Unrted States Air And Radiation EPA 402-R-93-004
Environmental Protectcn (6603J) September 1993
Agency
SEPA Analysis And Evaluation
Of A Radioactive Waste
Package Retrieved From
The Atlantic 3800-Meter
Low-Level Radioactive
Waste Disposal Site
-------
-------
REPORT NO. 402-R-93-004
ANALYSIS AND EVALUATION OF A
RADIOACTIVE WASTE PACKAGE
RETRIEVED FROM THE
ATLANTIC 3800-METER LOW-LEVEL
RADIOACTIVE WASTE DISPOSAL SITE
. Prepared for the U.S. Environmental Protection Agency
by
P. Colombo, M.W. Kendig, M. Fuhrmann
Environmental and Waste Technology Center
Department of Nuclear Energy, Brookhaven National Laboratory
Upton, New York 11973
August 1993
Robert S, Dyer, Project Officer
Office of Radiation and Indoor Air
U.S. Environmental Protection Agency
Washington, DC 20460
-------
-------
FOREWORD
From 1946 to 1970 the United States used designated ocean sites for disposal
of radioactive wastes. In 1974 the Environmental Protection Agency's (EPA) Office of
Radiation and Indoor Air (ORIA) initiated a program to determine the potential for any
adverse human health or environmental impacts resulting from previous ocean
disposal of radioactive waste materials, or from radiation contamination due to other
events such as the Chernobyl nuclear power plant accident. This program of wide-
ranging monitoring and assessment studies characterized natural environmental
parameters in and near disposal sites or depositional areas and identified and
evaluated concentrations and distributions of radionuclides.
Additionally, beginning in 1976, the ORIA initiated studies to determine whether
existing technologies could be applied toward assessing the fate of radioactive waste
packages that had been previously deposited on the seafloor in disposal sites. After
successfully locating clusters of radioactive waste packages in three deep-ocean
disposal sites, one package was retrieved from each site to evaluate the performance,
with time, of past packaging techniques. Under an interagency agreement with ORIA,
the Brookhaven National Laboratory (BNL) has performed container corrosion and
matrix analysis studies on the three recovered packages. The results of previous BNL
analyses of two waste packages were published in EPA Technical Reports "Analysis
and Evaluation of a Recovered Radioactive Waste Package from the Atlantic 2800-
Meter Disposal Site" (EPA 520/1-82-009), and "Analysis and Evaluation of a Recovered
Radioactive Waste Package Retrieved from the Farallon Islands 900-Meter Disposal
Site" (EPA 520/1-90-014). This report details the results of the analysis of a radioactive
waste package recovered from the Atlantic Ocean 3800-meter disposal site in 1978.
These three reports and the previously published EPA Technical Report
"Recovery of Low-Level Radioactive Waste Packages from Deep-Ocean Disposal Sites"
(EPA 520/1-90-027) may be particularly helpful at this time because of recent
information made public by Russia regarding past dumping or disposal activities by
the former Soviet Union of radioactive waste and radioactive equipment/materials in
the Kara and Barents Seas.
Readers of this report are invited to send comments/suggestions to Mr. Eugene
Durman, Acting Director, Radiation Studies Division (6603J), Office of Radiation and
Indoor Air, U.S. Environmental Protection Agency, Washington, DC, 20460.
