ML20012D154
| ML20012D154 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 03/14/1990 |
| From: | Roche M GENERAL PUBLIC UTILITIES CORP. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 4410-90-L-0019, 4410-90-L-19, NUDOCS 9003260539 | |
| Download: ML20012D154 (49) | |
Text
7,;
],{ y NN f p y.
.g-M ea GPU Nuclear Corporation Nucle.ar
- ,omareo Middletown, Pennsylvania 17057 0191-717 944 7621 ~
TELEX 84 2386 Writer's Direct Dial Number j
(717).948-8400 March ' 14, 1990 4410-90-L-0019/0356P l
I
'US. Nuclear Regulatory Commission Washington, DC
.20555 l
Attention: Document' Control Desk Three Mile Island Nuclear Station, Unit 2 (TMI-2)
Operating License No. DPR-73 E
Docket No. 50-320 j'
SNM Accountability i
Dear Sirs:
~'By'NRC letter dated October 17, 1985, GPU Nuclear was granted exemption from certain rNuirements for periodic inventory-and reporting of the.special.
. nuclear materials (SNM) balance for Three Mile Island Unit 2 (TMI-2).
As a condition of 'the exemption, GPU Nuclear is reouired to conduct an assessment of the;SNM remaining at TMI-2_following the completion of the defueling
. effort. This assessment.is referred to iri the exemption as the
" post-defueling survey." GPU Nuclear letter 4410-88-L-0162 dated 1
. September 30, 1988, submitted the initial Post-Defueling Survey Reports'
.(PDSRs)..
.{
As stated in that submittal, the PDSR ducuments the GPU Nuclear assessrent of the amount of residual SNM and describes the methodology utilized to determine l
.the cuantity of SNM in each case. The attached PDSRs transmit the
?
post-defueling~ survey results for the Reactor Vessel Head Assembly and the 'A'
- and 'B' Once-Through Steam Generators.
l.
iAdditional POSRs will be submitted as they are completed. A compilation of the individual PDSRs will form the basis for the final assessment of the L
"ouantity of, residual SNM at TMI-2 for accountability purposes.
L S h rely, L
9003260539 900314
(}
F 3
F;DR ADOCK 05000320
?-
PDC ll yh/
o p
M. B. Roche Director, TMI-2 g\\\\
JJB/ emf l'
t GPU Nuclear Corporation is a subsidiary of the General Public Utilities Corporation
\\
l
Q e l'l,,
p-y.;-
.. Becquent Control. Desk-a March 14,:1990
- ~
4410-904. 0019" c:
Page 2-s o
1 L
Attachments I
cc: % T. Russell'- Regional Aaministrator, Region I J. F. Stolz - Director, Plant Directorate I-4 L. H.1 Thonus - Project Manager, TMI Site l
F. -14 Young - Senior Resident Inspector, TMI-r f
I~
i I
I t
.2.
b b.
1
?
t s
- t I,
h c
t
(
in 1
p t
.I.'
- l p
t b
h.'
r
(
)
L s_
[.c
(;
('. '.
)
E kb' i
(
i
?
?
r t
I -'-7 t
i 6
- ~,.... -
1 I
s' 4
i TMI-2 POST-DEFUELING SURVEY REPOR1 j
FOR Th7 'A' AND 'B' ONCE-THROUGH STEAM GENERATORS.
6 t
t l
l
SUW4ARY The estimates of record of the amount of uranium oxide (UO ) remaining in 2
the Three Mile Island Unit 2 (TMI-2) 'A' and 'B' once-through steam generators (OTSGs) are 4.1 kg + 22% and 51.4 kg +_16%, respectively.
All statistical uncertainties are expressed as + one sigma limits for combined increments calculated or taken to be one sigma (defined as one standard deviation). The UO is distributed as follows:
2
'A' OTSG
'B' OTSG Upper Tube Sheet 1.4 kg + 21%
36.0 kg + 18%
Tube Bundle 1.7 kg +_ 48%
9.1 kg + 48%
Lcwer Head 0.29 kg + 48%
0.46 kg + 48%
RCP-1 J-Leg 0.40 kg + 48%
1.5 kg + 47%
RCP-2 J-Leg 0.27 kg + 48%
4.3 kg + 49%
TOTAL 4.1 kg +_ 22%
51. 4 kg +_16%
The 'B' upper tube sheet was characterized by neutron activation measurements.
Fuel estimates of the ' A' upper tube sheet, the ' A' and 'B' OTSG tube bundles, the lower heads, and the J-legs were projected using gamma radiation measurement data.
Earlier fuel estimates based on independent transuranic analysis of surface scrapings (Reference 1) and on direct alpha particle measurements of a limited portion of the ' A' OTSG tube bundle (Reference 2) indicate that the present values are conservatively high.
The cuantities of UO remaining in the ' A' and 'B' OTSGs are ( 0.5% and 2
( 5.5%, respectively, of the anticipated residual 00 inventory
- for i.he 2
entire TMI-2 facility in Mode 2.
Independent foil activation measurements were performed by a Battelle Northwest Laboratory group on both OTSG lower head and J-Leg locations.
These measurements confirmed the relatively low fuel deposition in these locations (Reference 4).
The anticipated residual UO inventory is as defined in the PDMS Safety 2
l Analysis Report and the PDMS Programmatic Environmental Impact Statement and is based on the assumption that the defueling program goal to remove more than 99% of the original core inventory of U0 is achieved.
2 l
l F
r-c 3
TMI-2 POST-DEFUELING SURVEY REPORT FOR THE 'A' AND 'B' ONCE-THROUGH STEAM GENERATORS
1.0 INTRODUCTION
i This report presents the analysis of the residual inventory of uranium dioxide (V0 ) in the Three Mile Island Unit 2 (TMI-2) ' A' and 'B' 2
once-through steam generators.
It is one in a series of reports generated to fulfill the requirements of the TMI-2 SNM Accountability Program (Reference 5). All statistical uncertainties are expressed as
+ one sigma limits (defined as one standard deviation).
Section 2.0, " Background", describes the physical attributes of the
' A' and 'B' OTSGs and their relationship to the accident and subsequent cleanup activities.
The boundaries for this report are also discussed.
Section 3.0, " Methods", describes how fuel measurements and sample data were used to produce the estimates of record.
Copper activation foils were primarily used to detemine the estimate of record of the amount of fuel on the 'B' OTSG upper tube sheet.
Separate copper foi1' activation calibrations were perfomed in the ' A' and 'B' OTSG upper heads with a known neutron source (Am-Be).
The calibration was used to compensate for uncertain scattering and neutron absorption (Reference 6).
Fuel measurement data for the ' A' upper tube sheet, the OTSG tube bundles, the lower heads, and the J-Legs consisted primarily of gamma radiation measurements, which were correlated to actual reactor coolant system (RCS) sample analysis results to produce the estimate of record.
The results of using gross gama measurements to characterize fuel present as films and as tube blockages was compared to earlier fuel analysis by direct alpha measurements and by scrapings taken from the 'A' and 'B' OTSG upper head manway and inspection hole cover plate inserts (Reference 1).
Direct alpha film measurements of the top 20 feet of nine ' A' OTSG tubes also were used for comparison (Reference 2).
l t
~~
l I
Section 4.0, " Analysis", explains how the estimates of record of fuel in f
the ' A' and 'B' OTSGs were calculated based on the fuel measurement and I
sample analysis data and compared to calculated exposure rates for known
+
activity concentrations modeled with the Microshield computer program (Reference 7).
Section 5.0, " Conclusion", presents the estimates of record for the amount of UO remaining in the ' A' and 'B' OTSGs, and states supporting 2
rationale leading to the conclusion that the estimates are reasonable based upon the available sample analysis data and the fuel measurement techniques employed.
2.0 BACKGROUND
The OTSGs are vertical, straight tube and shell boilers in which the reactor coolant (the heat source) is on the tube side and the secondary coolant is on the shell side (Figure 1). They are made of high strength carbon steel and all surfaces that contact the primary coolant are stainless steel or inconel.
The TMI-2 'A' and 'B' OTSGs were used for transferring heat from the reactor coolant system to the secondary system.
