ML20125D040
| ML20125D040 | |
| Person / Time | |
|---|---|
| Site: | National Bureau of Standards Reactor |
| Issue date: | 10/26/1979 |
| From: | NATIONAL INSTITUTE OF STANDARDS & TECHNOLOGY (FORMERL |
| To: | |
| Shared Package | |
| ML20125D027 | List: |
| References | |
| NUDOCS 8001130086 | |
| Download: ML20125D040 (92) | |
Text
._.
(op y b
/sf.4)7R f
Report on the Nature and Source of Radioactive Material Found at the NBS Reactor Site Prepared by the Ad Hoc Committee on Radioactive Material 90018093 w
(
Table of Contents Chapter 1.
Results of the Survey to Find and Remove Radioactive Material 1.
Introduction 3
2.
The Surveys Carried Out 2.1 Introduction 4
2.2< Definitions 5
2.3 Equipment 5
2.4 Site Survey 2.4.1 Description of Reactor Site 6
2.4.2 Site Survey Strategy 7
2.4.3 Results 2.4.4 Positive Result Areas 8
2.4.4.1 Reactor Lawns 8
2.4.4.2 Reactor Roof 8
2.4.4.3 Holding Basin 8
2.4.4.4 Outfall 9
2.4.4.5 Drain Exit and Beyond 9
2.4.4.6 Enclosed Area 9
2.4.4.7 Samples Beyond Drain Exit 10 2.4.5
' Negative Result Areas 11 2.5 The Monitoring Survey 12 2.6 Sample Assay Results 13 3.
Conclusions 14 Chapter 2:
The Analysis and Source of the Particles 15 1.
Introduction 16 2.
Chemical and Metallurgical Analysis 2.1 Particle Analysis 17 I
2.2 Wear Rings 17 2.3 Discussion 19 3.-
Investigation of the Radioactivity of the Particles 3.1 Introduction 22 3.2 Cobalt-60 Specific Activity 22 3.3 Iron 55 Activity and the Age of the Particle 25 3.4 Discussien 27 i
4.
Conclusions 27 i
i Chapter 3:
Possibilities for Particle Release 28 1.
Introduction 29 2.
Heat Exchanger Removal
2.1 Background
30 2.2 The Removal 31 3.
DiscusskonandConclusions 35 Acknowledgemerit 36 Tables I-VII Figures 1-9 Appendices A-D 90018095 9
11
J J
m Chapter 1 Results of the Survey to Find and Remove Radioactive Material 90018096 l
l 2
o Results of the Survey to Find and Remove Radioactive' Material
-I.
Introduction
-On August 28, 1979, an inspector from the Nuclear Regulatory Commission (NRC) came to NBS on a routine, unannounced inspection of the NBS Reactor (NBSR)
Health Physics. As part of the inspection, he conducted a routine survey of the area around the NBSR. The inspector's survey. instrument was a Ludlum model 16 survey meter (0 to 500 cpm, ranges x 1,000, x 100, x 10, x 1; he was on the x 10 scale, or 0 to 5000 cpm, on the recollection of NBS personnel, and the ambient background was about mid scale) with a Ludlum model 44-2 probe, which has a 1" x 1" NaI(TA) detector.
This survey meter probe, thus, is sensitive only to photons above about.100 kev; it reads counts per minute and would not read in mR/hr unless it was calibrated with the same photon energy as that measured.
Using this sensitive detector he found a spot of radioactivity on the lawn at the front of the reactor.
In cooperation with NBS personnel, he subsequently found other spots of radiation in the rear of the reactor adjacent to a fence enclosing a controlled area in which a heat exchanger that was removed from the reactor in 1974 was stored, and a more extended spot under and beside the inactive heat exchanger. Further investigation led to the detection of two more spots on the roof of the reactor building.
Samples taken from these radioactive spots were taken into the NBS laboratories and assayed with analytical equipment (see Sec.
2.3.for a description of the equipment used).
All the radioactivity was shown to come from 80Co. As a result of these initial finds, a thorough survey of the NBS grounds was begun immediately, along with other investigations.
On August 31, Dr. Ernest Ambler, NES, Director, appointed an ad hoc committee consisting of R. S. Carter, A. Schwebel, K. Bell, and E. Passaglia (chairman) to:
1)
" Track the complete survey which is now in progress so as to insure that such a survey includes a careful examination of the Bureau's grounds, relevant equipment and the shoes of those NBS employees who may have walked in the area of the nuclear reactor.
940018097 L
3
1.
Results of the Survey to Find and Remove Radioactive Material I.
Introduction On August 28, 1979, an inspector from the Nuclear Regulatory Commission (NRC) came to NBS on a routine, unannounced inspection of the NBS Reactor (NBSR)
Health Physics.
As part of the inspection, he conducted a routine survey of the area around the NBSR.
The inspector's survey instrument was a Ludlum model 16 survey meter (0 to 500 cpm, ranges x 1,000, x 100, x 10, x 1; he was on the x 10 scale, or 0 to 5000 cpm, on the recollection of NBS personnel, and the ambient background was about mid scale) with a Ludlum model 44-2 probe, which has a 1" x 1" NaI(TA) detector.
This survey meter probe, thus, is sensitive only to photons
]
above about 100 kev; it reads counts per minute and would not read in mR/hr unless it was calibrated with the same photon energy as that measured.
Using this sensitive detector he found a spot of radioactivity on the lawn at the front of the reactor.
In cooperation with NBS personnel, he subsequently found other spots of radiation in the rear of the reactor adjacent to a fence enclosing a controlled area in which a heat exchanger that was removed from the reactor in 1974 was stored, and a more extended spot under and beside the inactive heat exchanger.
Further investigation led to the detection of two more spots on the roof of the reactor building.
Samples taken from these radioactive spots were taken into the NBS laboratories and assayed with analytical equipment (see Sec.
2.3 for a description of the equipment used).
All the radioactivity was shown to come from 80Co. As a result of these initial finds, a thorough survey of the NBS grounds was begun immediately, along with other investigations.
On August 31, Dr. Ernest Ambler, NBS, Director, appointed an ad hoc committee consisting of R. S. Carter, A. Schwebel, K. Bell, and E. Passaglia (chairman) to:
1)
" Track the complete survey which is now in progress so as to insure that such a survey includes a careful examination of the Bureau's grounds, relevant equipment and the shoes of those NBS employees who may have walked in the area of the nuclear reactor.
90018098 3
l
- 2)
..." undertake a complete review of the Bureau's safety and operating.
procedures with respect to ionizing radiation in general and radioactivity in particular...and to staggest... revisions...as may be needed...to prevent future occurrences of this nature."
Implicit in Dr. Ambler's assignment is a directive to find the origin of the radioactive material, and to find how it got to where it did. Accordingly, a
the ad-hoc committee took the following to be its tasks:
1)
Carry out a survey of the total Bureau grounds to define the limits of the spread of the radioactive material and to clear up any radioactivity found..
2)
To seek the source of the radioactive material.
3)
To determine how the material got to where it was found.
4)
To recommend procedures to prevent any recurrence.
This Chapter is the report on the first of these talks;Eh 2.
The Surveys Carried Out 2.1 Introduction It is important to recognize that three surveys were carried out.
These were 1)
A site survey.
This was made to define the limits of the radio-activity and to determine if any hazard was posed to the workers of NBS and the citizens of the surrounding areas. This survey was completed by September 4, 1979.
All radioactive spots found in this survey were removed and cleaned up by August 31, 1979.
2)
A survey of the streams external to the NBS site.
This was carried out with an inspector from the Maryland Department of Health, and will be called the "Off Site Survey." This survey was carried out on September 7, 1979.
The objectives were the same as in (1).
6 3)
A monitoring survey of the area immediately, surrounding the reactor building. This survey was completed on September (ll,1979, and all radioactive spots were removed by that date.
This was carried out with very sensitive instruments (see below, Sec. 2.3) and had two purposes: (a) to discover any extremely low-level radioactive spots that might have been missed in the site survey and (b) to assure that no further radioactive material was spread during investigations carried out in the enclosed area.
90018099 4
l This report will describe the results'of all three of these surveys.
7 2.2 Definitions Four terms will be used in ' describing the results of the survey.
They are:
1)
" Area".. This denotes the. general region of the survey, e.g., " reactor roof," " storm drain," etc.
80Co was found.
It.
2)
" Spot".
This denotes a place where radioactive might denote a very localized region or a'more extended region.
In every case, the activity was dug up and removed to a safe storage area.
3)
" Sample".
This denotes a sample (usually a clump of dirt, mud, or sand), removed from a spot. A spot might yield one or more samples.
4)
" Particles".
The cause of the 80Co radiation was small metallic particles (from fractions of a microgram to as high as 300 micrograms).
Hence
" particles" refers to radioactive particles recovered from the samples (often a tedious process).
A " sample" might contain one or more particles.
2.3 Equipment This section describes the instruments used in the survey, and those used in assaying the level of radioactivity in the samples taken to the laboratory.
The survey instruments used were of several types.
The most sensitive instrument used was a Nuclear Enterprises Model MARK IV Scintillometer, which uses a 3" x 3" NaI(T2) detector.
This was used for the monitoring survey.
It has a low range of 0 to 0.025 mR/hr, with additional ranges of 0 to 0.1, 0.25, 1.0, and 2.5 mR/hr.
A variable time constant switch is provided; for all the surveys, it was used on the most rapid, i.e., 0.25 seconds.
This gave an indi-cation which fluctuated rapidly, but quickly showed any increase in ambient backgrounds.
An estimate of the minimum detectable activity is 20 percent above ambient, which was approximately 0.005 mR/hr.
There is no audio output'with this instrument.
Other survey instruments included 1" x 1" NaI(T2) probes and thin end-window G-M probes.
The scintillators were less useful as no audio output was available and the sensitivity was no greater than with the inspector's
(
detector. The G-M devices were used primarily for pinpointing radioactive spots when a more sensitive, but less directional instrument showed a high reading.
The G-M meters were useful in the initial stages of the survey, as the radioactive spots were enough above ambient background that responses 90018100 5
from the G-M instrument were positive.
The scintillator and the G-M probes were used with survey meters that have a scale of 0 to 0.2 mR/hr., with ranges of x l, x 10, and x 100.
Ambient background with these was also 0.005 mR/hr., with wide fluctuations, since both are event counters.
All surveys were conducted by experienced Health Physics or Reactor Operations personnel.
The assays were carried out on equipment of the Analytical Chemistry Activation Analysis Group by personnel of'that Group for particle assays and Health Physics personnel for sample assays. Various detectors and multi-channel analyzers (MCA) were used for qualifications and quantification of activity.
The detectors were, for nearly all assays, Ge(Li) from 63 cc to 80 cc.
The MCA's included several older Nuclear-data units, including the Health Physics Model ND4410, and the multi-user minicomputer-based ND 6600.
The Low Energy Photon Spectrometer (LEPS) was used for typifying metal particles, and for screening those which could profitably be sent to the X-ray Fluorescence group.
The 63 cc Ge(Li) and ND-3, a model ND 100 MCA, were used for quantifying the 60Co activity in a sample.
For samples of greater than about 1 pCi, an aluminum plate was mounted at 1.43 m above the open top detector shield.
For samples of between 0.01 pCi and 1 pCi, a distance of 0.32 m separated detector and sample.
Total activity per sample was measured, as sample geometries varied.
Lower level samples were counted directly on the detector in a plastic bottle, with activity per unit mass reported.
The minimum activity which can be detected, with a 95 percent confidence level, is 0.00008 pCi for soil samples and 0.00005 pCi for water samples.
This limit is based on the work of Lloyd Currie, as reported in Analytical Chemistry, 40, 586 (1968).
2.4 Site Survey 2.4.1 Description of the Reactor Site In order to make clear the topography of the area and the location of the radioactivity found, it is necessary to give a description of the reactor site.
A plan view of the reactor building and the immediately adjacent site is shown in Fig. 1.
To the rear of the building and at the left on the diagram is a fenced in area restricted to authorized personnel.
In the south east corner of this area is an entrance to the annex to the reactor building used to store low-level radioactive waste from all of NBS prior to shipping.
This area con-tains the heat exchanger removed from the reactor in 1974.
It was stored beside a ramp that leads into the reactor building.
This area will be referred to as 90018101
the " Enclosed Area." This Enclosed Area is paved with concrete and asphalt and contains a storm drain.
The pipe from this storm drain runs in a south-westerly direction.
Approximately 60 feet from the enclosed area (150' from the storm drain), there is an access vent to this storm drain.
This will be called the
" Holding Basin." Approximately 375' from the enclosed area, the storm drain exits onto the NBS field.
This will be called the "Outfall." The NBS site boundary is approximately 310 yards from the Outfall.
About 10 yards from the
' site boundary there is a marshy area and a small stream.
A topographic drawing of this site is shown in Fig. 2.
An aspect of the survey was to test the hypo-thesis that birds were the mechanism of transfer.
The Committee believes that the results of this survey have discounted this theory.
2.4.2 Site Survey Strategy The objective of the site survey was to define the limits of the radio-active material and to determine if any hazard was posed to the workers of NBS and the citizens of the surrounding areas.
With this in mind, the following areas were checked.
Each will be described in later sections.
1)
The area immediately adjacent to the reactor building, including the enclosed area and the lawns around the reactor building.
2)
The Holding Basin, the Outfall, and the area between the Outfall and the site boundary.
3)
Lawn equipment and storage areas for equipment.
4)
Shoes of NBS plant wcrkers who may have had access to the area around the reactor building.
5)
The total length of the NBS site boundary fence.
6)
The solar house, including interior and bird nesting area.
7)
Building 245 (Radiation Physics) including roof and interior.
8)
Trees south of South Drive.
9)
SEBA garden 10)
Lawn immediately adjacent to all NES buildings, building roofs, and all trees near buildings.
11)
The grove of woods north and north-east of the reactor building.
12)
Engineering Mechanics high bay area.
2.4.3 Results For clarity of discussion, the results section will be separated into those areas that gave positive results and those that gave negative results.
90018102 7
2.4.4.. : Positive Results Areas-z r; :All:the areas,that gave positive results are shown in Table 1.
,This' table shows the areas, the number of radioactive spots in the area, the
. number of samples taken, and (when available) the number of particles from the samples.
It.is to.be noted;that all positive results were obtained from three locations:-
1)
- The_ Enclosed Area.
2)5 The immediate vicinity of the reactor building and its roof.
3)
The Holding Basin, the Outfall, and the draining area for about 150' beyond the Outfall.
All the samples external to the Enclosed Area were assayed in the laboratory.
Details are in Table II.
The samples from the Enclosed Area were i
not assayed, except for three particles.
While it is not possible to draw any general conclusions about radiation levels because of absorption by surrounding material, for the worst case (i.e.,
in air) one microcurie corresponds to a radiation level of 0.00132 mR/hr at one meter (0.014 mR/hr at one foot).
2.4.4.1 Reactor Lawns These are shown in Fig. 1.
The front lawn of the reactor showed one spot (GF on map) (the original one found by the NRC inspector); the rear lawn adjacent to the enclosed area showed 9 spots (G -Gs); and one single spot was i
found on the north lawn (GN).
Exhaustive survey as part of the monitoring survey found eleven more very low-level spots in the lawns.
This will be des-cribed in the section under " monitoring survey."
o 2.4.4.2 Reactor Roof The reactor roof yielded three spots of radioactivity.
These are shown in j
Fig. l.
One of these spots (R, near the reactor stack) was broken into three j
3 samples, yielding a total of five samples.
One particle was separated from R,
y Rj,andRj.
Each contained one particle.
