releases from th? plant, routine ciposurr- of wofr>s inside

the fusion plant to t r t t t u n Is j i v e a iwienltal mnci-rn. The oitjpat tonal *hl

for IITO i n a i r is b ..Ci/m . and U.S. IRDA -juldelines '.all for ysinq levels of

?ui HPt as dcitnn guideline*. I t Is nol dear -hethT (hi-, irjtilellnr -.an I*

r-vX. in ,\ f Iftj if i t tannot, worllno ' . h ' f t \ n a h^.i- (o Ur rcMrt'.t*"! lo li'\-.

f

than 40 hours per c t or protective \u\\\ (to prcvem \ H n .iliior(,t inn) aPii

i^i^lii ojy nave l'i be used (7fl).

otl.ir*. tfnt jrr --.ui - n r r n'j'.'lr J ! o ^ - f j ! 1.if] !rTrr,i!'r*. ' N " !-.r d*.""! a 1
k

fur-'l, Hour,IT, 4fisJ ,- r i ! r n l to M h ( I <),*.*. 0* thf.r tjrti'llr -.-'ilvv tt'H

(omntKiir t o r.-lrj>.. o' r*.jtj<.1 v I'.* * rvr- tio t>.jftt 40,1 t o IK 1 j j - j t > n . t l D

fti-iUfr; Itut.lr itr-.J\ : l c r r 1 n>c%t * 1 * t m n "-., 1 n t r f u f t i f ' J (M.* 1 J I.' "<i

jt.rif.pMTi> in > i : 4 ' ! IT' *' 4 ' t t v o t n ! '.'.'AlgT -1;*U f- w i r - . ^ r , I:

, t
r.-jvl j l ' . u r-c rsvlis) ^ P I J T . r v r f i *Cf>- M I P I ! ,oin!!r In ;!.>(, . in'.c 41 *.11 j l 1.1"

p r t x l u . t*. t h * t *-it ;<T>rIntini j-inr.1 f j i F i"* " - - J ! ' 0 " \ ' " H - n,.'n rrj(t-or.\)

t>tl\r J potifilut I4,'.nj Ip u r l r - \ I n jtvJ n r . f r t hi> |>l.in!

Mutt- I T uli'nl'n' >n : " 'mint- *. I " hjt-ftf.-. !rir 14->*V, ' j M o n wu!rr-^

for iMttn:* nrntlinn J n i | ihi-rrrji r n r,|,

f u'nvrrMon, th.-rr ". -.tronij i m f n t nrp

tp c t i . i n l . T fVi-lpt- of np ir n\ u P '- '*r the M a n t c t rp.jior It U also f.-.i>Mlj]

ti> [ f i . l f i t ! " ' . u f r r , . omJm t tut) r-.i<)nr!\ 'rr" htrjn n r u l r n n fl.,.t'\. NrHTthclrw .

tit- -|H'.it pv'-rlNli-ni |.mvr 1:' * . V 'if.ion w u t n i i - . .in.I t-n- . i w - f l c . .;.-onrt'T '

fu'.'0''Tt'<t< t o r trl<Tiar% i I . <",, kc l o t - ^ l f l c ,hlrldmt) d i f n u l t in thr I'ru-tli.ttc

irr.i of l i * . r r j . tor ' j f ' 1. i.-nl \fnrt<Hn) t j n i*> invOn'Or.itiMi into i n * r.-jttor
 l
butiding itself to reduce the neutron dose outsjjte *"> desired level, however.

Thus neutron exposure will be an occupational hazard, but presumably not a pub!'c

one.

Kagnets for typical Tokanak reactors would produce fields that extend far

beyond the reactor 500 meters to drop to 1 gauss in UWAK-1, for example (10).

Prolonged exposure of 'uslon workers to magnetic fields of some tens of gauss

is to be expected and brief exposures to much higher fields Jre possible. The

physiological consequences of prolonged exposure of humans to strong magnetic

fields are not known. At exposures that will be encountered routinely in fusion,

such effects may be nonexistent or negligible or they may be significant, and

more research on the question is needed (3). At some expense, possibly signifi

cant, the magnetic field intensity outside the reactor building could be reduced

by means of shielding with magnetic materials or partial cancellation with addi

tional magnets arranged for this purpose.

Accidents

The risk from accidents at nuclear facilities is the integral or sum, over

all possible accidents, of the probability of occurence times the consequences

of these accidents, As the debate over fission power has shown, significant

uncertainties and even controversies atteit'J every step uf the assessment of

this risk: the Identification of the possitle accidents, the estimation of

their probabilities, and the evaluation of the potential damage;. Since there

is no operating experience nor even a firm engineering design for fusion power

plants, the difficulty of accident analysis is even greater than Is th* case

for fission. At most one can Identify some polPhtidl problem areas, fitimate

very roughly the consequences of hypothetical events, and make sort crude

comparisons with fission.

 Holevant to the Identification of possible accidents and assessment of

:heir probabilities are the amounts of stored enerqy In various parts of

fusion reactor systems and the oathwavs o o t e n t l a l l * available for the release

of such energy.

The amounts of energy stored in a Takamak reactor based on recent con

ceptual designs are swtmariied in Table 3.