As a result of the TMI-2 accident, fuel was transported through the hot legs into the OTSGs by a variety of pathway mechanisms which include drain and refill of the steam generator and inlet pipings, forced c'irculation from the coolant pumps, natural circulation, and fluid movement resulting from energy releases into the reactor vessel (Reference 8).
Fuel fines and sediment material were transported to the OTSGs by reactor coolant through the hot legs to the steam generator upper he.id and tube sheet. Much of the larger particulate material settled out on the upper tube sheet with the finer material being transported through the tube bundles to the lower head and outlet piping..
i
. 7
i
.4 This Post-Defueling Survey Report (PDSR) includes characterization of the
' A' and 'B' OTSG upper tube sheets, tube bundle regions, lower heads, and the associated coolant outlet J-legs.
Radiological Environment As a result of the March 1979 accident, fuel debris and large quantities of fission products were released and oeposited in the ' A' and 'B' l
OTSGs. Most of the larger debris material settled out on the upper tube sheets with smaller material being transported to the outlet piping.
Duri19 late 1987, the debris material on the upper tube sheets was removed by vacuuming and pick-and-place operations.
The total person-hours and person-rem expended for defueling the ' A' upper tube sheet was 108 and 4.3, respectively.
The total person-hours and person-rem expended for defueling the 'B' upper tube sheet was 153 and 9.4, respectively. Currently, radiation levels in the ' A' OTSG upper tube sheet range from 1-3 R/hr.
In the 'B' OTSG, levels range from 30-100 R/hr. The general area exposure rate in the ' A' and 'B' D-rings range from 50-800 mR/hr.
In general, higher exposure rates and significant loose surface contamination exists in the 'B' D-ring area.
The total person-hours and person-rem expended for the OTSG residual fuel measurement, program were:
OlSGLowerHeadandJ-Legs Person-hours Person-rem
'A' 164 7.4
'B' 140 8.5 Upper Tube Sheet Person-hours Person-rem
'A' 160 7.1
'B' 348
- 21. 9..
I 1
e i
The methodology covered by this report is considered to be optimized toward keeping radiation exposure as low as reasonably achievable (ALARA), due to the nature of the 'A' and 'B' OTSGs.
Due to the radiologically hostile nature of the OTSGs and their environment, further refinements are not considered to be beneficial in light of the cost of j
exposure and labor.
]
3.0 METHODS i
As part of the Special Nuclear Material (SNM) Accountability Program at TMI-2, gamma measurement surveys and neutron activation measurements were performed for fuel characterization of the ' A' and 'B' OTSG upper tube -
sheets, tube bundles, lower heads, and associated cold legs during September 1988 to mid-January 1989.
SNM accountability of the OTSGs presented a difficult task due to system inaccessibility and high radiation levels, requiring the selection of indirect fuel measurement techniques.
Each entire OTSG, including the upper tube sheets, tube bundle region, lower head, and associated cold legs, required fuel characterization.
Due to the inaccessibility and presence of primary coolant in the tube bundle region, lower head, and cold legs, an alternative method for estimating fuel was developed.
This method consisted of obtaining gross gamma measurements, correlating this data to predicted gamma exposure rates for fission products sequestered in surface films, and for clustered fuel debris blocking tubes with a composition assumed to be similar to samples collected f rom the OTSG upper tube sheet, and estimating fuel content.
Different types of debris in various geometries were modeled using the Microshield computer program to generate gamma exposure rates based on known isotopic quantities.
The actual in situ gama radiation measurements were then compared with the computer-generated values to estimate fuel content.
l l
4
i o
OTSG Characterization
)
The OTSGs were characterized in two phases. The first included measurements of the lower heads and associated outlet J-legs.
The second I
phase included measurements of the upper tube sheets and the tube bundle
)
region.
)
Phase One i
During phase one of the measurement program, CoPhysics Corporation, under contract to GPU Nuclear, fabricated several fuel measurement strings to obtain gamma measurements in the ' A' and 'B' OTSG lower huds and associated J-legs. The fuel measurenent strings consisted of 100-ft long by 0.5-in, diameter polypropylene tubes which contained, at the front ends, two Geiger-Millier (GM) probes spaced at 6-ft intervals.
The fuel measurement strings were deployed through a guide tool mounted on the steam generator upper tube sheet and were pushed down through a 56-ft long steam generator tube (0.56-inch inside diameter) to the lower head.
They were then pushed an additional 15 ft further into the associated J-l egs.
Additionally, miniature lights and a Welch Allyn videoprobe were deployed through adjacent tubes to document placement of the strings within the respective RCS component and within any observed debris.
Each fuel measurement string ala contained copper foils for independent measurement of fuel-related material by a DOE national laboratory.
The foil analysis was perfomed and reported by the Pacific Northwest Laboratory (Reference 9)
During deployment of the fuel measurement strings, stabilized assay meters (SAM-2s) were connected to the GM probes to obtain dual gamma radiation readings at every 2 f t in the tube bundle region and at 1-f t intervals in the lower heads and J-legs.
The fuel measurement strings were visually inspected to verify the string positioning and contact with debris material within the steam generator lower heads and J-legs.
The debris in the lower heads and J-legs was a l.
i.
Iow-density material and was easily suspended while moving the videoprobe l
which resulted in poor visibility. This low-density material was uniformly distributed with a few areas containing granular material with a higher density and greater fission product content that corresponded well to the higher exposure rates.
Phase Two Characterization of the OTSG upper tube sheets in phase two of the measurement program involved placement of copper foils inside the ' A' and
'B' OTSG upper tube sheets to estimate the amount of residual fuel remaining by neutron activation of copper foils. Activation is generated by neutrons emitted from residual fuel.
Following activation, the foils were removed and counted in a coincidence system.
This system detemines the quantity of
- + emissions from the activated copper foils.
This process was repeated with an Am-Be neutron source to calibrate the foil and detector system for the scattering and neutron loss environment of the OTSG upper heads. Using the foil activation data, the average neutron activation flux was calculated to determine the fuel estimate of record for the upper tube sheets.
Characterization of the tube bundles for fuel debris blockages involved the deployment of a single measurement string into a total of 52 evenly-spaced steam generator tubes out of 15,531 tubes.
The probing locations were selected based on the effective radius of sensitivity of a gross gamma detector probe to a 1-inch long debris blockage.
An 8-inch effective horizontal radius was calculated by extensive modeling wit 5 the Microshield computer program which resulted in 52 deployment locations needed for full coverage characterization of the 9.5-ft diameter tube l
bundle region.
The one inch blockages were assumed to be vertically displaced 5.5 inches from a given measurement point and always at a horizontal radius of 8 inches.
This geometry will maximize fuel estimates by this method since the assumed location is roughly midway between measurement points.
The Phase 2 fuel measurement string contained six Gti probes at the front end, spaced at 1-f t intervals.
The I
string was deployed into a steam generator tube similar to that used in l,.
l l
,-w
phase one, and gamma measurements were recorded at 5-ft intervals down through the total 56-ft tube length.
The gamma readings from all GM probes were collected simultaneously at each 5-ft interval in the tube using 2 IBM personal computers, computer-operated Counter / Timer Boards, LabTech Notebook
- software (Reference 10'), and preamplifier-amplifier-discriminators (PADS) connected to the GM probes.
Two computer systems were installed to enable remote data collection from outside of the TMI-2 Reactor f
Building.
The host computer system was installed in the Reactor Building along with a nuclear instrument module regulated power supply (NIM BIN) containing six PADS and a high-voltage power supply.
The remote computer system was installed in the TMI-2 Command Center and connected to the host cuputer system by standard twisted pair cables and two Black Box Short Haul modems (Reference 11).
PC-AnywhereE (Reference 12) software was used to provide remote access to the host computer to operate the data collection system.
The data collection system installed in the host computer included E
LabTech Notebook DAS Sof tware and four Metrabyte Counter / Timer Boards * (Reference 13).
LabTech Notebook provided easy-to-use menu-driven data acquisition capabilities along with real time analisis, display, and process control.
