AN ANALYSIS OF TRITIUM RELEASES TO THE ATMOSPHERE BY A CTR*

3 - 8 . - -. . ' C ? ' ,.:, . - , sn,

by

David S. ~enn;, William F. Sandusky and M. Terry Dana

Atmospheric Sciences Department Battelle, Pacific Northwest Laboratories

Richland, Washington 99352

August 1975

For Presentation at the

3RD ERDA ENVIRONMENTAL PROTECTION CONFERENCE Chicago, Illinois - September 23-26, 1975

Battelle Memorial Institute Pacific Northwest Laboratories Richland, Washington 99352

- . *This work was sponsored by the Encrgy Research and Development. Administration under ERDA Contract No. AT(45-I)-1830.

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency Thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

DISCLAIMER

Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.

AN ANALYSIS OF TRITIUbl RELEASES TO THE ATMOSPHERE BY A CTR

David S. ~ e n n 6 , William F. Sandusky, and M. Terry Dana

. . . . . ATMOSPHERIC SCIENCES DEPARTMENT B a t t e l l e , P a c i f i c Northwest Laboratories

Richland, Washington 99352

ABSTRACT

Remova1.b~ atmospheric processes o f rou t ine ly and acc iden ta l ly

re leased t r i t i u m from a con t ro l l ed thermonuclear r e a c t o r (CTR) has been

inves t iga ted . Based on 'previous s t u d i e s , t h e assumed form of t h e t r i t i u m

f o r t h i s ana lys i s was HTO, o r t r i t i a t e d water vapor. Assuming a CTR

opera t ion i n Morris, I l l i n o i s , su r face water and ground-level a i r concen-

t r a t i o n values of t r i t i u m were computed f o r t h r e e space (or time) sca les :

l o c a l (50 Km of a p l a n t ) , regional (up t q 1000 Krn o f t h e p l a n t ) , and giobal .

Results of t h i s a n a l y s i s show t h a t wi th in 50 Km of t h e p lan t atmos-

pher ic concentrat ions of tritium w i l l be t h e l i m i t i n g f a c t o r f o r r.outine

re leases . On t h e regional and global s c a l e , su r face water concentrat ions

tend t o become t h e l i m i t i n g f a c t o r . However, both a i r and su r face water

tritium concent.rati.ons are est imated t o be below e x i s t i n g standards during

normal commercial CTR operat ions.

An ana lys i s f o r an accidenta l r e l e a s e o f t r i t i u m shows sur face water

canecntra t ions i n t h e v i c i n j t y o f t he p lan t t o be q u i t e high a f t e r a r a i n -

f a l l . Concentrations c m i l d be 1 o r 2 o rde rs of magnitude above accerted

standards i f t h e amount of acc iden ta l ly re leased tritium now postula ted

(10 Kgms) occurred.

~ u t u r e work should be d i r e c t e d toward b e t t e r understanding of t h e

atmospheric scavenging processes o f t r i t i a t e d gases, t h e regional and g lobal

t r a n s p o r t of t r i t i u m i n t h e hydrometeorologic cycle , and b e t t e r d e f i n i t i o n . . . . . . . .

of t h e p r o b a b i l i t y and magnitude o f extreme c l i m a t i c events t h a t would

r e s u l t i n excessive concentra t ions o f tritium i n t h e atmosphere and su r face . .

water supp l i e s .

1. INTRODUCTION

The Controlled Thermonuclear Reactor (CTR) i s a new technology i n

e l e c t r i c a l p o w e i production t h a t could provide f u t u r e generat ions with an

almost unlimited source of e l e c t r i c i t y and a t t h e same time produce only a

minimum of environmental impact. With our present supp l ies of f u e l s used

i n power generat ion dwindling a t a r ap id r a t e , t h i s new.technology, which -

con t ro l s t h e fus ion process and uses f u e l s t h a t a r e i n abundant supply,

warrants in tens ive i n v e s t i g a t i o n and development. Nevertheless t h e environ-

mental r i s k s associa ted with t h i s technology a r e poorly understood, and

t h e s e must be s tudied along with t h e development of t h e technology. This

paper p resen t s an ana lys i s of t h e est imated concentra t ions i n t h e environ-

ment of t h e main e f f l u e n t expected t o be emitted from t h e CTRgs -- tritium.

. This i so tope o f hydrogen was first discovered by D r . Luis Alvarex and h i s

col2eagues i n t h e l a t e 19301s, and has. s ince found wide app l i ca t ion by

many s c i e n t i s t s i n t r a c i n g components of t h e hydrologic cycle. The i so tope

undergoes a r eac t ion favoring t h e formation of t r i t i a t e d water vapor once

re leased i n t o t h e environment, and then behaves e s s e n t i a l l y a s any o the r

water vapor molecule.

The ana lys i s o f tritium concentrat ions i n t h e hydrosphere expected

from CTR power p l a n t s follows t h r e e phases, o r time and space frames, based

on an ana lys i s by Machta, e t al . (1973), of r ad ioac t ive e f f l u e n t s r e leased

t o t h e atmosphere from a proposed Liquid Metal ~ a s t ~ r e e d e r Reactor located

i n Morris, I l l i n o i s . Phase 1 s t u d i e s ground-level atmospheric and su r face

water tritium concentrat ions within 50 Km of t h e proposed CTR under normal

operating conditions. Phase 2 studies regional ground-level atmospheric

concentrations (up to 1000 Km) at any given time during the normal operation

of the plant. Phase 3 examines the expected long term concentrations in

the air and surface water supply throughout the northern hemisphere.and

globally during the normal operation of the plant. In addition, a n . . .

analysis of surface water concentrations near the plant after an accidental

large release of tritium during a rainfall episode is presented. Before

this analysis is presented, however, it is appropriate to present a qualita-

tive discussion of the behavior of tritium when released to the environment

to set the stage for the rationale used in the computational procedures.