Rapid release of the nuclear energy repiesented by the fuel contained In

the plasma at any given t i - e seems exceedingly unlikely on the basis of present

knowledge of plasma behavior. I t is conceivable that malfunctions could produce

a temporary increase in reaction rate by virtue of inceased temperature or

magnetic f i e l d , but the delicate balance of conditions required for plasma con

tainment means that the end result of any departures from normal operating con

ditions would be rapid quenching of the reaction due to loss of plasma to trie

walls. If somehow the entire Quantity of fuel In the plasnu did roaic, the

less th*n hundred gigajoules evolved would raise the blanket temperature only

about 1MC (6).

The kinetic energy of the hot plasma Is about two orders of maanitude

snaller than the potential nuclear energy. If an i n s t a b i l i t y were to cause

th? entire hot plasma to be deposited on a small section of the vacuum w a l l ,

a local Own-through could result (3, 6 ) . 5uch an event would naturally be

disruptive anil expensive for the operators of the reactor, but i t could only

produce serious external consequences i f i t led to a major lithium f i r e , as

discussed below.

A vry iiihstanti<*1 nuflmltv of enerav. hundreds of aiqaioules. is stared

in the superconducting magnets that confine the fusion plasma. Concern i'.

sometimes expressed that the sudden release of this energy owing to a magnet

transition from the superconducting to the normal st^te would 'epresenl an

important acr nJpnt pithway 'or fusion reactors, and considerable analysis of
the question has been done I t seems a relatively straightforward matter,
however, to design magnets incorpora:ing norrajl conductors as alternate
current paths, along with adequate energy sinks for the associated joule
heating, in such a way that an explosive energy release from a magnet "ijoimj
normal" is precludsd (21}.

I t w i l l rema'n true, of course, mat large superconducting magnets in

operation are subject to very large forces (on the order of 1C tans), so
that catastrophic structural f a i l u r e perhaps cannot he e n t i r e l y ruled nut.
Such a f a i l u r e could procuce energetic missiles, raising the possibility of
severe damage to other reactor components, penetration o* contdtnn*'nt structure',.
iinO i n i t i a t i o n of lithium f i r e i .

Energetic missiles, with these same possihle consenuences. cculd also In-
produced by sudden vacuum-system f a i l u r e . t>y helium overpressure in the nvi.|net
cooling system, by "external" events such as earthquakes, loriados, rtir.-nifi
impact and sabotage, ana of course by combinations nf these pathways

An important accident pathway for a l l nuclear systems is W of ..uo'inl

. loss of coolant flow. Calculations made for the UUMAK I fotamaii ..criiceptu.il
design indicate that complete loss of coolant Mow during themonuc lear burn,
accompanied by f a i l u r e to shut down the fusion reaction, would cause the f i r s t
wall to reach a temperature of 600i ;n about In seconds; emtin tt lenent from
U
formation Of helium bubbles in the n * t d l . which oi'.urs arnynd fjS0 (.. cnuM
tnen lead to f a i l u r e of the w a l l , release of lithium into the vatuuti <-ham|>er,
.ind consequent quenching of the fusion reaction (3, TO). I ike SD many other
fusion accident pathways considered here, this one wpuld be disruptive and
expensive, but not catastrophic unless se<onrtary events prodded a major f i r e
ijnd/or breach of containment.

Loss of the coolant i t s e l f (as opposed to mere loss of f l o w ) , as could

occur owing to pipe breaks, would produce the same result as loss of flow
SDftewhdt 'norp rapidly, aga'n assuming thai the "usinn riMituin were -im ' m c a i -

atoly shut down. 1 ' . on thr otner hand, thr reaction is shut down, the concern

becomes the radioactive afturheat that results fro^ A llvatwin products in the

f i r s t Hall. The i n i t i a l af:erheat power densities in thp f i r s t wall at shutdown

f
f a l l 1fi the range of '1 i to 1 /cm for the main alternative structural riatc-riaK
1
[??). jnd the i n i t i a l rtdfatsatu temperature rise in j ToKimjk systc'" ha , be*n

calculated to tie (in the o^ler of H.IC/set (10). These values are no re than an

order of T-agmtude 'ower thjn the < orresponding figures for fission f u e l , and

they suggest that heat revival by radiation, conditi t inn ami -ijlur.it (timet tier,

w i i l suffice In prevent im-li.ing of the .ill

The largest coifi e o' -..[cprect energy in fusion reat tor designs relying on

1 iquid li*niiii" for tooling anil breeding of t r i t i u m is the ihernial energy

re: resent.'ii by the lithium U s e i f . and for vuth realtors 1 lithium ' i r e --
(

whether initiated bv internal or eiternal events -- may well represent the

"nuilmun hypothetical accident", t'thium reacts vigorously and eiot hermita l l y

both with a i r and with water; like sodium, i t remits also with concrete (actually,
1
with the water I i tier j ted fron tone re te by endoihermK dehydrat ion Althouuh

the kinetics of thes.- 'e.iilions are not well established I'lpff iment J 11 y. i.tKula-

ted mjitnum tljme te**per,itur >-. 'nr hn'.h th.> 11 thium-an- ,md 1 i thiur-t on,-ete

redcl.uns .ir- in the r,im)f r>* . .MW to .'.V)<).. l.V). 'hese temperatures .ire

below the melting points o' r,.fr tir,- metals, s.i.i as motnuf an.] '.'1 (titanium-