In our application, LabTech Notebook
- was used to collect data using two of the four Metrabyte Counter / Timer D
Boards The Metrabyte Counter / Timer Boards are short slot boards with 5 independent,16 bit programmable counters which operate up to 7 MHz for event counting, timing, and freauency counting.
The input voltage levels used by these boards are standard TTL level 0 to 5 volts.
A universal terminal panel was used to interfere with the PADS.
The Canberra PADS Model 814A (Reference 14) were.aternally modified to provide an output pulse amplitude of 3 volts.
The counter / timer setup combined with the PADS worked very well.
The fuel measurement strings used in phase two of the measurement program were fabricated, assembled, and calibrated in the Rad Instrument Shop at TMI.
The GM probes were connected with RG178U ;oaxial cable and.
installed into polypropylene tubing.
At the end of each string, the coaxial cables were terminated into a utility box which provided strain relief for the smaller diameter cables, signal output connection, and a
{
common high-voltage input.
Each GM probe in a fuel measurement string j
was calibrated on a range using a 600 uC1 Cs-137 source and a calibration
]
curve was generated correlating exposure rate to counts per minute.
4.0 ANALYSIS The data collected during phase one of the measurement program was reduced and evaluated and fuel quantities in the ' A' and
'B' OTSG lower heads and J-legs were estimated by comparing the actual in situ exposure rates to calculated exposure rates for known activity concentrations modeled with the Microshield computer program.
The gamma radiation profiles for both the ' A' and
'B' OTSG lower heads and J-legs are shown in Figures 2 and 3.
In addition to the gamma radiation measurements performed, visual inspection of the lower heads and J-legs was used to confirm string positioning relative to contact with debris material and for estimating debris depth.
The debris observed in the ' A' OTSG was a very low-density material and was easily suspended while moving the videoprobe which resulted in poor visibility.
This low-density material was basically uniformly distributed with a few areas that were fluff zones containiiig material with crust-like surfaces.
These areas (the 8 foot locations in Figure 2) corresponded well to the higher dose rates (Reference 15).
The debris observed in the 'B' OTSG lower head was also a very low-density material of minimal depth (less than 1/8 inch).
The debris material in the J-legs was fairly uniformly distributed.
A few areas exhibited variable debris depths in the RCP '2B' J-leg (the 9 foot l
location in Figure 3) which contained small gravel-like debris and corresponded to the highest exposure rate (Reference 16). 1.
1
t The amounts of fuel in the ' A' and 'B' OTSG lower hea'ds and J-legs were estimated using Several Microshield calculations to model the debris material and geometry in these areas. Based on the output from Microshield, it was derived that the material in the J-legs was most similar to pressurizer debris material except in two one-foot locations in RCP J-legs '1B' and '2B', where it is most similar to a mixture of pressurizer and OTSG upper tube sheet debris material.
Additionally, for calculation of the source volume, it was assumed (based on Phase 1 visual inspections) that the debris material in the OTSG lower heads was a uniformly distributed low-density material of minimal depth within a 6-foot diameter source.
In the J-legs, the debris material was taken to be uniformly distributed (based on Phase 1 visual inspections),
with a few areas varying in depth throughout a 9-foot segment of piping (References 17 and 18).
Based on these assumptions, the fuel estimates for the ' A' OTSG lower head and J-legs were 0.29 kg and 0.67 kg, respectively, and for the
'B' OTSG lower head and J-legs were 0.46 kg and 5.79 kg, respectively. Tabulated fuel estimates are shown in Table 1.
The foil activation data collected during phase two of the measurement program for characterization of the ' A' and
'B' OTSG upper tube sheets were evaluated.
Four copper foils were placed in the 'B' OTSG and two l
copper foils were placed in the ' A' Ol'SG.
A copper foil was also exposed in the Reactor Building (RB) above the D-ring to determine the amount of background activation due to environmental neutrons.
The difference in activation flux between the 'B' OTSG upp" tube sheet and the RB background gives the activation flux due to residual fuel on the 'B' OTSG upper tube sheet.
The estimate of record of residual fuel on the 'B' OTSG upper tube sheet was based on two deteminations with resultant fuel estimates of 36.0 kg +,18% and 35.3 kg + 16% (Reference 6). The 36.0 kg value was chosen to be the estimate of record.
This compares to previously determined values by visual-inspection of 28.9 kg (Reference 20) and gross gamma of 53.9 kg (Reference 21).
w 4
4 From the ' A' OTSG foil calibration by the Am-Be source, an effective flux t
was determined.
Based on a comparison of exposure rates between the ' A' and 'B' OTSG upper tube sheets, 3 R/hr and 126 R/hr, respectively (Reference 39), and the 36 kg fuel estimate for the 'B' OTSG upper tube sheet, the ' A' OTSG upper tube sheet fuel estimate of record is 1.4 kg +, 21 %.
This value is based on correcting the exposure rate measured at 2 ft above the ' A' OTSG tube sheet to the exposure rate I
predicted at 1 f t for disc source geometry of the 'B' OTSG tube sheet.
The calculated value is 5 R/hr.
The ratio of the calculated ' A' OTSG value to the measured 'B' OTSG value of 1 ft separation was used to I
provide the estimate of record (1.4 kg) from the assayed quantity of 36 kg for the 'B' OTSG.
The measurement data collected for characterization of the ' A' and 'B' OTSG tube bundles was reduced and evaluated to detemine the possibility of debris blocking the OTSG tubes.
Fuel quantities were estimated by comparing the actual in situ exposure rates to calculated exposure rates based on a one-inch long debris blockage and known activity concentration modeled with the Microshield computer program.
It was assumed that the debris material present in a potential blocked tube was most similar in l
activity and density to the 'B' OTSG upper tube sheet debris material.
l As previously mentioned, gama measurements were obtained in the OTSG tube bundles using fuel measurement strings which consisted of six GM probes installed in polypropylene tubing and connected to associated electronics. Calibration response curves were generated for each GM probe in each measurement string.
The response characteristics of each GM probe were very uniform and linear within the 100-5000 mR/hr range.
Therefore, the response curves for each GM probe in a single string were averaged and calibration surveys and response equaticns were generated.
l The use of the averaged calibration data, along with the Lotus 1-2-3 Computer Program (Reference 22), facilitated the data reduction of approrimately 7500 data points.
l l
Probe number one in all fuel measurement strings was shielded with cadmium to eliminate any response to high energy beta and excess response to low energy gamma. Fuel measurement strings were deployed into the l
i l
.O c'~
tube bundles in intervals to overlap measurements with probes 1 and 6.
The GM probe count data for probes 1 and 6 were reviewed and correction
)
factors for over-response were calculated. These correctier..~ actors were applied to probes 2-6 and incorporated into the Lotus 1-2-3 spreadsheet.
The ' A' and 'B' OTSG GM probe measurement locations, mean measured exposure rates, and standard deviations are shown in Figures 4 and S.
Analysis of the ' A' and 'B' OTSG fuel measurement gamma probing data i
indicates that there are no significant bright spots (i.e., high exposure rates above background) within the tube bundle region that are attributable :o fuel blockages.
In comparing the radiation profiles for both steam generators, the exposure rates in the dry region of the
'B' OTSG were approximately a factor of 4 higher than the ' A' OTSG.
Similarly, the exposure rates in the wet region of the 'B' OTSG were approximately a factor of 2 higher than the ' A' OTSG.
High radiation areas within the upper 6 feet of the 'B' OTSG tube bundle are largely due
' to the contribution from the upper tube sheet.
Additionally, high exposure rates were also associated with the wate interface, possibly corresponding to radioactive cation capture in the form of a " bath tub" ri ng.
Exposure rates for other areas were relatively uniform and within the calculated deviation.
Previous measurements of fission products and transuranics were performed (Reference 1) on the inside stainless steel surfaces of one upper head nominal 16 inch manway insert for each OTSG. The results (Reference 23) l were used to predict that 0.2 and 2.2 kg of UO could be evenly plated 2
out in films in the ' A' and 'B' OTSGs, respectively.
