ML20077P929
| ML20077P929 | |
| Person / Time | |
|---|---|
| Site: | Fort Saint Vrain  |
| Issue date: | 08/15/1991 |
| From: | PUBLIC SERVICE CO. OF COLORADO |
| To: | |
| Shared Package | |
| ML20077P923 | List: |
| References | |
| NUDOCS 9108200134 | |
| Download: ML20077P929 (19) | |
Text
.
VERIFICATION OF TSV ACTIVATION ANALYSIS August 15, 1991 Work Performed by:
/)Mff,,Epc I/ /,ftR B4W /
E 9,g, -
Approved by:
7/\\N) '/dth AA//?
ft////9/
f Z'f;
/ Date 9100200134 910815 PDR ADOCK 05000267 P
~ _ _ _,. -
t TABLE OF CONTENTS Page l
1.
Introduction........................
1 II.
Neutron Flux Verification 1
A.
Wire Specimens.....................
I B.
Charpy Specimens....................
2 l
C.
FCRV Tendon Wire Samples................
3 Ill. Material Composition Verification 5
A.
Sensitivity Study of Trace Element Abundances in Concrete........................
5 B.
Assumption of Homogeneous Concrete /Rebar Mixture....
6 i
C.
Concrete Sample Data..................
7 t
IV.
Results and Conclusions 8
}
V.
References.........................
9 l
t Tables figures Appendix A - Tritium Higration in PCRV Concrete t
1 i
t P
.-m..
~,--
- ~ - -.
-n,.
. m
VERIFICATION Of FSV ACTIVATION ANALYSIS 1.
INTRODUCTION Although there are many potential sources contributing to the uncertainty of the results from any activation analysis, the two most important sources are the accuracy of the thermal neutron flux predictions (the majority of all activation reactions occur at the thermal flux level) and the accuracy of assumptions used for material composition, including any trace elements.
PSC developed an action plan to assess the accuracy and sensitivity of these parameters and the impact to the overall activation analysis results.
This action plan, which was designed to validate the neutron flux and verify the material composition assumptions, provides additional confidence that the activation analyses results are very reasonable in terms of component activity and resulting dose rates. The efforts were divided into two sections: 1) the verification of the neutron flux predictions through comparison of predicted activities and measurements taken from activated samples of known composition, and 2) verification of the trace element assumptions used for the PCRV concrete and the sensitivity of the trace element abundances on activity levels.
- 11. NEUTRON FLUX VERiflCATION PSC has removed wire and Charpy specimens from the PCRV top head which were used to verify thermal neutron flux predictions from the ANISN (Reference 1) calculations and to assess tha conservatisms of cobalt impurity assumptions used in carbon steel.
At the suggestion of the NRC, PCRV tendon wires have also been removed from one vertical PCRV tendon (located 8" inches beyond predicted concrete removal depth), analyzed, and the results compared to analytical activity predictions.
A.
Wire Specimens The data available to validate the neutron flux came from the three wire specimens, located in the top head directly above the core, adjacent to the PCRV liner (See figures 1 and 2).
The wire specimens were originally designed as radiation monitors to be used in conjunction with the Charpy specimens which were to be used to monitor neutron irradiation effects on the liner material.
The wires were removed from the top head and were radiochemically analyzed for their specific isotopic activities.
The results of the radiochemical analysis were compared to the analytical predictions from the 1-0 neutron flux and activation calculations (Reference 2) and are summarized below.
One wire, composed of 99.5% aluminum and 0.5% cobalt, was judged to provide the best data for validating the thermal neutron flux.
The Co 60 isotope, produced from the thermal reaction of Co-59 + n yielding Co-60, was selected for comparison since its cobalt composition was known with accuracy.
The radiochemical analysis resulted in an average Co-60 specific activity of 14.5 microCi/ gram.
The 10 ANISN/ REBATE (Reference 3) calculations predicted a Co 60 specific activity of 14.0 microCi/ gram.
The comparison between the 1
4 experimental data and analytical predictions is excellent (within 3%),
i indicating that the estimates of the thermal flux in the top head are quite reasonable.
Similar results are expected for the thermal ocutron flux i
predictions, at the same distance from the core, in the radial atrection since the same analytical methods and assumptions were used to generate the radial neutron flux.
Better results may also be expected due to the homogeneity (less streaming paths, simple geometry) of materials in the radial direction.
l i
The two remaining wires were used for comparing the fast neutron flux.
The isotope Mn 54, produced from the fast reaction fe-54(n p)Mn-54, was used as the comparison isotope.
The two wires were composed of: 1) 99.55% Vanadium +
l 0.45% fe (88% enriched fe 54) and 2) 100% natural fe.
Trace element compositions for elements such as cobalt and tantalum were not known, although their presence was determined during the radiochemical analysis.