/ i r t oi.i-in molybdenum), that might be used in fusion realtor strut lures, t'ut

above '.he melting points o' other potential Struttural rn.itefi.il \ s,;i as series

J()0 siair.less steels Thr r,.fri lory tnet.ils. although they would not melt,

wmli! be rather IMP idly lO'iScmcd at such temperatures bv fOrn.lt ion jnd v o l a t i l i

zation of their ontles 1.7). 'he high temperatures and lame energy releases

potentially associated wit'i lithium fires therefore puse the two-edged possibility

of (a) breathing multiple lonl-nnment b.irners between the re.tt tor rore and the
public and (t>) augmenting the v o l a t i l e t r i t i u m inventory that could escape
through such breaks by converting activation products and tonic nonradloailive
metals to v o l a t i l e form.

The value of increasingly detailed safety analyses at the early stages

of design, of course, is that the designs can then evolve to cope with the
most serious accident pathways that are i d e n t i f i e d , the apparent f l e x i b i l i t y
of fusion in this respect is considerable, especially si.ice passive systems
seem capable of handling most i f not a l l of the stored energy sources. The
use of liquid lithium as & coolant md breeding medium for i n l n * . wMch
produces the largest stored-enerqy threat in m,jr,y T'tniun realtor dc.igns.
has of course been questioned, unfortunately, the alternatives .ire tiot without
their own d i f f i c u l t i e s . Use of helium as a coolant tn conjunction with oreeii-
itig in lithiuni solids, for example, v i r t u a l l y requires tlif u\>- cl t u T . l ' i n r .)>
a neutron m u l t i p l i e r ; this nuteridl is t>tremoIy t i m e , thus iirodui m>j addi
tional threats to occupational and public health, and i t i \ stance and fipensivc.
1
Use of f I ourine-lithium-beryl I lu- . TO 1 ten salt (M ihe) as an a I lefn.it M - radiant
and breeding material, on the .ilhi'r hand. lead-, to rn.iteri.iH -lospat itn 11 tv
problems and to the production of a part I C U I J H / .1.m jeruus (or- i>f i>*iiref lourli
acid (wherein the hydrogen is t r i t i u m and the flour mo is /-hour half lift*
flourine-lB). Clearly, much additional work w i l l In- required to iVterr.ini-
how much the apparent f l e x i b i l i t y of fusion with respect to tool.int ,iil tTeeii-
ing media can actually be exploited to maxir.we safety.

Notwithstanding the p o s s i b i l ' t y of maim.j the frequent* i'f OIA idi-ms very
low. i t is important to understand the consequences that uould w m r ih the
most severe events (including acts of war, sabotage, " i t u r j l .li'..i-.t.T'.. *<!
accidents that exceed the desion upabi 111 les of the s y . H ^ j '< an i n i ' t j l
itep in such a "worst-case" analysis, I have used r s M - n l i j t ' : , (hi- i nini.i phic M

'Ridel 0' the Reactor Safety ^tudy l"SS) i,f the .', W I N I ' ae-mUl^r , d n - U U i ' "
[?i) '.o estimate Ihi> " c r i t U . i l <luse' to bone rirr<,, revtiHiTuj from

d r i ' l f j M ' of 111 cuMes of t r U l y r . c i d e under adverse x-U-oroloyii a I tomlUions

The i-Mltc*! dose, as the concept ts used in the H*)S, means . i l l the dose received

In the f i r s t eight days and half the dose received f r w the eighth through the

t h i r l t v l h day, U is said to tie the most relevant figure for estimating the

Incidence of early f a t a l i t i e s and, to .* e i l e n l , early m j y r t e s . Qnv hundred

megacijrleS represent* 4U percent of the I r i l i u n inventory tonsitfered in Table '

Whether the release of so large a quantity (if t r i t i u m M one t i n e , and .ill as

!he !,-, c, n t u d l l y credible is debJlable, t.ut H provides a U \ i ' , f r cunpar-

ison ,|'h calculated fusion-reactor .in 1 dents releasing J similar percentage

of rutiy of the f u s i o n products. Figure 1 shows the c r i t i c a l dose lo t>,ue marron

plotted versus area over which .1 given dose is exceeded for tne 10 f. 1 <UQ release

anil f r the f usion-reactor ace idem .i,-^.U-i Pkft- I .n -he FT.V (Pitf-l releases

'.!) neneM of trie I T M - W , > .1 1.1 i.jr-.. i J (el lur iV"S . ''1 ,.rru-nt () U.e iddines,
Jf