2. THE BEHAVIOR OF TRITIUM IN THE ENVIRONMENT

2.1 Emissions from Controlled Thermonuclear Reactors

Several CTR technologies have been proposed, and each differs

slightly in the fuel 'cycle, plasma confinement, and overall .powerplant ' . . .

design. A summary of each design has been published by Young, et al.

(1975). It is expected that, due to stringent plasma conditions, the

fuels used in the initial plants will be deuterium and tritium. The

reaction equation is given by:

D + T + 4 ~ e + 0 + 17.6 Mev

The tritium fuel used in this reaction will need to be fabricated

since adequate natural supplies do not exist. Tritium can be produced

by bombarding lithium with neutrons, and indications are that blankets

can be designed for D-T fusion reactors which will breed tritium

(Young, et al.; 1975). ..( ,

The major effluent from virtually all CTR designs will be tritium. .

Most designs use a liquid metal cooling blanket, which is expected to

contain large'quantitles of tritium. Although many techniques have

been proposed to contain the tritium, it is expected that large amounts

will permeate through the heat exchanger and on through the heat trans-

fer system. One proposed design uses a gas cooling system, and here

large quantities of tritium are expected to penetrate the heat

exchanger and enter the steam system, and from there be released into

the environment through steam blowdowns. Although many tritium con-

tainment systems are proposed for the reactor, they become economically

unfeasible for the powerplant steam turbine system. Thus it..is

assumed that all tritium entering the steam generation system will be

released into the environment through steam wastewater streams and . .

ventilation.air. These emissions are expected to.be several orders

L

of magnitude greater than with fission powerplants during normal

operation, and can constitute a possible environmental hazard in an

accident situation, particularly if a liquid metal fire breeches the

reactor containment walls.

Tritium releases into the environment from nuclear power plants

are expected to exceed the natural rate of tritium production by 1990.

A pressure water reactor can release up to 20 curies per year of

tritium (9600 curies equals 1 gram of tritium), and it is expected

that CTRts will emit several orders of magnitude inire.than this into

the atmosphere and surface water.

2.2 Transformation Processes in the Atmosphere

Tritium can be released from a CTR either in its elemental form,

s11r.h as HT, nr as tritiated water vapor, HTO. The mass acti~n

equilibrium coefficient for the reaction:

HT + H20 2 HTO + H2 (2)

is approximately six, thus favoring the formation of tritiated water

vapor. Jacobs (1968) cites many references which show that the

predominant form of tritium in the atmosphere is in tritiated water

vapor, and that tritium rel.eased to the atmosphere in its elemental

form is quickly converted to tritiated water vapor. Therefore in this

study it is assumed that all tritium released from a CTR into the

environment is in the form of tritiated water vapor, HTO. . .

. ...

2 . 3 Global Transport of Tritium and the Hydrometeorolbgic cycle

Once' tritium is released into the environment and becomes

tritiated water vapor it's properties are essentially the same as those

of atmospheric water vapor. Thus tritium becomes a component of

the hydrologic cycle, and behaves the same as water vapor in the

hydrologic cycle. Jacobs (1968) notes through several references that

. the majority of tritium released in.the troposphere is limited to the

general latitude of release, although some lateral mixing is to be

expected. Almost half of all the tritium released into the strato-

sphere during the nuclear bomb tests was deposited in'the northern

hemispherical mid-latitude belts, and since these are the latitudes

in which most of the CTR activity is expected to take place, these

are the latitudes in which the highest air. and surface water concen- ' \

trations of tritium are expected to occur.

The removal of tritium from the atmosphere, as noted in the

following section, occurs primarily through the precipitation process,

although near the point of release some tritiated water vapor can be

diffused to the ground and taken up directly by vegetation. Once

deposited the tritium can infiltrate into the groundwater, remain

as part of the surface runoff and storage, assimilated into plants,

o r t r ansp i red and evaporated back i n t o t h e atmosphere. Because of

i t . s c l o s e as soc ia t ion with t h e hydrologic cycle it has served a s a

use fu l i so tope i n determining t h e c h a r a c t e r i s t i c s of var ious water-

sheds. , .

' 3. SCAVENGING OF TRITIUM BY PRECIPITATION

. 3.1 heo or^ of Scavenging

The process of removal of mater ia l emitted t o t h e atmosphere by

. p rec ip i t a t i on is t r a d i t i o n a l l y invest igated i n terms of below-cloud . .

. ,.. . .

scavenging i n t h e form,

where:

Q(t) = amount of emitted source mater ia l remaining

a f t e r some time t (c i /yr)

Qo = source term (c i /yr)

A = constant scavenging coe f f i c i en t f o r t h e

emitted mate r ia l (sec -')

t = time period (sec)

This approach has been used f o r t r a c e q u a n t i t i e s . o f gases assuming

t h a t t h e gases a r e pe r fec t ly soluble (Engelmann, 1965). However it

has been shown t h a t t h e s o l u b i l i t y of gases - is important (Hales, 1972a).

Since A i s not necessa r i ly constant and can vary considerably i n

time (Dana, e t al., 1972; Hales, 1972a) it i s not useful t o u t i l i z e

t h i s d e f i n i t i o n of washout f o r gases. Assuming a steady s t a t e

condit ion with non-interacting drops and no chemical r eac t ions , t h e

r a t e of change of liquid-phase concentrat ion with height z can be

expressed as

where :

. . - .

'K = overa l l mass t r a n s f e r coe f f i c i en t Y

(mass-ci-?-sec-')

Vt = terminal ve loc i t y of f a l l of t h e raindrop

of rad ius R, where downward i s negative (cm-iec")

x = l o ca l atmospheric gas-phase po l lu tan t concen- - 3 t r a t i o n (mass-cm )

H1 = s o l u b i l i t y (Henry's law) f a c t o r ( in u n i t s

appropr ia te t o those of C and X)

C .:= gas-phase concentrat ion i n t h e raindrop

( m a ~ s - a n - ~ )

r = raindrop rad ius (cm)

I f pe r fec t s o l u b i l i t y of t h e mater ia l emitted t o t h e atmosphere is

assumed (HI = 0), it c a n be shown t h a t t h e i n t eg ra t i on of equation (4)

, leads t o t r a d i t i o n a l forms f o r A.