The radiochemical analysis of the two wires resulted in Mn 54 specific activities of 3.43E 02 microci/ gram and 2.68E-03 microCi/ gram, respectively.
The 1-D ANISN/ REBATE calculations predicted a specific activity for wires 1 and 2 of i
1.81E-02 microci/ gram and 5.50E-04 micro C1/ gram, respectively. Although the l
predicted versus the actual results were not as close as for the thermal flux case (the fast flux prediction varying by a factor of 2 to 5 lower, while the thermal flux was within 3%), the results are reasonable with consideration of g
the possible variables involved:
i 1)
The fast flux varies more rapidly as a function of position than the thermal flux and the precise location of the specimens (within the holding tray) was not known.
i 2)
The effects of fast flux neutron streaming in the core / reflector are not easily determined in a 10 analysis and can therefore introduce j
more error than for the more isotropic thermal flux, l
l 3)
Any error in the assumed material composition would affect the analytical predictions.
It should also be noted, as mentioned previously, that the majority of all activation reactions occur at the thermal flux level; the contribution of the fast flux toward overall activation is minimal.
B.
Charpy Specimens In addition to the wire samples, carbon steel Charpy V-notch specimens, from the same top head location as the wires, were analyzed using gamma spectroscopy.
The exact composition (trace elements) of the specimens was not known, and could not be used to judge the accuracy of the neutron flux, but does provide additional evidence that the flux level is reasonable.
Additionally, the experimental and analytical predictions were compared to help judge the conservatisms of trace element assumptions used in the activation analysis.
The trace composition of cobalt in carbon steels in the activation analysis was assumed to be approximately 200 ppm.
This is almost twice the average level found in the rebar and vessel steel samples surveyed in NUREG/CR-3474 (Reference 4). The average Co-60 specific activity measured in the Charpy specimens was 2.48E-02 microCi/ gram and the analytical predictions resulted in a value of 5.59E 02 microCi/ gram, which, as expected, 2
.,~
is almost twice as high as the treasured.
Although the sample could not be used to judge the accuracy of the thermal neutron flux, the results do provide evidence that the flux predictions are reasonable and the the assumptions of the cobalt levels are conservative.
C.
PCRV Tendon Wire Samples The activation analysis (Reference 5) predicts that at five years after shutdown, approximately 24" of concrete would require removal from the PCRV sidewalls to achieve-a (contact) dose rate of 3.4 microR/hr. The 24" depth is based on the assumption that the dose rate in the PCRV would be due to contributions from the top head, core support floor and sidewalls.
The top head and core support floor will be removed during decommissioning, leaving only the PCRV sidewalls cor tributing the dose rate inside the PCRV.
In this case, a (contact) dose rate of 5 microR/hr is predicted to be achieved when approximately 23" of concrete is removed (Note: Although the above dose rates are predicted on contact, they are representative of dose rates at one meter from the surface since the exposure was calculated inside the cylindrical PCRV).
In -order to assess the accuracy of this dose rate prediction, the accuracy of the neutron flux solution was determined through comparison of predicted-versus actual-specific activities of the tendon wires.
The closest set o' vertical tendons to the inside of the PCRV lie at a distance of 32" from the inside of the PCRV liner (31.25" concrete depth) (See Figure 3).
Wires were selected from a tendon which was located radially next to region 22, a high powered region during Cycle 4 (See Figure 2).
Due to the hexagonal geometry of the fuel elements, the radius of the active core varies from approximately 275 cm to 305 cm.
Therefore, actual activity levels may vary somewhat in the azimuthal direction.
The location of the selected tendon corresponds to a maximum active fuel radius, i.e., where the tendon wires are closest to the active fuel.
The activity in the selected tendon. wires is considered to be a maximum activity in the azimuthal direction.
The selected tendon is contained in a 4.28" I.0.
carbon steel tut. and contains 169, 1/4" diameter, carbon steel wires (Figure 4).
The tendon diameter is approximately 3 1/2 inches.
Six wires were pulled at equidistant positions around the circumference of the tendon.
The tendon wires spiral down-through the tendon tube.
Therefore, selecting. six' circumferential wires provided confidence that at least one wire would be located closest to the core (having the highest-activity).
Four test sample segments, each 12" in length,-were -selected from each wire.
Each test segment was divided into four 3" segments for use in independent verification.
The sample locations correspond to elevations of: 1)_the top fuel _ block, 2) slightly above core midplane (location of highest axial _ peaking factor during operation), 3) bottom fuel _ block and 4) at the bottom of the core support floor- (CSF) (See figure 1).
The sample segments at different
- axial locations provide axial _ profile activation information.
The sample segment below the CSF was intended to provide verification for the assumption that there is negligible activation of the concrete below the core support floor.
4 3
.. m
a.
In addition to the test samples, an unirradiated control sample was taken from the bottom of each wire.
The control samples were analyzed for the exact composition of the wires.