'*! percent u' the noble gases, ', percent of the stnjntiuS and (i.srlum*, and
 A few further iual (f k a t ions are in order, f i r s t . release of ').'. ;ier<cnt
of th<* actinides in the fission-reactor accident considered in u\'-. n,i<ic <i very
small fractional contribution to the calculated earl* f a t a l t t i e i but a l.ir..
fractional contribution to calculated long-term cancer incidence l - ' i . f>. ]i).
This result suggests thai cm p r i s o n of long-term casualties friw. "isMtm jiu]
fusion releases (which I have not ypt done), mnjht snow a l.r.jcr itivrintaqe to
fusion than the factor of 100 computed hern for early fatal H i t " . . Second, the
release fractions used in the present comparison could be unreal ist u m spverdl
ways. Tritium inventories in fusion reactor', mi.iht be n-dm r>l M, lever <le', i .jus .
reducing the conceivable consequences of a worst-cose release Hut , it .u>,
significant fraction of fusion activation products could be released in ,m 01 t
dent, the calculated consequences could be worse than given here tor trittur- jlonp
(This is reasonably clear from the BHPs in "sble ?.) If a r t m t d r rri, ',<-s in IMlflH
M

accidents could be greater than the RSS assumed for LtfBs. t h i s , too, inuld i tia'"|t'
the comparison. These masters need further tnvc ' ' l a t i o n .

Radioactive .Wastes
Jtte presence of activation products of lomj h a l f - l i f e (V.ms that fusion
w i l l not be entirely free of the problem of radioactive-waste mana-jemem , with
which the fission power community has long been struggling, the major advantages
that have been claimed for fusion with respect to radioactive wastes arc: (a)
tlie f l e x i b i l i t y to t a i l o r the design and the choice of < t r m t u r . i l material-
exposed to neutrons wfti. the aim of minimizing production of long-liv-d activa
tion products (to be contrasted with the cast of f i s s i o n , where the s ' ; r und
character of the waste burden 1s lartjely determined by t>.i> unf.han3e.1l1h' physhs
f
of the i s s i o n process i t s e l f ) , and (b) the immob'M i.'al.on of the ad'v.i'.cd
material in the solid strucU-re of '.omponents made of refractory < 1 lays (reduc
ing tne chance of escape into the environment). It is possible, however. *'.at
f l t - H l n l H f in rm-fi.lls i'.flii", a l l ! ffv.f i!: vr,
u .it li-.is; for viy-' : i-v

t'j 4-Offr). i t |T<:Me wit*- 'iCj'.ron sl.v-yijv, los* O* M r w u l i .. niqn t c T f r , . -

W'PV. t i b r i c a S i l i t y , m-MaM I 11,, and si> c- -in' " , .ol.d] for the Um-,1. f .
nation n J U ' r l j K . -hi'trvr j i H i j l i i i a iirodmis in '..:U n-,ninr strm t J f f .
w i l l prgvp less notiili- that! fission i.ro'1ui.ls that * wi- : ; - i ' . pirlrt'.tiled for .];..
pPSJl In .1 tiTdmli r j l r u r<"Min\ to In- vi-ri'iec! 1 n . l t ' t j i l . i !(-,> sirm , y [*<
stqiilf ti.int u n d T t j i n t I v i ! txith i'.)<.!, ; f o ' uti.-.itt'i1 'u'.'ur s t rut t ut'.'s (li.it
Jo not i j w r to t'C ryii rot r*. - t"i ."o rc< ->' ' . u n r r\i 11" i , i ! \ .In pft , [ r.is t r J
F;

- l i r i f|s-,ian hri-i'il.-rs . who'..- pmnrt-ii t M M l u i i-,.., . r,.-. < ,.l , r p , .mini] i ,

u Y P

I hi- stvilli-r nci'J tn h.tru]) - th< rj.ln,.H I i vf r v i t i - M l - I I I ; ,ir. ,.;1v,int .|<U'

(

Tafitr i st-O-s the Br-Ps with -ff,.r,.ni|. to -" asi". t', . J U T for r j d i o . K t ' v c

.)sti-\ fftwi d Tol JIM* fusion reactor anil .in 1 "' H. no'inal I/rM Hi I i,Wr- f of
c l c t t r u U y (irodui t ion Thr n^j,, vtru.tur.il '.UiTi.t! -n the fusion re.ii tor is
!!! stainless sti'el. 'to (.ri>.!U is :a*,m m tn*> 1 MI i J' (or r#*, y i l i n g of .ittimJl's
Dttici" tiijn uranium ant! plutonium. fusion has an .uNiMrtijr in his t,oo.[>drisnn
that v j r i r s between 1 and .' iinli:r\ i.f ndnni t-i.1i' Jurwi.) t >>r S|ian of I imp nut
10 110 yvv-.. brlwrpn HKK1 ji<1 1C.00U ^i'or\ l . u o n .HII fission do not d i f f e r
S i i j n i f u j n t l y , hut .iftrr ;,>i,i)fl0 yiMrs fusion's rn.ni"' i n i w . f N attain. If
f
v.ltiadlum-tlianlLiin alloy (-if. tn- usiM in [il.Kc o .'l!> tjinlt-ss stt'ot in Pit*
fusion rpattor. one .jam-, jo j.lili t uitu 1 t.i tor of MOd rnlui t inn in the !IHP
'jy ^O years, anil by 100 years thi> "later i.i'. is iimn.ucus (less radio.H 11 ve than
Jvcr.jije rotk). This advanta.j'- tit V - ' i in dei.iy tir.- is -n tin] th.)t i t i^ust t>c
1
tonsiderrd q u a l i t a t i v e . ami no ,.ihir Ictf-ai t i v.l t ioi> i utuliitr nkitpri.il lost *,
its hazard .is rapidly l/hotht'r f-Ti . i ncct tht> o t T r ilcrtkitiriinfj r*>i i ro"H--ilS
llr )u

of fusion anplimtioii thcroforc ilt'-.vvcs. tfip i l o s f s t * . r - t i n v . (

