ML19289G124
| ML19289G124 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 06/07/1979 |
| From: | Read J Office of Nuclear Reactor Regulation |
| To: | Mattson R Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 7906260496 | |
| Download: ML19289G124 (8) | |
Text
-
4 oat.
ROUTl?iG AflD TRAfiSMRTAL SLIP gm 1 3 7y, 70: (Name, of" Ice symbot. room number, initia!s D:te building, Ag:ncy/ Post) 3.
f4RC PDR
/
Assessions Unit (P-50) [
2 3.
4, S.
Action Flie Nete and Return Approval For Clearance Per Conversation As Requested For Correction Prepare Repfy Circufate For Your informat;on See Me Comment Investigste "Jgnature Cecrdination Justify REMARKS TO BE PLACED IN f4RC PDR O
DO NOT use this form as a RECORD of approvals, concurrences, disposais, j
clearances, and similar actions FROM:(Name, org. symbot, Agency /Pcst)
Room No.
n;dg.
P-ll22B 2
Robert L. Tedesco, L Task Force,
rhone No.
Tt11-2 i X28090 A 1-102 o;iTioNAL FORh4 41 (Rev. 7-76)
AC 3
II #
- c.s.cro:ts77-o-241-s30/322e 252 ir7
/
7906260/'fg
>/
c j!
l C_
~
- p,e v e c,,' <,
U NI I L t> s r/i r t s
[ }... (/[ h NUCLE AR REGULA TORY COMMISSION
^
' $'. c w asm no r ou. o. c. ms
,".'.s' i~
r s
j La s
- .,l5 (f c g. <. r
. h (:n
%w-r
.../1 Q,, e,\\ c i 4,'-
JUN 7 U19 n
[ d }"4\\
MEMORANDUM FOR:.40ger J. Matt;on, Director, Divisian of Systems Safety fi 77 /J.7" '
THRU:
R. W. Hot.ston, Chief. Accident Analysis firanch, DSE
/v
FROM:
J. B. J. Road, Secticn A, Accident Analysis Branch, DSE SU3 JECT:
CHEMICAL LESSONS LEARNED FROM TMI-2 The I:11-2 incident doncnstrated that tuo elements, iodine an<J hydrogen, are core easily released from the reactor vessel than had been supposed.
Cheaically, the simultaneous control of both volatile iodine species and hydrogen gas within a containment atmosphere contains inherent contradiction, since hydrogen is a reducing agent and iodine is an oxidizing agent.
The dangers from both hydrogen and iodine arise because they may be oxidized by oxygen gas in the containment atnasphere.
The attached discussion revic.s the role of oxygen and hydrogen gas in iodine chemistry.
Its thesis is that it is nore advantageous to control oxygen in a post-accident containment atmosphere than it is to attempt to control hydrogen, and that it nay also prove to be easier.,
q
/
M/
, g an %
J. B. J. Read, Section A Accident Analysis Branch Division of Site Safety and Environmantal Analysis cc:
R. !!. Houston J. B. J. Read W. E. Kreger J. T. Collins T. Murphy L. G. Hulman AEA kI
CONTROL 103INE REUDX POTENTIAL Mathods for the control of radio-iodines after accidantal release at present invohe adsorption on charcoal or chemical reduction to the soluble iodide ion.
Other raatho'Is, useful in the laboratory but not feasible following an LWR acc-lent, are reaction with starch, the for:ution of cupreus iodide, and ti'e form-ation of transition retal iodo-co plexes.
To scca extent these other mtl.c is will re.ove some iod?ne in a real iodin $ release (" plate-oJt"), but the rathods do not lend tiier.selves to use in engin ered s af ety feat ures.
Reduction of iodine to iodide in aqJeous solution M3y be assured eilher by supplying a reducing agent, such as hydrazine or dissolved hydrogen, or by raising the solution pH such that water itself is able to reduce iodine.
The table below lists the relevant standard reduction potentials.
.n.