The exact isotopic composition of the wire must be known to make an accurate assessment of the flux level solutions at the tendon location.
The predicted specific activity of the tendon wires from the activation analysis is extremely low, in the picoC1/g range for both Co-60 and Hn-54 (the only isotopas that were anticipated to be detectable).
The fact that the neutron population in the area of the tendons was low (very low flux) makes the analytical prediction of the neutron flux in that location difficult.
All wire samples were examined in the FSV radiochemistry laboratory for reittive activity levels to determine which wire was the closest to the reactor core at each axial location.
The wire having the highest relative activity at each axial location was assumed to be the closest to the core.
The highest activity wires were quantitatively examined in the FSV radiochemistry laboratory using gamma spectroscopy to determine the specific activity (microC1/ gram).
Duplicate samples were examined in an outside laboratory for independent verification of the specific activity.
A comparison of the measured and predicted specific activity at core midplane, as well as measured data from other axial levels, are presented in Tables la and Ib.
The results of the quantitative analysis indicate that the thermal flux (based on the Co 60 activity at core midplane) at the tendon location was under predicted by the activation analysis by a factor of about 2.8.
This is considered to be good agreement in light of the distance from the core, and the number of mean free paths (about 12) traveled by the neutrons through the concrete before reaching the tendon wire.
The under prediction of the thermal flux may also be, in part, due to the azimuthal location of the tendon wires.
The location of the wires, as previously discussed, corresponds to a maximum active fuel radius of approximately 305 cm.
The activation analysis assumed an average active core radius of 297 cm.
Therefore, the tendon wires would be closer to the active core than assumed in the analysis and would have seen a slightly higher thermal flux.
In comparison with available activation data in NUREG/CR-5343 (Reference 6),
the accuracy of the thermal flux prediction is reasonable.
A comparison of the predicted and actual measured activities found in NUREG/CR-5343 resulted in agreement between 10% and a factor of 2.
However, the predictions and actual data measurements were made for areas near the reactor core, where the activation flux levels were much higher (and much better characterized) than in the area of the PCRV tendon wires which are relatively far away from the active core.
The Charpy/ wire specimens taken from the PCRV provide the same degree of accuracy for those areas closer to the core, in one case in _ NUREG/CR-5343, the predictions were made for in-core fuel assemblies and the agreement was quite good (except for end fittings where the geometry was complex).
The activity of the samples was in tne Ci/g range, 12 l
e
.m
--~+m.
,r_-
w
orders of magnitude higher than the specific activity predicted for the tendon wires.
In a second case, for the Gundremigen reactor, samples of the vessei steel were analyzed.
The actual and predicted specific activities compared within a factor of 2, which was considered good by NUREG/CR-5343 since the neutron flux varied by over two orders of magnitude from the reactor pressure vessel wall.
In comparison, the neutron flux in the FSV PCRV drops three orders of magnitude from the PCRV liner to the tendon wire location, and ten orders of magnitude from the reactor core / reflector interface to the tendon.
It was concluded in NUREG/CR-5343 that very geod agreement can be obtained near fueled regions as evidenced by the samples provided from the fuel assemblies and the ssure vessel steel.
PSC's Charpy/ wire specimen data also confirm this coru
- sion, predictions further away from the core become more difficult.
No dat. from NUREG/CR 5343 was available at large distances from the reactor core, but the results from the tendon wire analysis appear to be reasonable, based on the accuracy of near core predictions in NUREG/CR-5343.
As indicated above, the predictions nearer the core should be more accurate.
The neutron flux at the 24" concrete removal depth is predicted to be higher by about an order of magnitude than the flux at the tendon wire location.
It is anticipated that the agreement between the actual and predicted flux at the location where 24" of concrete would be removed should be more accurate.
If the flux (at a depth of 23" of concrete removed) were under predicted by a factor of 2.5 to 3, the resulting dose rate (5 years after shutdown) would be approximately 12 microR/hr, compared to 5 microR/hr.
This increased dose rate would require approximately 4" of additional concrete (from the original depth of 23") to be removed to obtain a dese rati in the 5 microR/hr range, which is 4
still within the first row of tendon tubes.
Ill. MATERIAL COMPOSITION VERIFICATION Analytical sensitivity studies have been performed for the homogeneous rebar/ concrete mixture to assess the effect of trace element abundances on the amount of concrete required for removal in the radial direction.
The assumption of a homogeneous concrete /rebar mixture was also studied to determine the conservatism of the assumption.
Surface concrete samples have been taken to help determine the actual trace element abundances in the PCRV concrete.
A.
Sensitivity Study of Trace Element Abundances in Concrete The sensitivity of the trace element concentrations to the amount of concrete required for removal in the radial direction was investigated using the available trace element data in NUREG/CR-3474.