Misuse, of r-uclear. materials

The advent of commercial fission power and the associated l.-.irr e (both

real and forecasted) in accessibility of plutonium and enriched uranium has

engundered considerable concern about the possible misuse of these materials

as weapons. To understand whether fusion power w i l l have any analogous l i a b i l

i t i e s , i t is important to c l a r i f y several aspects of the general thrtat by

distinguishing (a) production of weapons by governments vt>rsu<, production by

subnationa'' groups, and (b) explosive weapons versus radiological wvapcns.

Lack of suitable f i s s i l e nyfprials and the d i f * i c u I t y of tjettimi them

have been major factors in lim;tin-j the international si-rr.id ,,f f i v . m n homiis,

the knowledge of how to produce a fission bomb om.e the material is "> hand has

been more accessible than the material i t s e l f <<?$j. ant) this n<ejns that the

spread i f material (or the technology for getting i t ) as a lonsrtjuvnti* of the

spread of fission power has changed the situation in a ')udh t a l i vc way. Hut

fusion bombs arc more d i f f i c u l t to construct t?.in are ' i s s ' o n bombs, su '"uct.

so that obtaining the fusion fuels (say, tritium) i s . relative to tin- fission

case, only a small part o* the task (?). Hence the spreao of t r i t i u m bv fusion
1
power presumably does not contribute much to th<? spread M fusion bomb?.

There is a cross-connection between fusion and f i s s i o n , however- the neutrons

produced by fusion reactors could be used to product- f i s s i l e material for fission

bombs. This cannot be done without the consent Ai.a eiti's;vc uioiK"'.". '0 of the

operators of the fusion reactor, so i t is a problem at t j level of (."desirable

activity by governments or very large corporations, not snviilfr groins

Tritium could be useti as a radiological weapon t>y terrorists c gangsti-r\

who acquired i t , just as plutonium from fission fuel tvclev coutd be used.
Sump of the ilrttd rt-i>uirfd to asicis tno r i ' l d l i v * nvi.|i.i .u<Ji-' of thi>\? *'i:"<tt\

Jre ^uimuri/od in Tdblv ^>. the MPCs I'OI' plu'.Mium l'.otopf, "" ruK.fi I U M T
1
thdn thtlt of t r i t i u m [mediur d v d S Ci/rt. ), hut the -..petlfu . u . , v i t y of t r i t i u m

[Cl/'jroi") is mucn hiqhi-r t"dn for plulnniuiti. The r e v i l " i*. that the fliil",

js^OLldtt'd wild ttic inventory per (.Me or Mow ptr ',Hc ,r tpf I r i t i u n ' ami

reai-ior-ijrjile Plutonium work out .i*. follow*, t r i t ' o n * is li-'i*. i)jtu|i-n, -. man
u

plutonium by ,' to '. in (Jcri of rwnjnl tudf with rty.pvi t It' i unt jmnLit IC*I of i i i r ,

but the two are du'iru'inwlvly i-judl f l u r^\p?tl lu i on Id*'i nation uf natrr.

Son*.* Jijrt*i' of KtnsefiMjs .tppe.irs to in- TT(JH*I f u t tfu* HfH *. for p l u l u ' " ,tn
1
art* too tiiijh by j ' u t o r of ' tii I'.i l i d o r^l,)t**t' '.*'M , th.- i omic i'.wi

uci: j tn thi-, idsi- t t h r.idlumj I ,'<i), ,ir*il U>i-rr i . .. r ,ison to think the
H r

MI'Ls for t r i t i u m rryy Uv too low r- f J. tor\ of .' t i , i '.> tine ..'.'.ion .ii.uvi".

JfiiOrpor.it Mm O* -.Jin .t,rr,. ! ' n i . t a u , I* .-/ arc *.! i I dat fil . .uu M i.',!")"
l 0

I lie foM'i|oini| i.ort'(i,irnu<i to -utt- ! n l i ' '-.. ]oni;rr(.u\ t'un ; luUiiiiu" [.^ 'JW
u

.irJiT of luHjrmuoV .th n*\jfi ! t.i . mit j * i h j * jn of ^ J t f r .titd tr. -ji *i- I n t i j o -

in ,nJvd"t.njc uf I lu t orOi-r., ,.' -,.* ). 11 - > i . / r rspet t tf> < <"' -Vim I i on of ,nr.
u

In tiTtn*. of (lulr'it i,l .H.C.* ti. " M k r u l !,, -o Irf .1. -,IT% . V-LV* .,), ,ir,

jdntnt.Hlt' t)i-i,|.,\. '<ieri* i . rid fii-.-J to t r pf-.v^nf T r 1 l ur tyyonij tut '..fled