The other samples have not been 1
separated.
2.4.4.3 Holding Basin This yielded one spot and one sample.
This sample has not been separ-ated for particles.
90018103 8
r-p-
e
-m
i
,,e e
2.4.4.4 0utfall~
This area and the next-merit discussion as to-their topography.
Immediately at the point where the underground pipe that-drains the Enclosed Area exits onto the NBS field, there is a pool' approximately sir aet in dia-meter. The area of this pool is' denoted the "Outfall".
This area yielded four spots-and four samples.- One'of them (D ) was separated to yield one particle.
1 2.4.4.51 Drain Exit and Beyond l
Diametrically. opposite to the e'xit of the pipe described in the previous section,.and across the small pool, the terrain rises slightly then falls away gradually to the site fence approximately 310 yards away.
For a distance of I
.about 150' from the Outfall a' region of increased activity was found in a shallow channel several' feet wide.
This is denoted the " Drain Exit Spot".
At sixteen points in this Spot, the activity was somewhat higher than in the immediately adjacent vicinity.
TheseyieldedsamplesDfothroughD$o.
Oneofthese(Dfo) was separated to yield a particle.
The. total activity of radioactive cobalt in I
.these. samples was low, being only 3.8 p Ci.
In addition to these more active samples, seven more samples were taken for laboratoryassay.'One(Df2)was50feetfromtheOutfall(i.e.,fromthebeginning ofthedrainexitspot)andtheother(D!1)130'fromtheOutfall(i.e., twenty feet from the end of the spot). These samples did not show any increased activity over the immediate vicinity and were taken'in order to estimate the general t
level of radioactive 80Co of the spot and how it varied with distance from the Outfall.
Midway between the Outfall and the site boundary (i.e., about 150 l
yardsfromtheOutfall)'sampleDf2wastaken;andattheedgeofthestream (about20yardsfromthe.siteboundary)'sampleDkiwastaken.
A sample of I
sedimentfromthestreamwastakenanddenotedDk2 These Dit samples are collectively' called "Beyond-Drain Exit" despite the fact that two of them were from the Drain Exit ~ Spot.
One-sample of grass was taken adjacent to the Outfall, and a soil sample was taken from the corner of.the SEBA garden nearest to the Drain Exit Spot.
The, activity of all these samples is discussed in Sec. 2.4.4 7.
2.4.4.6 Enclosed Area The Enclosed Area contains the heat. exchanger'that was removed from the reactor in 1974 and a large number of concrete blocks.
A rope approximately 1
10 feet from the heat exchanger defined the 2.5 mR/hr. line.
The radiation 90018104 t
9.
came principally from the heat exchanger, and this masked that coming from the j
spot beneath it.
All these materials were exhaustively tested.
The heat exchanger was exhaustively swabbed, and all the concrete blocks were surveyed.
The only spot of radioactivity that was found was under and adjacent to the front end of the heat exchanger.
This spot was about 6' x 25'.
A large amount of sand was in this spot.
All the sand and radioactive material was removed and put into barrels.
The heat exchanger was then moved, and a large number of pieces of
' asphalt that showed activity were chipped'out of the ground and stored in a barrel.
The total activity of the samples recovered has been determined to be 2
0.25 mci.
The Enclosed Area is and has been restricted to authorized personnel.
All isolated spots of activity have been removed.
There is, at yet, a general area of asphalt beside the ramp which shows up to four times background with no definable hot spots.
2.4.4.7 Samples Beyond Drain Exit A
E As described in Sec. 2.4.4.5, these are samples D11 through D11, the sample of grass at the Outfall, and a soil sample at the corner of the SEBA garden. The activity of these samples followed a consistent pattern.
The results A
are given in Table III. The first sample, D11, was 50' from the outfall.
It 80Co could be showed an activity of 14.8 0.23 pCi/g.
Radioactivity from detectedinsampleD!1,'(130'fromtheOutfall)butitslevelwasnothigh enough to permit quantification with an accuracy of 10%*.
Radioactivity from CoinsampleDki(150yardsfromtheOutfall)couldnotbereliablydetected.
80 Thefinaltwosamples(DkiandDht)showednodetectablelevelof Co activity.
80 The sample of grass taken from the Outfall showed no reliably detectable 80Co activity, and the soil sample from the edge of the SEBA garden showed no detectable activity.
The significance of these results merits some discussion.
First there is an obviouci decrease with distance.
While at the beginning of the Drain Exit 88 40 Spot the Co activity is nine times the average K activity, toward the end of 4
the spot it has decreased to about 10% of the K, and near the site boundary is 40K level.
Second, it is instructive to compare these a maximum of 1/50 the concentration with the maximum permissible concentrations (MPC) regulations for
- See Table III for the quantitative definition of the terms " detected", "not reliably detected" and "not detected".
90018105 10
. soCo.of the NRC.
Regulations for MPC do not exist for soil.
For water they are 50 pCi/ml for soluble cobalt and 30 pCi/ml for insoluble.
Equating one ml. of waterwithonegramofsoil,theactivityofDf2, the most active sample, has a concentration of about 1/3 the MFC for 60Co in water.
The minimum detectable D
E concentration is about 1/1000 the MPC concentration.
Samples D21 and D12 are at or below this minimum detectable concentration.
2.4.5 Negative Result Areas All areas were checked with various of the instruments described in Sec.
2.3.
All the surveys were conducted by experienced personnel.
A positive
~
result (namely, the location of a radioactive spot) was taken to be when the reading on the survey meter in use rose to at least 20 percent above background and then decreased to background as the detector was passed over the spot.
The lawn equipment survey was made on all the lawn equipment of NBS, and on the shoes of all NBS plant workers who might have had an opportunity to work on the grounds near the reactor.
The NBS site fence was surveyed by two teams of two workers each.
One team went clockwise for a full circuit, and one team counter clockwise. In each team, one man was on the outside of the fence and one on the inside.
The solar house and building 245 were thoroughly checked inside and outside, paying special attention to bird nesting areas.
The area under all the trees south of South Drive were checked, as was the entire grove of woods north and north east of the NBSR.
The areas around all NBS buildings and their roofs were checked, as were the areas under all trees near the buildings.
The SEBA garden was checked, and the road exterior to the NBS site at the gates.
Particular attention was paid to the drainage area from the active spot from beyond the Outfall'to the site boundary.
From this area, water, soil, and drain sediment samples were taken for laboratory analysis. (See section 2.4.4.7 for.results).
A survey was made of the high bay area in the Engineering' Mechanics Building, since trucks used for shipping spent fuel stop in this area to have fire shields removed.
Particular mention should be made of the fact that in the site survey, all radioactive spots around the NBSR building were limited to 35' of the building.
(During the monitoring survey, one low level spot was found 75' from the building, as discussed in the next section.) A thorough survey to somewhat beyond 200' of the reactor building did not locate any further radioactivity in this area, except, of course, that associated with the drain system for the Enclosed Area.
90018106 11
Three samples of soil were taken for laboratory assay from the region beyond the rise South and East of the NBSR.
These three samples (denoted O, 0 y
2 and 0 ) showed no 80 3
Co at the minimum detection level of 0.040 pCi/gm.
2.5 The Monitoring Survey The results of the site survey described in the previous section were completed on September 4, 1979.
However, since extensive work was being carried out in the Enclosed Area (primarily associated with moving and opening of the heat exchanger), and to assure that any rhdioactive spots that might have been missed during the site survey were found, the area around the NBSR was continuously monitored.
This was done with the first of the instruments described in Section 2.3, which is considerably more sensitive than any of the other survey meters used in the site survey.
Use of this instrument detected eleven more spots of radiation, plus some residual radiation at two of the spots previously found in the site survey.
The assay results from these spots (one sample per spot only was necessary) are given in Table IV.
Seven(012-Gf2, including two at previously found spots G2 and G ) of these new finds were on the rear lawn near the reactor 3
nearthepreviousspotsinthisarea,twowereonthesouthlawn(G}3andG{s),
one on the front lawn (G 2), and one across the road from the fenced Enclosed Area (Gl3).
This spot was 75' from the reactor, which is the farthest of any spot except those associated with the Storm Drain (see Sec. 2.4.4.3, to 2.4.4.5 and 2.4.4.7).
As can be seen from Table h k ese spots were very low level, the maximum being only 2.1 pCi. The total amount in these spots is 13 pCi.
Two random samples of soil were taken from the reactor lawn behind and in front of the NBSR.
The results are discussed in Sec. 2.8.
2.6 Sample Assay Results The detailed results of the assay of the samples exterior to the enclosed area are given in Tables II and V.
The total activity of the " lawn" samples is 118.9 pCi including those from the monitoring survey.
For the roof samples it is 132.4 pCi, and for the drain samples it is 59.9 pCi.
Hence the total activity outside the enclosed area is 311 pCi.
The most intensely radio-activespotwasRjontheroof,withanactivityof47.0pCi.
On the ground, the most active was G, from the lawn behind the reactor, with an activity of 3
24.5 pCi.
This most active spot would produce a maximum radiation level of 0.34 mR/hr at one foot.for the source in air.
The actual level measured from the spot in the ground was about 0.2 mR/hr.
90018107 12
l The description, analysis, and source of the particles recovered from the samples is the subject.of the second chapter of this report.
The isolation of particles from the~ samples is a laborious and tedious task.
As a result only a small number of particles have been isolated.
These are indicated in Table 1.
The mass of the particles varies from fractions of a microgram to 300 pg.
2.7 Off Site Survey The off site survey was conducted in cooperation with an inspector from the Maryland Department of Health.
Two samples during this survey were taken on site, one esch on the front and rear lawns of the NBSR, but not, of course, includinF any spots.
The other samples were as follows:
. Water Samples (each 500 ml.)
East inlet to Izaak Walton Pond West inlet to Izaak Walton Pond Pavilion in Izaak Walton Park Exterior water sample from caretakers house The well of Mr. George Lees (at his request)
Exit of NBS stream Sediment Samples East and West inlets to Izaak Walton Pond NBS stream outside of site boundary.
While the site survey was being conducted, Mr. George Lees, who owns a house near the NBS site, approached the NBS employees carrying out the survey.
He requested that his well water be tested.
This request was, of course, honored.
All the water samples showed no detectable 80 Co activity at a level of at least a factor of three below the limit of detection of the equipment used, which is 0.1 pCi/ml at a 95% confidence level.
The sediment samples showed no detectable activity with the limits shown in Table III.
The NBSR front lawn Co; the rear lawn had a detectable but no quanti-sample showed no detectable 80 fiable amount of 80Co.
90018108 13
1 3.
Conclusions The radioactive material found on the Bureau grounds was determined to be of such a level and dispersion that it did not represent a health or safety '
hazard to the general,public or to NES employees. No radioactive material (beyond that occurring naturally) was found off the Bureau grounds.
On the Bureau grounds, the most extreme area of activity found was still 780 feet from the site boundary, and concentration of material at this point was not high enough to permit a measurement of activity.
More specifically, the conclusions drawn from the survey results are best described with reference to the schematic 80 map shown in Fig. 3.
This figure shows the only sites of radioactive Co found.
It is evident from this map that these sites are immediately adjacent to the reactor building, at the holding basin, and at the exit of the storm drain.
No other tites of soCo radioactivity have been found during the exhaustive survey conducted.
This pattern indicates that the Enclosed Area was the source of radioactive material found outside that area.
The drain from the Enclosed Area provides a natural means for this material to be transported from that area to the Holding Basin, the Outfall, and the area immediately beyond.
The remainder of the radioactive materials was found in the immediate vicinity of the reactor building.
The evidence therefore clearly indicates that the radioactive material was restricted to the immediate vicinity of the Enclosed Area, and several other areas that are connected to this one by natural means and in no other areas.
There are two mechanisms that could account for the local spread of radioactive material from the enclosed areas.
These are local swirling winds (which could lift particles onto the roof) or tracking by feet, or both.
The Enclosed Area, as can be seen in Fig. 1, is a courtyard with open sides facing the south and west.
The prevailing strong winds in this area are from the west, and this courtyard forms a natural trap for them.
Swirling winds are in fact. often observed in this area.
The sand associated with the Enclosed Area spot (see Sect. 2.4.4.6) is in a location next to the loading ramp that forms a quiet nook in these swirling winds, and where loose material is expected to be deposited.
No other evident mechanisms beyond wind and tracking are available to explain the localized spread of radioactive material.
90018l09 14
Chapter 2 The Analysis and Source of the Particles 90018110 l
15 l
The Analysis and Source of the Particles I.
Introduction The objective of this phase of the investigation was to determine the source of the particles found in the survey described in the previous chaptet.
More specifically, the objective was to determine if the particles were formed in the reactor or had some other origin.
The evidence on the location of the particles available from the survey, namely that they seemed to have been distributed from the spot under the heat exchanger by natural drainage and tracking and/or wind made it plausible to hypothesize that they were released from the heat exchanger when it was removed from service in 1974.
However, the nature of the winds in the Enclosed Area is such that loose material collects at the location where the heat exchanger was stored.
Indeed, the sand found under the heat exchanger very likely collected there by just this mechanism.
Hence another (but less likely) hypothesis was that they were somehow released into the Enclosed Area from some other source, and then were collected under the heat exchanger by winds.
Possible other sources considered were SOCo radiation sources, or one of the trucks that came into the Enclosed Area to pick up radioactive waste. For this reason it became important to determine if the particles could have been formed in the reactor.
In order to make this determination it was, at a minimum, necessary to do the following:
a)
Analyze the particles thoroughly as to composition and morphology.
b)
Find components in the primary stream of the reactor that have the same composition as the particles.
Because. cobalt alloys are commonly used for corrosion and wear resistant surfaces in bearings, pumps and valves, attention was focused on these components in the reactor.
c)
Develop at least a plausible explanation of how particles are formed from these components, and check this explanation if possible.
A point of departure was the well-known fact that particles of various sizes and shapes are formed during the wear of metals, and the nature of the particles can be an indication of the nature of the wear process.
d)
Determine the specific activity of the cobalt to see if it could have been irradiated to that lev 61 in the reactor core.
At the same time, the activity of any other isotopes (e.g., Cr and Fe) when compared to the Co activity, could give an indication of the age of the particles, i.e., the 60 time since activation.
90018111 16
In addition to this, strongly corroborative (if not conclusive) evidence would be available if particles could be located within the heat exchanger and/or the reactor and shown to be of the same nature and type as particles found outside.
With this in mind, the primary side of the heat exchanger was opened by removing the head (Fig. 4).
Extremely fine, almost colloidal material showing GoCo activity was found.
Three particles were separated from this colloidal material.
Two were found at the flange connecting the head to the body of the heat exchanger.
One of these, found at the seven.o' clock position when facing the heat exchanger was labelled HEX 7-1.
In addition, two particles were found in the water of the pool directly under the reactor core.
This pool is used to store the spent fuel elements prior to shipping.
These particles were analyzed along with a number of representative particles from outside the heat exchanger.
The analysis of these particles, analysis of pump components, and the specific activity of the particles are discussed in subsequent sections.
2.
Chemical and Metallurgical Analysis 2.1 Particle Analysis The particles analyzed and their source are shown in Table VI.
As can be seen from this table, particles were obtained from samples from each of the principal areas in which radioactive spots were found: the Enclosed Area, the reactor lawn, the reactor roof, the Outfall and trie Drain Exit Spot.
In addition, the particle from the heat exchanger and one of the particles from the spent fuel element storage pool were analyzed.
The Report of Analysis of these particles is attached as Appendix A.
A discussion of these results is given below in Sect. 2.3.