/ (_Yj
/MaKgo T. Oge, Director sf
~nice 0T Radiation and Inda^ivAir
iii
-------
-------
TABLE OF CONTENTS
Page
FOREWORD iii
LIST OF FIGURES vii
LIST OF TABLES ix
1. ATLANTIC 3800m RADIOACTIVE WASTE DISPOSAL SITE 1
2. WASTE PACKAGE SURVEY AND RETRIEVAL OPERATIONS 5
2.1 Survey Operations 5
2.2 Retrieval Operations 9
3. ANALYSIS OF THE CONCRETE WASTE FORM 15
3.1 Description of the Retrieved Waste Package 15
3.2 Concrete Coring 15
3.3 Radiochemical Analysis 19
4. CORROSION ANALYSIS OF THE METAL CONTAINER (STEEL DRUM) 29
4.1 Visual Inspection 29
4.2 Dimensional Analysis 42
4.3 Micro-Analysis 46
4.3.1 Protected Regions 46
4.3.2 Rapidly Corroded Region on the Sediment Side 51
4.4 Chemistry and Metallurgy 55
4.5 Discussion 55
4.6 Summary and Conclusion 61
5. REFERENCES 63
v
-------
-------
LIST OF FIGURES
7 Page
Figure 1 Major U.S. Radioactive Waste Disposal Sites in the Atlantic Ocean .... 2
Figure 2 A radioactive waste package observed at a depth of 3,970 meters ... 6
Figure 3 A close-up of the package in Figure 2 appears intact with
signs of corrosion and blistering on the exposed bottom rim 7
Figure 4 A waste package having a distinguishable marking was located
at a depth of 3,970 meters surrounded by a scour moat 8
Figure 5 Retrieval operations using ALVIN's mechanical arm to place a
metal harness around the waste package 10
Figure 6 After clearing the surface, the waste package was subjected
to a radiation survey prior to be taken aboard the R/V Advance II ... 12
Figure 7 Following visual examination the package was transferred to a
sealed storage container to inhibit further corrosion prior
to laboratory analysis 13
Figure 8a An exposed mild steel pipe which ran the entire length of the
container and the weathered concrete surface 16
Figure 8b Overall view of the container showing the bottom side and the
"markings" on the sea side 17
Figure 9a Orientation of the waste package 18
Figure 9b Position of cores taken from the concrete waste form for analysis ... 18
Figure 10 Modeled leaching of Cs-137 from a cement waste form; release
after 20 years in water is 65.3% 24
Figure 11 Schematic of mud-line on the container 30
Figure 12 Close view of severe general attack on sediment side
compared to the more protected sea side 31
vii
-------
LIST OF FIGURES (continued)
Page
Figure 13 Severe general pitting below the sediment line 33
Figure 14 Blistering of an apparent "coating" on the sea side
of the container 34
Figure 15 Accelerated attack near chime on sea side of container 35
Figure 16 Close view of chime attack with buildup of orange product 36
Figure 17 Close view of rim perforations 37
Figure 18 Upper rim perforations at the sea and sediment buried
sides of the container 38
Figure 19a Photograph of inner side of sheath after removal from concrete.
Perforations appear near rim of the chimes and at the mudline 39
Figure 19b Inner side of sheath showing perforations at the top of
the container, at the chimes and at the mudline 40
Figure 19c Inside bottom of container showing perforations below the
sediment line 41
Figure 19d Plot of thickness of sheathing material vs. longitude 43
Figure 20a Optical photomicrograph of the metal/concrete interface
from the well-protected portion of the container 47
Figure 20b SEM corresponding to the same area as Figure 19a 47
Figure 20c Energy Dispersive Spectroscopy (EDS) of deposits
(Ca, Al, CI, etc.) on the inner coating of the container sheath 48
Figure 21a SEM photograph of a pit initiated on the sea side of the
container at R=11.25, x=21 cm, 0=270° 49
Figure 21b EDS of scale above 5-7n film or "coating' showing Al and
Ca compounds 49
viii
-------
LIST OF FIGURES (continued)
Page
Figure 22a SEM showing the distribution of elements in the scale formed
on the sea exposed side of the well-protected region of the
container 50
Figure 22b X-ray DOT mapping of magnesium layer 50
Figure 23a Optical photomicrograph of an oxide-filled pit on the sea
side of the container 52
Figure 23b SEM photograph of the same area showing metallic "Ghosts" 52
Figure 23c' SEM photograph of inclusions and precipitates in the metal 53
Figure 23d EDS analysis of inclusions and precipitates of the metal and
the MnS inclusion 53