The activation analysis assumed that the trace element constituents in concrete were at the average trace element impurity level found in samples from the bioshield concrete of twelve different nuclear plants, lh order to assess the effect of varying the trace element concentrations on the amount of concrete requiring removal, a worst case scenario was considered, lhe activation analysis (Reference 5) was rerun using all the 5
original assumptions except the elements whose activation products are most important to dose rate were assumed to be at the maximum values given in NUREG/CR 3474 for both the rebar and the concrete, The exception to this was the cobalt level in the rebar, which was conservatively set at 200 ppm, a value higher than the maximum value given in NUREG/CR 3474.
The elements considered to be most important, in the short and/or long term, were cobalt (Co 60), niobium (Nb-94), silver (Ag 108) and europium (Eu 152), where the isotope in parenthesis indicates the isotope of concern.
The increase in number densities of the above elements varied from a factor of 1.5 to 2.2.
The * : sults of the study are as follows:
Dose rate 5 years after shutdown (microR/hr):
i Max Trace Previous Calculation Elements Trace Elements 24" Concrete 6.0 3.4 4
Removed 26" Concrete 3.2 n/a Removed Dose rate 60 years after shutdown (microR/hr):
Max Trace Previous Calculation i
Elements Trace Elements 8" Concrete 7.4 3.5 l
Removed.
i 10" Concrete 3.7 1,7 Removed i
The above results indicate that the effect of using the maximum value for all trace elements is ar. increase in the required depth of concrete to be removed of approximately 2" in both the long and short term.
In the short term, the dominant gamma emitting isotope is Co-60 from both the rebar and concrete.
l The increase in the cobalt number density was approximately 1.66 and the corresponding increase in dose rate at 5 years after shutdown was 1.76. In the long term Eu-152 is the dominant isotope.
The increase in the europium number i
density was 2.2 and the corresponding increase in dose rate at 60 years after shutdown is 2.1.
These results indicate that the increase in the required depth of concrete to be removed is related to the increase in the levels of cobalt (short term) and europium (long term).
It is unlikely that the cobalt and europium concentrations in - PCRV concrete are widely different than those sampled in NUREG/CR-3474.
In that respect, the depth of concrete requiring removal would i
be. bounded by an increase of not more than 2" in a worst case scenario for trace element concentrations.
l 6
l 5
B.
Assumption of a Homogeneous Concrete /Rebar Mixture The activation analysis modeled the PCRV concrete and rebar as a homogeneous mixture.
In the actual rebar configuration, two sets of rebar lay just outside the PCRV liner and the remainder of the rebar lays at least 1.5' beyond the inside ring of tendons (See figures 3 and 5).
The two inner sets of rebar will be removed during decommissioning and the outer sets of rebar will not be activated. Therefore, the assumption of a homogeneous mixture at a depth of 23" of concrete is conservative.
A sensitivity study of the contribution of the rebar to the overall dose rate was performed to assess this conservatism.
The rebar density values were removed from the homogeneous mixture and the activation analysis was rerun.
The resulting predicted dose rates at 5 years after shutdown, with 20" and 22" of concrete removed, are 6.15 microR/hr and 3.14 microR/hr, respectively.
This indicates that approximately 2" less concrete would require removal than originally predicted.
C.
Concrete Sample Data Surface samples of the PCRV radial concrete were analyze / to provide confidence that the material composition assumptions for the concrete were conservative.
Six samples, representative of the concrete mixes used in the PCRV sidewallt, were taken at the same elevation as the active core. The sample size required for a trace element analysis was only a few grams, but large enough samples were taken to ensure a good mix between the aggregate and other concrete constituents in the sample.
The cylindrical samples were approximately 4" in length and 1.5" in diameter.
Samples were crushed to provide a homogeneous test sample.
l Table 2 shows that there is no significant variation of the trace element abundances among samples.
The only element with a large variation in abundance is cesium.
However, cesium is relatively unimportant from an activation standpoint due to its relatively low cross section and abundance, and the short half life of the Cs-134 isotopa produced.
The data in Table 2 provides support for PSC's assumption that PCRV concrete trace element abundances do not vary widely throughout the different mixes of concrete.
Table 3 compares the average values of the PCRV concrete trace element abundance with the average values from NUREG-3474 (which were used in the activation analysis).
The trace element abundances in the PCRV concrete all fall within the range listed in NUREG 3474 for concrete from 12 different nuclear facilities, with the exception of nickel, which falls below the range.
The two most important isotopes, in terms of dose rate, are cebalt (short term) and europium (long term).
The average cobalt abundance of 3.93 ppm found in the PCRV concrete is less than hal f the average value of 9.8 ppm listed in NUREG 3474, and is significantly less than the maximum value of 31.0 ppm.
The average europium abundance in the concrete was found to be 0.75 ppm.
This is slightly higher than the average value of 0.55 ppm found in the NUREG, but is still less than the maximum value of 1.2 ppm.