2.2 Wear Rings Water is circulated through the primary circuit of the reactor by three centrifugal pumps.
The impeller of these pumps contains a ring approximately nine inches in diameter whose convex surface mates against a concave surface consisting of another wear ring on the pump housing.
The purpose of this assembly and pair of rings is to seal the pump chamber and to prevent water from flowing back into the intake side of the pump.
This is not a normal load-bearing assembly; in normal operation there is a clearance of somewhat less than imm between these two surfaces.
The pump specifications call for 90018112 17
'the impeller wear ring to be coated with a hard facing material, but not the wear ring on the housing.
Such an approach is common practice in wearing situations.
The manufacturer of these pumps (Allis-Chalmers) notified us that the surface of the impeller wear ring was coated with a cobalt base wear alloy by a flame spray process.
Allis-Chalmers does not manufacture the wear ring.
Our pumps are 15 years old and the records of Allis-Chalmers do not identify the supplier of the wear ring.
When the reactor was constructed, and during the testing phase while still containing light water, one of these circulating pumps began running rough, and was incipiently binding.
It was determined that the piping to which the pump was connected was not accurately aligned, causing the binding.
The pump was removed from service, and when disassembled it was found that the wear rings were indeed heavily worn.
This impeller wear ring was preserved, but the mating wear ring was discarded.
This series of events occurred in 1967.
Spare wear rings were purchased along with the pumps.
The worn and spare wear rings were extensively investigated.
A photograph of the worn impeller wear ring is shown in Fig. 5.
The worn surface can be readily seen.
This surface is shown at higher magnification in Fig. 6.
Other pictures are given in Appendix A.
Sections were made of the worn wear ring and of a rew wear ring. These cross sections were extensively analyzed, and the analysis is reported in Appendix A.
A discussion of these results will be given below in Section 2.3.
Here we present some metallurgical observations on the wear ring.
Cross sections of both the new and the worn ring at a magnification of 6X are shown in Fig. 7.
From these it can be seen that the wear surface itself was deposited in a groove approximately Imm deep in the body of the wear ring.
The worn impeller wear ring, as~ described in Appendix A, has a composition that is high in cobalt, chromium and nickel, with regions of a cobalt-chromium type of alloy.
The Co-Cr-Ni phase has a composition very close to Stellite alloy SF 6, while the other phase has a composition corresponding to Stellite 157.
Both these alloys are made for flame spraying.
- Stellite type alloys are frequently employed in wearing situations l
90018113 1
18
The microstructure of the spare impeller wear ring is unusual and bears out the results on the worn impeller wear ring.
The compositions as described in Appendix A again are very nearly those of SF-6 (darker phase) and Stellite 157 (lighter phase).
The dark spots are porosity.
It appears that the ring was formed by flame spraying two different' Stellite alloys, probably during the same operation, one powder from one side and the other powder from the other side.
An expert on flame spraying contacted at the Stellite Corporation said he had never heard of this being done, and knew of no reason why it should be done.
He ventured the opinion that the amount of porosity is excessive.
Although the wear mating ring from the housing was discarded a new one purchased at the same time as the pump was available.
Analysis showed it to be 316 stainless steel with no coating on any surface.
Examination of the cross section of the worn impeller wear ring at high magnification (Fig. 8) leads to an interesting observation.
There is a thin (10-20 pm) surface layer that appears to be a different material from the base material.
It in fact appears to be material transferred from the opposing ring surface, as would occur during wear.
This surface layer was analyzed extensively, and the results are given in Appendix B.
The composition at a j
I representative location'is 39% Fe,13.1% Co,18.5% Cr and 25.7% Ni.
The significance of these results is discussed below in Section 2.3.
There is also a very thin composite layer on the back side of the impeller wear ring.
An analysis of this layer is given in Appendix A, Table 5.
This layer is probably designed to prevent fretting or seizing between the wear ring and the surface on which it is mounted.
It has no bearing on this investigation.
2.3 Discussion The chemical analysis of the particles and the wear impeller ring as reported in Appendix A, and the metallurgical observations as presented in Section 3 of this chapter lead to the following set of observations:
1)
The hard facing of the wear ring consists of stellite 157 (66% Co, 21%
Cr, 4%W, 2% Fe, no Ni) and very likely stellite SF-6 (19% Cr, 14.5% Ni, 7.5%
W, 55% Co).
Because the latter is a multiphase alloy, the composition varies markedly from point to point.
Significantly, the Fe content of all portions of this hard facing is very low (about 2%).
90018114 19
2)
The surface of the worn impeller wear ring is very variable in composition (Table 4, Appendix A; Table 1, Appendix B; Sec. 2.2 of this chapter).
Signifi-cantly, this surface contains iron in significant amounts (up to 40%) along with Cr, Ni and Co as the principal constituents.
The iron cannot have come from the alloys used to make the wear ring hard facing, since they contain low concentrations of iron (* 2%).
3)
Fig. 8 in Sec. 2.2 very clearly shows a layer on the surface of the wear ring that appears quite separate from the matrix.
Indeed, in the process of putting hardness indentations in this layer, it separated by fracture from 1.he matrix.
This is clearly seen in Fig. 1, Appendix B.
It is'high in iron content, and contains Cr, Ni and Co as the main alloying constituents.
(See Table 4, Appendix A, Table 1, Appendix B).
4)
AlltheparticlesexceptDho are hetereogeneous.
They consist of a rather large flake of stainless steel with cobalt rich surface inclusions on one side.
These inclusions have the composition of stellite 157, (lines E, Table 1, Appendix A) very nearly stellite SF-6 (see table II, line F, Appendix A), and the composition of the worn wear ring surface layer (in Appendix A, compare Table 1, line C, D, and Table 2, line C, D, with Table 4, lines A-H), noting that this layer is of variable composition.
These particles also contain Al in sometimes significant quantities, and often as oxide rather than Al metal.
A Particle D o is a homogeneous particle with a composition corresponding 5) i to the surface of the worn impeller wear ring and spots on particles S and y
HEX 7-1 when account is taken of the variability in these compositions.
To account far this series of observations, the committee fully subscribes to the conclusions given by the analysts in Appendix A.
It further suggests the following as a reasonable explanation of the source of the particles.
As the pump from %iJ the impeller wear ring was recovered began to fail, severe abrasian between it and the mating surface began to occur.
This severe abrasion between the wear surface and its mating 316 stainless steel surface caused the surface layer seen in Fig. 8.
This severe abrasion caused surface alloying giving the very variable Fe-Co-Cr-Ni alloy found on the wear A
surface, and on rt; ions of particles S) and HEX 7-1, and particle D o.
The i
A Particle D o, the very small l
particles were very likely f: med as follows.
i homogeneous particle with the cenposition of the surface alloyed layer, very l
likely chipped off from the surface.
As seen from Fig. 1, Appendix B, this 1
90018115
A Particle D o has the approximate dimensions surface layer can break off easily.
i of this surface layer.
The remainder of the particles are hetereogeneous, consisting of a stainless steel flake with occlusions on one side consisting of material of compositions found in the worn impeller wear ring.
These heteregeneous particles could have been formed in two different ways.
The first way is that during the final stages of the failure of the pump, signifi-cant flakes of stainless steel were abraded from the mating surface.
This clearly would form particles with the observed morphology, although the particles are rather large (HEX 7-1 is about 2mm long; see Fig. 4, Appendix A).
An alternative explanation is as follows.
During machining in manufacture, particles and small burrs are left on the machined surfaces.
Such particles are often.found in wear debris (A. W. Ruff, private communication).
Any such particles or dislodged burrs carried through the region between the impeller wear ring and its mating wear ring could form heterogeneous particles with the morphology observed, including the scratches on the "back" side of the particles (cf Appendix A, page 2).
The committee feels that probably both of these mechanisms operated.
These particles were then lodged in the reactor and activated by neutron radiation for varying lengths of time and in locations of varying neutron flux, as described in the next Section.
The above explanation explains all but the aluminum content of the particles.
While this must be somewhat more hypothetical, the committee believes that there are at least two reasonable explanations for the high aluminum content of the particles.
First, the piping of the reactor is aluminum, giving many possibilities for aluminum particles or corrosion products.
Second, the corrosion of holes in the heat exchanger would have produced a significant amount of corrosion product which could deposit on the particles to provide the observed aluminum content.
As already noted, this is very oft.a the form of aluminum oxide.
Third, a number of traverses of the particles through the tubes of the heat exchanger and the reactor piping could have provided the abrasion necessary to coat the particles with aluminum as observed.
The committee feels that the normal wear process of these wear rings would have caused the very fine, almost colloidal material found in the heat exchanger, and that the relatively large particles found in the heat exchanger and outside it were caused by the severe wear conditions that caused the failure of the wear ring, 90018116 n
The next Section discusses the evidence from the activity of the particle.s.
3.
Investigation of the Radioactivity of the Particles 3.1 Introduction In order to determine whether or not the particles could have been formed in the reactor, it is necessary to determine their specific activity.
The distribution of neutron flux and the maximum neutron flux in the reactor are known.
If the particles were indeed' released when the heat exchanger was removed, then they would have been in the reactor flux for a maximum of 1210 days.
In addition, they would have decayed for five years (the heat exchanger was removed in August 1974).
Hence the decay time (the " age") of the particles cannot be less than five years.
To carry out this analysis it is necessary to measure the 66Co specific activity, (for the formation) and that of one other isotope (for the age).
For the second, ??Fe was used.
To carry out this analysis, six particles were analyzed to determine their specific activity.
TheywereG1,G2,RF,Dl, HEX 7-1,andDfo.
Their 60Co y-ray activities were entirely from decay of the 5.27 year half-life isotope.
Low energy x-rays are emitted subsequent to electron capture in the iron-55 isotope with a 2.7 year half-life, and other low-energy x-rays are generated by the fluorescence of other elements in the particle by the cobalt 80 decay radiation.
Only the Co activity from the particles is significant from a radiation safety point of view.
3.2 Cobalt-60 Specific Activity The activity of the six samples was determined using a Ge-Li detector y-ray spectrometer calibrated with a standard cobalt source by the Activation Analysis Group in the Center for Analytical Chemistry and reported in Appendix C.
The results are shown in the third column of Table VII.
The mass of cobalt in each of the six particles was estimated from the electron microprobe analysis but could not be determined accurately because of the limited range of the probe electrons and the heterogeneous nature of most of the particles.
A Only D o appears to be homogeneous and even there only half of the volume 2
could be probed.
(See Appendix A).
Therefore, the cobalt mass was determined by neutron activation analysis. (See Appendix C).
The facilities at Oak Ridge National Laboratory were used because the NBSR was shut down.
The high 8 Co made it activity already present from the decay of the ground state of 90018117 22
-~
7 5
. impractical-to use the activation of 98Co to the ground state of soCo to determine the mass of 99Co present.
Fortunately, however, about half of the neutrons' absorbed by !8Co activate an isomeric state cf soCo which decays to the ground state'with 10.5 minute half-life emitting a low energy y-ray.
This' low energy y-ray can be detected by a thin germanium detector which is relatively insensitive to the much higher energy y-ray from the ground state of 80Co.
Thus, accurate determinations of the EsCo present in each particle could be made and are given in the second column of Table VII.
The specific activity.of the cobalt has been calculated from the data in columns two and three.
It is given in column four in ma,emocuries per gram of !$ o and in C
column five in enesccuries per gram of total cobalt.
It is also possible to determine the ratio of the mass of 80Co to 99Co (N /N59) (and hence'the specific activity) directly because 80 60 Co itself can be activated to 81Co which decays with a 99 minute half-life emitting a low energy y-ray that can also be detected by the low energy detector.
The specific activity in amesocuries per gram of total cobalt based on these direct measurements was determined for four of the' particles and is shown in TheagreementisquitegoodexceptforDho[notvisibletothe column six.
naked eye because it is so small.
This made it difficult to locate properly in the counting system rssulting in a significant uncertainty in the value for the mass of 99Co.
Consequently, the value for the specific activity of Dkogivenincolumnsixismorereliablebecauseitreliesonlyonratiosand uncertainties such as sample location in the counting system cancel out.
If the particles were removed from the reactor along with the heat exchanger, they must have been out of the neutron radiation field for at least five years.
Their maximum time in the reactor flux would have been 1210 days.
The 80Co specific activity expressed as curies per gram of cobalt can be calculated for varying particle histories in the reactor, and for times out of the core.
Two cases will be calculated.
The appropriate equation is:
N6o o, n 1-e-l e-Aso' 90018118
,59 g
A'T T = Irradiation time in reactor i
23-
t
.T.= Decay time since removal from radiation field-58Co l
59 = Neutron activation across section of c
$ = Neutron flux Agn = Decay constant for " Co A
=A
~#
60 59' @ # "60 4 80 60 = Neutron absorption across section for Co e
1 0 60 specific Activity =
ci/g 1+ 60 59 59 Case 1: Maximum activity that could be generated.
This assumes that the particle was in the maximum flux in the reactor for the full 1210 days, and released when the heat exchanger was removed (five years ago; 80Co half life of 5.27 years used for convenience).
14 2
$ = 1.9 x 10 n/cm _,
8 T = 1210 days at 10 MW = 1.045 x 10 s.
T = 5.27 yrs (soCo half-life so e "A60T = 1/2)
Then N60-
=.312 59 and Specific Activity = 342 Ci/g Case 2:
Particle circulated with primary cooling water until trapped in heat exchanger just before heat exchanger removal.
90018119
~24
,f I
t
~
12 2
$'= 4 x.10 n/cm,
i t = 1210~ days 1
i T = 5.'27 yrs.
Then N60
=.00645 N59 and specific activity = 7.1 Ci/g l
The experimentally determined activities are shown graphically in Figure 9 which also includes markers indicating the two cases evaluated above.
The above comparisons assume a decay time of about five years.
(The Co-60 half-life of 5.27 years was selected for convenience in the above calculations).
3.3 Iron 55 Activity and the Age of the Particle If the decay time were less than five years, the particles could not have been removed with the heat exchanger which was removed in August 1974.
Therefore it is important to determine the decay time if possible.
This gives the time the particle was out of the flux.
This can be done if the specific activity of two cifferent isotopes with appropriate half-lives can be determined.
All the particles contain iron.
One of the iron isotopes, Fe-54, activates to.Fe-55 which decays with a 2.7 year half-life.
This is a convenient half-life compared to 5.27 year Co-60 half-life.
The Fe-55 decay, however, does not emit a y-ray but decays by electron capture emitting a low energy x-ray.
The specific activity requires knowledge not only of the activity but also of the mass of iron.
Although all the particles contain iron, only one is homogeneous enough to provide a reliable iron mass determi-A nation.
This is particle D o.
The micro probe analysis of half its volume i
showed it to be homogeneous and this was confirmed by the neutron activation Se analysis of the co which agreed with the micro probe results.
Therefore, Dho was counted by the Nuclear Radiation Division and the ?!Fe activity determined. (See Appendix D).
This was_ combined with the iron mass determi-nation to give the specific activity in terms of the ratio of the number of
??Fe nuclei to.the 54Fe nuclei (N /N54).
If we define:
55 90018120 25
'8 N
55 60'.and R w
R
=
N h
N 59 54 then the decay time T can be determined from the equation:
m
-A' T
)
1-e l
" 1-1 54 Co 55 "A
T 55 60 59 Fe 60 y,,
I A' 'T
\\
\\
.where the various sym o s have the same meaning as given previously.
bl The term in the square brackets is a slowly varying function of I and $ so the equation for T must be iterated with the equations for When this the ratio of g and g to obtain a consistent set of T, I and $.