Figure 24 Optical photograph of severe attack on a sample taken from the
sediment side atx=38.1 cm, 0=90° 54
Figure 25 Optical photomicrograph of the concrete/metal interface at
x=38.1 cm, 0=90° 54
Figure 26a Schematic of metal loss vs. time assuming a constant corrosion rate . 59
Figure 26b Schematic of generalized corrosion kinetics 59
ix
-------
-------
LIST OF TABLES
Page
Table 1 Primary United States Ocean Disposal Sites for Low-Level
Radioactive Waste 3
Table 2 Cement and Aggregate Content of the Waste Form Retrieved
from the Atlantic 3800 Meter Site 20
Table 3 Activities of Cs-137 and Co-60 in Cores taken from the
Waste Form Recovered from the Atlantic 3800 m Site 23
Table 4 Activities of Pb-212 and Bi-214 in Cores taken from the
Waste Form Recovered from the Atlantic 3800 m Site 25
Table 5 Corrosion Rate Data for Waste Packages Retrieved from
Deep Sea Disposal Sites 45
Table 6 Position and Description of Container Scrapings and Samples 56
Table 7 X-ray Diffraction Identification of Surface Scrapings 57
Table 8 Metallurgical Analysis of Waste Container 58
xi
-------
-------
1. DESCRIPTION OF ATLANTIC 3800-METER RADIOACTIVE DISPOSAL SITE
In June, 1978, the Office of Radiation Programs, U.S. Environmental Protection
Agency, conducted a monitoring program at the Atlantic 3800-meter radioactive waste
disposal site. A major objective of this study was to locate radioactive waste
packages and to identify and recover a typical package to determine the effects of the
ocean environment on the chemical and physical properties of the package over the
period of time it was on the ocean bottom. Similar surveys were conducted at the
Atlantic 2800-meter radioactive disposal site in 1976 [1], and at the Pacific Farallon
Islands 900-meter site in 1977 [2].
The Atlantic 3800-meter site is located approximately 320 kilometers off the
mainland coast on the lower continental rise, 260 kilometers seaward of the edge of
the continental shelf, near the main channel of the Hudson submarine canyon system
(Figure 1). This study was conducted in the area of coordinates 37°38'N and 70°35'W,
covering an area on the lower continental rise near the confluence of the main
channels of the Hudson and Block submarine canyon systems.
In this region, the channel of the Hudson Canyon is characterized by an
approximately 1 -kilometer wide canyon floor that contains a narrow, deeper,
meandering thalweg or axis. The walls that bound the canyon floor vary from gentle
sloping to vertical and are about 200 meters high. In this area, water depths vary
from 3700 to 4100 meters [3].
The 3800-meter site was used between 1957 and 1959, during which time it
received an estimated 14,500 waste packages having a total estimated activity of
2,100 curies of mixed fission products (Table 1). Most of the disposed wastes were
generated by government and commercial nuclear facilities along the eastern part of
the United States.
1
-------
75'
73"
74'
Woods Hole
* ^
New York.
•36°
km
100
nm
76'
75°
74°
71"
70"
72'
Figure 1 Major U.S. Radioactive Waste Disposal Sites in the Atlantic Ocean
2
-------
Table 1 Primary United States Ocean Disposal Sites for Low-Level Radioactive Waste
Site
Coordinates
(Center of Site)
Depth
(m)
Distance
from Land
(km)
Years Disposal
Site Used
Estimated No.
of Disposed
55-Gallon Drums
Estimated
Activity in
Drums at Time
of Disposal
(Ci) (T Bq)
Atlantic
38°30'N
72°06'W
2800
190
1951-1956
1959-1962
14,300
41,400(a) (1530)
Atlantic
37°38'N
70°35'W
3800
320
1957-1959
14,500
2,100 (77.7)
Pacific
Farallon
Islands
37°38'N
123°08'W
900
60
1951-1953
3,500
1,100 (40.7)
Farallon
Islands
37°37'N
123°17'W
1700
77
1946-1950
1954-1965
44,000
13,400 (496)
(a) This does not include the pressure vessel of the N/S Seawolf reactor with an estimated induced activity of 33,000 Ci.
(1220 T Bq)
-------
-------
2. WASTE PACKAGE SURVEY AND RETRIEVAL OPERATIONS
The recovery of the waste package from the Atlantic 3800-meter disposal site
was a coordinated effort involving the DSV ALVIN, its mother ship R/V LULU, and the
R/V ADVANCE II which was designated for shipboard recovery of a radioactive waste
package. The R/V ADVANCE II also served for collecting biological, geological and
radiochemical data relevant to the characteristics of the disposal site.