Direct alpha L
measurements (Reference 2) of the top 20 feet of nine roughly evenly-spaced 'A' OTSG tubes were performed. Based on the alpha measurements, approximately 0.09 kg of UO were predicted to be held on 2
all internal surfaces of the ' A' OTSG tubes. This value is in reasonably good agreement with the earlier projection based on the ' A' OTSG manway insert plate.
l The second potential fuel compartment, namely fuel debris in blocked OTSG tubes, was evaluated using gross gamma measurements within approximately 52.
1 O
(
evenly-spaced tubes in each generator.
The gama field within a tube is taken to be the sum of background and of possible fuel debris located close enough to be measured.
Background includes cosmic ray, contamination external to the OTSG, contamination on the probe string (Figure 6), Cs-137 activity on the inside surface of the tubes, and the 0.07 uti/cc Cs-137 activity in the primary water wetting the lower portion of the tubes.
The Cs-137 activity per unit stainless steel surface (Reference 1), reduced for the lower corrosiveness of the inconel tube material, was used to predict the general exposure rate within the
+
tube bundles from the tubes (Reference 23).
The values are approximately 670 and 375 mR/hr for the ' A' OTSG dry and wet parts and 1185 and 705 mR/hr for the 'B' OTSG dry and wet parts, respectively.
Using these background values and the contamination on the probes and in the water, no statistically significant positive values in the ' A' and in the
'B' OTSGs were indicated.
The inferential method adopted for determining fuel is a modification of the conventional calculation for the lower limit of detection (LLD).
Normally, the LLD is calculated for 95% confidence. A smaller confidence level of approximately 60% was used to partly compensate for two geometry assumptions that lead to unlikely maximum values. They are:
- 1. Blocked tubes only occur at the maximum effective radius of 8 inches.
- 2. The point of blockage is always vertically displaced from a measurement location by 5.5 inches.
t.
From earlier discussions, all or nearly all of the gross gama fields within the OTSGs were due to Cs-137 activity on the primary side surfaces.
A leadscrew taken from the TMI-2 reactor vessel head was characterized for adherent films (Reference 24). Variation of factors of 2-3 are reported within distances of a few feet. Due to this expected variability, the OTSG gamma measurements were averaged for each five-foot increment of insertion.
Standard deviations were calculated for the variances between the 52 tube results for given insertions,
'(A-L).
The estimate of record was based on converting the agg,t) values to 00 weight (Reference 25), multiplying by 52 to account for 2
all space at a given depth of insertion, and summing the results for insertion increments A to L as:
L U02 = 52
{04 = estimate of record.
i=A The amount of UO2 predicted to be present in the tubes as films was not added to the total since it was assumed that the gross gamma method already accounted for this increment.
A large percentage of the variation in the rate could be attributed to variations in con osion layer thickness largely containing Cs-137.
It was assumed that all variances are due to fuel.
Based on the previously stated assumptions, the modeled steam generator fuel debris blockage, and the corresponding dose rates, fuel estimates for the ' A' and 'B' OTSG tube bundles were calculated. The estimates of record of total fuel debris deposited in the ' A' and 'B' OTSG tube bundles are 1.7 kg and 9.1 kg, respectively (Reference 25).
5.0 CONCLUSION
The use of the fuel measurement strings for obtaining gamma measurements and the copper foils for neutron activation proved to work very well considering the rough handling and strenuous work conditions encountered.-
in the Reactor Building.
Overall, the fuel measurement strings j
facilitated the deployment of the GM probes into the difficult-to-access
)
areas of the OTSGs and were less dose-intensive to personnel deploying eoutpment and performing measurements than alternate methods such as probing statistically significant tube cuantities of approximately 7000 tubes per OTSG.
The estimates of record of the total amount of UO remaining in the ' A' and 'B' OTSGs are approximately 4.1 kg i 22%
2 and 51.4 kg + 16%, respectively.
The uncertainty associated with these fuel estimates (except for the 'B' OTSG upper tube sheet foil activation fuel estimate of 36.0 kg i 18%) is based on a modeled geometry error of i16%, a source calibration error of 12%, a GM probe response error of i 10%, and a cesium-fuel ratio error of 144%.
The total uncertainty is based on the square root of the sum of the squares of the individual uncertainties associated with each :,omponent of the analysis.
The gamma probe measurement uncertainty aa.sociated with the above fuel estimates is 1 48%.
This estimate of record is derived from existing measurement and sample analysis data.
It is expected to remain static since it is not expected that additional quantities of water from the reactor vessel will be added or cycled through the OTSGs.
i.
r
.o c
o j
6.0 REFERENCES
i i
l
- 1. C. H. Distenfeld.
Deposition of Fuel on the Inside Surfaces of the j
RCS.
TB 86-37, Rev. 1.
Middletown, PA: GPU Nuclear Corporation.
29 November 1989.
- 2. B. Brosey.
OTSG ' A' Tubes Uranium Film Quantity and Alpha Probe Efficiency. Calculation No. 4530-3224-89-009. Middletown, PA:
GPU Nuclear Corporation.
31 January 1989.
P
- 3. DELETED
- 4. Neutron Measurement of the Fuel Remaining in the TMI-2 OTSGs.
Pacific Northwest Laboratory Operated for the U.S. Department of Energy by Battelle Memorial Institute.
PNL-6807. January 1989.
- 5. J. S. Schork. TMI-2 Core and Special Nuclear Material Accountability.
Administrative Procedure 4000-ADM-4420.02, Rev. 3.01.
Middletown, PA: GPU Nuclear Corporation.
10 October 1989.
- 6. M. H. Haghighi.
OTSG Upper Tube Sheet Fuel Estimates Using Copper Foil Activation.
TB-SNM-89-05, Rev. O.
Middletown, PA: GPU Nuclear Corporation. 23 May 1989.
- 7. Microshield Computer Code. Version 3.12.
GPU Nuclear Corporation.
Registered #109.
Grove Engineering Corporation.
- 8. Location and Characterization of Fuel Debris in TMI-2.
TP0/TMI-051, Rev. O. Middletown, PA: GPU Nuclear Corporation.
April 1984.
i i 6
9.
Dr. B. Geelhood, K. Abel. Calculation of Neutron Flux Seen by i
Copper Coupon.
Pacific Northwest Laboratories.
22 September 1988.
10.
Lab Tech Notebook Software.
Laboratory Technologies Corporation.
400 Research Drive. Wilmington, MA 01887.
11.
Short Haul Modem Model B - Asynchronous. Black Box Corporation.
P.O. Box 12800.
Pittsburgh, PA 15241.
12.
PC Anywhere III.
Dynamic Microprocessor Associates, Inc.
60 East 42nd Street.
Suite 1100.
New York City, NY 10165.
13.
DAC-02 Counter / Timer.
Metra Byte Corporation.
440 Myles Standish Bl vd.
Taunton, MA 02780.
14.
Pe Amp / Amp / Disc Model 814/A.
Canberra Industries. Inc. One State Street. Meridan, CT 06450.
15.
M. E. Greenidge.
' A' OTSG Lower Head and J-Leg Fuel Measurements.
Memorandum 4730-88-6057 to G. R. Eidam. Middletown, PA: GPU Nuclear Corporation.
23 September 1988.
16.
M. E. Greenidge.
'B' OTSG Lower Head and J-Leg Fuel Measurements.
Memorandum 4730-88-6058 to G. R. Eidam. Middletown, PA: GPU Nuclear Corporation.
6 October 1988.
17.
A & B OTSG J-Legs & Lower Head Fuel Estimates.
Calculation No.
4530-3221-88-028. Middletown, PA: GPU Nuclear Corporation.
20 October 1988.
l 18.
J. T. Horan.
' A' & 'B' OTSG Lower Head and J-Leg Fuel Estimates.
L TB-88-19, Rev. O.
Middletown, PA: GPU Nuclear Corporation.
26 October 1988.
l l
19.
GPU Nuclear Radiological Survey. File Code No. RB600-3841-87.
Middletown, PA: GPU Nuclear Corporation.
September 1987.
l l
l~
20.
P. J. Babel.
Calculation of OTSG 'B' Residual Fuel.