59 54 is done using the measured values of RCo(T) and RFe(T), one obtains T=4.5 years
- with an estimated standard deviation of 157,and $T =
22 2
1.3 x 10 n/cm
- To determine this number exactly, the neutron absorption cross section of
$ is needed.
Te-55 which appears in the expression A'55
- 55 ~ "54@
- U55 This-cross section has not been measured, but an estimate can be made.
The 5?Fe isotope is produced at ORNL.
If the 0 cross section were large enough 55 to effect the production rate, it would have been noticed.
The cross sections i
for all the other iron isotopes are small, being of the order of two barns.
Therefore, c was; assumed to be similar to the cross sections for the other 55 f r convenience.
The calculations are iron isotopes and taken equal to c54 not sensitive to the exact value.
90018121 26
1 3.4. Discussion TheageofparticleDfo and the time at which the heat exchanger was removed are well within the one standard deviation error estimate assigned to the age ^of the particle.
Thefluence($t)experiencedbyDfoisless 22 2
than the maximum available (2.0 x 10 n/cm -s) and the decay age shows that the particle left the radiation field at about the time tie reactor was shut down for the heat exchanger removal.
Figure 9 shows that all of the particles could have been generated by irradiation in the NBSR. The two with the lowest specific activity, D1 and' G2, could have been irradiated while circulating with the primary cooling. water.
The others must have been retained for various periods of time'in higher flux
~
regions of the reactor.
A The large variation of the specific activities and the decay age of D o i
show that the particles could.not have resulted froci the disintegration of a radioactive source left in the heat exchanger.
4.
Conclusions The investigations carried out demonstrate the following:
a)
The compositions found on the particles is the same as that of various regions on the worn impeller wear ring.
~
b)
The composition and morphology of the particles found outside is the same as one found in the heat exchanger and one found in the pool water.
c)
The morphology of the particles shows that they could have been formed by the wear process that the wear ring underwent.
d)
The flux in the reactor and the residence time in the flux can account for the specific activity of the particles.
e)
The age of the one particle on which it was possible to measure the age is consistent with the time the heat exchanger has been removed from the reactor.
f)
The wide distribution in particle specific activity shows that they did not come from a soCo radiation source.
t The Committee considers this to be overwhelming evidence that the particles were formed in the reactor by the mechanisms described in this Chapter.
90018122 l
27
l l
I i
Chapter 3 Possibilities for Particle Release 90018123 28
Possibilities for Particle Release I.
Introduction As described in the previous Chapter, the evidence that the radioactive particles were formed in the reactor is convincing.
Any other origin for the particles is so exceedingly unlikely that its probability is negligible.
Hence,'since the particles were formed in the reactor, it becomes important to try to determifle how they were released.
Because they were found under theheatexchanger,andbecausethetimethatparticleDfo was out of.the radiation field corresponds closely to the time the heat exchanger has been removed from the reactor, the most likely time for the release of the particles was when the heat exchanger was removed.
The only other components that are removed from the reactor primary system which might contain significant amounts of particulate activity are the fuel elements and the pre-filters in the water purification system.
The fuel elements are sealed in a shipping cask before placing on the shipping truck.
The cask and truck are carefully surveyed before leaving the building.
Although the loaded shipping truck passes through the enclosed area it does not stop there.
It does, however, stop at the Engineering Mechanics Building where a crane is used to place a fire shield over the cask.
No trace of activity has ever been found in or near the Engineering Mechanics Building.
The filters, after being allowed to decay for a long period of time are packaged and removed directly to the waste storage building.
They are never stored in the Enclosed Area, although they pass through it on the way to storage.
90018124 29
Hence, it is most likely that the particles were released during the removal of the heat exchanger.
It becomes important, therefore, to try to reconstruct the events that occurred during the removal of the heat exchanger.
Any indication of how the particles might have been released can give an indication that changes in procedures might be necessary for the future.
2.
Heat Exchanger Removal
2.1 Background
Heat exchangers serve to cool the water that moderates the reactor core.
In the NBSR, the moderating water is heavy water (D 0) of 99.6%
2 purity.
This increases the efficiency of the reactor as compared to light water moderated reactors.
The D 0 is pumped through the reactor core and 2
through the primary side of the heat exchanger.
Ordinary water (H 0) is 2
pumped through the secondary side and then to a cooling t wer in a closed loop, and is never subject to the neutron flux of the reactor.
As shown schematically in Fig. 4 of Chapter 2, the D 0 is pumped 2
through the more than 1000 tubes of the heat exchanger, which are in turn cooled on their outside by ordinary (light) water.
In 1971 the first of 15 tubes developed leaks with the last one occurring in 1973.
The reason for the leaks was mostly due to mechanical failure where vibration caused the tubes to rub against the baffles.
There was one leak that could be positively traced to corrosion.
As a result of these leaks, small amounts of heavy water were transferred from the primary to the secondary systems.
After the first leak, the end bell was removed, so that the tube could be plugged and the heat exchanger thoroughly tested including eddy-current measurements of tube wall thickness.
The tests however, were limited in that they could not provide indication of the condition of the tubes in the vicinity of 90018125 30
. baffles.
The leaky tube plus 32 other suspect tubes were plugged by fitting tight aluminum plugs into their ends.
The heat exchanger is constructed
)
i totally of aluminum.
Subsequently during the period 1971-1973, 13 more l
leaks involving 14 tubes developed.
These plus 34 other suspect tubes were also plugged.
In all 81 tubes including 15 identified leakers were plugged from 1971 to 1973.
Because of these problems, a new heat exchanger, this one constructed of stainless steel, was ordered.
The leaky aluminum heat exchanger.was removed in August of 1974, and replaced by the new stainless steel heat exchanger.
Removal of heat exchangers from a reactor is an uncommon operation.
The removal of a major piece of equipment such as this is a very complex 1
task that needs to be handled with great care.
First, the D 0 in the 2
primary loop needs to be preserved in an uncontaminated condition.
- Second, the level of radioactivity present (principally in the form of tritium oxide and radioactive particulate matter) is high and every precaution gust be taken to insure the protection of workers and to prevent any unauthorized release of radioactive material to the environment.
2.2 The Removal The actual removal of the heat exchanger was accomplished by personnel of the NBS Reactor Operations Division, with the help of private contractors.
All of the radiation health and safety aspects.were under the supervision of Health Physics personnel.
As part of its investigation, the An Hoc Committee attempted to recon-struct the actual series of events that occurred.during the removal operation by interviewing the actual personnel involved.
The sequence of events was gone over extensively, and a recapitulation is as follows.
90018126 31
In the, period from 8/15/74 to 8/22/74, the heat exchanger (E) was.
_ disconnected from the. reactor system and readied for removal from the building.
This necessitated draining all the D 0 from the E and blowing 2
each tube several times to. remove remaining traces of D 0.
The plugged 2
tubes were not drained or blown out.
Great care was taken to minimize exposure to tritium, especially when the flanges that connected the E to the system were opened.
Finally, on 8/22/74, the E was wiped down thoroughly, and all the openings in the primary and secondary sides were sealed by bolting steel plates over them.
It is to be noted that the primary head ball (see Fig. 4) was not removed during any of these operations.
While there were possibilities for radioactive particles to be released from the E during these operations, the very stringent monitoring conducted mtkes their undetected release extremely unlikely.
Moreover, even if a release had occurred, the particles would have been released in the building, and not outside, where they were found.
The Committee concludes that release did not occur during this phase of the operation.
On 8/23/74 the completely sealed and wiped E was moved to its final location beside the ramp.
A rope defining the 2.5 mR/hr isodose line was put into place.
Because the E was completely sealed and had been thoroughly wiped, the Committee concludes that release of particles did not occur during this phase of the operation.
This is further substantiated by the fact that no contamination was found along the path travelled by the E from the interior i
of the building to its location beside the ramp.
During the period from 9/4/74 to 9/27/74, the E was thoroughly flushed 4 with water.
This flushing was done to remove as much tritium oxide as possible from the E.
Under continuous monitoring to assure that the 90018127 32
amount of activity released was well below the maximum permissible concen-t' ration, the flush water was drained down the storm sewer.
On 9/27/74 the flushing was' discontinued because personnel had to be used to connect the new heat exchanger.
On 10/22/74 flushing was re-commenced and concluded on 10/25/74.
Because of the continuous monitoring of the effluent, and because no HE orifices were opened other than the small ones to which the flushing hoses were attached, the Committee concludes that particle release did not occur during these flushing operations.
On 10/31/74, the new heat exchanger was found to rattle.
It was decided that it might be necessary to use the old HE temporarily, and hence it was readied for possible re-installation.
To do this, it was necessary to remove all traces of light water from the primary side to prevent contami-nation of the D 0.
To do this, on 11/6/74, the covers to the access ports 2
were removed and the primary tubes blown.
All water was vacuumed out.
There was concern that some more tubes mi ht be leaking.
Therefore, 3
on 11/7/74 the HE was hydrotested.
The secondary side was filled with water and raised to a pressure of 110 PSIG.
Five tubes were found to leak or were suspect, and these had to be plugged in order to be able to use the I
HE.
To plug the tubes, the end bell was removed from HE on 11/9/74.
This was the only time the end bell was removed during the HE removal operation.
On 11/11/74 the newly discovered leaking tubes were plugged, and on 11/14/74 the end bell was replaced and HE covered with polyethylene.
The new heat exchanger was subsequently fixed, so the EE did not have to be used.
It was left with the primary side completely sealed and stored in its location until 9/5/79.
90018128 33
On that date, the end bell was again removed to look for particles, as described in Chapter 2, Sec. 2.
When it was opened 15 of the plugs on the tubes were found to have fallen out, and there was evidence in the form of stainsein the bottom of the end bell that water had been released from them, although there was no water in the end-bell.
These tubes were those that had been originally plugged, not those that were plugged in the operation just described.
Some or all of these tubes could have been filled with water by it being forced into them through their leaks during the hydro-testing on 11/7/74.
About half of the tubes from which the plugs had I
fallen out were known leakers and the others could have developed leaks during the pressure testing.
Very likely the plugs were forced out by freezing of the water.
Since.these tubes were plugged, they were not blown out when the HE was prepared for possible reuse, and hence may have contained some particles, which could then have drained out along with the water in them.
By the nature of the configuration, this water would have drained into the primary outlet port, (see Fig. 4) which was capped with a steel plate.
These may have leaked at some time (perhaps due to freezing water) since water was not found in it when opened.
The water would have drained, in fact, under HE in the spot where radioactive material was found.
The possibility of this occurring with no radioactive particles remaining in the outlet port is very remote, but cannot be completely discounted.
This possibility is made even more remote by the fact that subsequent swabbing and blowing of selected tubes that had remained plugged revealed no significant amount of loose contamination and no particles.
90018129 34
Discussion and Conclusions, The removal of the HE from the reactor and all the operations performed on it-as described in the previous section were performed under carefully laid out procedures.
Unknown release of radioactive particles would have been very unlikely.
Of all the operstions carried out, the Committee feels that two are the most likely for particle release.
The first is the removal of the end bell.
The only time this occurred after reactor shut-down for the HE removal (with the exception of our investigation)'was on 11/9/74 when it was removed to plug the leaking tubes after hydrotesting. -The joint where the end-bell joins the body of the HE forms a natural crevice for the collection of particles, and two were found there during this investigation.
This joint was wiped when the end bell was removed, but particles could have fallen out and have been missed, because of the level of activity emanating from the HE as a whole.
The possibility is remote, but must be considered.
The other possibility for release is along with the water that
' drained from the tubes that had their plugs fall out as previously described.
The Committee considers these two as being the most likely of the possibilities for. release, with the release at the time of removal of the end bell the more likely of the two.
90018l30 l
35 l
Acknowledgement A great many members of the NBS staff carried out the work on this report, so many that individual acknowledgement is not, practical.
How-cver, the Committee would like to single out the following organizations and individuals for particular thanks.
Health Physics:
Mr. Thomas G. Hobbs Reactor Operations:
Messrs. Tawfik Raby and James F. Torrence Center for Analytical Chemistry:
Drs. Harry L. Rook, Dale E. Newbury Robert L. Myklebust, Rolf L. Zeisler, Ronald F. Fleming; Messrs. Eric R. Lindstrom, Stephen B. Carpenter.
Center for Radiation Research:
Dr. Francis J. Schima Center for Materials Science:
Dr. A. W. Ruff, Mr. Charles H. Brady Publication Information Division:
Ms. Sharon A. Washburn, Mr. Matt Heyman 90018131 1
36
I 1
k i
i a
Tables I-VII 90018132
Tabis I Areas with Positive Results Total Number of Number of Number of_ Samples
- Activity, Particles from e'l Spots and Designation u Ci Samples ent lawn of cctor _
1 1, G 11.5 One present 7
ar. Lawn of setor -
9 9, G -G 72.2 One in G ;
9 1
others not separated eth lawn of actor 1
1, G 22.2 One present N
~ f of Reactor 3
L1 ding 3
5; R, R One in R ;
p y
p R,R,R 132.4 S
S S
S One in R ;
2 R,R not y
g separated Lding Basin 1
1; D 24.0 Not separated 2
- fall 4
4; D, D, D, D 3
4 5
1 Others not separated tin Exit 1
16; D 0 0;
One in D 3*
~
0 Extended E
R D
-D
"#8 "
10 10
- losed Area 1
Two barrels of sand; 2 + 0.25 mci Four removed:
numerous asphalt chips S{*,S' 3'**
2 3
S.
& ny o bers 4
present J
- kground samples; Dg, A through E See text.
See text. Sec. 2.4.4.7
'ond drain exit NA Sec. 2.4.4.7 iss Sample of
- fall IA Garden Soil NA 90018133 Ltis is one extended spot of low level activity extending to about 150 feet from the outfall, iaea were taken.between the end of the previous area and the site boundary.
They are
,...scussed in Section 2.4.4.7'on " Samples Beyond Drain Exit".
1 3
t -
l
T;blo II Detniled Assay Risults: Activa Samplos Exterior to Enclosed Area
.9 Sample Activity, u Ci a (%)
Sample Activity, u Ci a (%)
Df0 G
11.,5 2.6 2.4 7
G 2.7 B
1 D10 0.03
_..14.5 G
12 ' 4 ~ ~ '-.
2.5 2
Dfo 24.5 2.0 3
0.2 4.8 G
3.7 4~. 4 4
D D
G
- 0. 9 8.8 io - -..
P M
5 4.1 4.2 E
6 D
0.05 11.0 10 G
0.2 4.5 D
G 1.4 7.1 0
8 D;,
G, 22.3 2.0 0.14 e.0 U
N D
0.08 7.8 R
8.7 10 p
7 D
gy 28.4 1.8 10 Df0 R
1.9 0.04 11.9 3
K R
47.0 1.5 D
0.09 7.4 S
10 a
s 1o o g,-
,S o.4 0.04 10.7 a
D, 0.03 13.5 D
24.0 1.9 Dfo 0.10 7.0 D
0.8 9.1 3
D,0 D
3.3 4.6 0.2 4.9 4
D5 D
0.008 27.4 10 Standard deviation expresses only counting uncertainty 90018134 e
Table III Activity of "Beyond Drain Exit" Samples (Units are pCi/g)
~ ~ ~ ~ ~ ~ ~
'{,
'137 40 g
K Co D
1.72 1 0.066 0.77 1 0.0045 14.8 + 0.23 D
0.474 1 0.028 1.3 1 0.056 D
l D
.0.559 1 0.031 1.52 1 0.061 NRD.