7
. - - -. -. - -. - _. _ - - _. ~.. -
-. e it-is recognized that the abundance of trace elements in the concrete can vary
'and the~ results of a _ few samples may not ensure that all mixes of concrete -
would : have. the same trece element abundances as the_ samples.
We believe, however,thatitishighlyunlikelythattheabundanceswould'varyenoughfrom
.the sample data to have significant impact.
Additionally, it is unlike y that
-the abundances-would be dramatically different from those listed in
-NUREG/CR-3474.
To thi:: end it is reasonable to assume that the results of the surface' samples, when combined with the sensitivity analysis and NUREG/CR-3474 data, prcvide a high level of conf _idence in the activation analysis _ results.
1 IV.
RESULTS AND CONCLUSIONS In order to obtain a more accurate prediction rf-_the required removaf depth for the concrete, the activation analysis for the concrete was recalculated' utilizing the information obtained from the verification efforts.
The thermal-fi m was assumed to be underpredicted by a factor of 2.8, no rebar was assumed to -exist in t.he concrete at-the required removal depth, and the average measurec trace element abundances from the _ PCRV surf ace samples &:ere used.
.The analysis was also performed:for a worst case scenario using the highest trace element abundance: found in NUREG/CR-3474.
The results of the analysis are given in Table 4.
The predicted required concrete removal depth of 21" is considerably less than the prope;eu moval depth - of 27" to 32".
Even in a worst case scenario, where-th t v a-element abundances were at the maximum values found in
-NUREG 347. < se required removal depth (27") is still within 'the' proposed removal d,na.
Ihe dose rates for the proposed removal depth are pred' ted to be well % the 5 microR/hr release criteria.
Additionally, the ac evation analysi, :;as performed at approximately core midplane, the location of' the highest activity.
In reality, the activation of the concrete will decrease in g
both axial directions away from the core midplane.
L Th'e data obtained from the verification effor ts and the results of the-
-recalculated activation analysis provide strong e;idence that the predictions from the original.activatM analysis (Referenca 5.? are reasonable.
The results: of the C5arr tre ms. rement'. Gow : th;t _ closer - to the core, the thermal _ flux-is prtGicted quite accurately (w! thin 3%).
Even at large distances from the core, the thermal flux ' predictlons are considered - good
.(within a factor of 2.8) in light of the difficulty in= predicting the flux in i_
an area where the neutron population was low during power operation. The PCRV concrete data and the results of the--sensitivity studies for the trace element--
abundances _ show that-the assumptions used in the activation analysis ara reasonable, and that no ~significant variations of the trace _ element abundances-in: the concrete would be. expected.
Additionally, the comparison with the results from -NUREG-5343 indicate-that the-' accuracy of tne rest.lts are acceptable.
L Based t upon the measured data and comparisons with enalytical predictions provided in this document, PSC beliews that ' the calculational methods and material composit' ion assumptions for the concrete prcvides reliable ee.timates of the activation products within the PCRV.
8
s a
V.
REFERENCES 1.
Engle, W.W., "ANISN-P - Multigroup One-Dimensional Discrete Ordinates Transport Code with Anisotropic Scattering,"
Radiation Shielding Information Cent:., Oak Ridge National Laboratory,] Oak Ridge, Tennessee.
2.
PSC Internal Memo, NDS-91-0066, " Comparison of Upper PCRV Wire Specimen and Charpy Bar Activity Measurements to Analytical Predictions," February 8, 1991.
3.
Ebasco Services Incorporated, " REBATE - A Computer ?rogram for Calculation of Decay Gamma Source Strength for One or Tws Dimensional Gamma Transport Analysis," Ebasco Services Incorporated, New York, New York.
4.
NUREG/CR-3474, "Long Lived Activation Products in Reactor Materials,"
August, 1984.
5.
EE-DEC-0010, " Fort St. Vrain Activation Analysis," October, 1990.
6, NUREG/CR-5343,"Radionuclide Characterization of Reactor Decommissioning Waste and Sperit Fuel Assembly Hardware," January,1991.
t 9
i
k Table la.
Comparison of Measured and Predicted Specific Activity of Cc-60 for the PCRV Tendon Wires Axial level Measured Measured Calculated FSV Lab Outsida Lab (microCi/ gram)
(microCi/ gram)
(microCi/ gram) 1 8.23E-06 6.70E-06 nc 2
10.9E-06 10.0E-06 3.84E-06 3
4.80E-06 3.70E-06 nc 4
NSA NSA nc Table Ib.