For gases i n general H I 71: 0, and t h c r c ' i s a re luctance t o being

scavenged; i n regions where t h e second term of equation (4) predominates,

t h e gas can desorb from t h e drop. Equation (4) can be in tegra ted

ana ly t i c a l l y f o r t h e case where H t = constant and t h e r a i n f a l l is

vex t ica l (Hales, 1972a, 1972b) o r numerically f o r H1 = H t (C,x)

(Dana, e t a1. , .1973) .

Computer programs have been developed t o perform t h i s ca lcula t ion.

'These programs f a l l under two categor ies - l i n e a r and non-linear models.

The l i n e a r model can be e f f ec t i ve ly used f o r HTO due t o t h e na tu re of

t h e s o l u b i l i t y of HTO, which l i k e H20 i s a function of temperature

only (Hales, 1972b). ~ n t e g r a t i o n of equation (4) f o r HTO leads t o

ground-level concentrat ions i n t h e r a i n f a l l which depend on several

var iables : plume parameters, downwind d i s tance x (i.e., t ime f o r a

constant wind speed u = x / t ) , temperature, mass t r an s f e r , d i f fus ion

coe f f i c i en t s f o r gas phase t r a n s f e r , and H' f o r l i qu id phase t r an s f e r .

Using d i f f e r en t va r iab le values , plume cen t e r l i ne concentrat ions

of t h e raindrops, Ccl, e f f e c t i ve scavenging coe f f i c i en t s k, and source

s t reng ths Q(x) have been obtained (Figures 3.1 and 3.2). The input

condit ions a r e l i s t e d i n Table 3.1.

The values of k were computed by taking advantage of t h e f a c t

t h a t f o r l i n e a r l y soluble gases, t h e concentrat ion i n t h e raindrops

off t h e cen te r l ine C(y) is

where a i s t h e hor izonta l d ispers ion parameter and y is t h e d i s tance Y

from t h e center l ine . The cross-plume in tegra ted f lux, o r "scavenging -

rate", W, is:

Jdy = J C ~ , & ~ U

TABLE 3.1

Input Conditions Used i n EPAEC Linear Model

Temperature '273,303 K . . . .

source 'Height 30,100,000 m

~ a i n f a1 1 Rate 1,s mm/hr

Wind Speed

Plume Pasquill-Gifford with appropriate sigmas (Smith, e t a l . , 1966)

Mass Transfer Regime Gas-Phase Controlled (Well Mixed Drop)*

Raindrop Spectrum Lognormal ( typ ica l f r o n t a l r a i n )

*There is not much di f ference between gas-phase control led and

stagnant drop t r a n s f e r f o r HTO (stagnant drop l e s s i n magnitude).

Experimental work (Friedman, e t ' a l . , 1962; Booker, 1965) "has

shown HI'O t o behave pr imar i ly i n accordance with gas-phase

control led behavior.

.. -.

.. ...

..

Figure 3.1

Amount of Tritium

in Rainwater

for Equilibrium and Non-equilibrium

Conditions as a Function o

f

Distance.

Figure 3.2

Q(x)/Qo as

a Function o

f el ease

Height (h) ,

Rainfall Rate

(J), and

Downwind Distance

(x).

where J i s t h e r a i n f a l l r a t e . . The e f f ec t i ve scavenging coe f f i c i en t ,

k , is then

Since t he calcula ted values of k a t each x were s u f f i c i e n t l y h i g h , t o

dep le te t h e plume s i g n i f i c a n t l y during moderate downwind t rave l ' t imes,

Q was reduced numerically by

where kg i s . t h e average value of k ca lcula ted a t t h e loca t ion x and

t h e previous locat ion, and i s t h e mean wind speed. The curves f o r

Q(x) i n Figure 3.2 a r e approximate due t o neglect of t h e near-source

region. However, it is evident t h a t deple t ion of HTO by r a i n f a l l

within about 10 s tack heights downwind i s negl ig ible .

3.2 Belod-Cloud Equilibrium Scavenging

In Figure 3.1, C represents t h e raindrop concentrat ion which s q

would r e s u l t i f t h e KT0 i n t h e r a i n i s i n equilibrium with ground-

l eve l a i r concentrat ion:

Since Hs is a constant f o r a given T, C (x) exact ly follows t h e well- e q

known Gaussian behavior of X. I t is c l e a r from Figure 3.1 t h a t

CQJC beyond x = 10 km, f o r a l l reasonable values of h. The e f f ec t i ve e q

equilibrium scavenging coe f f i c i en t i s

where H* is a Henry's law f a c t o r f o r x and C i s wr i t t en i n u n i t s of

3 . niass/cm . he exponential term i s e f f ec t i ve ly un i t y f o t reasonable

values of h and plume parameters beyond x - 10 h.

An a l t e r n a t e expression f o r k a t g rea t d i s tances from t h e e q

source can be obtained by-assuming t h a t

where x i s a constant and L i s an e f f e c t i v e thickness of t h e g 1

l aye r i n t h e z d i rec t ion . I t is a l s o assumed t h a t x = Xgl

at a l l levels . With these assumptions k is equal t o J/(H*L). I f e q

it i s f u r t h e r assumed t h a t H* = pa/pw, where pa i s t he absolute

humidity of H20 and p t he dens i ty of water, and il is t he r a i n f l u x W

-2 -1 i n grams >cm -sec , then

- - R k = - R and J = -

eq paL D

"w

This form of k reported by and Dickerson and Crawford, .(1972), i s eq' .

s imi la r t o equation (10) f o r ground-level r e l e a se s a s long a s d i s tances

from t h e source a r e considerable.