Calculation No. 4550-3234-88-002, Rev. O.
Middletown, PA: GPU Nuclear Corporation.
15 January 1988.
21.
P. J. Babel.
OTSG Upper Tube Sheet Reactor Fuel.
Calculation No.
4530-3224-88-019, Rev. O.
Middletown, PA: GPU Nuclear Corporation.
27 July 1988.
l 22.
Lotus 1-2-3.
Lotus Development Corporation.
55 Cambridge Pa rkway.
Cambridge, MA 02142, 23.
Calculation # 4800-3224-89-1E OTSG Tube Bundle Dose Rate.
30 November 1989.
24.
K. J. Hoffstetter, et al. Chemical Analysis and Test Results For Sections of the TMI-2 H-8 Leadscrew.
TP0/TMI-103.
February 1984.
25.
OTSG Tube Bundle Fuel Estimates. Calculation No.
4800-3224-89-006. Middletown, PA: GPU Nuclear Corporation.
15 May 1989, l
l t
1 i
y_.
g e
)
s o
i l
TABLE 1 ONCE-THROUGH STEAM GENERATOR FUEL ESTIMATES OF RECORD i
'A' OTSG
'B' OTSG i
Upper Tube Sheet 1.4 kg i 21%
36.0 kg + 18%
Tube Bundle 1.7 kg i 48%
9.1 kg + 48%
Lower Head 0.29 kg + 48%
0.46 kg + 48%
RCP-1 J-Leg 0.40 kg i 48%
1.5 kg + 47%
RCP-2 J-Leg 0.27 kg + 48%
4.3 kg + 49%
TOTAL 4.1 kg + 22%
51.4 kg + 16%
O
e FIGURE 1 TMI-2 ONCE-THROUGH STEAM GENERATOR Reactor Coolant (Tg)
Inlet o
Upper plenum j
- ==
Tube sheet I
r9 h
Emergency feedwater inlet (auxiliary feed nozzles)
E.: Seat (Apet. at Tubes rated load l
s e
o Baffle y
Film boiling e
Steam outlet region at l
rated load Bypass steam for A
gg gE o
feedwater heating e
Feedwater inlet o
y Downcomer section g
n Nucleate Shell boiling region at rated load u
n if
=
Tube sheet Lower plenum l
Reactor coolant outlets (T )
C 1
~
FIGURE 2 l
' A' OTSG LOWER HEAD AND J-LEG EXPOSURE RATE PROFILES gg. ~ "'sgg&Pb NOTE: MEASUREMENT LOCATIONS IN THE J-LEGS IN FEET ARE DISTANCES FROM THE x
LOWER TUBE SHEET.
O 12 11 (NOT TO SCALE) 2.0 in 2.3 8
4.2 8
3.8 W
7 3.5 Y
p 6
8.4 3
4,75 4
0.8 M '#
RCP - 2A RCP - 1 A J - LEG J - LEG Z
LOWER HEAD-PtAN VEW y
I I
I I
I I
I I
I I
I I
I I
I I
I I
I i
i 1
15 14 13 12 11 10 9 8 7 6 5 FEET 5 6 7 8 9 101112131415 37.58415.310.2184031.54.9 1.71.50.7 0.70.81.63.916.57.15.35.816.076.015.0
i l
FIGURE 3
'B' OTSG LOWER HEAD AND J-LEG l
EXPOSURE RATE PROFILES NOTE: MEASUREMENT LOCATIONS i
IN THE J-LEGS IN FEET 4.,, gg;ggr.
I ARE DISTANCES FROM THE LOWER TUBE SHEET.
GM PROBE 1 FAILED IN y
RCP-1B J-LEG. NO DATA O
BEYOND 9 FT LOCATION.
11 0
(NOT TO SCALE) 9 8
7 W
17.0 Y
5 15.2 d
I Q3 7.8 N$N U#
RCP - 2B RCP - 1B J - LEG J - LEG Z
LOWER HEAD-PtAN VEW i
l k
/
I I
I I
I I
I I
I I
I I
I i
i i
i i
i i
i i
15 14 13 12 11 10 9 8 7 6 5 FEET 5 6 7 8 9 101112131415 DOSE RATE 24517.51.8 0.9 0.6
^
~
(FlHR) 4
FIGURE 4
'A' OTSG TUBE BUNDLE 1
GM PROBE MEASUREMENT LOCATIONS A
UPPER TUBE SHEET B
8 FT
+
0 13 FT GM PROBE MEASUREMENT DATA LOCATION MEAN DOSE RATE (mR/hr)
D 18 FT A
350 + 59 B
320 7 37 C
207 7 26 E
D 204 7 25 Water Level E
228 T 24 23 FT F
387 7 19 G
384 T 22 p
H 334 7 23 i
= 28 FT I
215 7 19 J
187 T 17 K
164 T 13 G
L 123 T 21
~
33 FT t
H 38 FT l
43 FT J
l' 48 FT K
53 FT LOWER TUBE SHEET 58 FT L
NOTE: Numbers in feet are the distances below the top surface of the upper tube sheet.
FIGURE 5
'B' OTSG TUBE BUNDLE GM PROBE MEASUREMENT LOCATIONS A
UPPER TUBE SHEET l
1 B
l 8 FT C
i 13 FT GM PROBE MEASUREMENT DATA U
LOCATION MEAN DOSE RATE (mR/hr) 18 FT l
A 1290 + 295 B
1270 7 244 E
C 1213 I 247 Water Level D
679 7 150 23 FT E
1311 T 288 F
771 7 124 F
G 606 T 50 28 FT H
540 7 46 I
354 7 27 J
352 7 25 G
K 359 T 24 33 FT L
318 T 28 H
38 FT I
43 FT J
48 FT K
53 FT LOWER TUBE SHEET i
58 FT L
NOTE: Numbers in feet are the distances below the top L
surf ace of the upper tube sheet.
1
s 3
9 (e
5 n
t t._-
'2 o
i T
t a
9 a
. /
8 4
= C o
e e
t
/
c
- l a
o 4
C R
v Y
\\,
f e
/
l x.
/
/
\\
r d
=
e e
=
ml b
r
/
}
/
m a
i
\\
A u w l
[\\
M 3
/
e d
8 lI o
C a
s c
/
k e
i r
/
i t
a l
a (n
F c
e o
R l
e S
x 9
/np'W a!s o
L f.
m 4
4 W
i d 4
e m
8 e
1 ll e
aet 2
r 1
T- "
tro p
l
=
i t
o s e
Y V
?.
p
' na a
c s
4 S
(
csM a
f 8
t S
f
=
u
.i
[* au r
e y8 i
1 ie a
A t
r g-y_
=
,/~@,.n awc,/Ee. a/5m rd%
vl s
N m
1 n m i W,.
eip O
o s
o t tm I
o a ca T
o D AS etI i
v A
N
/
\\)
d I
v
\\
e M
t f
l r
c
=
A f
r 6 T r
N i
e,r N
5 e
C m.
P[
s*
E O e.
v 2,*
r u
R C a
i U
D. F.
a C
t G E s M 1 C.
i I
c n
n/a e
B o
o I5C8 w
F O t
i e
R o
t m.
r a
P o
i d
u, u
A
- n. u 1 2 a
F M
2 m
n o
e L
A R r. u.
D. F.
A G
a t
t h
s M l C.
c t a
N
\\l D
n/a e o E "A I5C8 T N t
I r
G ne p
m I
)
t r T /
\\
/
R u
f r
O
{
s e
f n
v N I
r u
e p
S u
D.
n t
o sMlf 9 n/af i
t I5CE a
i j
W0-n
)/
m T /
\\
a l
t
/
n e
o t
C e
te u
D.
h tsNlf 9 c 8
n/af e
8 I 5CE T
9 1
2 r
2 a
n e.
f C
n o
c s
e i
z o
t l
t
'e.
r E
a r
o W
e.
D n
c o e c#
o Y
m f
u p
=
o m
n i
T
=
I N
=
y s a.