Comparison of Measured and Predicted Soecific Activity of Mn-54 for the PCRV Tendon Wires Axial Level Measured Measured Calculated FSV Lab Outside Lab (microCi/ gram)
(microCi/ gram)
(microCi/ gram) 1 3,01E-0E 2.76E-06 nc 2
4.75E-06 3.46E-06
- 0. 93 E- 06 3
1.33E-06 1.13E-06 nc 4
NSA NSA nc
- Core Midplane NSA - :.9 significant activity ( < 0.44E-06 microci/ gram) not calculated nc 10
i l
Table 2 l
Result of PCRV Trace Element Analysis j
f Element Sample 1
2 3
4 5
6 i
Ba (ppm) 63 56 54 56 60 63 i
Ca (%)
12 10 10 11 11 10 Cs'(ppm) 1.20 0.97 1.40 1.40 1.80 2.20 l
t Co (ppm) 3.90 3.55 4.45 3.90 3.40 4.35 Eu (ppm)-
0.68 0.64 0.74 0.96 0.66 0.83 Fe (%)
1.3 1.1 1.1 1.2 1.1 1.1 i
ti (ppm) 11 12 12 12 12 Mn (%)
0.052 0.053 0.052 0.056 0.053 0.050 f
Ni (ppm) 7 7
7 8
10 7
K (%)
0.31 0.29 0.22 0.26 0.24 0.25
[
Ag (ppm)
<1
<1
<1
<1
<1
<1 l
i l
i I
t L
t u
s l
I i
l l
l 11
4 Table 3 Comparison with NUREG-3474 NUREG-3474 Average Measured Average Value Range Concrete Data Ba 950 ppm 20 - 7060 58.7 ppm Ca 13.3 %
8.3 - 34.7 10.7 %
Cs 1.3 ppm 0.32 - 6.2 1.50 ppm Co 9.8 ppm 1.1 - 31.0 3.93 ppm i
i Eu 0.55 ppm 0.11 - 1.2 0.75 ppm Fe 3.9 %
0.5 - 24 1.15 %
Li 20 ppm 11.8 ppm Mn 377 ppm 56 - 990 527 ppm 1
Ni 38 ppm 11.9 - 87.0 7.7 ppm K
0.75 %
0.047 - 2.5 0.262 %
Ag
< 0.2 ppm
< l.0 ppm l
l-l-
I 12 i
t
Table 4 Resultslof Activation Analysis Using Verification Data 1.
Predicted depth of concrete required to be removed to meet the 5 microR/hr at 1 meter release criteria using the data obtained from the verification efforts:
21" 2.
Predicted depth of concrete reqeired to be ren.oved to meet the 5 microR/hr at I mrcer release criteria from the orig.nal activation aialysis (See Table 3.1-3.from the Proposed Decommissioning Plan):
23" 3.
Worst case scenario (maximum trice elements) depth of concrete required to be reme.ved to meet the 5 microR/hr at 1 meter re". ease criteria:
27" 4.
Planned removal depth (per the Proposed Decommi-ssioning Plan):
27" to 32" 5.
Predicted dose rate inside the PCRV, 1 meter from the concrete surface with 27" concrete removed:
0.8 microR/hr 6.
Predicted dose rate inside the PCRV, 1 meter from the concrete surface with 32" concrete removed:
0.2 microR/hr l
l i
13
,y
=
W9-- --
,Q., ~,g;,. 6--
, i; j
=
L.-
5
~
.t
.m
,, w.a.
we r-w
_~
..v w m n.
m
-.... s( f - d h f -l.-C'."J*,.h -..J.*f.'..E *.'* W_[
/~~ Z.M 5
L,- f, p.
T f.60'f.*Lji,M1p4 4{W - c 4l*3l.
I1 d':::ld
.I _l WJ._. J 7 L
1_.f_.
.T' l
- rr,-Qd Mja tr d
.f,.""*4 t#
w
+;d y a'
-J
' - - -a
=p-
_= g'
{
~_ _ -,=. w 8
1 g-
. _... @g w_s.
g L;-
, Ya ;
Ekf}
h A 4 -. C:.
- 4L g i 0
4 r
s x
x
~
', l
-+
e' Y
ll
.n P4','-d-d"I M
i W
d- -l 4t -
y.i._%y 1
.l.
+,
m.~'~i'1
...t,#
e j
e
'., t
_ 9
'-k g
M
-<i T_J ;
~
T y
~
? '.
,.=-k '*
n 4., *
~..
'[;).,
.<3. M, i
a w_,.,
,y,
te g
- l y
w i
.e -
,1-0 I
' N c.
g
/
$d---g *I
\\
'e Q,.l <
- e i-W y
V n
p a g i
2 0~.
j.7 -
4-
)
-[
1-
- L..
s.
e
.{i e
/
i
. ;..)
. /' A M N
i
-l-
.h.-a aes:::Q4 S,
A u1 l
a r. < u u *"-_,
a.y W.' =.
,e
.)4.e s.y;
,.==,***,
.t
- TT',
,1,
,1 2: t-w
\\
f,.
w
= r, p.
_ - _q _ _p yg m
,, _. a_ _ amt.,,
._3, s
,n s -. T-e 3_,,,_ -
1_- w si fM
. su..'