The assumption of H* HTO = H* i s probably q u i t e good, H,O

L

except t h a t it must be recognized t h a t t h e absolute humidity p f o r a

water vapor and HTO is a funct ion o f temperature. A p l o t of H* a s ' a

function of temperature. i s given i n Figure 3 . 3 .

Due t o t h e low f a l l speeds and small s i z e o f cloud d r o p l e t s it

can be assumed t h a t these which e x i s t i n a plume w i l l have HTO con-

cen t ra t ions which a r e i n equil ibrium with t h e l o c a l a i r concentrat ion

of HTO. Therefore, in-cloud scavenging i s character ized p r imar i ly

by t h e pickup of cloud d rop le t s by raindrops. Thus, an add i t iona l

term t o equation (4) is added:

- 3Eb.I (LWCIx az dc I - - 4 r H"'p

cloud drops

where (LWC) i s t h e l i q u i d water content of t h e fog and E i s a

c o l l e c t i o n e f f i c i e n c y which de f ines the . mass t r a n s f e r of cloud 'drops

t o raindrops. By comparison of t h e cloud d rop le t term t o t h e gas

scavenging term, a c r i t e r i o n f o r t h e importance of cloud drop

scavenging can be developed. For t h e cloud drop component t o be

i n s i g n i f i c a n t :

4K H1pw Ratio = >>I o

E (LWC)Vt

For mO, assuming gas..phnse t r a n s f e r , Vt = 500 cm/sec, T = ? R ~ ' K , and

E = 1, t h e values of t h e r a t i o and t h e f r a c t i o n of t h e concentrat ion

TEMPERATURE. OK

Figure 3.3 Henry's Constant (H*) as a Fmction of Temperature.

contributed by cloud drop scavenging can be estimated. Some a r e l i s t e d

i n Table 3.2.

TABLE 3 . 2

Values of Ratio and Fraction of Raindrop Concentration of 'HTO 'Contributed 'by ' Cloud-Drop 'Scavenging

(LWc) Ratio Fract ion 3

0.1 g/m 4 020

For r a i n which forms from t r i t i a t e d water vapor by condensation,

it i s probably s a f e t o assume t h a t C = C because. of t h e time e q

invo lved . in t h e process, a s long a s x > 10 Ian o r so. For t h e case

o f t h e convective shower ac t ing r i g h t a t t h e source point , it is

probably s a f e t o assume t h a t t h e e f f i c iency with which t h e shower

removes water w i l l a l s o apply t o HTO. A very heavy r a i n f a l l r a t e

w i l l a l so deple te t h e plume qu i t e e f f ec t i ve ly by washout alone, a s

Figure 3.2 ind ica tes .

3 . 4 Alternat ives and Limitat ions on Models

The l i n e a r model is not very accurate f o r d i s tances l e s s than

about one s tack height downwind, due t o t h e neglect of t h e angle of

raindrop t r a j e c t o r i e s . The deple t ion wi thin t h i s region w i l l be very

s l i g h t , although t he value of k may be q u i t e high i n a small space

between where desorption predominates ( i . . , with t h e plume well above

t h e ground, so t h a t C a t ground-level i s very low) and where t h e e q

concentrat ion begins t o decrease toward C eq'

The curves of Figures 3.1 and 3.2 can be used t o eitirnate ra in -

water concentrat ions o r scavenging coe f f i c i en t s f o r o ther input va r i -

ables.. The concentrat ion C is d i r e c t l y proport ional t o Q(x) and , c 1

inve r se lyp ropo r t i ona l t o u; k is d i r e c t l y proport ional t o J. here-

fo re , concentrat ions f o r any a n d ' a l l values o f Q(x), J and u can be

predic ted from one curve. To account f o r deple t ion, however, Q(x)

must ,be calcula ted fo r each point x based on t h e scavenging l lhistory ' l

of t h e plume. The scavenging coe f f i c i en t may be determined a t any x

s ince it i s not a function of t h e source term. .

4. ANALYSIS OF TRITIUM CONCENTRATIONS IN THE HYDROSPHERE

4.1 Phase 1: Estimated Average Ground-Level Concentrations Near the Plant . . . . . . . . . .

The technique for computing' average annual ground-level concen-

trations of tritium.within 50 km of a CTR power plant involves evalua-

tion of the relation:

where :

- 3 X a = atmospheric tritium concentration (Ci-m )

Q = source term.to the atmosphere (ci-year-.')

a = 0.5t = horizontal standard deviation of the Y

plume (meters)

- 1/2 a = 2Kz(x/u) = vertical standard deviation of the z plume (meters)

t = travel time (sec)

2 - 1 K = vertical diffusion coefficient = 5 m -sec . Z - u = mean wind speed through the layer 300m to 2000m

abovc ground

k = equilibrium scavenging coefficient e q

x = downwind distance from the source (meters)

y = plume crosswind distance (meters)

The term in the curly brackets in equation (15) was evaluat?d by

Heffter, et al.; (1975) for the LMFBR environmental analysis using a

climatology of wind trajectories near Morris, Illinois. In this

analysis it is also assumed that the CTR will be located near Morris,

. . Illinois. . . . . . . . .

The second exponential term in equation (15) represents the

depletion of the 'plume due to scavenging of tritium by precipitation.

Under equilibrium'conditions, where the in-cloud and below-cloud

tritium concentrations are equal, and where the height of the release

is at ground-level, the equilibrium scavenging, k coefficient is e q

obtained from equation (10):

where

o = Smith-Singer dispersion coefficient = 0'.63x~'~~, 64.4 km Z

= 859 m, x>4.4 km

By assuming equilibrium conditions below the precipitating cloud,

the values of k are within 10% of the k values determined by e q e q

Dickerson and Crawford (1972) for use in the LMFBR analysis at

Morris, Illinois.

The results of the ground-level atmospheric tritium concentration

estimates are shown in Figure 4.1 for Phase 1. The concentrations in

- 1 t.his f i g u r e have heen normalized to a 1 ci-year release. To

determine estimated concentration for any source rate the normalized

concentrations in the figure must be multiplied by that source rate.

. .. . .