(Or H i l t U t Wi.'n!u. ie\ i,f i f . ..-.t. ' i n ,

fldvaneed' fuel, .eyeJ=;
Fusion reactions other than D-" require higher i-i.-actant nnr.|ii, ,in:1
better confinement to achieve energy breakeven. I f adv.jm.ed terMHilcly ;>er>-.u
attainment of the needrt co'-Ji tions. these re d l f f u u l t rea<tii"'. K M I I I <>"-r
! -,
important environmental advaitafjes. Of the adv.m<e;f r w i l i n - w *-m I idm-i - l i ,
J 1
D-D, D-He , and H V ) . the k<ast d i f f i c u l t i \ O-Li. The I -.') tv,uti.i <hjm
(which includes 0-r and O-He reactions on the I ,inri He pr.i.iui ",i ., .,!> twin
products Of D-D) produces fewer and lev. cnergetw neut run-. *ti<m purr : - ' .
although the neutron activation does not ' . i l l <t\ mini -T. nnc MTJM . , i><>. t -

t>y 15 to 75 portent, depending on th '.true u.-,11 - . t t r r i . i ! l</..'" " i. *''

Greatest advantage of the 0-0 evele '*> that t i would rrrrw the n n v . - . i t ^ '
use neutrons to breed tritium from l i t h i u n . Thi*. means that the i nm|> !*> 11,
Of the equipment surrounding the plasm* <nulrt he < nnMiier.ilil y r-dm i-d, -i<l ''
array of cardidate materials for use there would be !iroad'ed hy n-1 ,i ..it 10" "t
the neutron-economy constraints present when t r i t i u m must !* l>red ' i total
t r i t i u m inventory ',r\ the reactor could tie redtned di-Iow that in !)-f i c m l i i r s
by a factor of perhaps 3 to 10 [?). Other "\.tred r c u h f i i n n,w<* tin- initi-nTl.il
for greater reductions in neutron activate and t r i t i u m invrnti -y, ; .t r t n ..larly
in 'two-component" systems -here the rate^ < th*> neutron-free reattioir. i .in
be enhanced by t a i l o r i n g the energy distributions of the d i f f r r m t x-.u. t.ints.
It is problematical whether confinement ter'mology w i l l ever he <pM enm>'|h
to make these reactions practical, however.
 .tfMM'i-, f I'll f,Hi .. I . . .M.-f:ifl.n \^v r.-,ii ;,..--, .in' - l . - r ; > t><:

.c. *' i ,in tfvrt fi^ !< I m ! . . . J I , ,il f ...jui r e - n * m \ i'f '.,S'->" >* . i " * - s * u ' t i * " . '4r

1
.trp ) S t . I'd the d t H - .itw!. \ >\ '-1^'", i-.w'<i fi.lT l"\\U ,.-'.^'"t,|l . 1.|r,i,

I'.lliN Mill I.U|. i , i r r o' ,,..-.,.l,.^,

f or [.. ij.r. ...r, ,-i ;, .....I >n , > ' ! . U.r

SM*,M fiMttor M-i I W K ,i.-M ,..:-}, ti f r \ U ! . - j i . ! . i ' , ,' * j \ n " ''.. U ^

I n wHKft tMrtV <>' t h i ' (> m< > M ' '"-' !.*>< *n> IfW'StsfUM; < *.ii>*i ,iT . i l l
i( >,*%( S-fen

utiftttri'd 1" ( i u r v . 1 ? ' ..[niT ,-H-| i M - o r i n . , .i-ul r. ivi<v*u . i'h""i,t

Notes

1. Much of the information upon which this analysis 1" based was assembled as
part of a workshop on environmental characteristics of fission and fusion
carried out in 1975 and 1976 under the auspices of the International Institute
for Applied Systems Analysis. The principal participants included the author,
G. L. Kuldnskl (University of Wisconsin), G. Kessler (Kernforschungszentrum
Karlsruhe), and W. H&efele ( I . I . A . S . A . ) . The complete output of this work
shop, including some contributions by Soviet scientists noc available at this
w r i t i n g , w i l l be published as an I.I.A.S.A. report. The responsibility for
the present formulation is the author's.

2. J . P. Holdren, T. K. Fowltr, and R. F. Post, 1976. Fusion power and the

environment. In R. Karam and K. Morgan (eds-h Energy and the Environment:
Cost-Benefit Analysis. Hew York: Pergamon, pp.
3. J . R. Young, B. F. Gore, D. D. M2hlum, J . A, Strand, R. C. Thompson, and
J. K. Sadat, 1975. Information Requirements for Controlled Thermonuclear
Reactor Environmental Impact Statements, Gattelle Pacific Northwest Laboratory,
Report BNHL-1083.

4. H. Haefele and C. Starr, 1974. A perspective on fusion and fission breeders.

Journal of the British Nuclear Energy Society, Vol. 13, pp. 131-139 ( A p r i l ) .
See also the critiques by F. von Hlppel and by J . P. Holdren, Vol. 14, pp.
US 122 (April 1975).

ii. R. F. Post and F. L. Ribe, 1974. Fusion reactors as future energy sources.
Science, Vol. 186, pp. 397-407 (1 November).

6. U. S. Atomic Energy Commission, 1973. Fusion Power: An Assessment of

Ultimate Potential. HASH 1239. Washington, D.C.: U.S. Government Printing
Office.