D 0.263 1 0.015 2.27 1 0.048 ND D
NRD 1.68 1 0.045 ND Outfall Grass D
1.38 + 0.084 NRD SEBA Garden 0.286 1 0.023 2.45 + 0.067 ND Notes Standard deviation includes only counting statistics.
ND = Not detected NRD = Not reliably detected Can be reliably reported as " detected", but quantification to 10% is not possibl D
=
90018135
Applic?bla Msesur e nt Limits for 95% Confidenca Leval', p Ci/g Table III (Cont'd)
Marginal Detection Reliable Detection Minimum Necessary (ND is less than this)
(D is greater than to quantify 10%
this)
~ error Cs 0.032 0.066 0.253 40 0.032 0.069 0.
g 60 0.040 0.083 0.354 Co Note:
NRD indicates a level between the first and second columns D indicates a level between the second and third columns 90018136 e
f
.=
F
~
Table IV Areas Showing Negative Results Area Date Completed Results i
^
Lawn Equipment 9/1/79 Negative
'"~
~!
J Workshoes 9/4/79 Area to 200' from Reactor 8/31/79 Site Fence 9/3/79 Solar house, including interior and bird nesting area 9/2/79 Building 245, including interior 9/2/79 Trees South of South' Drive 9/2/79 Groves N and NW of Reactor 9/4/79 All NBS Buildings, including roofs and trees 9/3/79 SEBA Garden 9/1/79 Road Outside of Fence at Exits 8/30/79 Area 150' beyond drain exit to site boundary (see Table 1) 9/1/79 High Bay Area in Engineering Mechanics Building 9/3/79 i
- Three samples over rise North and East Reactor 9/3/79
- These were soil samples taken far away from NBSR.
The results showed no 60Co at a detection sensitivity of 0.040 pCi/gm (see Sec. 2.4.5 and Table III).
90018137
Table V Activity of Samples in Monitorine Survey Sample Activity, u Ci e (t) y G
0.7 2.7 2
2 G
1.6 2.0 3
3 G
0.7 2.8 2
4 G
2.1 1.8 12 5
G 1.2 2.2 8
12 6
G 1.6 2.0 12 7
G 0.6 3.0 12 8
G 0.2 4.9 12 9
G 1.3 2.2 12 y
G 0.3 4.0 g
G 1.2 2.2 13 2
G 1.2 2.2 y3 3
G 0.3 4.1 y3
?
Standard deviation expresses only counting uncertainty e
90018138 S
e I
l i
TABLE VI Particles Analyzed
. Particle Identification Source
,sy Enclosed Area; Sand under heat exchanger 3
ei
~
ie Gy Rear Lawn of Reactor R
Reactor Roof y
D Outfall y
Dfo Drain Exit Spot HEX 7-1 Interior of Heat Exchanger P00L-2 Spent Element Storage Pool 90018139 1
- i I
Table VII Particle Specific Activities
- 59 60 sP* A '1"I'Y Wt Co Co Activity Sp. Activity Sp. Activity (From Activation Ratios)
Particle (lig)
(pCi)
(C1/g (59))
(C1/g (Co))
(Ci/g (Co) )
G2 14.0 12.8 0.91 0.91 D1 6.95 9.3 1.34 1.34 1.36 Rf
.88 9.0 10.2 10.1 10.3 G1
.48 5.58 11.6 11.5 11.5 IIEX 7-1 13.1 257 19.3 19.3 D
.0092 2.70 293 234 259 10 O
O e
CO
- Based on data contained in Appendix C Based on the average of the results from Appendix C and Appendix D.
m.
__m=--
. - = _ _ _. _ _ _ _. - - -
---m_.-=__.=___._m.__-__
---m--., -, _ _
8 e
J Figures 1-9 90018141 i
l e
d Figure Captions Fig. 1 A plan view of the reactor building and adjacent area showing where radioactive' spots were found in the site survey.
Fig. 2 A general map of the reactor building and surrounding area out to the nearest site boundary, showing the location of various areas discussed in the text.
r Fig. 3 A schematic map showing the location of all areas where radioactive 89Co material was found.
Fig. 4 A schematic diagram of the cross section of the heat exchanger.
Fig. 5 A section of the worn wear ring.
The diameter of the ring is nine inches.
Fig. 6 A view of the surface of the worn wear ring.
Note the severe wear X4.
Fig. 7 A cross section view of the worn wear ring (left) and an unused wear-ring (right), showing the hard-facing inset.
In each case, two sections of the wear ring are shown face to face. X6.
Fig. 8 Two views of the cross section of the worn wear ring, showing a deposited layer on the surface.
This layer is l
variable in composition but contains a high concentration of iron (up to 40%).
The matrix material is multi phase, but contains a low concentration of iron (s2%).
X500 upper, X1000 lower.
l i
Fig. 9 Distribution of specific activity of six particles.
Arrows indicate specific activity corresponding to the maximum achievable for a particle circulating with the primary water and for a particle residing in the maximum flux region of the reactor.
90018142
N r-------
7 m
i 1
Q--Q l
l
\\
REAR LA WN l
s L_____ A l
N Gg-Gg
\\_
g s
\\
N j
w________
x I
\\O c.
i s
l NEAT EXCHANGER
~
eG;S
(
~
O RS*
I STORM l
WASTE DRAIN i
ANNEX l
l l
M 8.NY I
a M
e O
O O
T-
- "r si O Gy REACTOR BUILDING SITE Figure 1
\\
/
STREAM AND.
MARSHY AREA f
pppgoxlMATE COURSE 570' 0F DRAINAM 39 7
~
S 1
GARDEN <
OUTFALL) f j
l V
/
4'O' f
SOLAR H0 k }-(i
\\
[
4to'
{
\\
\\
f-
~
HOLDING 235 BASIN
/,
EN LOSED JN 90018144 Figure 2 I
SCHEMATIC MAP OF RADIOACTIVE AREAS FOUND l NORTH LA WN (map not to scale)
.22.2 NUMBERS ARE ACTIVITY INpCi REAR LA WN gggy 82 e-132 FRONT LA WN 11.8 1."2_A SOUTH LA WN HOLDING BASIN 1.5 24.0
' SOUR HOUSE
, gg,
OUTFALL pCilg
. DRAIN EXIT Soll ASSAY 3.8 O
.*ND SEBA GARDEN I
l
- NRD i
- ND CONCENTRATIONS
\\
D: Between 0.08 and 0.35 pCilg NRD: Between 0.04 and 0.08 pCilg ND: Loss than 0.04 pCi/g 90018145 Figure 3
t V
2 Fig. 4 Secondcry Outlet Primary Inlet h
t Double Tube Sheet o
Shell
<r,,[A:n
/
o lJ l
j
/l l
s-
^
=
i
[
l l l
l l
l l
l l
l N
End Bell l l Baffles d l
Tubes I,
y_
_ _]
- - _ -- _ - - - - - - j - - - - - - _ _ _ _ _ - - - --- - - _ _ _ y _ _ _ _
I s
i i
Divider I
i e
i e
i l
l i
l
/
Plate 8
i
/
I I
t l
F - +/
l l
-i
-4 '_ ;
- ---2 i
i mi i
l
.g n>
==
o O
j i
Steel Backing Rings a
Primary Outlet
[
End-Bell to Tube Sheet joint Secondary inlet c3 Ch
_ m.,
. m=.-., m,,2 -- m mm,,, m -,. w w,% -#,,,
, _ _ __ _._,.. __. mymppip;qpgg;c
e-
)
y l
l 5'
j
.:. 3,
/-
f a,
49F -
/.
s s,
r
'/
S('
90018J47
, /
f~-
4 0
Figure 5,
- v
s,.
O g.
g (;
pa
' a 9
1; D'.t M y pqy g /a;W fi,r g
i s
d a'ty;.
p
,r
- yin :1 %
.p
! )i r - ",; p:
- W
- t
.q. I p.g
._7
,5 b3 q s#.j+f, [' fi'jf Il sygdN;)
t ld hf.k5 f k lh.lp!
Li l ;rNPmpf a'h Q
,4l. R
- .i Q l y l u r1 st s
l p{'h$)g: jdpf j"6
,li l[:
- f$
f j-'
Ii
)
i
- 1 fl Figure 6.
90018148
l
\\
\\
1, aM.,
.si!
f g
t
- j 3
S
.4
$2:
L.?
e 1
a $s/n S,r ?3 h
t g.
t1 l
.P F.r)- n 1.
.v.
- ' e:.
p 94 N, ;.9@..
iq 3
~
uo q u.;
qn s.,
t O
lp f
h s' f-t wr I
e.;'.-
i i
IW e.~.
a.
e m,
e.-
j l
I
- .1-a.,
u il t:
b c
L M f,3 4
- v. t sec. g J,;.'.,,y k, h l!N $
v:w,
I'6.'.,*.
+
t i
1 ' ' '.
e
' r, l i,
i
,s' 9
w& w n,>y l
~
w
,.,s,.
-n il.
1 a :.,
n,
""" 7 -
90018149
m-<-
y e
4
.s
- k p< g,*
t _q\\)
/
R.
^
o h
a
?
s h m... p,, f 5 L Q ). f s ',y4) 0,
~
- .13$(<
c V.'4 -<L L \\ A f l
l i
\\
sf )
., tj
,,s.
.,??~ n, s.:
4
. or e
e q,
I k
'M 4
g
[ >, n y
pg
\\
i z.ge:
. t g g g g.. y f
&95su W t% ai:w Figure 8.
90018150
l i
Number of Particles o
m 03 A
G
,c i
G2
-g Di
<.n ta 0.
5 Max. value for circulation la 2
in primary water. (4= 4xlO n/cm -s) o pO RF GI-w i
HEX 7-1 i
9.
N C
b o
~
O D'A 14 2
'il
- Max. value possible (4 =l.9 x lO n/cm -s) 3-i O-O O
90018151
l Appendix A s
90018152 i
)
.'condues.us Apptndix A - piga 1 (18 48)
U.S. DEPARTMENT OF COMMERCE NATIONAL BUREAU OF STANDARDS WASHINGTON, D.C. 20134 REPORT OF ANALYSIS Particles From NBS Reactor Site by Scanning Electron Microscopy and X-ray Microanalysis Submitted by:
Elio Passaglia, Center for Materials Research Laboratory No:
553-39045 Date Submitted: August 30, 1979 Analysts:
R. L. Myklebust, D. E. Newbury, J. A. Small, E. B. Steel Request:
Determine the composition and structure of certain radioactive particles recovered from the NBS reactor site.
Cobalt is of particular interest.
~
Description of Method and Results:
The Cameca electron probe micro-analyzer equipped with an energy dispersive x-ray spectrometer was employed for the analysis.
The x-ray spectral data were reduced to compositional values through the application of the NBS theoretical matrix correction procedure FRAME C.
For the analysis of rough surfaces, the NBS special matrix correction procedure, FRAME P, which has recently been developed, was employed.
Because of the uncertainties in the analysis of rough surfaces, the estimated errors in the analysis values must be presumed to be as high as 10% relative.
Note that only elements with atomic numbers of eleven (sodium) or greater are detected.
The following samples were received from'the NBS reactor division:
S-1, G-1, RF, D-1, HEX 7-1, S-3, POOL 2, D10A, and pump corings (PUMP 1,2,3).
An abraded wear ring and new wear ring from the system pumps were also provided.
The particles were mounted for analysis in carbon dag on a carbon substrate.
All particles were analyzed in the as-received condition..
In addition, particles S-1 and HEX 7-1 were examined after cleaning by ultrasonic agitation in an alcohol-acetone mixture (10:1).
Particles S-1, HEX 7-1, and 010A were analyzed on both sides, which required cleaning to remove carbon particles adsorbed from the carbon dag.
Metallographic cross sections of the wear rings were analyzed.
The analysis of the particles reveals two distinct classes: Class 1 (samples S-1, G-1, RF, D-1, HEX 7-1, S-3, and P00L 2) consists of heterogeneous particles which contain regions of aluminum, stainless steel (approximately Fe - 18Cr - 10Ni), and cobalt-chromium regions.
Class 2 (sample D10A) contains a single particle which is nearly homogeneous and which consists of a stainless steel (Fe - 18Cr.- 15Ni) with about 13% cobalt.
The details of the analyses of these classes 90018153
Appandix A - paga 2 are as. follows:
Class 1:. In the as-received condition, all particles.
in.this class produce x-ray spectra which indicate the presence of aluminum and stainless steel (Fe-Cr-Ni) in significant amounts (10%
or more) and cobalt as a minor element.(<10%).
The particles were found to be heterogeneous, with the composition varying markedly from point to point. Morphologically, the particles appear to be flakes, with the thickness dimension typically about 10% of the length or' width.
.Two samples, S-1 and HEX 7-1 were chosen for detailed examination.
Micrographs of-these particles [e.g., Figures 1(a) and 1(b)] reveal that the particles have distinct structures on their " front" and "back" sides.. The "back" side has distinct linear features which appear to be scratches. The " front" side _ has no such features, appearing instead rough and irregular.
Detailed x-ray microanalysis (Table 1) of the back surface of S-1 reveals that the dark areas "A" in Figure 1(a) contain principally aluminum and iron, while the bright areas "B" consist of stainless steel.
The " front" side of S-1, Figure 1(b), contains similar Al-Fe ("C") and stainless steel ("D") regions but the microanalyses (Table 1) reveal that a significant cobalt content is also observed in these regions.
j A cobalt x-ray area scan of the front surface of S-1, Figure 2, reveals the presence of several high cobalt content regions, as well as a general cobalt distribution.
Close examination of the largest region, Figures 3(a), (b), reveals a flat, approximately triangular particle.
X-ray microanalysis at three points (Table 1,.El, E2, E3) indicates that this cobalt-rich particle is nearly(21%) - tungsten (3.3%) with virtually homogeneous in composition and consists of cobalt (65 wt%) - chromium no nickel.
The particle is surrounded by an aluminum-rich matrix.
Note that the stainless steel regions contain about 10% cobalt which is in marked contrast with the stainless steel on the back surface.
Other cobalt-rich particles in Figure 3(b) gave similar compositional results.
The cobalt-rich' particle E (designated S1-SPSB) was subsequently extracted mechanically for specific radioactivity measurements.
During the cutting operation, particle S-1 fractured.
Analysis of the fresh edges of the S-1' fragments indicated that the interior of particle S-1 was stainless steel.
Sample HEX 7-1 in class 1 was also studied extensively.
This particle also had distinctly different morphology on the front and back surfaces, Figures 4(a), (b), with very little cobalt found on the.back side in either the Al-Fe (Table 2, A) or stainless steel regions (Table 2, B).
The front side of HEX 7-1 is similar to the front side of S-1, containing stainless steel /Co regions (Table 2, C)
Al-Fe regions (Table 2, D), and Co-Cr-W regions. Again, a cobalt x-ray area scan reveals areas of high cobalt concentration, Figure 5..
This cobalt-rich region viewed at high magnification, Figure 6, contains two structures E and F which are cobalt-chromium-tungsten particles.
One of these particles (Table 2, E) contains. low nickel (1.9%) while the other (Table 2, F) contains significantly more nickel (11%).
The matrix between these particles is principally aluminum (Table 2, G), probably in an oxidized state, as indicated by.the low total composition (oxygen is not directly measured in the x-ray spectrum).
90018i54 l
2
. y.
l Appsndix A - page 3 i
Class-2: Sample D10A, Figures 7(a), (b), was determined to b' e a particle which was markedly different from the particles of Class 1.
010A was found to be a stainless steel enriched with cobalt, (Table 3).
The particle had 'a similar composition on the front and rear surfaces, as well as-on the sides. ' Locally, the cobalt composition was similar point to point, Figures 7(c),.(d), with one exception.