6 l -
_ TABLE I reduction half-cell reaction to, stcmf ard Elf NH2S
+ 3H + +2e-4 2 ;H +
+1.24 Volt 4
4 02 + 4H+ + 4e -; 2H O
&1.229 2
10
+ 6H+ + 5e 4 hI2 + 3H O
+1.20 3
2 I3
+ 2e 4 31-40.5355 I2 + 2e -+ 21-10
+ 3H O F G e ->
l
+ 60H-
+0.26 3
2 2Hf + 2 e -> H 0.00 (by definition}
2 N H4 + 40H-
-1.15 2 + 4H O + 4e ->
N 2
2 n.
Much as the conduction electrons in metals possess an electrostatic potential, so the electrons in chemical species possess chemical potentials.
This poten-tial expresses the ability to perform or accept chemical work by electron trans-fer, i.e., to undergo chemical reaction, When a chemical species accepts an
'l hen a species loses electrons electron or electrons it is said to be reduced.
it is said to be oxid'ned, even if the process does not involve oxygen. Although free electrons usually are not involved in redox, or oxidation-reduction reactions, the convention is to write these reactions as half cell reactions With the electrons explicit. Thus the observed reaction, A + B 4 C + D, in $hich A is reduced and B is oxidized can be written as the difference between two reductions.
A + n e -> C
- ( D + ne
--)
B)
- t. + B, 9 C, + D '.
The relationship between the concentrations of species and the redox potential was derived by Nernst in 18S9.
1.98 X 104(M)
E=E l0910 ETM y
LJ>
lJU
-3_
Eicautal iodine,1, is only 4 3ringly soluble in '.;ater, and is present as the 2
very soluble tri-iodide ion, 1
,..henever iodide is also present.
3 1-12 + l~
--)
3 To keep iodine non-volatile, the redox potential of a soluticn must be maintain-edsufficientlyreducingthatthefactor(I- / h-3 in the applicable f:drnst 3
equation is as low as possible.
For convenience, we may take the sump solution tc.;arature as 50 c (about 120 F). To appreciate the relative inportance of pH 0
0 and dissolved oxygen, it is instructive to write do..n the :ernst equation for the oxygen-iodide reaction.
~f E = 1.T B V.
O + 'tH 4'l i -1 1\\\\ o o
t A~
3 I.t 9 2~ % GI~
LQ : 0.5$5g S
og t{pt (>l M A N O + U-O. @ W.
E t
t 3
o:
g t
_t
+ o, 6c3 __ O. OI b lo g ' n O) [I u
tr :
- ]Od %
0 At equilibruim, E o, and in equilibruim with air, the partial pressure of oxygen is 0.2 ata.
In water, the concentration of water is 55.5 miles per kg. The resultsent iodine partitions are markedly dcpendent upon pH.
[Wl/DT
- o 14 q.ly/g3 k[/d 5
as ff 'b % / O b
9.1% / b l
j r, >
b Q.\\ 'ilO In brief, the addition of base to a boric acid soluticn can reduce the volatility
~
of iodine by a factor of a oillion.
Dissolved Hz gas is even more effective at reducing iodine.
-t T 3 + ll a 3 I + 2 H E = 4 o,5355 V.
t o
4-partial pressure was of the order of 1ithin the TMI-2 primary ccolant the Up 70 atm, and even at 600 K and pH = 5, iodine is very strongly reduced.
0 3
- 'l Ylb W
[r]S It follows that in the presence of steam-zirconium reactions, even under d
and high teiperature conditions, the 6ydro[en produced will do.ainate the re of iodine species.
As can be seen from the table, hydrazine nay act as either an oxidizing or reducing agent, depending upon pH. Hydrazine is a very,,eak base,
~l 9,1l3 + Fi
'> Mjg k'6 8,5y/D kg/%q i
t T
f the oxidizing and, except in very strong acid solutions, the concentration o For the reaction of hydrazine and iodine.
form, f1 H *, is negligable.