7. 0. Steiner and A. P, Fraas, 1972. Preliminary observations on the radio

logical Implications of fusion power. Nuclear Safety. Vol. 13, pp. 353-362
(September - October).

8. R. J . Budnitz and J . P. Holdren, 1976. Social and environmental costs of

energy systems. Annual Review of Energy. Vol. 1 , pp. 553-580.
G. L. K u l d t i s k i , 1374. Fusion power - an assessment of i t s potential
impact in the U.S.A. Fiergy Policy. Vol. Z, pp. 104-125 (June).

B. Badger, K. W. Conn, G. L. K u l d n s k i , e t . a l . , 1973. UHMAK-I, A Wisconsin

Toroidal Fusion Reactor Design. University of Wisconsin Nuclear Enginesring
Department rteport FDH-112 (NovenSer).

See, e . g . , H. Wild, 1974. Radioaktlvc Inventare und deren zejtllcher

Verlauf nach Abschalten des Reaktors. Kernforschungsientr-im Karlsruhe
Report KFK 1797 (November), and note 1.

F. H. Pigford, 1974. Environmental aspects of nuclear energy production.

Annual .teview of Huclear Science. Vol. 23, p. 515-559.
United States Code of Federal Regulations, T i t l e 10, Chapter 1 , Part 20
(Standards f o r Protection Against Radiation), Washington, U.C.: U.S. Govern
ment Printing Office (26 December 1975).

See the discussion by Victor P. Bond, 1971. Evaluation of potential hazards

from t r i t i u m waste. In Environmental Aspects of Huclear Power Stations,
USAEC CQNF-700810-6, pp. 287-300.

P. S. Rohwer and W. H. Wilcox, 1976. Radiological aspects of environmental

t r i t i u m . Hue tear Safety, Vol. 17, No. i' (March-April), pp. 216-223.

J. Darvas, 1976, Tritium technology in fusion devices. In 0. S. Watson

and F. w. Wiffen (eds:). Radiation Effects and Tritium Technology for Fusion
Keacton,, COhF-750989, Vol. l i l , Springfield, Va.: National Technical Informa
tion Service.

T. J . Kabele. A- B. Johnson, and L. K. Hudge, 1976. Definition of source

terms for t r i t i u m evolution from CTR systems. In Watson and Wiffen (eds:)
(see note 16).
V. A. Haroni, 1974. An analysis of t r i t i u m d i s t r i b u t i o n and leakage character
i s t i c s for two fusion reactor reference designs. In Proc. F i f t h Symposium
on Engineering Problems In Fusion Research. IEEE Publication No. 73.

J. R- Powell, 1976. Tritium recovery from fusion blankets using solid

lithium compounds - I : Design and minimization of t r i t i u m inventory. In
Watson and Wiffen (eds.) (stt note 16).
T. R. Galloway, 1976. Tritium Control in a Mirror-Fusion Central Power
Station. Lawrence Liver-more Laboratory Report UCRL-78586 (25 August).
 -t4-

21. See. e . g . , 0. Okrent, W. Kastenberg, T. Botts, C. Chan, H. F e r r e l l , T.

^rederking. H. Sehnert, and A. Ullman, 1976. On the safety of tokamak-
type, central station fusion power reactors. Nuclear engineering and
design, i n press; and II. T. Yen, 976. Considerations qf Coil Protection
and Electrical Connecting Scheme: i n Large Superconducting Toroidal Magnet
Systems. 0RNL-TM-5043, National Technical Information Service, Springfield,
V i r g i n i a , March.

12. D. Steiner, 1975. The technological requirements for power by fusion,

Wudear Science and Engineering. Vol. 58, pp. 107-165.
23. U.S. Nuclear Regulatory Commission, 1976. Reactor Safety_.S,t.udy: An Assess
ment of Accident Risks jin U.5. Commercial Unclear Power plants; Appendix VI
C a l c u l a t i o n s ^ Reactor Accident Consequences. WASH-1400 {NUREG-75/014).
National Technical Information Service, Springfield, Va.

24. See e . g . , U.5. Environmental Protection Agency, Office of Radiation Programs,

1976. Reactor Safety Study (WASH-1400): A Review o_f the Final Report.
EPA-52O/3-76-009. USEPA, Washington, D.C., June.
25. H. Willrlch and T. B. Taylor, 1974. Nuclear Theft: Risks and Safeguards
Cambridge, Mass.: Ballinger.
26. R. C. Thompson, 1975. Transuranium element t o x i c i t y - dose-response relation
ships at low exposure levels. Summary and speculative i n t e r p o l a t i o n relative
to exposure l i m i t s . In 0. Nygaard, H. Adler, and W. Sinclair (eds.) Radiation
Research. New York: Academic Press, pp. 1278-84; J . Edsall, 1976.Toxicity of
Plutonium and some other actim'des. Bulletin of the Atomic Scientists, Vol.
32, No. 7 (September), pp. 26-37.

27. G. L. Kulcinski, M. Abdou, and 0. G. Doran, 1975. Comparison of displace

ment ard gas production rates in current fission and future fusion reactors.
In Priperties of Reactor Structural Alloys after Neutron or particle Irradia
t i o n . 5TP-570, American Society f o r Testing Materials, Philadelphia.