3 r s r
e a n-i d
M.
v n
E r
o r
e u
iz r
w o~
S t
t t
r t
a i. s e
h v c r 'v x
n oo
?
c abt L a eue I
L E
D TSR f!
ll l
l
W.
p,.
4 e,
3
[
i i
s TMI-2 POST-DEFUELING SURVEY REPORT FOR THE REACTOR VESSEL HEAD ASSEMBLY i
i l
9 9
g.
Lm T-q
{[
A,.
r SUMARY J
The' estimate of record of.the amount of uranium dioxide (UO ) remainingin
)
2 the reactor vesse1 ~ (RV)' head assembly is 1.3 kg, with an uncertainty of +80%
j and -65%.
The U0 was distributed as follows:'
)
2 Films and Loose Contamination Leadscrew Motor Housings 0.12. kg l
Leadscrew Support Tubes 0.46 kg Dome and Flange 0.28'kg.
0.39 kg (Subtotal) 1.25 kg Gravel-like Materials
- 0. 01 3 r
. TOTAL 1.3 kg The-estimate of record of the quantity'of UO remaining in the RV head 2
assembly was determined by the extrapolation of dato reported on the gamma.
L scanning of residual fuel measured on sections of the E9, H8, and B8 L
leadscrews. These data were combined to create a composite 149ure of a i
l l:
The. quantity of UO remaining in the RV head assembly is (0.2% of the 2
j anticipated resitaal UO inventory
- for the entira TMI-2 facility in Mode 2.
2 h
i The anticipated residual UO inventory is as defined in the PDMS Safety 2
Analysis Report and the PDMS Programmatic Environmental Impact Statement and is based on the assumption that the defueling program goal to remove l
more than 99% of the original core inventory of UO is achieved.
2 r
.O---
n
-u
'l. N N
--- ------ -..a
f]
'[
s TMI-2 POST-DEFUELING SURVEY REPORT FOR THE REACTOR VESSEL HEAD ASSEMBLY.
3
1.0 INTRODUCTION
This-report presents the analysis of the Three Mile Island Unit 2 (TMI-2) reactor vessel (RV) head assembly uranium dioxide (UO ) inventory.
It-2 is one in a series of reports generated to fulfill the requirements of the TMI-2 SNM Accountability Program (Reference 2). All statistical data uncertainties are' expressed as + one sigma limits (defined as one standard d?viation).
Section 2, " Background", ' describes the physical attributes, location, ~ and l
intended functions of the RV head assembly.
The relationship of the RV j
head assembly to the accident and suusequent cleanup activities is' L
discussed as well as its current status.
Section 3, " Method", describes generally how the restdual UO in the RV j
2 head assembly was calculated using leadscrew data.
Section 4, " Analysis", explains in detail how the estimate of record of the amount of fuel in the RV head assembly was calculated, and discusses j
supporting data, assumptions made, and the assigned uncertainties.
l Section 5,'" Conclusion", cments the estimate of record and uncertainty
~
i 1
for the amount of Fesidual UO remaining iit the RV head assembly and
~
2 supporting ratlonale leading to a conclusion that the estimate is i
r reasonable based upon the available data and analysis performed.
I,
+
1
2.0 BACKGROUND
1 The RV head assembly covers the reactor vessel and serves as a containment barrier for t' e reactor coolant.. It has a hemispherical n
segment top portion attached to a cylindrical flange bottom portion.
The
~
hemispherical segment accommodates a protrusion of leadscrew motor housings (LSMHs), leadscrew support tubes (LSTs), and leadscrews (Figure 1)..The surface areas of these components are listed in Table 1.
The RV head assembly, as considered for this report, consists of 69 LSMHs that extend from the thermal barrier into the underhead area, 69 LSTs, 66 'leadscrews that extend from the thermal barrier to the tip of their bayonets (three leadscrews were removed for analysis prior to head-remova'l), a dome, and a flange.
The boundaries of the RV head assembly covered by this PDSR extend from the thermal barrier to the end of the LSMHs and LSTs, and the inside surface area of the hemispherical segment and flange.
This report considers the RV head assembly as currently stored in the Reactor Building.
During the 1979 TMI-2 accident, fuel relocated to the RV head assembly when the water boiled and steam rose within the vessel. The RV head assembly was removed from the vessel in July 1984 and is currently stored at the 347'-6" elevation of the Reactor Building on its storage stand.
The radiological condition of the RV head assembly after head removal, at a point approximately 120 cm beyond the inner diameter of the reactor vessel and 150 cm above the flange plane, was between 8 and 20 R/hr (Reference 3).
The measured dose rate in proximity to the plenum was in the 400 to 1000 R/hr range (Reference 4).
Radiological Environment The reactor. vessel head is resting on the head storage stand in the southwest section of the 347'-6" elevation. The stand is a heavy steel
_?_
i
..g.
I frame which includes a 2.54 cm-thick cylindrical steel skirt that eliminates access to the. interior of.the head.
Interlocking sand-filled j
shield cylinders are oriented on end to fom an overlapping helical
)
shield wall designed to preserve the moderate general area radiation level of-30 to 50 mR/hr endemic to the 347'-6" elevation.
Measured Cs-137 film values on surfaces establish a general area exposure rate of 5 to-10 R/ha under the head.
Further measurement efforts are not considered to be ALARA, since the work would only tend to decrease the uncertainty of a fuel value that is small when compared to the residual fuel residing in the reactor vessel.
This condition, together with the hostile environment under the head, suggests that the cost would outweigh the benefit.
-3.0 METHOD The total fuel Quantity was derived as the. sum of three general increments:
the fuel contert on the leadscrews, the fuel on the exposed
-underhead surfaces, and the fuel on the hidden or inside underhead surfaces.
The leadscrew increment was based on direct assay of the fuel content of various sections of two leadscrews.
This and other data was smoothed to define a " typical" leadscrew which,'when extended over all remaining leadscrews, provided the total for this increment.
Since no other direct fuel-related data exists for the remaining surfaces, radiochemical assays of a piece of leadscrew support tube and l
of the previous sections of leadscrew were used to infer fuel content for tne remainder.
Radiochemical results for the inside and outside surfaces l
of the leadscrew support tube were used as the means to relate measured i
and inferred surface fuel values for the leadscrews to the other surfaces. The inside leadscrew support tube Cs-137 activity was assumed to be proportional to the average fuel content of the leadscrews.
The ratio of Cs-137 activity on the outside to the inside of the leadscrew l
support tube was used to establisn the fuel content for the external j
surfaces. 1
s.
i4l in November 1982, the E9, BB, and H8 leadscrews were removed for analysis.
In July' 1984, a small section (approximately 9 cm long) of the H8 LST was removed and sent to Battelle Columbus Laboratories for radioisotopic analysis.
To date, fuel analyses have been performed by Scientific Applications, Inc. (SAI) and Babcock & Wilcox (B&W) on other sections of' these leadscrews.
The estimate of record of the amount of UO remaining in the RV head 2
assembly was determined by extrapolation of data reported on residual UO2 (areal density) measured on the E9, 88, and H8 leadscrews.
These data were combined to create a composite figure of a leadscrew (Figure 2).
The 61 leadscrew.; connected to the fully extended control rods were extended through the plenum, and the 8 leadscrews connected to the axial l
power shaping rods were only 75% extended.
Extrapolating fuel
. measurement values for the control rod leadscrews, which were in more intimate contact with the fuel, to the RV head assembly is a conservative approach to determining the amount of UO in the RV head assembly.
2 Film values from sections of the E9, B8, and H8 leadscrews (Figure 3) represent values for the RV head assembly surfaces because of the proximity of these leadscrews to the RV head assembly during the accident.
The RV head assembly contains two types of surface areas: the inside surface areas and the outside surface areas.
As indicated in Figure 4, the inside surface areas are considered to encompass the leadscrews, the inside of the LSTs, and the inside of the LSMHs.
The outside surface l
areas are considered to encompass the flanges, the dome, the outside of l
the LSMHs, and the outside of the LSTs.
Since the outside surfaces are characteristically vertical tube surfaces, the data extrapolated from the l
1eadscrews, which include the threaded portions, to those surfaces will probably result in high estimates.
l' l
1 i
c 1
Fuel sample examinations and code calculations at-THI-2 ' support the use of Ce-144 and Eu-154 as fuel tracers (Reference 6).
