ML20058N784
| ML20058N784 | |
| Person / Time | |
|---|---|
| Site: | University of Michigan |
| Issue date: | 08/06/1990 |
| From: | Fleming R MICHIGAN, UNIV. OF, ANN ARBOR, MI |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 13, NUDOCS 9008150073 | |
| Download: ML20058N784 (4) | |
Text
y
> bl.
~ Nuclocr Roccter L baratery
@[ ' '
Tha Univoralty of Michigtn-
.w
. ;.[ '
Ford Nuclear Reactor i
M Phoenix Memorial' Laboratory M
2301 Bonisteel Boulevard U
. Ann Arbor, Michigan 48109-2100 I
(313).764-6220 August 6, 1990-i Docket 50-2
.bicense R-28 5
United States Nuclear Regulatory' Commission t
l.
. Attn: Document Control Desk Washington, D.C.
20555 Reportable Occurrence No. 13 l
Tritium Content in.the Ford Nuclear Reactor Heavy Water Tank in" Excess of 50 Curles Gentlemen:-
This letter is submitted as required by Ford Nuclear. Reactor Technical Specifications, Section 6.6.(2).a.
Heavy Water Tank Analysis.
j Ford Nuclear' Reactor Technical Specifications require that the tritium content of the facility's 46 gallont heavy water tank, located adjacent to the north face.of the reactor ~ core, be maintained less than 50 curies.
Tritium is produced by-neutron absorption in deuterium when the reactor is operating.
At i
l intervals of two to three months, five gallons of tritiated heavy water are removed from the heavy water tank ~. After.the removal, five gallons of fresh heavy water are transferred into the-tank.,
Thus, with each transfer procedure, the tritium activity in.the l
tank is reduced by a factor'of 41/46.
t At the time of transfer, samples of the tritiated-water are taken and analyzed to provide an evaluation of the total tritium activity in the tank.
On July 21,.1990, the analysis.of six samples of tritiated heavy water resulted in an averageLyalue of 50.4
+/-.1.4 curies in the tank before the transfer.
The result j
for a heavy water tank analysis following the previous heavy j
water transfer, performed on' April 13, 1990, was 42.5 curies. The.
Jump to 50.4 curies was unexpected and-inconsistent with the effective full power hours of reactor operati'on between the two heavy water transfers.
I i
s q
- 8a*188aadjig2 (pz S
./.
t 0
- ?
United States Nuclear' Regulatory Commission August ~6, 1990 Density Value. For Tritium Concentration Determination In th'e process of evaluating the beavy water problem, 1
a review of the heavy water tank analysis procedure, Health J
physics Procedure 107, was initiated by the reactor operations-staff to determine the cause of,the large apparent-increase in j
measured tritium activity.;
A-second prob.'em was uncovered involving the density of heavy water that-resulted in an 11 percent systematic error in-the calculation of. heavy water tank l
tritium values which invalidated data back to April, 1985.
In April,-1985, theLheavy water analysis procedure u s modified.
The sampling-of tritiated heavy water was changed from a volumetric to a gravimetric basis to obtain greater precision.
1 One step near the end of the procedure converts the mass of heavy l
water-to volume.
This is done to determine the tritium-l concentration in curies /m1'which'is then used to find the total i
tritium content in the 46-gallon heavy water tank.
The procedure specifies use of the density of light water;.1.0.gm/ml.
The density of heavy water, 1.11 gm/ml.should have been specified; a non-conservative, systematic error of 1 11 resulted.
Thus the 50.4 curies in the July 21, 1990 analysis should have been
= 55.9 curies. The same calculational error resulted-l (50.4*1.11) in greater than:50 curies in nice of 25 tank analyses dating.back to April 12, 1985'.
All of these tank analyses were-less than 50 I
curies based on original measurements and calculations.
After the heavy water transfer of five gallons on July 21=,
- 1990, the net activity in the tank was reduced to (55.9*41/46)
= 49.8-curies.
The density error in the procedure was detected after-July 21.