2 5 3
t 10H + g NJh ~3 3[+{Ng11(O P,2 (, G9 V.
t t
t;ominal In equilibruja with air, the nitrogan partial pressure is 0 8 atm.
hydrazine concentrations are 50 ppm, and at pH=5 contairaent spray Er~
--30 kf/4v-tg-1-
1 Iy/o
=
-]3 i
d
~
Briefly stated, the addition of small amounts of hydrazine is orders of d many core more effective than hydrogen gas in keeping iodine non-volatile, an ilibruia with air.
orders of magt,1itude more effective than simple pH control in equ43 In a 3 X 10 m
';ote, ho'..ever, that iodine is not the only oxidizing agent present. 2 5 males of oxygen, and at the very most only 10 contair, ment there are 2.7 X 10 is J I-
roles of 2
[ J l
t O g 4 H++'(O HUNj't-) 6M 0a M e = 7,g g U, 1
_3 It is interesting to note that the oxidation of hyJrazine yields one role of nitregen for every nole of oxygen consumed.
The chemical kinetics of the hydrazine-iodine reaction are very rapid, due to the low dissociation energy of iodine.
The dissociation energy of oxygen, hc.iever, is very large, 50 the hyJrazine-oxygen reaction is usually quite slow.
The equilibruim concentration of hydrazine it. a solution in contact with air is 2 ppb.
To be at equilibruim 6
with a E0 ppm hydrazine solution, the nitrogen / oxygen ratio must be 10.
liydrazine containment spray solutions are used in P'.G's in conjunction with soluble bases in the sumps.
Du' ring the injection phases, spray droplets of boric acid and 50 ppm hydrazine are used to wash iodine from the contairment atmo;phere. The boric acid is stored under air, and hence contains about S ppm of dissolved oxygen, and is similarly saturated in dissolved nitrogen. Within c
a few seconds after hydrazine is mixed with the boric acid solution, the oxygen is re oved by reaction with the hydrazine, and the sp.ay solution appears at the spray nozzles at the top of the containment depleted in oxygen but super-saturated in nitrogen, and with its hydrazine concentration reduced by 8 ppm.
The spray droplets, on f alling through the containment atmosphere to the sump, are 'iarr;ed by condensing steam, evolve nitrogen and absorb iodine vapor and oxygen.
The warm droplets arrive at the sump in equilibruim with the containment atmosphere, containing about 6 ppm of dissolved oxygen.
During the residence of the solution in the sump, the dissolved oxygen is depleted and nitrogen is evolved, qn,,an,l,;y,
-s-
~
equal count.
At a spray flow rate of 3,500 gpm (12,400 kg/ min), the half-life of volatile iodine in the sprayed contairmant atmospbere is only a few minutes.
- Oxyg2n, ho',;ever, is rcoved from the containmant atnosphere at the rate of about 2 6 kg of hyJrazine-boric acid, the iodine has
~
noles/ min. After injecting 2 X 10 been effectively removed from the contairmnt at:.;osphere, and 12 kg of oxygen cs well, or 375 moles.
The solution, now slightly alkaline from dissolved base in the sump, is recirculated, on each recirculation the cycle of resaturation in oxygen, oxygen dcpletion and nitrogan evolution in the sunp
, and further reduction of the hydrazine concentration is repeated.
After about 10 cycles, or one day of operation, the hydrazine concentration approach 1s equilib ruim at about 2 ppb, and the contain unt oxygen would be reduced by about 3 X 10 3 coles. By this tirae, of course, the soluble base will be controlling th e iodine redox equilibruin.
The volatility of iodine will increase, hc,; aver,,;han oxygan destroys the hydrazine and redoainates the redox potential of th e containment For long-term iodine control, the hydrazina ast be protected f supp.
rom air axidation.
To grossly simplify:
in controlling iodine releases, hydrogen is a friend and 5:
oxygen is the ennay. Rather than attmpt to maintain H2 partia! pressures balow.
.04 atn by mains of Icw flow-rate recombir,ers, it would be preferra the oxygen partial pressure to below about 0 04 aim as quickl y as possible.
1l2ter radiolysi; ahtays produces hydropn gas In the presence of a reducing agent, such as hydrogen, the production of oxygan is inhibited by the rapid destruction of hydroxyl free radicals.