The activity of-a radioisotope that is chemically similar to fuel can be used to calculate i
the amount of UO.
A measurement of the isotopic activity should be 2
directly proportional to the amount of UO material present.
Ce-144 2
has been shown to be an analog for transuranic material.
Cerium oxide compounds are not water soluble and Ce-144 was transported through the reactor coolant system during the accident.
However, the radiochemical analysis that was performed did not always report Ce-144 activity thereby reauiring the use of a different isotope.
The quantity of residual UO on the RV head assembly was obtained by 2
characterizing surface activity on the LST section and by assuming that Cs-137 activity follows the same trend as Ce-144.
This effort is ba'.ed on an analysis performed by Battelle Columbus Laboratories (Reference 5) which indicated that the outside activity of Cs-137 on the LST is approximately twice that of the inside. Therefore, an assumption of the calculation is that outside fuel activity on the tube is twice that of the inside.
The fuel content was calculated as follows:
A. The fuel content of the leadscrews was taken directly from the composite leadscrew.
B. The fuel content of the remaining RV head assembly components was calculated by:
o inferring the amount of fuel per unit surface o multiplying by the corresponding surface area of the RV head assembly component
,=
o summing the results of each component
e
- r. -
x I
r 4.0 ANALYSQ The areas of the RV head assembly components were calculated by representing each 'section by a geometrical shape (Figure 5).
The sum of the areas of these geometries resulted in a ; tal surface area of 1,44 E6 cm2 (Table 1).
An effective average fuel area density of 628 ug/cm2 (Reference 7) was calculated for inside surfaces of the reactor vessel head assembly areas. The details are shown in Table 2.
Since the outside surface activity is conservatively estimated to be L
twice that of the inside activity (Table 3), the effective average fuel 2
areal density for outside surfaces is 1256 ug UO /cm,
2 Activities sequestered in surface films consist of two layers (Reference 8):' A surface layer which is loosely adherent and prohably represents deposited aerosols, and a deeper tightly-bound layer probably chemically bound to the surfaces.
The loosely bound layer includes mixed oxides which probably contain uranium, cerium, and europium in tha standard relationship (R( ference 6).
The deeper tightly-bound layer is probably due to physical and chemical adsorption. The latter would be strongly affected by solubilities of the referenced isotopes, and would not follow the ORIGEN code prediction for bulk dry fuel.
For this l
reason, radiochemical analysis of or near the objects of interest are I
important to characterize the surface deposits.
The fuel value for each RV head assembly component was then calculated.
The results for the components were summed to detennine the fuel content of the RV head assembly.
The methods used for the calculations performed for each component are discussed in the following paragraphs.
The possibility of gravel-like deposits being trapped in RV head assembly components is highly unlikely due to gravity and RV head assembly orientation. Therefore, contributions due to gravel-like material (estimated to be less than 0.01 kg based on observations in Reference 3) are insignificant when compared to the films and loose contamination contributions for these calculations. -
rf--
m m.
.m
. ~:
- l 4.1 L'eadscrew Motor Housings 1
The estimate of record of the amount of UO remaining in the LSMHs is 2
0.12 kg (Table 1). This estimate of record was determined.by multiplying.
the total surface area of the 69 LSMHs by the corresponding fuel value.
per square centimeter. This area consists of inside surfaces and outside.
surfaces.
The remainder of the outer areas of the LSMHs were not in contact with reactor coolant system water, and.therefore, are not included-in the calculation.
4.2 Leadscrew Support Tubes ll The estimate of record of the amount of fuel remaining in the LSTs is L
0.46 kg (Table 1). This estimate of record was determined by multiplying the surface area of the 69 LSTs by the corresponding fuel value per square centimeter. This area consists of inside surfaces (inner surfaces from the thermal barrier to the end of each LST and the LST outer surfaces not covered by the LSMHs) and outside surfa:es (outer surfaces of each LST.from the end of the LSMH to the end of the LST).
The'effect of the area of the 9 cm-long section of the H8 LST cut for analysis.is negligible compared to the total LST areas (the area of the analyzed' section is 0.07% of the total area of LSTs). A corresponding small overestimate of fuel was accepted since all of the LST surface area was used.
4.3 Dome and Flange
- The estimate of record of the amount of fuel remaining in the dome and flange is 0.28 kg (Table 1).
This estimate of record was determined by multiplying the total surface area of the dome and flanges (inner surfaces only, since these are the areas that could have been in contact with reactor coolant system water) by the corresponding fuel value per square centimeter. The total surface area for the LSMH holes is subtracted from the spherical segment surface area representing the dome.
~
4.4 Ceadscrews The estimate of record of the amount of UD-remaining in the leadscrews is 0.39 kg (Table 1).
This estimate o:
'n x s determined by using available fuel value information repe iections of the leadscrews, treating the remainder of the leadscre-uside surfaces, and 4
calculating an average fuel value for these sections by using the average cerium-to-fuel ratio for the threaded sections of the leadscrews.
This cerium-to-fuel ratio is determined by measured values for the leadscrews (Reference 7).
The average cerium activity of these locations was divided by this cerium-to-fuel ratio to obtain the fuel value.
Using average fuel values reported for the threaded section of leadscrews (Reference 7), fuel value information of films for the leadscrews was combined to produce a composite figure of a leadscrew (Figure 2).
The fuel value contained by this leadscrew was calculated to be 5.85 g (Figure 2).
There are 66 leadscrews currently parked in the RV head assembly (since three leadscrews have been removed). Therefore, the estimate of record of the amount of fuel contained by the leadscrews in the RV head assembly is 0.39 kg.
'4.5 Analysis Conclusion The total fuel value captured on surfaces of the RV head assembly is 1.3 kg.
A range of values were calculated based on differing models.
Comparing the values to the quantity of record provided the error estimates of + 80% and - 65%, which were adopted as the estimates of error..
'l g
r
. - 3.
- , 7 5.b donclusion
't 1
The estimate of record of the. amount of uranium dioxide remaining in the RV head assembly. is 1.3 kg, primarily in the form of surface films and loose contamination.
This estimate has an associated uncertainty of +80%
and -65%.
Any additional analysis of leadscrew support tubes or RV head components is not expected to significantly alter this estimate.
The relatively small amount of residual UO remaining in the RV head assembly, when 2
viewed in the context of the maximum allowable plant inventory of residual UO2 ( <0.2%), does not warrant the additional occupational exposure.
Additional measurements of the RV head assembly are not considered to be justifiable based on ALARA considerations. Analysis of additional leadscrew samples would result in significant radiation exposure to technicians perfonning the measurements.
The radiation exposures-could be justified if the additional measurements were expected to significantly enhance the current estimate of record.
However, review of the current measurement data indicates that additionel sample analysis would not be-expected to result in a significant change to the overall estimate of record.
Additionally, the relative ~ amount of residual UO 2
in the RV head assembly compared to the projected total plant inventory of residual UO does not warrant the added person-rem burden.
2 The RV head assembly will remain on its storage stand in the Reactor Building; therefore, fuel quantities are not expected to change.
L l
1 i
-l grERENCES
-I r,
-.1. -
DELETED
-i
- 2..
J. S. Schork. TMI-2 Core and Special Nuclear Material Accountability.
'j Administrative Procedure 4000-ADM-4420.02, Rev. 3-01.
Middletown, PA:
GPU Nuclear Corporation.
10 October 1989.
3.
V. R. Fricke.
Dose Modeling of Underhead Source.
GPU Nuclear Corporation. TP0/TMI-042, Rev. 2.
April 1984.
4.
R. Rainisch.
Underhead Data Acouisition Program. GPU Nuclear Corporation.
TP0/TMI-110, Rev. O.
March 1984.
5..
M.P. Failey et al.
Examination of the Leadscrew Support Tube from Three Mile Island Reactor Unit 2.
GEND-INF-067.
Battelle Columbus Laboratories. March 1986.
6.