A small (ca. 5 pm) cobalt-chromium-tungsten inclusion (Table 3, A) was found in the matrix of cobalt-enriched stainless steel.
The mass of the particle was estimated as follows:
Calculation of the Mass of Particle D10A (1)
On the front side, the image.of the particle is bracketed by a rectangle 29 pm x 58 pm (scale calibrated from a Leitz stage micrometer).
..(2)
On the back side, the image of the particle is bracketed by a rectangle 32.5 pm x 58 m.
(3) The thickness is 6 pm.
(4) The volume of the particle is 10,700 pm or 1.07 x 10-8 3
3 cm,
(5)
Considering the density of the particle as 7.8 g/cm3 (the density of iron), the mass of the particle.is 8.3 x 10 8 g.
(6)
Considering the average cobalt concentration in the particle is 13 weight percent, the cobalt mass is 1.1 x 10 8g, The particle was analyzed at beam energies as high as 25 kev, which provided a sampling. depth of 1.5 pm.
Since both surfaces were analyzed, approximately one-half of the mass of the particle was sampled.
If we consider the center-of the particle to be composed of the 65% Co - 20% Cr - 4% W alloy, an unlikely hypothesis in view of the analyses of the side of the particle, then the possible cobalt content of the particle increases to 3.3 x 10 ag, Analysis of Pumo Comoonents (1)
Corings:
Corings were provided from the base metal of the pump housing, impeller, and wear ring.
The analyses of the coring flakes (Table 4) show the base metal to be Fe-Cr-Ni-Mo with virtually no cobalt or tungsten.
(2) Worn Wear Ring: The abraded surface of the wear ring, Figure 8, was found to be heterogeneous. A series of eight analyses (Table 4, A, B, C, 0, E, F, G, H) in a local region, Figure 9, reveals substantial differences in the content of the principal elements (Cr, Fe, Co r Ni) from point to point.
The wear. ring was also examined in a polished cross-section, Figure 10.
At higher magnification, Figure 11, at least three distinct phases are 90018155
~
1
~
Appsndix A'- page 4
~ '. -
~
l observed:. a) a continuous fine phase, b) a discontinuous fine. phase and l
c) a coarse spherical phase. The analyses for these-phases (Table 4) show that the.two-phase matrix surrounding the spherical phase consists of a cobalt-nickel-chromium alloy while the spherical phase is a cobalt-chromium-tungsten alloy.
.(3) New Wear Ring: A new, unabraded wear ring was examined in cross section. The front surface.of the wear ring, Figure 12, which corresponds-is found to consist of to the abraded surface of the worn wear ring,(Tables 5, A, B), consists of four distinct layers. The outer most layer, a high cobalt-nickel alloy, with about 10% chromium and 3-5% tungsten.
The next layer, (Table 5, C), is multiphase, Figure 13, with the individual phases yielding the analyses listed.
The third-layer, (Table 5, D), is the 65 Co - 20 Cr - 4 W 1ayer.
Beneath this high cobalt layer is another cobalt-nickel-chromium layer,- (Table 5, E).
A multiphase 1
layer, Figure 14,'(Table 5, F), is found at the interface with the stain'less steel.
The back surface of the wear ring, Figure 15, consists of three distinct regions.
The outermost layer, A, consists of an iron-chromium alloy.
The innermost layer, B, consists principally of nickel.
In addition inclusions,'"C", of high molybdenum content are observed.
Discussion: With the exception of aluminum, the compositions of the two classes of particles can be derived from the components of the hard facing
~
For alloy of the wear ring (and the stainless steel of the pump housing.S-1, HE the class 1 particles inclusions correspond closely to the composition of the coarse spherical phase found in the. wear ring hard facing alloy (65% Co - 20% Cr - 4% W).
The' stainless steel matrix of these particles is quite similar.to th'e-stainless steel base of the wear ring (60% Fe - 18% Cr - 11% Ni - 3% Mo),
but enriched with cobalt (10%) and additional nickel (4 to 6% extra), which could be derived from the cobalt-nickel-chromium content of the wear ring. The aluminum which is frequently found as a coating on the class 1 particles must originate from a source other than the wear ring or pump components.
The class 2 particle (D10A), which has the composition of a cobalt-enriched stainless steel and which is similar to certain regions of the class 1 particles, could also be created from the wear ring and the base metal.
Region A from the surface of the wear ring (Table 4) has a 3:1 Fe to Cr ratio similar to the base stainless steel, with an enrichment in cobalt and nickel which could be obtained from the Ni-Co hard facing alloy.
010A, in fact, appears to be a subclass of the class 1 particles rather than i.
a member of a separate class. That is, D10A is similar in composition
]
to the cobalt and nickel enriched stainless steel regions of the front surface of the class 1 particles, The formation of these particles during the severe abrasion of the wear ring against the stainless steel surface of the pump housing is an attractive 4
90018156 p
~.
X Appendix A - paga 5 hypothesis.
This mechanism explains the curious structure of the class 1 particles, e.g. 5-1 and HEX 7-1, which have a coba^ t-free stainless steel rear face and a cobalt-enriched stainless stiel front face, with occasional inclusions of the 65% Co - 20% Cr - 4% W aiioy.
The analysis of the surface of the wear ring itself reveals patches of stainless enriched regions on the cobalt-nickel-chromium hard facing alloy whereas iron is missing in the analyses of the cobalt-nickel-chromium facing examined in cross section.
The stainless steel on the surface must have been deposited during the severe abrasion between the wear ring and the stainless steel pump housing.
The further observation of the 65% Co - 20% Cr - 4% W alloy inclusions in the hard facing alloy cross section explains the occasional appearance of these inclusions in the reactor particles.
The aluminum which is observed on the reactor particles may have been picked up during abrasion of the aluminum heat exchanger by the stainless steel - hard facing composite particles as they passed through the system in high velocity water flow.
/
2 Ro ert L. Myklpbust Research Chemist Gas & Particulate Science Division Cente for Analytical Chemistry 90018157
, gg' 0) 4 Dale E. Newbury Research Metallurgist Gas & Particulate Science Division Cen f r An 3 -iia,1 Cherdstry
\\
ohn A. Smal' Research Chemist Gas & Particulate Science Division Center for Analytical Chemistry M 6.~.saB7 Eric B. Steel Research Chemist Gas & Particulate Science Division Center for Analytical Chemistry fNM Harry L. Rook Acting Chief Gas & Particulate Science Division Center for Analytical Chemistry September 28, 1979 5
Appandix A - ptga 6 Table 1 X-ray Micreanalaysis of Particle S1 Al Si Cr Mn Fe Co Ni Mo W
- Rear, scratched A
23 1.4 13 0.5 45 0.4 7.0 3.8 B
5.8 1.3 16 0.4 61 0.2 10 4.3
- Front, rough C
30 1.1 12 0.1 21 12 7
1.8 D
9.3 1.3 17 0.3 38 9.3 13 3.3 El 1.0 2.7 20 0
2.5 66 0.7 0.8 2.8 l
E2 0.9 2.2 21 0
2.2 65 0.6 0.7 2.9 E3' O.4 1.9 21 0
0.7 67 0.7 0.2 3.9 90018158 d
J
,~
Appandix A - piga 7 Ta bl e, 2 X-ray Microanalysis of Particle HEX 7-1 Al Si Cr Mn Fe Co Ni Mo W
Rear A
36
~0.4
.4.3 14 1.0 2.3 1.1 0
B 2.4 1.3 14 51 2.1 9.0 2.0 0.6 Front j
C 4
0.8 15 30 11 10 1.5 0.9 D
23 0.6 9.7 20 8.5 7.0 1.1 0.6 i
E 9
1.7 15 2.4 44 1.9 0.3 2.9 F
0.6 2.4 16 1.5 46 11 0.1 2.8 G
42 0.6 1.7 7.5 2.0 0.4 0.7 0.1
- Not detemined 90018159
..~--.....~...c
} :.: e..
'; s.
Appsndix A - page 8 Table 3 X-ray Microanalaysis of Particle D10A Al Si Cr Mn Fe Co
,. Ni Mo W
Front
- 1.1 1.3 18 41 14 16 1.6 1.4 Rear
- 1.4 0.7 18 42 13 15 1.8 1.7 Side l '. 3 0.9' 19 39 14 16 1.6 1.5
- Average of 8 analyses A
1.2 1.2 21 15 37 9
0.9 2.7
- Not determined i.
90018160
)
l l'
I I
a
I-I Appundix A - pega 9 l
i Table 4
/
X-ray Microanalysis of Pump Components 4
i l
Al Si Cr Mn Fe Co Ni Mo W
Base metal l
(turnings)
Wear Ring 0.5 1
18 1.7 65 0
11 6
0 i
Impeller 0.3 1.6 20 0.3 69
.0 11 5
0 Housing 0.2 1.7 19 0.4 70 0
10 4
0 1
Surface of i
wear ring l
A 2.5 1.3 9.6 30 9.2 17 1.6 1.2 i
B 0.3 1.4 18 28 17 33 1.0 1.3 C
0.3 2.2 16 11 26 46 0.4 1.7 E
D 0.2 1.0 19 5.6 38 32 0.3 2.0 E
'4.7 1.5 22 21 12 29 1.0 2.0 t
F 0.3 1.6 19 22 14 40 0.8 1.2 i
i G
3.4 1.1 7.5 29-7.7 15 1.6 0.6 H
0.9 2.1 19 20 17 36 1.2 2.3
{
- Wearing, i
polished cross section base metal 0.2 0.8 18 62 0.8 11 2.8 0
i continuous 1.3 ' 47 11 0.3 1.3 free phase 0.3 0.9 27 discontinuous i
' free phase 0.3 4.8 10 1.3 50 30 0.2 1.6 8
spherical phase 0.1 2.3 20 0.5 64 0.6 0
3.4
- Not determined 90018161
.J.
Appandix A - piga 10 Table 5 X-ray Microanalysis of Ne'w Wear Ring Al Si Cr Mn Fe Co Ni Mo W
Front surface A
0.2 3.3 9.9 2.0 43 35 0.2 3.0 B
0.1 2.8 13 1.4 46 25 0.2 3.3 C1 0
3.3 9.0 1.2 43 29 0.2 5.2 C2 0
0.4 25 1.0 43 10 0.8 1.3 C3 0.1.
1.4 13 0.8 51 12 0.3 8.3 D
0.1 2.0 18 0.5 65 0
0.3 5.0 E
O 3.6 8.3 1.8 39 36 0.1 3.6 F1 0.2 0.5 41 0.8 16 9
0.4 14 E
F2 0.1 4.2 8.3 4.5 11 63 0.2 3.7 F3 0
0.2 70 3.3 0.6 8.7 0.2 0.1 Rear surface A
0.1 0.3 11 80 0
0 0.3 0.3 B
0.2 0.2 0
0.1 0.2 91 0.2 0.4 C
0.5 0.5 0
0.1 0.1 0.2 96 0.9
- Not determined 90018162 1
/ '_. ; ;.
Appendix A - prga 11 I
' Figure Captions 1(a)
Particle S1, rear surface (b)
Particle 5-1, front surface 2 Cobalt x-ray area scan corresponding to Figure 1(b) l 3(a)
Particle S-1, inclusion SP-5B
-(b)
Cobalt x-ray scan of inclusion SP-5B 4(a)
Particle HEX 7-1, rear surface (b)
Particle HEX 7-1, front surface 5 Cobalt x-ray area scan corresponding to Figure 4(b) 6 Structure containing high cobalt regions in Figure 5 7(a)
Particle D10A, rear (b)
Particle D10A, front (c)
Particle D10A, front; "A" marks cobalt-rich region-(d)
Cobalt x-ray area scan corresponding to (c) 8(a)
Optical micrograph of wear ring showing transition from fresh to worn region (b) Wear ring, showing high wear area 9 Scanning electron micrograph of wear ring showing high wear region 10 Abraded wear ring cross section, optical micrograph 11 Abraded wear ring cross section, showing fine scale continuous phase, fine scale discontinuous phase, and coarse spherical phase 12 Front surface of new wear ring in cross section 13 High magnification of area C, Figure 12 14 High magnification of area F, Figure 12 15 Rear surface of new wear ring 90018163 i
p
~
F-C ww s, 4,:m. -. s
< ?1 1
21.
s pgg.u.
u.'
o q.
O
.,s.
C
/.
y,, g,
s @Cp4*.,82 1. T.. ;; w
. c y JcN sv s
4 ;.b g.p f r.ga.,i g..;s;,. @e h. m p-
=
~. 3.
.'j.%,p n..,,4.,.
v.e
. J
a'g. I '.
'kB
.h%.7 -
)f,.;.-.;'.
- 1 f,
j yd y! :.a *** 'L:.-;49.cg 'c%:' %.x.
.n i
v. '..
,in::'; k.. L6
[t e
at,'<; :
a n!;N,. s,w..t,%.y., w,*.,..,;..q
. c N,
.:m.
..:...m y..
M.;. *
'.,
- 1Y g..Mp..
.,,w e g[.s.N.i'8$
e J
- '.v., -,.. e -
a ' 4, r.. '
%,.3,s f,s~..W.y :.. c ~,'.
,1 g.% p** f.
.p o
- .q:,-
, e.q.-
>u...
p
,. c. ~..:.
n
.~
S e%
w:,s-w
- ,, e.%o. -
y
,p#..
i 1
m i O i d d
.. a l
.1,
.a W.-
g f-- - _-
g-_-----
'ia
,,g,,
..y-
~ '*
i i
_ w _.,gy/
o s
O
- .~.. A-._a ca. ]:-
dM;-
o 90018164
. ~.. *
.# J...
a*.
-g.
y
,e
,.. =
- T
- C, l
w.'
- p. m. '-
c 2
s,
- c...
.,,,; ),
n
.+..
- J'...
. a.w..,s.
- . uC. w ^%.,e.i s
s ;
.,se
..,.,. r -,.: <'r
'p.
L&. '
N,. *1" 5*m(t.k%%,*
- , n c.
'y
'....q.*/ *,
- M.,11
'Q y *C n
- o' s.,,
,Lw:6-1.
y.
q
- % n....
a -
r-d ' e =*
.a.
Y s
l O ='
- <s
.e.**M? f,,,ot*" ~
..',, ] r - f,<A Y
J.. - >- ~
3
.. * ** g. y,, L. <#f_.~-
%*'* d 1r=.
J J
OC m
m.. !
l l
i l
q m
- y qp n g -
.7.,st.(,..
.g>y y
- 7
..n.
e
,a.-
.e-y,,
q w
4 r'.
- f is u,:
..,.,. e ?.-
,.v-
^
e, p.-
..s s.
~a.
(.
\\.
'Q *
%'qi, J' '..
4
- N.,.i if j
7 i
,3,
e..
\\
n
.E 3
j iG e
i.,,,. y,<.. /.,.
1 a
%.. r..
i i.
i n -
- c..
3;
.,,,.\\
dj
- I'-a.
..1
,-e s.
.~.. a.
i n
[
1'.\\
N. l.. ' '
AR...t. 3',,
7/M, u
a,,.
~
.a..p
~
.n 1
- " ~ -g -.3 ->
5 ci
, ;. x.,.-
i
,4",l '.
.)'.,.34
- d. cf.
l V
s
.p
.c a
). uv.n.,.--:.
s.Y..:,
l
.w 1
a g
1.,.sw A
.q
, 4,,
\\
-p s
=m_
w i
,y-
-u'--
2
' ' ^ ' ' ' * ^ ~
g m...,
,m x
. - =,.,
,l
- I e-
[l a
l 1'
1
.4
l
.)
l s'e 9
l
. J[ f.,
.I
=,
-*g.