28. Bon Steiner, 1971. Neutron i r r a d i a t i o n effects and t r i t i u m inventories

associated with alternate fuel cycles for fusion reactors. Nuclear Fusion,
Vol. 11, pp. 305-306.
 Radioactive Inventories of Fission and fusion Keactors
Megacuries per GKe (2.5 GWt) versus time d f t e r shutdown

L*FBR 3165S V-T1

Fission Fusion Fusion

at shutdown
fission products/tritium 11,200 250 250
structural activation 100 2700 3100
1
coolant activation* 110 c75 b
<75
actinides 4000
"
10__sec. (2.77 h) after shutdown
fission products/tritium 5,200 250 250
structural activation 83 1500 150
coolant activation 96 ,60 b
<60 b

actinides 1800

1 day a i t e r shutdown
fission products/tritium 2,900 250 250
structural activation 70 1200 130
toolant activation 37 so' <50 b

actinides 1400
-" ""
30 days after shutdown
fission prodocts/tritlim 920 250 250
structural activation 53 600 11
coolant activation 1 ,40 b
<0 b

actinides 70

Assumes pool-tvpe LMFBH and l i q u i d L1 coolant i n CTR. Looj>-type LHFBR has 4-fold
smaller coolant activation. Helium cooled CTR has negligible coolant activation
1 nee after shutdown.
d e t a i l e d calculations for fusion coolant activation not complete.

Source: IIASA Workshop (see Note 1 ) .

 Table 2
Comparison af Biological Hazard Potentials with Reference to
Accidental Releases to Air. Million cubic kilometers of air
per GWe (2.5 SWt) versys time after shutdown-

UIF8R 316SS V-TI

Fission Fusion Fusion

at.shutdown
FP/tr1t1um >zm" 1.25 1.25
structural activation 55 390 OS
coolant activation 26 m W
actlRides 13,500
- --
10 sec {2J7 h) after shutdown
fP/trit1um
structural activation
coolant activation
acti nides

1 day after shutdown

FP/tritium 2300 1.25
structural activation 54 3?0
coolant activation 12 tiA
actinldes I.J00

30 days after shutdown

1
Ff /tritium
structural activation
taolant activation
actlnides

hA = wt available
a
35 isotopes (omits much short half-life activity)
Stwrtc; {1ASA Workshop
 a
Stored Energy *ft ^ 5U& Tpkamatt, fusion Jteactpr
(1 significant figure)

Energy Form Sigdjoales

Chemical energy i n l i q u i d lithium 60,000

Magnetic f i e l d energy 300

Complete fusion of fuel i n plasma 70

Pressure-volume work in vacuum 20

Kinetic energy in plasma <1

Sources: 6,10
 Biological. Hazard Potentials of Lon^-Hved Wastes
with Reference to Releases to Water. Cubic kllomett;rs
of water per GWe-yr versus time after shutdown.

Fusion LMFBfi

3165S fission LHFBR

activation products a<: t i n i d e s a

1 year 1400 10.000 100

10 years 49 3600. 28

30 years 9.0 2100. 24

100 years 2.7 330. 20

1000 years 1.1 0.020 4.7

10000 years 0.9 0.018 0.b2

I t of Plutonium and uranium and 100% of other actinldes from discharged fuel

Source: 11A5A Worksdoj)

Table b. Radiological Hazard^ of Plutonium and Tritium (Quantities
normalized where appropriate to 1 GWe of capacity)

Plutonium Tritla

Inventory outside blanket (kg) 900 25

Annual flow outside reactor (kg) 1500 32*

MPC,, (Ci,W;
*-al r

-insoluble <**?u, HT or T gai 2

0.001 40.000
2 3 9
--soluble P u , HTO vapor 0.00006 200

BHP {km of air) per gram of:

pure Pu, elemental T 63 to 1000 0.25
- r e a c t o r P , T In HTO u
b
300 to 5000 50
c 6 3
BHP/GWe-yr (10 km of a i r par year)
--best case* 3
450 0.00!
worst case e
7500 1.6
f 6
BHP/GWe (10 ir,r f a i r ) 0

best case 2/0 0.00

won.t case e
4500 1.25
3
HPC water (CIAm for soluble forms) 5000 3 000,000
t

3
BHP {rti of water) per gram of:
--pure Pu in soluble compound 12,500
reactor Pu in soluble compound 62,500
pure T in HTO 3,300,000
c
BHP/GWe-yr (km of water per year) 3 94 110

56 aa

a. At breeding r a t i o 1.26.
b. Contains Pu-238,239,240,241,242.
c. Based on flow outside reactor.
d. Reactor-grade Pu dispersed i n insoluble form, t r i t i u m dispersed as T 95s.
e. Reactor-grade Pu dispersed i n soluble form, t r i t i u m dispersed as HTO vapor.
f. Based SR inventories outside blankets.
rigure 1

CRITICAL DOSE TO BONE MARROW VS AREA FOR

HYPOTHETICAL FISSION AND FUSION ACCIDENTS .L9
10000C <=1 1 1 1 ini| ||I| 1r-rrm3100,000
Fission release PWR-1, RSS-

10,000 - 10,000

2 1000? -1000

I
-100
i
I
0.01 0.1 1.0 10
2
Area receiving this dose or greater, km