In order to further' reduce the tritium activity in the-heavy water tank, another five gallon-transfer was made on August 1
3, 1990.
Discounting the small amount of. tritium buildup during the operating period from July 21, 1990 to August 3, 1990, the tritium activity in the tank on August 3 was (49.8*41/46) or 44.4 4
curies.
j Corrective Action The' apparent increase in tritium activity between April 13, 1990 and-July 21, 1990 from 42.5 curies to 50.4. curies was operationally inconsistent.
Tritium, with its 12.33 year.
l half-life, builds up almost linearly with operation at 2 Hw.
The I
analysis performed prior to the-April-13, 1990 analysis was done i
on February 2, 1990.
The' tritium activity was measured as 48.0 euries at that time.
The transfer process, including the addition of 5 fresh gallons of heavy water, would have reduced the heavy water tank tritium activity to (48.0*41/46)
= 42.8 curies.
Subsequent operation would have increased the net curie content of the tank, not reduced it to 42.5 curies as indicated by the April 13 analysis.
The April 13, 1990 analysis had to be in error.
A L
Page 2 i
i w
United States Nuclear Regulatory Commission August 6, 1990 trend study over the last five years shows that tritium buildup averages approximately 4.4x10-3 curies /2Mw-hr of reactor operation.
This general guideline will be used in the future by
- ,c!or operations e rsonont to track triLlum buildup as a l
tenwerot totei a!
w.,J.
to serve as a general check on the heavy water tank tritium analysis performed by the llealth physicist.
i llcal th physica procedure 107 has been corrected to reflect the I
correct density of heavy water, 1.11 gm/ml.
In an attempt to check the accuracy of our analytical results, samples of tritiated heavy water have been sent to the National Institute of Standards and Technology and to KMS Fusion in Ann.
Arbor for independent analyses.
Safety Implications The activity limit in the heavy water tank is based upon a hypothetical rupture in the tank, total mixing of the heavy water with the reactor pool volume, evaporation of tritiated pool water.
and discharge of that tritiated water to the environment out the-reactor exhaust stack.
I If the 50 curies in the heavy water tank were mixed with the 48,000 gallon pool volume, the average concentration in the pool water would be 0.28 microcuries/ml.
The measured evaporation rate from the reactor pool is 3.33 gallons /hr.
On the assumption that tritiated water evaporates at the same rate as light water, the R
tritium evaporation rate from the pool would be 58.8 microcuries/ min.
The reactor building exhaust stack flow rate is 10,000 cfm which is 2.83x106 ml/ min.
The average tritium concentration in the reactor building exhaust stack, assuming the i
tritium uniformly mixed with building air and was allowed to exhaust unrestricted from the reactor building, would be 2.07x10-7 microcuries/ml.
The Technical Specification stack dilution factor is 400.
Therefore, the maximum ground level concentration would be 5.19x10-10 microcuries/ml.
From 10CFR20, Appendix B, Table II, Column 1, the maximum permissible tritium concentration (MpC) for the general public is 2x10-7 microcuries/ml.
The full release of-the heavy water tank to the reactor pool could ultimately result in maximum concentrations outside the reactor building of 2.6x10- 3 MpC.
Exceeding the 50 curie tritium limit indicates lack of administrative control, but does not pose a significant safety hazard to the public.
Sincerely, Ronald F.
Fleming Director 1
page 3
d.-
"! l '^
+
v.
' ' Qlg '
?-
. +
ae x
United States'NuclearJRegulatory Commissi'on=
~~ A u g u s t' 6, 19905 xc tc i_Dr. -William-C.
Kelly,nVice President for-Research 1 Hark ~Driscoll,'-Radiation Safety; Officer; KenSchatz1e,-. Director,: OSEH
-Director,1 Reg' ion JII.nUnited' States' Nucle ~ar Regulatory Commission < _
Safety Review. committee = Members.
(
.i i
1
- }
..3 1
4
'k i
4 ill z
Page 4
_ l,;,
4 y
'. _ -