-..t.
~ y. Q.n W a
I 4=}..
U
?d5
- i1 Yg, N
jl 1.5 i
4 t. g' '
1,'
e
!I I
.I l
A m
a e
,l 50Jm i
iI 38) 1 90018165 I
l
a l
. f..
- a,.,
.y
...a c
...-).
~
- fle,oS'} r ),....., ~. -
'4
- y 8
- ~..
e...
O 1
.:t s.
c.
..* a.
- 2.,x..,.1..
y.
. v.r,
,s.
.O O.
$h,.,k: M.bg G-%. ' -
i p,
N.
g 1
O e.' '2 %w. eo., i g u.;g, g a,.
'l
-(N) o
.s
...n; ldr.v s.
.,.I.;*
(T', "p,C.. '
e 6
7
. c' w%.
.. ~~..r:. u r....,s, a
3.,t.,,'.e..p&,,sR.s....
.. '. v..,.. ~. ;.,
. ~
,\\.
p
~.n,.
..4...,.
.? -
'%' ';."j',.,i 7.. 4 s'y*
e a
- p.,.+. %,.s :. ~;....,
)
m', v.
. m,...
... l 1 a
W,.,~C..,.
,..; ~.r.. :y.. w. 7. '..
. l'
+ m_,A,y_
V
.n..
u.
i
~.; :...-
.m
~
..s g
s' q-,.>
- ... s.
?1 4.5Q <'.
... e l
-~s 1
- .m
.. 5 7..
. *:' w.v4 o-. \\ W *.
,;... lA s...p y s>;.. +. :. w f, s
.a.mtwe
.::r; m.
.,.,.. 1,J >.
- c... ~,
P. p.< M vh n..+%
jp r.,.
.;,d a
8
,.. v.s ne L ;l; Q l;.g g,61,.h o
__ j
.~....
n.
~
~
I l
w:.-
-_w-w = gn ;,-~..:.. : ~.:~..,. m....=...g. ; =. 3- ;.p, u__._
~
c w
..e
. =_ _ + r 1
N m.g t -5 1
- .. J 1
J
~
i r....
i C
t,.
s y / ' ~.,.1
.. ?...',,%c....
O l, '+.n
.w.
.f
- CN g
<jM.
q
. r
- e.., s.,. D 3..o
,y
.f,,. y,,..C,*,' s y. g./ ys ' ~* r3
') j&
.f g
.., m'
,;3 %..jf.,.
,.,A.".' 9
.; p
. ',y k.:;.%..
,,e 2
?,$p. - m.y go.,, '..
. >q. s -
o lls, 'h * ~'W.l
. I%4* ':
t?&'.a.W ',? 6" M.~..%"., ).
~Y 2
f
, 2. s % ;
- 4. %,;,i '" N. z.'*g.,,'..
j.* '?
'~
... F. -
g,~..
(
+/
,... e. s
.. w s.
x...
y.
1
- 9... el9. m..
r... %pe., s.,.,,. ; ;.,.
yry.
-?
u
-. ' @,:..n..
y t, /,.y '
.:a
- y. y...,,,'
,e..
a--
. a.*
.p4 ft Q.T Q..:,.,. -fy..%s.r % % J
$.Q'-
@N Ek
,p' 4
.. $:(..
l
- 4r - <;
' 6j3;f. Q(
e y
..?1."M..i.*L,f(Mr.yp;.f:%%'.i.st g
l
.]i
@ f1
, i.
.W.
,E'.,,t,.f! '. ~
~.,. ' u /sp,y
%3' C;. /A...I w ;p../
,;,h..,(V.,%. g -, N.... l j
.,1 A
,'I.
3.
ar
.. A n
y.
6 f
A
'4.9 i. 7,1..c..l
.s
.j d
- .. 4 3,M..fD..:.A
- k. ;.*
- M.
,.h.
a.
- v. g.
- 4 S.A4.
4 s
,, 3 i
- P W [
(
.....qs
. v..;.s,
t
. e. n,
.a c.
r, -
-j
(
'40'0}f?f66,f
_ m r ~.' L 4N,'
f,.
lass.s
~ i ' - - l P' g.
d g
y....
l
j
_d
=>
...-,4
.., 1 a
,,, s.
se t
2
' o a o
.AN s.
es..
i e
g
/'i 7.r9
.H j
r
~
a; j
.a
,9-
'%,' n'.
s :L., a
..%... M.*. '
. q.,.: 4.'.
'., - e.,p. [.h, <-;.,
y r
- s;
,.m..
\\.
y; a
r.
e ib.,h".
,t l
....s b
r
., e.
.c.-
V M ' 2
?h.-.2.,
- n. v., * *.., ',. - i.
.e
- ?.,.... ; --
.e,
, - x A,,
, r,
.,2,J
..e
. L......
eac.;.
.. y w ;;:. n.
.c.. w' * -
}
~
. J. e. (.-l,8.s., v.e : s., :.
-* r " ~
1
- s. sy.
g
.c W
. / "~
~
- 1F
- 4 s
U,.V 3
J i
==
_x d
2.e.
~
a.:
Un i
_ _ _ _ _ _ _ _ '~
^^
. '. c.< :L. _ A :^.; :, ~ ' L -
w,w--
4
~'.,+e
~gr%=
%~
u w-
,.vvww--
,7,,g-
'~r' m % ^ ' A ' X-
?
z i
I s
-x y
v --
.J.
a-a x
.,.m -
V. c.
~.
- r..
C L
. i,f' 'aQ (<. j'b.
(,,j$, f
'.,- *,, s,,;.
h;.
t 1
' ~.
1 qL q'
- c..
u..
Q
, s.,-Ps..,.
1 0
&..,+:.,&Q -
s 1
u.
, N sv.v.) > r. t...
t
.,e.
.~
c m. \\,,.'.
m a
.W, c.
. C'N a
'f c
-e C.
p
. Q. 9 7 y *:%. '?: %.<
, sp*nn
- r n
c&. g-
.,.' 7 k-i * 's P. \\+.
A 1;
u s
r n
,. 6
.. +,
3.
- s. p
% L j.
(t3. ;/y'ar".O.
<,,,, -- / s.'s; 'pTA%y,.7 t"*
t'
- - an y.., f.,w.,*. y bg.
'),;
i
..,,.,.:r4, e.
\\_
' y.s.
-w r
.s..
rn.,,.
~. 3.v ;
. y w: :q,.
.., p l T..
.,';.q !
.s
....,cp,. 9 9
m
+.
e..s.4 (n s..
i {s;.,,. - 'f.. e
. &.m',.
.,.. :?.l5,,,..,,..f. 5;\\. %', 4ll.i),.; Y. s,'* % k s Y
c: 4. ;..
i I
gc
} :.* W,f*'i u,,g. &w.i. * -h 5.*.S.hlr,.~)~)t 4.h I
%'&.h
?. v. k..$ E.c w<.v,.1,
- 3' l
~
t e
) * * ',n ' k'.,{s.. y.,'~ ' + Ww, 4 ;.,, l
,[.
.n'Qy'.,..1,, i j,'7}' i 1
'.)
m
. r.
f
.i
>d,,
- 4. l 9..,,,
3,
.s pt
.,.. ;p.a
. M. g a.>f--
v r
- . )
.y.$.,.,
., ~.,
n:
I
., a.rn <..- :
n M, *; F.s.p,.,
n, t
e
'C f'
f.
4 m
d q Cl M, 3, g.,, _
i g.*;
~.y J.Q
-y j S; 1
.W
- f f r-*.
'4 pg */ g n
"f
[)
- ...r,-
J
' =
. Ja.,c -
'M..,
p
es -
h/ -
m,;
' - ~ '
i 4
d f'.,.,,.J
}
j r '.e
,.J7~"'"*T T.,f{,..'1,.' <f. {,.h
~
4
- , ' G
..l at,. * ; r.';,e..s*'* Q 3
* ..? *.] ?
'? ':,s Y..,.,)
.h,. '. s. '.
s l
~~')
- . ~
.h}. f;,$l-
-Yy ff.k,Q~
ft' Q
/,Fe{s,..,'(.
i, f$,/,
. 4 n
L 'N'*.' f,o.'i *T '-
.J,;/.'/g. fbi.f,.
k'i 4
a4 q
,fsp e gr-s. r.
[ *t N ',.
- u..c 1.e.,. '..+
}
. m...f'.
4.).
'p.
(l",;p s,o/ G.,.fr h...
.,r..'.
,L r.
' r:
'w.
1t %
$.r
- lt h.l F ~ f.../,s,s.' e.t-Qo*.
Q.-
C-f, ri,,., p,. ae.'<
m*
\\
9
/p.,
Lo t;
~ t. o / 1 g *.' ;
S'.,
o i L t','.f.[4,..
1 d.'
7***. ; *
., [.
W f.
W.
' X,.
fr
..p
,I.'ih... i.2..'O....I'.Z
-E.
'. < </. :,5 n *, '.., q.o-
$./,. J..o.!.
< 6'r CD. 4. p..p. x r.
,c
(
,.f
/
.kn v
.=..'.s.
.s,s.
.,.s. - s/'
. s.:.
e
.f..f..<.(
. (c *.,,
.]
a*p;t 4 t:
'} 'f o.
9 a
(# ej 1
.re d
d
- .h.. h..'
[s
.n, I' S-
. /~~ *. gl y *
.f/
.s
,f h
,' ',.&s.1.,~.,.,/= s.7..,
,',,[, ;g ;. 't'W9
.),
4** ' W. f.
"' o
. p
- s...
~
- f.,.t,.,jk,i;if,',(,:,;.ipw.,
..yd@en i
s' ;*,s. i,lr. 1 l %e ;') x ;y. ;l.*,,l*
. o.....
cr.
1 s.. I,. y,,.. l..
bA.
g,
,4 1g.*
A s,
j e
.i.
,.?e A
'! 0
$-GJ 90018168 v.r..-u.g.,,, -- er -.. -, --, i-.., - -.,. _....--._m._.
s y
.xa a
w
-- ~ -
= -
.,1,, w f:n***~m'D't,,a @.*
1
- ~;. ;.g..p";;io ;-., 7; ;,
, {.q ';,
r y
...s
. c..,
...~
l-
.~..
2' -
,;.)..pr '
V y: -l,.,, g.l;g.u.
'.f..,.< m.y2. +,3 e
i r
,.a s w;1,.
t;
~ f, '.',,;.'.
y }.',.
C C
-,.
- q.
- f..
@ ?,.;
- . :p L
. :.. y J,
., J
(,,
i r,,s.j t '
f c..'
.y'p..,Jp"Q,
.j%,.. - ;
. ' *,;'.' t..
- ' ~ M. T,,.
}<.,-V;';*6'
- ].,'
"' 'l
)
c: -'
i (./ \\M.,0,1
', j.
- ^
'.., l,
- j.
- f. 2
.'t','.
s.
,4 o.
s
.m.,
.s,>, a t-
, <* *Y.*b.: ',, ?
)Y U,
,a if :
.t j
e
=.,
w
- D*.,P Yf '. (~
L Y '.
' ker**tr'T,.y"M f'
.='%Y.$
. s.
.?.;) '., -.)
ty n ~~~.y
~ *
. ~...,$';f>..,
. s.r1.::.H' r';.
.t. -
i ~
~
>.9...
. j.'f'kf.?
~
/
w se..,.,..;
{9 </ s
. 'g
.~
.'., s n ~,^, a. ] ia *;'..
t s$
- 6 l
s 1-
.1,% '.~.,
' Q'
.. * * \\* %_'?
- ; sq,.;,..;r...,
.s
.l.f r
.,s
'n
. p y.-
', ~, * ' '
,,'u
. ',/
- <i*
.,,. i i < ^. ~
.4.3
, 8.J.
o
=. t e"*'
p
,,;.c,.a s. W..
j
..e 8 e -
,.'=r 4
,.1,,J
. A s..f
,4
>i o
g a
...rJ.* <,. :.
1 A..a
~,.3, f,...f,,
- r...
- g..,, - -
.n.y t
. I.
..f b.
,A.
x.g'
.r
. ~
.L d
,. ' semg f, V.
,f,8 #3.
.4
.<$,,1, s.-
, l,.
e g
' ' '. s. r# ~
.d.
,I"
.t'-
7 t,.,
.?
.s i*f-l
' f,..,...,
.a
%e o
..T
- d
)
... e '/., -
.. L..
.~
s t.
. e.
.. '.,.-)
w.i
'? :
4:l ;.
..i,.
y.-
J
. l, :,.>.. y,n t.-:
_,.. j.
,-j
.:. d
- m...
~<.,3c..
s
.A
.'~.,.
c;'
- 4,.....
y<
- s.e s
- q
,..y 1,...).iqg..,
..... c3,, v.
, x.
,,/
s'd 64 4,
p 4
..';4.s.: '. i. '.-
'9 Q. :.*e.
.p..,4 m
~.
. :m;3 'y'
,,l
>1 m m,.
,.., /.'. f E f N':', ~.'C[^~
L l
- hY'll;M'), Y&' C.#9,)' h. 2" l&
,K..i.
~
- J'::Q %
- r' 6*
v 3
.mc.e % - --.
- M mA A
I l
=-.
e, jl
.t
_n
- n.,
q l
4 b.
es s
(
s
~
I a
.e r
g i
n.
D...?._..?."7&
,3,.3
'~
s 1
% p
<..%.;.s.," 3, u
. '.,. pet..
,.g
, 5.}.,
4
- a......, -a. s y a
. o.
.s......*
t
..%;n.,;; f ':sy.,,
p
. t p:.;.1,,. r,6.y e.
-... ~
n 3
b
~ ^...
~w
. :',. r...
u
.. r....-
o
~-
- 1..
s 3
i 3
..-s ti "p
g j
6e
(
~
' 500Jm I
e m.
m _.
lc)
\\
1
_.m.-.
l
- l i
.I
[
m.-
9
-s..
,..p.s. e. eg;.wy:.?.:.p4. 4.,,,.7....c,y
...... e.
.. - r3'....
..... t.;. s... p.
.m. :
3,
...., ;,,. O,,i y r. *. s ;J. T. -. ! n*}., :
1
. r r..-
.. (,. <.. -....,.: : }.'
..? :9, g..
2
..e....
,. :;.,.y n. - +. 3 ' ~ u.74p., ?' a 0
y
- , \\,.,' ' g ; '. *..,,
s.
,o
.. /.. *.,
ic.::.i..
..,. 7.;.. ;y. s... 6..(.
/
. g s
l a
i 27 v.,. s,...,.
e
..s...
s
.. A
- i
.w, c-s..., i,. e
. s....,..
.a
...'t n
n
. p.'
..r...-
I s -.
g
.s
. a;..
.+'.
,' *.,* ;. % e *
< ;5. *. 4,. ' e, y,'.
.a
~'
. l. r'. '. :
.,,s' y
o.
e
.r.s. '.g'.
<a... '.
'.i..
.2=-
g,.
s*
g,,
?
s s.
=l'... ' '"t.r
. s.?:':.;p *,
~
.e %
g.
s v-
,'."h, 3 ' '
s
.d*
p,:,.
p*,
- i,-
, q. ~..
- r'. :
94'
- 6.,,e s
.S.
I
s,','
.7
' : q.' *
..J -r %
g
,1 e
,,c N.r,..' '. 3.y.s
.s
~
. *.1.. ic,,
,. i'
..,.C.*g, v-.
. '...a
,....J.
.... s,...c..n r... j.,,.... e. ;...