P.J. Babel. _ Ce-144, Eu-154 and Eu-155 as Tracers for Fuel Debris.
TB 86-41, Rev. 2. Middletown, PA: GPU Nuclear Corporation.
26 February 1988.
7.
M. H. Haghight.
Reactor Vessel Head Assembly Fuel Value Estimate.
GPU Calculation No. 4800-3221-89-098, Rev. O.
Middletown, PA: GPU Nuclear Corporation.
29 September 1989.
8.
K.-Atwood.
Characterization of Films Present on Systems Internal Surfaces.
TP0/TMI-150, Rev. O.
Middletown, PA:
GPU Nuclear Corporation. March 1985.
9.
R.E. Lancaster.
Calculation of the Reactor Yessel Closure Head.
4550-3221-87-042, Rev. O.
Middletown, PA: GPU Nuclear Corporation.
- 13 November 1987.
10.
K. Vinjamuri, D.W. Akers, and R.R. Hobbins.
Examination of H8 and B8 L
Leadscrews from Three Mile Island Unit 2.
GEND-INF-052.
L EG8G Idaho, Incorporated.
September 1985.
l l.
l
TABLE 1-FUEL YALUE OF'RV HEAD COMPONENTS-Effective Avg.
Surface Number Total Fuel Per Total Area (*)
of Surfact) Area- (vg UO2/cm2()f)
Unit Area Fuel Component (cm2)
Components (cmz (g 907)
-8.76 E 4 1
8 76 E 4 1256 110.1-Dome 1,39 E 5 1
1.39 E 5 1256 174.6 (minus LSMH holes)
LSMH Inside' areas 2.36 E 3 69 1.63 E 5 628-102.4 Outside areas 2.04 E 2 69 1.41 E 4 1256 17.7 LST Inside areas 3.10 E 3 69 2.17 E 5 628 136.3 Outside areas 3.70 E 3 69 2.56 E 5 1256 3 21. 6 Leadscrew ( )
8.60 E 3 66 5.65 E 5 386.1 Total 1,44 E 6 1248.8 1
l:-
l
(*) All areas obtained from Reference 9.
(f) Effective average fuel per unit area and total fuel area obtained from Reference 7.
(t) Three leadscrews were removed (H8, B8 and E9).
1
T,W.
e i8:
TABLE 2 RV HEAD ASSEMBLY FUEL SURFACE DEPOSITS
- v 2
Method **
M (g/cm )
2 1
4.97 E-4 2
5.78 E-4 l
3 6.95 E-4 4
8.07 E-4 5
2.15 E,
6 9.20 E-4 7
6.85 E-4 Average 6.28 E-4
.i
(
j
- For all surfaces except for leadscrews
- (1) Quotient of H8, B8, E9 UCi Ce-144/cm2 (at head elevation) and average l
mci Ce-144/g UO2 for H8 and B8 threaded section.
(2) Quotient of H8, B8, E9 pCi Ce-144/cm2 (at head elevation) and average-uCi Ce-144/g U02 for H8 and B8 unthreaded section.
(3) Same as (1) above, eliminating E9 data.
(4). Same as (2) above, eliminating ES data.
i j
(5) Average fuel density thickness measured for unthreaded sections of leadscrews.
(S)' Same as (5) above, except for threaded sections of leadscrews.
~(7) Same as (5) above, except for all leadscrew d.ta.
3
e w.
,e 1.
- -s
.. j L
TABLE 3 0
'1 H?, LEADSCREW SUPPORT TUBE SURFACE ACTIY1TY MEASUREMENTS *'
-l l
s L
Activity per Measured Unit Arta Ratio Sample Nuclide (aCi/cm )
Cs-137/Cs-13,4, r
6-1.1 Outside Cs-137 1250 Cs-134 53 23.5 l
6-1. 2 ' Insi de Cs-137 633 Cs-134 27 23.4 w
- Reference 5 r
I; 3
I.. f '
y
--g..
- e 1
FIGURE 1 TMI-2 REACTOR.- UPPER HALF (Simplified)
Thermal t
Barrier
.il 1 jl3 -
Head
~
o mp
~
'^
w Leadscrew y"C'"
4 Support Tube 7r Leadscrew s'
\\n e\\,,,,
p;",,
Y gl' m,
p m-minummmmmmmmmmmmmmmm g y.,.
11 N
,/
_qtg C. 3
\\ f'
' \\ ;r
- r 3lf '\\
Rf w
L' 4
("j i
W Upper Plenum (v / ("
(v/
i
/
$# Assembly
e i
/
Y:
L
~4;>
2 e
e Core Support
' ~
% Shield J.. a.. m.
n o
e
?% tw:
d ?j"
)
mz
- v,
@ (yjygold Leg e
e o
gi e
e L E.igy Nozzle g; /
\\ "c%
m a
a j
h.N,
~
L 4
i'
{ ore Barrel e
m
- f" M@[N' y
t.
,,a F
T-Fuel Assemblies pg
/s
'~l Reactor Vessel Wall (Drawing is not to scale)
'"W5-9' W
F e
w w
. ~
,, : a ; y g..
I '.-
FIGURE 2-i COMPOSITE LEADSCREW SECTIONAL FUEL VALUE DATA (Drawing'is not to. scale),
'; Thermal ' Barrier
!T6p of RV Head i
2 170.2 cm 939 pg UO /cm 2.75 g UO 2
2 2
22.9 cm 755 pg UO /cm 0.3 g UO 2
2 2
22.9 cm 503 mg UO /cm 0.2 g UO p
2 2
0 cm 1700 yg UO /cm 0.9 g UO 2
2 p
-Assembly 2
f.1 g UO 87.6 cm 719.6 pg UO /cm 2
2 2
229.9 cm 215.3 pg UO /cm 0.6 g UO 2
2 t
Total U0 for the leadscrew = 5.85 g UO2 2
Total of 66 parked leadscrews = 5.85 g U0 x 66 = 386.1 g U0 2
2
-B5ttom'of Leadscrew 2
Threaded section area of leadscrew/cm - 17.38 cm /cm.
2
- Unthreaded'section of area of leadscrew/cm - 11.97 cm /cm.
I ~ NOTk[: Data from. Reference 7.
en
-v.
-n e
e
~
i t-3p; j {-. _
FIGURE 3
.E9, 88, AND H8 LEADSCREW LOCATIONS IN THE TMI-2-CORE
- Assembly coordinates ABCDEFGHKLMNOPR
//H/
V
'f y
15
/"'00
~I9
/"'HO 1
9
- (
/
n
/
10
[
eI
/
~
A-outlet 8
l 6
4
- =
B-outlet h
)
l
~
g
-5 4
Thermal 2
shield 1-Core barrel f-Inlet niet 11
~
U///
- Reference.10 2
>I F2!GURE 4 RV Head Assembly Component Surfaces Leadscrew
=
~ Thermal Barrier No Fuel on Surface Outside Surfaces inside Surfaces i
USMH W eld LST
=
l
%m
=
]
=
y i
(Drawing is not to scale) j
\\.
, +...
,.-,.4 i
- FIGURE 5 GEOMETRICAL REPRESENTATION OF HEAD ASSEMBLY
- Component Geometrical Shape Area (cm2)
Diagram flange annulus cylinder.
+8.76 E 4 I
dome hemi-spherical
+1.44 E 5 segment holes for LSMH circles (69)
-5.59 E 3 l
l l
inside LSMH cylinders (2) and
+1.63 E 5
)
s l
cut-off cones (69) l l
- Reference 9
r.. ;.._.-
,........~m
- s'. 4 -
FIGURE 5 (Continued) e GE0 METRICAL REPERSENTATION OF HEAD ASSEMBLY
- 3-e Component l
' Geometrical Shape-Area (cm2)
Diagram outside LSMH '
annulus cylinders (69)
+1.41 E 4 g
-l F
inner LST cylinders (69)
. 2.17 E 5 q
+
N 1
L l-
\\
li outer LST.
cylinders (69)
+2.56 E 5 l
l L
l leadscrew cylinders (66)
+5.65 E 5 (threads considered)
- Total area 1.44 E 6 f
a
-