Rcaoving oxygen from tha containment, therefore, is permanent, <;hile remval of hydrogen nust be
';3ter radiolysis continues.
continued as long as The violence, or shock,; ave aptitude, of hydro pn-oxygen explosi ons depenis upon both the energy density and speed of reaction density. Since oxygen und not hydrogen doninates the rate of reaction, the violence of a potenti l a ex-plosion within the concentration range of interest is approximately pro r onal to the square of the oxygen partial pressure, while only lin ear in hydrogen pre-scure.
CO?l TROL rF OXYGEN
, t i, - ) )I'*
J l"~
Catalysts f or increasing the speed of the reaction of oxy;2n cnd h d y razine are available, but the rate of oxygen removal is limited by being a heteroganeous
- reaction, i.e., a gas must dissolve in the hydrazine solution.
A preliminary investigation indicat.j that a courter-current extractor, i.e., a packed colu an of high surface area, with da.;n.sard flowing 35% catalyzed hydrazine and upward fic. zing containment gas, could be built uiih nuch higher gas flow rates than cu-rrent hydrogen-reccmbiners, l'se of gaseous hy& azine or direct spraying of concantrated hydrazine solution into the contain: ant appear inadvisable for a c
variety of reasons.
These methods, t,hile faer than present recombiners, are still too slow to countei(act hydrogen release rates that can be postulated, ho.iever.
The most rapid oxygen-depletion nethod available appears to involve the follew-ing el ments:
1.)
jet turbine engines within contain: ant, fueled by hyd. ogen gas through small diamata piping fro:a a tank form.
2.)
nitrogen gas tankage for purposes of relieving vacuum caused by Oxygen depletion.
Jet turbines can operate at very low partial' pressures of oxygcn, have lar' flow rates, and do not require danisters or other prior intake treatments.
Large displacemant diesel engines could also be used, although intake demisting and po,;er input at low oxygen partial pressures would be required.
'lery rapid F.aans of oxygan r a val, for exampic by corbustion of magnesium, are problematic.
They cannot be safely tested, inadvertent act uation is dangccous, ad if delayed in an emergency use, or if ccobined with a nitrogen vacuun-breaker failute, they could endanger the containment.
An extra knefit of the capability of inerting the contain 2nt uuuld hbLde ability to use graphite as a structurpl natorial halow the cactor vessel.
Graphite would be preferred,,:_re it unable to burn, since its high to yerature reaction with water is endotherc,ic, and its high heat of sublimtion of fers an attractive cooling..echanism.
-E-The in!ustry and faC view that hydrogen generation is a hazard, inar se, in accident scenarios is raisplaced.
For a variety of reasons, outlined above, it is the presence of oxygin t.hich constitutes the threat under accident condition <,.
/,
}
..s-a
O P
V-
'I w
}
l s c-4 79 -1 b I b ty,ngo,i, 6/5/79 s.
NRC CLCRE T ARI AT TO O comm.ss,oner o ie XW c,,c o,, fog,,
O con counsei O cong t son O soie,io, O Public Aff. rs O secretary O
Americo Lazzari incom ng f rom Phoenix Arizona ggHendrie o,,,_5/29/79 To nuclear power sun,ec, e
d O
e,,p re reni, eo, s 9n.iure o, 0 cn..rm.n O comm.ss.oner O E oo. cc. co. sot. PA, sECY O s.9nrurebioc,om iieo O
O R, turn or.9.n. or.ncoming..in r,soons.
O ror e.reci reni,-
ror appropriate action O ror, formation nc e_ : h77 O ror rec.ommena.i.on C-
.A.< '.
Y JL )
j I.
5 1
-1 i
/_
~.J flH how 2 I hb&s. _ d Bem.rk s ia
_, g..I. y, -
3d*%
b Y )be L_
f,W H
I
^
M pCQDbS g _* y db f or the Commit enn 4{
- send three 13' copres of reply to secy M.il Facility nc e2 ACTION SLIP
.. J m)
) lo ]/
)L r