.. u-...,.. ~., =.. m
. m..
m....
.(==..
S g g
%.dTT...,4M, s 4. :a '<00 Jm
. C....
.m... s. >,,e..c..s.
ll 90018169
t, V
b,
,~)
1
~.;.
\\ 1
.. (
,e l...
3.;
.h '.
1
.a 9
.' 4
- V.
'_4 3_ j = ' ~p *g.
e i
e*
t s,
.I
- .~s_fo;;
t a
,o L
.dke (!:..
}
-if
\\
,8 4
N
,l '
u
~
A B-C
. ;D i
E ' #.'
l.,
i# -
.V.%.
i a*
-g
}
/.t c.
5Til i
. (..
a i
+
i
\\
s
}.,
~
'a 4
Q 4,
j/
1
/
J i
i h:(-
r.
- ,sh t
s:
-.y
.,T4 '
~
~. -
7 i
[
}
l 'e Ji u...t1 N
r i..I
'p.'%
- ?.,,.pG., -j W
I 6
\\
_xd g
(
g, l
}a.'d;.
y.
y 3
):
~.n.-... a a,,,.w a. a# ;.00 um l*
.ph.j.'Y 2am t-2 x
/Z l
ft
.. ~..._
I p
t, i.t. w t
i.
'd C12:
1
/^
y,7
.C3 i.
s 1
i o 1%.
t u t
g g
s
,d
{-
/.#
((
.c 1 W
.- 7 @9
\\
.h s 'I*
p v
[
4 iO B
J' I-p
- A#'T' e
i O
O 1
p?
4j q
1 A. C ': b\\
4 I:
.)
C r
i l
co l'
i x.f,,
l 1
l i
.-l.s!
.fg s.
[
)q f.y i
m h
's
'r 4.
o.
2 4.
Lz.
.n. r#. - ?. li i
r c
g'
%)
?m p-t 41%
}
l f.%. _ _
- l Q
,j 1
m q
n[..'
.e.,.,
c.
s 1
{7.
i
'l N,
3"*/
g g
f,
- (' I a
b
/
i i
jb
[
Jy
._,.mA.
. L.
=-
s M
1
/5 u.
l 1
Appendix B 90018171
"h)"
Apprndix B - paga 1 U.S. DEPARTMENT OF COMMERCE l
NATIONAL BUREAU Olr STANDARDS WASHINGTON. D.C. 20134 REPORT OF ANALYSIS Pump Wear Ring from NBS Reactor Submitted by:
Elio Passaglia, Center'for Materials Research Lab. No. 553-39054 Date Submitted:
10-3-79 Analyst:
D. E. Newbury Request:
Determine the composition of the surface layer on the outer edge of the wear ring, as marked by the hardness indentations.
Description of Methods and Results: The Cameca electron microprobe equipped with an energy dispersive x-ray spectrometer was employed for the analysis.
The NBS theoretical matrix correction procedure FRAME C was employed for the analysis. The results are estimated to be accurate within 5%
relative.
The surface layer on the wear ring was examined in two locations.
In the first location, Figure 1(a), the surface layer was separated by a crack from the substrate. This area was analyzed at the points indicated in Figure 1(b).
The analyses, (points 1, 2, and 3) Table 1, reveal that the surface layer consists principally of iron, nickel, chromium, and cobalt.
This layer is apparently a cobalt and nickel enriched stainless
- steel, Just on the inside of the crack (points 4 and 5), the composition is radically different, consist of a nickel-cobalt alloy, with minor amounts of chromium and iron.
i A similar result is obtained from the second surface layer, Figure 2.
In this region, the surface layer again has the composition of a cobalt and nickel enriched stainless steel (points 6 and 7), Table 1.
Locations examined at a depth at 10 pm below the surface (points 8 and 9) show a radically different composition.
Location 8 is a nickel-cobalt alloy, with minor amount of chromium and iron.
Location 9, which corresponds in the image to a distinct phase.which appears in relief due to polishing is identified as a chromium-cobalt-nickel-tuqten alloy.
'/n5aw Dale E. Newbury Harry L. Rook Metallurgist Acting Chief Gas and Particulate Science Division Gas and Particulate Science Division Center for. Analytical Chemistry Center for Analytical Chemistry Enclosure October 5,1979
t a
Table 1 9
Analyses of Surface Layer on Wear Ring Loca tion Al Si W
Mo Cr Fe Co Ni 1
0.4 1.2 0.8 1.8 18.5 39.2 13.1 25.7 2
0.5 1.1 1.1 1.7 18.4 38.1 13.3 25.3
~
I 3
0.4 1.0 1.0 1.7 18.1 37.5 13.5 25.9 1
4 0 '. 4 5.3 0.3 0.3 8.5 4.1 26.0 56.6 5
0.6 6.0 0.2 0.3 8.8 4.4 14.4 66.2 l
6 0.3 1.0 0.8 1.9 18.1 44.3 9.8 21.1 7
0.2 1.7 0.7 1.6 18.1 35.6 12.2 26.0 8
0.4 5.3 0
0.3 8.4 3.8 23.7 57.0 9
0.5 2.8 11.1 0.3 42.3 1.1 18.4 9.6 Ta x
a O
to i
i$
~
w V
u
j_
m m_
m t
~
.n 4
i s
.y
.a
+
sf,i
.s. -
~
,. c. _.-.. -
?.
~ ~. -
.._,, _...- s =...
m.-
?
y
_ _ _ ' ui n
. 50Jm
___._c
'M
)(a)
S
-.a-
. ~.
l i
l
.)
]3 9
,i L'
,.r.
e
..4, y 2s
.s.
5 4
s -w/yspwp 1
.. w w
y s
q m.
2 n.
x.:/..
m a'
~
~
l 1
/ 0Jm s
-u
) llp )
I 900i8174
a y
a a
h a
en,*
s.,.
e.-
W.rp I
\\%..
O
,z-T F
T
"]...o
,o.
' {,1l sT..
,y"-
4
.f,~
g g
~.
p n
J e.
\\
I e
a
]
- e g
O
=
e r
s \\,\\
s s.s.r.
- c *
- )
1
\\
<g.:
q s,.f 8
Y. )..
,y <]
,h.A f
y..
i 1
....~,,e n n
. m
- w?
g
..a.- 20 um'
- ~ - -
90018175
- ~
w 9
6 l
Appendix C 90018176
---n.
' roAe4 was.aas Appen D C - page 1 U.S. DEPARTMENT OF COMMERCE NATIONAL. BUMEAU Ofr STANDARDS WASHINGTON. Q.C. 10234 October 23, 1979 REPORT OF ANALYSIS To:
Elio Passaglia, Deputy D(rector Center for Materialt Science
Subject:
Cobalt and Cobalt-60 in Particles from NBSR Site Objectives: To determine cobalt contents and cobalt-60 specific activities (in Ci Co-60/g Co-59) in particles found in and near the NBS Reactor.
Methods and Procedures:
Six particles were assayed by neutron activa-tion analysis at Oak Ridge National Laboratory (ORNL), using the pneu-matic tube facility of the Oak Ridge Research Reactor (0RR).
The nuclear reactions whose products were observed are given in Table 1.
It is con-venient that the experimental parameters are such as to permit measure-ment of both stable Co-59 and radioactive Co-60 simultaneously with the same detector, and thus the relative specific activity of the particles can be derived with little manipulation of the primary data.
Two approaches are available for determining the Co-60 specific activity:
the direct method from the Co-61/Co-60m ratio in one spectrum, and the con-ventional method fr'om the ratio of separate measurements of the Co-60 gamma activity and of the Co-60m produced by neutron activation of stable Co-59.
The samples were assayed for Co-60 activity before and after irradiation with Ge(_Lt) detectors and with a precision ionization chamber.
The results from all measurements agreed (Table 2).
Each sample was ir-radiated in the ORR for 1 or 10 minutes at a flux of 5x1013 n/cm sec, 2
accompanied by a standard of known Co content.
A thin Ge photon detector was used to measure the low-energy ganima-and x-rays of Co-60m and Co-61 in the presence of large quantities of high-energy Co-60 radiation.
Each sample and its standard were counted at least twice; some were also counted with a conventional Ge(Li) detector.
The detector used was connected to a Nuclear Data 50/50 analyzer system, based on a PDP-15 computer.
Spectra were recorded on DEC tape, and peaks detected and integrated with the MONSTR program.
Corrections were made for random coincidence, dead time, and radioactive decay, including decay during this dead-time-extended counting interval.
All other parameters (cross sections, counting efficiency, reactor power varia-tions) were arranged to cancel in the calculation of results and were separately monitored.
90018177
Appsndix C - paga 2 The specific activity of the particles can be measured directly by counting the activity of both the Co-60m and Co-61 in the same par-ticle.
Then N
A I
60,
0 59 60m 60m N59 A
I 0
60 61 61 ace e
-1 where A
3 0
(1-e-Ky)(1-e-4) g with A = decay constant C = net counts in peak y = irradiation time L = live time of count i = dead time of count 4e = time from end of irradiation to start of count P = activation cross section F = gammas per decay 6 = detector efficiency 7
p 6
59 60m 60m In a separate measurement at ORNL the factor K = r60 F 61 6 61 was determined as 0.2292 (210%).
The specific activity is then given as i
SA (Ci Co-60/g Co-59) = (.1151.6)(0.2292) A0 60m A0
= 264 A
/A 0
0 Results, Observations, and Conclusions 1.
The results are given in Table 3.
2.
The same specific activities are given by the conventional and the direct procedures, with the exception of D10A.
It was necessary to repackage this particle at ORNL to remove Cl-containing mounting medium.
Since the particle was invisibly small, it could not be placed with assurance in the same position as the standard for irradiation and counting; for this reason the direct measurement is preferred.
The two particles of highest mass and Co-60 activity, G2 and HEX 7-1, were irradiated for 1/10 as long as the others, with the result that the Co-61 peak was not observable.
3.
Two different Co-59 standards gave the same counting rate per microgram of Co to within 15, and a blank standard showed no Co-60m peak. This assures that the relation between count rate and Co content is properly calibrated.
90018178
i Appandix C - page 3 4.
While six points do not define a distribution well, there is a cluster (G2 and D1) around 1 Ci/g Co, another (RF, G1, and HEX 7-1) between 10 and 20 Ci/g and particle D10A by itself at 335 Ci/g.
5.
Tungsten is present in particles 01 and G2, and not visible in D10A or Gl. Manganese is present in Di and D10A.
f rat
~~
R. F. Fleming, Research Physicist
) :-) W
'_Dk R. M. Lindstrom, Research Chemist
. P. P.
z g
J. F. Emery, Analytical Chemistry Division Oak Ridge National Laboratory
/:Sk (YV
. L.'Zeisler, Research Chemist m
/
. L. Garner, ief Inorganic Analytical Research Division Center for Analytica1 Chemistry Attachments cc:
C. W. Reimann R. W. Burke 90018179 J
Appendix.C - page 4 i
O Table 1.
Reactions Employed oo 00 i
- - - Product - - - -
Determined Reaction t
Radiations o.
w Stable (Co-59 Co(n,)"Co 10.47 m 58.60 kev 55 Co-60 8'Co(n,)Co' 99 m
67.4 kev Manganese Mn(n )5'Mn 2.576 h 846.6, 1811.2 kev 55 Tungsten lW(n, ) 2 "W 24.0 h
61.1 kev e
A
-t
e Appendix C - page 5 i
f Table 2.
Cobalt-60 Activities co
- - - - - - Co-60 Activity (pC1) - -
3 Ge(Li)
Ge(Li)
Ion Ch.
Ge(Li) b Sample NBS ORNL ORNL NBS O*
r ID 1 Oct 3 Oct 4 Oct 16 Oct Mean r.s.d.
13.10 13.4 12.85 13.12 2.1%
G2 I
D1 9.27
-9.18 9.30 9.25 0.7%
RF 9.00 9.02 8.92 8.97 8.98 0.5%
G1 5.57 5.50 5.53 5.58 5.55 0.7%
238 257 HEX 7-1 D 10A 2.63 2.54 2.60 2.59.
2.59 1.4%
3,40 3.39 1.2%
Co-60 std. 1 3.43 3.35 Co-60 std 2 3.65 3.55 3.58 3.62 3.60 1.2%
Co-60 std. 3 2.93 2.82 2.88 2.86 2.87 1.6%
.i.
-n-~---
-, -------- - -~
l Appendix C - page 4 l
t Nco Table 3 Results of Measurements co
-o O
5 8
Co '
Co
Co ' Specific Activity cb Weight Wt. Co
- conc, activity (Ci Co/g CoSS) ss Particle (pg)*
(pg)
(percent)
(pC1) conventional direct G2 300.3 14.0 4.7 12.8 0.91 01 66.2 6.95 10.5 9.30 1.34 1.36 RF 28.0 0.88 3.1 8.97 10,2 10.4 G1 15.7 0.48 3.0 5.58 11.6 11.6 HEX 7-1 134.3 13.1 9.7 257 19,6.
D 10A 0,083*
0.0092 11 2.59 282 335
- Weights by R. L. Zeisler.
Sample D 10A is not directly weighable; estimate is from micro-analysis groups' report of 8/30/79.
4 0
e D
i i
f
}
I O
.e M-l P
Appendix D 900:8l83 m--
r^
~ ~---
- "~
App;mdix D - pigg 1 1
UNITED STATES DEPARTMENT OF COMMERCE g
National Bureau of Standards
/
Washington, D.C. 20234
~..
October 23, 1979 t
MEMORANDUM FOR Dr. Elio Passaglio From:
F.J. Schima $..
~
Subject:
NBS Reactor speck labeled 010A The photon emissions from the chip labeled D10A have been examined with Ge(Li) and Si(Li) spectrometer systems.
The Ge(Li) detector measurement was straightforward as D10A made an excellent point source at the 25cm j
source position. A very well-characterized efficiency curve is available for this source position. Only the Co-60 radiations at 1173-and 1332-kev were observed.. The corresponding activity value was determined to be 0.1044 MBo, at day 264 of 1979.
The 1 o Poissonian error of this one i
measurement was 0.20 percent and the best estimate of the systematic error limit was 1.25 percent.
On the other hand, the Si(Li) measurement was not as simple. The Mn K-alpha x rays, believed to be from the decay of Fe-55, were observed, but j
in the presence of the x rays of Cr, Fe, Co and Ni.
These latter four x rays are most likely fluoresced by the Co-60 radiations.
The measure-ment was at a non-conventional source distance of 1.8 cm.
Also, 0.5 mm j
thick Be foil absorber was introduced to stop the Co-60 beta particles from entering the Si(Li) detector.
After suitable decomposition of the spectrum, the detection rate of the Mn K-alpha x ray was determined to be 0.4807 cps at day 264 of 1979, with a 1 o Poissonian error of 1.5 percent and an estimated systematic error limit of 7.0 percent, mostly due to the decomposition procedure.
This Si(Li) detector setup was calibrated with the Fe-55 x-ray emission standard 4260-B #13, which consists of carrier free Fe-55 (chloride form) evaporated on stainless j
Using P Wk = 0.2796, the Fe-55 activity for D10A, assuming j
steel foil.
k negligible source self absorbtion, was found to be 1194 Bq, as of day 264 of 1979.
The 1 a Poissonian error is 1.5 percent and the estimate systematic error limit is 10.6 percent.
A second Fe-55 reference source was used to calibrate thic Si(Li) detector setup and the result agreed
/
to within 2 percent.
These results are the summary of single measurements of D10A on both a Ge(Li) and a Si(Li) detector system.
Reproducibility of these values has not yet been established.
90018184 i
l