Figure 4.1 Ground-Level Air Concentration Values of Tritium for an Area Near Morris, Illinois.

4.2 Phase 2: Estimated Average Ground-Level Concentrations in the Eastern Half of the United States .

This phase of the analysis deals with plume trajectories from

the proposed CTR plant over travel time of several days, or distances

up to 1000 kilometers; Heffter, et al., (1975) treat this problem in

the same fashion as in Phase l,,utilizing a climatology of air trajec-

tories over.the eastern United States and then applying equation (15).

The results of the regional ground-level tritium concentration

estimates are shown in Figure 4.2.

4.3 Average Surface Water Tritium Concentrations near a CTR Plant

The average annual surface water tritium concentrations, Xg '

are determined from:

where :

W = average annual depth of the surface or groundwater

mixing layer, (wyear-l)

2 A = area of precipitation (m )

Since this analysis is intended to provide conservative estimates . ,

of tritium concentrations that would show the maximum expected impacts

for fiormal operation and accidental releases, the value of W was

assumed to be determined from the mean annual net rainfall:

Figure 4 . 2 Ground-Level A i r concent ra t ion Values of Tr i t ium f o r t h e Eastern United S t a t e s .

. . A W ' - = year P - E

where : . . . . . . . . . . . . 3 - 1 P = .amount of precipitation (m -year )

3 '

E = amount of evaporation (m

In other words, there is no groundwater reservoir into which

tritium would be mixed after tritiated rainfall percolates through

the ground. For Morris, .Illinois the annual rainfall averages 9.93

- 1 0.93 m-yr , and evaporation (determined from Sellers, 1965) is . . - 1 - 1 0.56 m-year . Thus the mean annual runoff is 0.37 m-year , and

!V = 0.37 meters.

Equation (17) shows that ,estimates of surface water concentrations

of tritium are made by determining the amount of the source term that ,

is depleted by washout processes. Thus, estimates of surface water

concentrations were made for each concentric areal ring around a

CTR located in Morris, Illinois from the relation:

where :

i = radial distance to inner circumference of a

concentric areal ring

j '= radial distance to outer circumference of a

concentric areal ring

This technique has the disadvantage of assuming that all precipi-

tation systems are distributed equally $round a CTR, along with the

distribution of HTO. This assumption obviously becomes poorer at . . . . . . . . . .

greater distances from the source. In particular, it is obvious

that the tritium washed out by precipitation systems will always be

downwind of the CTR, which in Morris, Illinois will always be in a

prefered section, and not uniformly distributed around the plant.

For this reason the analysis is limited to distances up to 1000 km,

using the hydrological conditions for Morris.

The result of the surface water tritium concentrations using an

it.eration of equation (19) and (17) are shown in Table 4.1. Again

normalized concentrations are provided, based on a normal release of

- 1 1 ci-year . Figure 4:3 shows the variation of surface water

concentrations with distance from the source using the hydrologic

conditions at Morris. The value of a during precipitation episodes z

were obtained from the LMFBR analysis. For Morris, the value of'uZt

increases to a distance of 4.4 km, and is constant beyond that point.

Figure 4.3 reflects this change in the vertical dispersion. It is

evident that the smaller dispersion near the plant acts to increase

surface water concentrations by as much as two orders of magnitude

above what would have been estimated from a constant dispersion.

. .

TABLE 4.1

Surface Water Tritium Concentrations (X ) . . . . . . g

For Various Distances from the Source

Radial ,

Distance (b)

Area

Af* (m2) - -~

*Note: 'Af = (1-f ) - ( 1 . (See equation (19)) x=j X= 1

Firure 4.3 Normalized

Surface Water Concentration

(xg/Q) as a

Function of Distance

(x).

4.4 Phase 3: Estimated Latitudinal and Global Average Atmospheric and Surface Water Tritium Concentrations

A technique was developed to analyze the estimated concentrations . . . .

df tritium'in the atkosphere and. surface water supplies in the north-

ern hemisphere based on a.continuous normal operation of CTR plants

distributed throughout the 30' - SO0 N latitude belt. The analysis

technique is based on the procedure presented by Libby (1963) with

modifications to account 'for sources near ground-level and for the

transport of hydrologic c'omponents across latitude belts. The

technique is essentially a "box model" approach which assumes that

the tritium follows the hydrologic cycle, and represents the simul-

taneous solution of atmospheric and surface water tritium concentra-

tions to establish the maximum values. Y

The balance equation for atmospheric tritium in.any latitude

belt can be developed from an examination of Figure 4.4 and is

written

where :

. .

Qa = source of tritium to the atmosphere, distributed

throughout the latitude be1 t (~i-~r-'),

x = concentration of atmospheric tritium entering the a" latitudinal boundary (~i-m'~) . -

3 -1 P = latitudsaveraged precipitation.(m -yr )

Figure 4.4 Schematic o

f the Box Model

Approach to

Computing Atmospheric and

Surface Water Tritium Concentrations by

Latitude Belt.

3 -1 E =, latitude-averaged evaporation (m -yr )

w = depth of atmospheric water (i.e., precipitable water), m

(w/18) x a = average. radioactivelife of tritium in atmosphere . . " " m3' H o

p = absolute humidity ;.a m Air

2 A = area of 'the latitude belt (m )

'in = .influx of water vapor across the latitudinal boundary 3 - 1 (in -year )

Iout = outflux of water vapor acorss the latitudinal boundary 3 - 1 .(m -year )

Likewise, the balance equation for surface water tritium i's written

where :

Qs = source of tritium to the surface water, distributed -3 -1

' throughout the latitude belt (ci-m -yr 3

= concentration of surface water tritium entering the Xgo

latitudinal boundary

W = depth of readily-mixed groundwater, meters

(W/18)xg = average radioactive life of tritium in the surface water

3 -1 Rin = surface run-off into the latitudinal "box" (m -yr )

3 -1 Rout = surface run-off out of the latitudinal "box" (m -yr )

Over a long period of time (for the atmosphere a.period of several

weeks and for the groundwater a period of several years) each of these

balance equations has the solution:

I .