N C%
Of32W i
a, w ca,,
o. --.
t/> f,
'o, m
.m A -
W
.m er. y e. <,
- o.v m
- re.
C
~
.g
--..amamme. w f
i s
a r
a
,. ~e
,- --g,
10-,,.,.,.
l
[
t
.t _g
.; f, 1
.a--o
., =
A
/ y.
~* _.
.-/
p y*,
g.,r,.J.:.
Inner Reb i
l
- '...--.d'.
7 _'.,
.,'lll' "::W; $
5 j
_....-.m 4>,.
+
w s.3
. __b
+
Qy Outer Reb
,...s.
.: w j -
es ear _'.e
.r a
i 1
mr
.1
.;l., g,y~~h'y.,
j,-+J
- e.-
.. a. <4-x=..
VerEiC8I I
%.._..-3,
, Tendon Tub-l.
?, _--e<.
.-- c..,,
e
- l r
+---c-e 1.-
-,'q 1
5 e
r
- 1..
-_')
~
e s
a :
1
- -...i_..
..?.1. < - *.J,!.*- h 1
e -
-e
. +
-f
-a...,
- m...
la
._I _ _, e M 6'-
"'d w-,
l
[.
. n w
4*
M 1
, l '
.' a' f,q sg Q->,3
._..f[-.4 6.
M
- =- 9 a w. e. rwe g-
,g y
d 0.c Q
m
'=
- ~6. 555
' f
?
f
.. y:::w 7.~.s.*
,=,
I-,,
a
- w f.
3 g
-.a.
- ]
.o 4
-.C." . " " ' '
,e,,,,,
p
-n-
]
i.
- ~
--1 3
, t-g L'
n p_.
~~~-
.' W -
_ _,.W....1M.
2
~'
.,.w.-..-ri,,
r p ~.w.-
.me s.t. _ _.
C.E.,
. GM
~
- - -y j-i t
~~v 9
.t
,.M.M; :4
'9
?,
d-N
, ~. +~.v g
g e
.+
.~
-di.
g q
e,
,i
}
o g
s L
-C**,._*.*'.*L
< p_.
_ =,
i 1
-e ~_j 1:.}-
t---S l,8'*4***'-
.e i
I
'^ '
g g -
[,,,,,,_
y "'".').
gfM 5"--
4m..<
d J
L-y i""*'l' ".
s..
sa.e
- e. a.e f., s A
,(
.m t.
q i
~
. GF S' s$'$'
00'4*
5ECTION l,,' l. 6 T0D w a t,8' --
QQ C_ &
__m.'
M I i-
m 4" I ihet e
i 1
,l..
3 l
.I..,*,. -
.l 3-
+
5
- y.._
--.a -..
.t.
..r E W**'- Y ~~~al M. ;,.-c.1 *A*""."L-9 D 1 :*f t f.:,_.on,.,
'N ' [,
c_c j'y ' ; M; -.ag % * ~ \\
h ' ',I j
l u..,4s..,. p - - g m.
wQu,.
P i
nrB y 7. s r.,j _ _ r;.r w.1 t= g - -
6,e
- v
- -. ~
a
~ --
.r x
e {
v:,u,.,,j q,4 '.:,.
s i
r
]
g.,,:.
.j g;; ---m
}
u r.y br m.. i,. i-w f
v, s..
-L M
1 4
6 t-2
.V i.
1 %.. -..,, -
s
- u. j,',::",,;_,, u. -
/,
...., at
,ji, t-m,
'=
i.
6' 1
1 s,.
a n
a e
c.-
1E<
I.
g.
e } -.,:.
j
. !,pk_-
,-4
'i 4-p 3
.r. - -.
M"-
.i n
=
-.,agy sm *L,g e.--n.
,i -
e i
1 m
y m
7 e.e
--..e..
. q
- r. ~,
m t.
q p
ew i
i p
3
[, a t._.
m I
, r' i,
s
- m.,
j
.[1i l;,
T-i
-- "~
t
--~a--
3, q
i i t.
To
!n' ya..~... -
t J.
b, b..
q vu y
.(i t
.t z
+
- i--
?m
'd,..
M.
M,',
u xmmge m,m n g
~
g(x'y~
_. ~
~
.a,..ae.
F
.M N s
?
~
~
I m
m, 4
1 L.. s. v -. :
4 a
- =
g e
~-
1 s.
er
. _ r
- g
~
o__
1 l
I
~
.a.
y
- ?
e b
_ P
.2
-2 k
Figure 5 t
3r -
m r-=-~w te Un. > Y l
y
."_.e.
- ;m.p g _, ::: :::.u '.