': Xa Qa + Exg + Iin xa max - - 0

'a. + Iout + Aw/18

If it is now assumed that:

- lout - 'in - (P-E)

then equations 22 and 23.have the simultaneous solution:

The solution of these equations for each latitude belt becomes

quite complex, since; due to the evaporation and precipitation processes,

an inflow of tritiated surface water into' a latitude belt also repre-

sents a source of tritiated water vapor to the atmosphere. Thus the

.equations would have to be solved simultaneously for each latitule

belt by establishing the values of xo and x from adjacent latitude go

belts. As seen from Figure 4.5 obtained from values published in

Sellers (1965), there is a mean northward transport of water vapor,

and consequently (based on the above assumptions) a mean southward

transport of surface water throughout the mid-latitudes of the northern

hemisphere. ,The reverse is true of the tropical latitudes, and the

subtropics,represent a major source of water.vapor and sink of surface I

water. To avoid the complexity of these solutions it is assumed for I

this analysis that

At latitude 30" - 50°N: Xa

= 0 0

All other latitude belts: Qa = 0

This analysis assumes in addition that tritium concentrations are spread

uniformly throughout the 30" - 50" belt, the latitude belt in which

all the sources occur, due to the north-south eddy meridional transport

properties of the traveling cyclones, but that outside this belt the

transport of tritiated water vapor and liquid water is performed

entirely by the mean meridional transport of water vapor and liquid

water. To avoid one additional computational difficulty, a southward

eddy water vapor transport term across the 30" meridian was used to

determine x for the 20° - 30" belt. Values of the parameters a 0

needed to solve equati~ns (28) and (29) are shown in Table 4.2.

Table 4.3 gives the results of the computation for two choices of

W: W = 0.5 m (land areas) and W = 75 m (oceans).

Fig

ure 4

.5

Mean

Merid

ion

al T

ransp

ort

of W

ater Vapor

and

Su

rfac

e W

ater B

ased on

Valu

es P

ub

lished

in

Se

llers

(19

65

)

TABLE 4.2

Parameters Used for Computation of Atmospheric and Surface Water Tritium Concentrations by Latitude Belt (from Sellers, 1965)

Latitude P E in T O C 'a I in

R -

2

ma H 0 0-10 O N 85.25~10'~ m31yr 54.44~10~~ m3/yr 30.81~10~~ m3/yr 0 i5.5 23.1x10-~ T? m air 10-20 4s. 24 59.42 4.13 14.3x1012 m3/yr 25.1 23.0

20-30 31.76 50.69 4.94 18.33 20.4 17.5

30-50 6C.32 ' 56.67 19.14 10.55 11.0 10.1

TABLE 4.3 .

Average Groundlevel Atmospheric and Surface Water Concentrations by Latitude Belt for Laqd Masses (W=O.Sm) and Oceans (W=75m)

Assuming a Ci-year-l Release in Latitude 30"-5OUN

Latitude W=O .5m ~=75rn (ON) x,/c., ',/pa xa/Qa 'g/Qa

The assumptions noted above which are needed to simplify the

complex it^ of the solution result in an underestimate of tritium

concentrations. This is a result of a failure to reestablish new . . . . . . . - .

values of xa and x as concentrations for each latitudinal belt go 0

are made. ' Thus a "backgroundt1 estimate of tritium concentrations

was established by solving equations (28) and (29) globally. Under

this condition I, x x , and R must equal zero and P=E. Thus: a ' go 0

Globally, E = 1.47~10 14m3 -1 -yr . Again, using the extreme values

for W of 0.5 m (land) and 75 m (oceans) the global tritium concentra-

tions can be summarized as follows:

Since these background concentrations are much larger than the

estimates shown for each latitude belt, it becomes evident that, based

on this technique, the gradients of tritium concentrations across the

latitude belts are quite small, despite the total source occurring

in only one of the belts. Thus, Table 4 . 4 summarizes the. estimated

atmospheric and surface water concentrations incorporating this . . . . . . . . . .

"background" value. . . ,

It can be seen from Table 4 . 3 and 4 . 4 that the selection of W,

the readily mixed depth of groundwater, is quite critical to the

computations in the range of 0.5 to 75 meters. This is particularly

true of the surface water. computations. Figure 4.6 shows a plot of

the variation of normalized tritiated water concentration per unit area

with W for the values of P, E, I aid R in the 30"-50" latitude belt.

The variation occurs mostly in the O.lm to lOOm range. Libby (1963)

used a value of W = 0.5m ,for the North American land mass, but

Begemann and Libby (1957) determined W to be 8 meters for the

Mississippi watershed. The value of 0.5 meters over'land was

selected for this phase of the study to obtain the most conservative

estimates of surface water concentrations. Jacobs (1968) shows that

the value of 75 meters for the oceans is acceptable.

A simple box model r .alr . i i lat . inn nver the 30"-50" latitude belt

was made to provide a check on the values given in Table 4.3. By

assuming mass inflow of clean air through the top half of a vertical

column bounded by the 30" latitude on one side and 50" latitude on the

other, and extending to the top of the atmosphere, and mass outflow

of. tritiated air from the bottom half of the column, atmospheric

concentrations of tritium throughout the column can be determined from:

TABLE 4.4

Average Ground-Level Atmospheric and Surface Water Concentrations by Latitude Belt for Land and Ocean

. . Areas Incorporating the Global "Background"

LAND OCEANS - . PCi-m-". pci-I-'

xa/Q ~ i - y r -I xg/Q ci-yr - 1 x,/Q xg/Q Latitude (ON)

* Source latitude belt ** "Background"

Figure 4.6 Variation

of Normalized

Tritiated Water Concentration

Per Unit Area.

. . . .

where:'

C =.average wind vector normal to the latitude belt . n . .

for the outfldw

L = length of the latitude boundary

H = height of column

The value of C was determined from palm& and Newton, (1969) to n

- 1 4 7 be 0.5 m-sec . Using H = 10 m and L = 3.07. x 10 m (40°N):

which is in close agreement with the value shown for the 30"-50" belt

in Table 4.3 (no "background"), particularly for W =.0.5m.

' 4 .5 Summary of Normal Operation Impacts

The information in Figure 4.1 and 4.2 and in Tables 4.1 and 4.4

allow computations of the maximum allowable emissions (MAE) of a CTR

based on the analysis prbcedure used here and the maximum permissable

concentrations (MPC) that have been developed for HTO. For atmospheric

- 3 tritium the MPC is 5 x Ci-m , and for tritium in the drinking -

water supply the MPC is 1 x Ci-liter. Thus a value for the

maximum allowable emission for a CTR is the ratio of the MPC to the

normalized atmospheric or surface water concentration estimate:

, : MAE = MPC xa/Q or xg/Q

A,suMary of maximum.allowable emissions based o,ri atmospheric

and surface water estimates is given in Table 4.5. It is evident

from this table that atmospheric concentrations represent the limiting

factors for a CTR operation near the plant, while surface water concen-

trations represent the limiting factors at great distances from the

plant. However, it is also evident that 1arge.quantities of tritium

releases could be tolerated according to this analys.is - quantities

that far exceed the emission currently expected by a CTR technology.

TABLE 4.5

Summary of Maximum Allowable Emissions [MAE) of Tritium to the Atmos~here by a Normal CTR

Operation Based on Atmospheric (x,) and

Phase

Surface Water (x,) Concentrations

4.6 ~ccidental. ele eases of Tritium

One of the.major concerns in a CTR accident situation is that

large quantities of tritium will be released into the environment due . . . . . .

to a liquid metal fire burning through the containment vessel. The

. impact of'such an accident on the atmosphere and surface water environ-

ment can be analyzed in two ways -- by describing the worse (i.e., most

stable) mete'orological conditions and thereby maximizing the ground-

level atmospheric tritium concentration calculations, or by describing

a situation'where a large portion of the accidental release will be

scavenged by a precipitation system and mixed in with the surface

water supply almost immediately. Since surface water concentrations

could ultimately affect a large part of the general population, this

aspect of an accidental release will be analyzed here.

The analysis procedure involves the same basic approach that was

used to estimate surface water concentrations during routine operations.

This approach, assuming equilibrium conditions, will produce conserva-

tive results for distances near the emission source (See Figure 3.1).

The relation for determining surface water tritium concentrations, xg,

is given by

where :

Qt is the total amount of tritium released in the

: accident. (grams or, Ci)

k is defined by equation (16) e q . .

J -is the rainfall rate during the accidental release

W=Jt (no mixing with surface or groundwater systems)

t is the length of the storm period

The value of J was determined from rainfall frequency curves

published in Linsley, et al., (1958). For Phase 1 the maximum

expected rainfall rate within a 100-year period in the vicinity of

Morris, Illinois is 75 mm-hour-'. For Phase 2 the maximum daily

rainfall expected within a 50-year period near Morris, Illinois is

-5 125 mm-day-'. For this analysis, H* = 1.05 x 10 (T.286.7 OK),

- 1 o = 859 meters, u = 4.05 m-sec , and t = 3600 sec for Phase 1 Z

5 and 8.64 x 10 sec for Phase 2.

Results of the Phase 1 and 2 accident analyses are given in

Table 4.6. Based on these rainfall rates any accidental release of

tritium would be scavenged almost immediately within a distance of 10

kilometers for Phase 1, and within a distance of 150 kilometers for

Phase 2. Assuming the values of x /Q in Table 4.4 are representative g a

of normalized concentrations that would be found in the drinking water

in the plant vicinity after an accident, the maximum permissable.

concentrations would be exceeded if an accidental release of greater

than 68 grams under Phase 1 conditions, and 128 kilograms under Phase 2

conditions. Thus a postulated 10 kgm accidental release of tritium

could result in severe environmental impacts, particularly in the

immediate vicinity of the plant, under these meteorologic conditions.

The effect of rainfall rate on surface water concentrations at

various distances from the site is shown in Figure 4.7. Since no

groundwater mixing is assumed a large rainfall rate actually represents

greater dilution, with highest concentrations occurring for, light j

rainfalls. However many other factors enter into these computations

including the manner in which the scavenging coefficient is computed

(See Section 3), and the depth of surface water available for mixing

with rainwater. The figure also shows the concentration estimated

for a postulated 10 kgm release of tritium.

. Although the results of an analysis such as this emphasize the

need to develop every possible safeguard against a CTR accident, there

is also a need to develop estimates on the probability of such

accidents occurring under the conditions described above. In addition

there is a need to investigate further environmental conditions that

might e x i s t which could result in even greater impacts from a CTR

accident. These investigations require a detailed statistical

evaluation of climatological data, as well as a complete analysis of

potential CTR accident situations.

TABLE 4 . 6

Surface Water Concentrations of Tritium During Occurrence of Maximum Rainfall Rates Following a CTR Accident

Phase 1 (5.75 mm-hr") Phase 2 (J=125 mm-day")

* See Table 4.1 for explanation

Figure 4.7 The Effect of Rainfall Rate on Normalized Surface Water Concentration.

5. CONCLUSIONS AND FUTURE RESEARCH NEEDS

This analysis has shown that the normal operation of a commercial CTR

in the vicinity of Morris, Illinois should have a small impact on the

environment in terms of increased tritium concentrations in the air and

surface water. This, of course, assumes that CTR operations do not emit

quantities of tritium exceeding lo3 kgm per year. In addition, the limits

to these normal tritium releases to the atmosphere appear to be controlled

by atmospheric concentrations in the vicinity o f the plant (Phases 1 and 2).