Upper PCRV
.gge u. e a -w..;-
xia V ew o
-3,
,3
=
-_..e
%,.o-
, _ _. _ _ -,42"
..a i
h P
ME_r-g' i
{
~..
n -
. -~
m..-
-.k
.b
$,.4 0
phe
- m. _
.m
~
_.._m,,,.,..
n s...
.- r
- 3. :.
4 -- - -
j.
3 X kW:
>x-b CAUTION
- n. st ma.
gT l
- >-4 y.-
wv.,HE DESIGN DOCUMENT CONTElfR.
L_;,
- ~;: j
_.___s STATUS INDEX (00SI) FOR 0(TUMENT
~.!
.=
- m..%m,
.w -u
.i d,
, y.
ANYONINCORPORATEDm ur.
~~-
~ ~~l M~h
]
MODiflCAfl0NS AffECTWG E
7 k,._+m'I E*--g..
' 4_==...-
THIS DOCUMENT.
W t__f, -..., _.
-f g,
-, > ' -".L i
- b. E>---3-.
r:-,, a e.
- C**
-.a.N4 m,
") t
.s s-r-.
v i'n
'+
se n.
e.
g; r
e
.ar e, ma*
, re.o La.
- ~
=.
q
' h.
- a'. ;--.
e.,.e
,,,J i
- -*.aa av a
- A.,
r L
- et
,.p_ r p <.,... '
e,,.n.
te v s a
- s.,rw..-
-ei w.
m '7 p._.r-.--
w.
~*
.c, s o.r 6 -e
.s.y.
ee e
.,e.. e,.., sea...e n<w=
- l. -.
a e
_n
.-~.
- _MN,
_.<6
^
a
.W
.A; i
.y o,
/
t
=
'e s-
- a.,.
hnurure.w,qart.h,4i&% ,""$rG:W.5m
- r,Qi,G @M em e unau p"" *"* ""* ** " "'" 4.=;
r4 gpayj? * 'J';M -M.t;g3,i,,.,/y,ur,,c,,
qc 1 i i i i i i tm
..s -
-.e ~ Ci + ^ i m;'
=
g,!
n.",
mv L
+.
%-~ ~.
p_s.' 2v 4Ti'D-:J
- ae s.'ia( 5 SM
,,,o [ e _n_o.,,...o 10
..u..
v._
4.- !
i. i r % -.
1-r- i-r- +.m at o cact -
~
3. e -'-._
..s-
,-g
.1 1.
. -n >,- L-c <
6_.
7
~
1 Appendix A Tritium Migration in the PCRV Concrete j
~
j l
l t
- i Tritium Migration in the PCRV Concrete k
i Tritium activity in the concrete is not a factor in determining how much concrete must be removed to achieve the release criteria of 5 micro R/hr. Tritium is a beta emitter, ar.d the 5 microR/hr criteria i
is based on gamma emitting isotopes.
However, the total enount of
(
tritium in the remaining concrete must be considered in pathways analyses performed for site release. The release criteria for 1
pathways analysis is proposed to be 10 mrem /yr.
l Tri:ium generation in the concrete is due to the thermal neutron i
reaction with Lithium or other impurities in the concrete.
In i
discussions with the British (P. Woollam, Nuclear Electric), tritium
-does-not diffuse from the primary coolant through the PCRV liner and 1
into the concrete.
Tne initial predicted tritium profile tends to decrease exponentially from the inside of -the PCRV, following the l
thermal flux profile directly.
This -is consistent with the initial predict ons made by the British.
Over time, the tritium will l
diffuse through the concrete, flattening the profile.
From British-
- ?
experience, it may. take up to 130 years for the profile to become i
flat.
The total amount of tritium in the concrete will remain l
constant during this time, except for-decay and a small ' amount of i
tritium that will diffuse into the air.
Assuming no-tritium migration, the specific activity in the first 6"
)
of concrete at five _ years after shutdown predicted from the activation analysis is approximately 5.0E 03 microci/ gram.
The actual specific activity in the first 6" would be somewhat less due i
to the tritium migration.
l The tritium activity, regardless of its final profile, is not I
~ anticipated to-be a dominant dose contributor in a pathway analysis for FSV.
If a
conservative specific activity of 5.0E-03 l
microCl/ gram is assumed ~ for the concrete, the total effective dose eouivalent--(TEDE) (for-the expected tritium activity would be less than 0.01 mrem /yr 1).
This _ value is only a small fraction of the i
10 mram/yr proposed release criteria. Based on the low ~ contribution i
I of the expected tritium activity to a total effective dose, it is
[
reasonable to assume that the removal of additional concrete will not-be required to meet the proposed release criteria for a pathway i
analysis..
l t
i (1)
Based on the unit equivalent TEDE for tritium from Table 3.1,
{
from NUREG/CR-5512
" Residual Radioactive Contamination From t
Decommissioning," January, 1990 (for comment).
- l I
t
)
r E
,m
=m ts--,-e-11--wy,&--
e
-w-
,aes=-