However, as tritium becomes an integral part of the hydrologic cycle (Phase

3) surface water concentrations become the limiting factor. Although the

total body dose is determined both from atmospheric exposure and ingestion,

the maximum permissable concentrations in drinking water are much more

stringent than in the atmosphere. Furthermore, the assumption that all

tritium releases will be in the form of tritiated water vapor places a

higher constraint on the allowable emissions, since the ICRP-recommended

occupational maximum permissable concentration for elemental tritium in air

is 400 times higher than for tritium oxide in air (Young, et al., 1975).

The analysis also considered accidental releases of tritiated water

vapor into the atmosphere coinciding with an unusually severe precipitation

event, Assuming no mixing with groundwater, this analysis showed that .

maximum allowable concentration standards of tritium in the surface water

(and thus drinking water) supplies would be exceeded by 1 to 2 orders of

magnitude if the estimated accidental release of 10 kgm of tritium is

postulated.

Several topics requiring additional research have been identified . .

during the completion of this analysis. More sophisticated modeling

techniques for .transport over several hundreds of kilometers are required,

particularly if complex terrain and land-sea contrasts are to be included.

For example, some .success has been obtained using numerical boundary layer

models to estimate tritium concentrations in the vicinity of the Savannah

River Laboratory (Kern, 1974). The manner in which HTO enters into the

hydrologic cycle, and the similarities between HT0,and H20 need further

quantification. This becomes particularly important when examining the

scavenging process of HTO by precipitation, and the subsequent return to

the atmosphere via the evapotranspiration process. Research into the

probabilities of large acc.identa1 releases during unfavorable meteorological

episodes needs to be undertaken.

Despite the fact that this initial analysis showed normal releases of

tritium due to a commercial CTR technology to have a relatively small

impact on the environment, the need for stringent control of CTR tritium

releases must be emphasized. In addition, developing a CTR technology

could produce large and perhaps serious secondary impacts on the atmosphere

due to massive construction programs, industrial development, and iarge

influxes of population into unpopulated regions. In addition, considerations

must be made to avoid serious emissions of tritium when a'commercial CTR

facility is finally decommissioned. Above all, stringent controls must be

pursued to avoid large releases of tritium during an.accident situation.

6. REFERENCES

Begemann, F. and W. F. Libby, 1957: Continental Water Balance, Ground. Water Inventory and Storage Times, Surface Ocean Mixing Rates and World- Wide Water Circulation Patterns from Cosmic Ray and Bomb Tritium. Ceochim. Cosmochim. Acta 12(4): 277-296.

Booker, D. V., 196.5: Exchange Between Water Droplets and '~ritiated Water Vapor. Quart. J. Roy. Met. Soc. 91: 73-79.

Dana, M. Terry, J . M. Hales and M. A. wolf, 1972: Natural Precipitation, Washout by Sulfur Dioxide. ~ ~ ~ ~ 1 3 8 9 , Battelle, Pacific Northwest Laboratories, Richland, Washington. j

Dana, M. Terry, J. M. Hales, W.G.N. Slinn, and M. A. Wolf, 1973: Natural Precipitation Washout of Sulfur Compounds from Plumes. EPA-~3-73-047, Battelle, Pacific Northwest Laboratories, Richland,'Washington.

Dickerson, M. H. and T. V. Crawford, 1972: In-cloud Precipitation Scavenging of Tritiated Water Vapor. UCRL-51283, Lawrence Livermore Laboratory, Livermore, California.

Engelmann, R. S., 1965: The Calculation of Precipitation Scavenging. BNWL-77, Battelle, Pacific Northwest Laboratories, Richland, Washington.

Friedman, Irving, Lester Machta and Ralph Soller, 1962: Water-vapor Exchange ~etween a Water Droplet and Its Environment. J. ~eo~hys. Res. 67(7): 2761-2766.

Hales, J. M., 1972a: Fundamentals of the Theory of Gas Scavenging by Rain. Atmospheric Environment 6: 635-659.

Hales, J. M., 1972b: Scavenging of Gaseous Tritium Compounds by Rain. BNWL-1659, Battelle, Pacific .Northwest Laboratories, Richland, Washington.

Heffter, Jerome L., Albion D. Taylor and Gilbert J. Ferber, 1975: A Regional-continental Scale Transport, Diffusion, and Deposition Model. NOAA Technical Memorandum ERL ARL-50, Air Resources Laboratory, Silver Spring, Maryland.

Jacobs, D. G., .1968: Sources of Tritium and Its ~ehavior Upon Release to the Environment. U.S. Atomic Enerev Commission. Division of Technical ". Information, 90 pp.

Kern, C. D., 1974: A Simple Model to Determine Mesoscale Transport of ~irborne Pollutants . proceedings of the ~ym~osium on Atmospheric Diffusion and Air Pollution, Santa Barbara, California, September 9-13, 1974.

Libby, W. F., 1963: Moratorium Tritium Geophysics. J. Geophys. Res. 68(15) : 4485-4494. .

Linsley, Ray K., Max A. Kohler and Joseph L.H. Paulhus, 1958: ~ ~ d r o l o ~ ~ '

for Engineers. McGraw-Hill Book Company, Inc., .New York. 340 pp.

Machta, Lester, Gilbert J. Ferber and Jerome L. Heffter, 1973: Local and World-wide Pollutant Concentrations and Population Exposures from a Continuous Point Source. Air Resources ~aboratories ,- National Oceanic and Atmospheric Administration, Silver Spring, Maryland, 20910.

palme;, E. and C. W. Newton, 1969: Atmospheric Circulation Systems. I

Academic Press, New York. 603 pp.

Sellers, William D., 1965: Physical Climatology. The University of Chicago Press, Chicago. 272 pp.

Smith, M. E .. and I. A. Singer, 1966: An Improved Method of ~stimatini Concentrations and Related Phenomena from a Point Source Emission. J. Appl. Meteorl. 5: 631-639.

Young, J. R., et al., 1975: Information Requirements for Controlled Thermonuclear Reactor Environmental Impact Statements. . BNWL-1883, Battelle, Pacific Northwest Laboratories, Richland, Washington:.