On June 23 and 24, 1978, two dives were made with the DSV ALVIN (dives
#812, 813), to locate a region containing radioactive waste packages. The first
radioactive waste package was observed at a depth of 3970 meters (Figure 2). A
prominent scour moat was present on the up-current side of the package and a
mound of granule-sized sediment had built up on the down-current side of the
package. The package appeared intact with signs of corrosion and blistering on the
exposed bottom rim of the mild steel container (Figure 3). Currents, flowing
southwesterly, were estimated at a velocity of 25 to 30 cm/sec. These strong currents
have kept the upper surface of this container free of sediment accumulation. Soft marl
talus littered the local area of this package. The water was turbid but slightly clearer
than the water higher on the channel wall [3]. Identification marks on the container
were not detectable, probably due to corrosion.
During the course of the survey, a package having distinguishable markings
was located (Figure 4). The number 953 was painted, in yellow, near the upper rim of
the container. The package was surrounded by a well-defined scour moat similar to
that found around the package described earlier (Figure 2). The bottom side of the
container had a wedge of sediment sloping away from it, which was probably formed
when the package slid into its present position, plowing up the sediment as it moved.
The drums located during dive #813 appear to be lying near the base of the eastern
side of the Hudson submarine canyon channel, on the lower continental rise, at a
depth of 3970 meters.
5
-------
Figure 2 A radioactive waste package observed at a depth of 3,970 meters.
-------
Figure 3 A close-up of the package in Figure 2 appears intact with signs of corrosion and blistering on the exposed bottom rim.
-------
Figure 4 A waste package having a distinguishable marking was located at a depth of 3,970 meters surrounded by a scour moat.
-------
2. WASTE PACKAGE SURVEY AND RETRIEVAL OPERATIONS
The recovery of a waste package from the Atlantic 3800-meter site involved
using DSV ALVIN, its mother ship R/V LULU, and R/V ADVANCE II. The ADVANCE II
would, receive the recovered package from the ALVIN, and would also collect
environmental data to characterize the disposal site.
On June 23 and 24, DSV ALVIN made two dives (#812, 813), to locate a region
containing radioactive waste packages. The first package was observed at 3970
meters depth (Figure 2). A prominent scour moat was present on its up-current side
and a mound of granule-sized sediment had built up on Its down-current side. The
package appeared intact with signs of corrosion and blistering on the exposed bottom
rim of the mild steel container (Figure 3). Identification marks were not detectable,
probably due to corrosion. Currents, flowing southwesterly, were estimated at 25 to
30 cm/sec. These strong currents kept the upper surface of this container free of
sediment accumulation. Soft marl talus littered the local area of this package. The
water was turbid but slightly clearer than the water higher on the channel wall [3].
Another package with distinguishable markings was also located (Figure 4). ¦
The number 953 was painted, in yellow, near its upper rim. This package was
surrounded by a well-defined scour moat similar to that described earlier (Figure 2).
Its bottom side had a wedge of sediment sloping away from it -- probably formed
when it slid into its present position, plowing up the sediment as it moved. Drums
located during dive #813 were lying near the base of the eastern side of the Hudson
submarine canyon channel, on the lower continental rise, at a depth of 3970 meters.
It was decided to recover waste package "953" due to its identifiable markings
and its apparent physical suitability for a successful retrieval. The position of the
waste package was fixed, utilizing the ALNAV navigation system operated from the
R/V LULU [4].
-------
Figure 5 Retrieval operations using ALVINs mechanical arm to place a metal harness around the waste package.
-------
2.2 Waste Package Retrieval Operations
In the morning of June 25, 1978, ALVIN was launched to recover the marked
(953) package located by the submersible during dive 813. Preparations were made
for a direct lift recovery, similar to that used for the waste package recovery at the
Farallon Islands 900-meter disposal site in 1977 [2,4]. This included the removal of
non-essential scientific equipment on the ALVIN, installation of releasable lift equipment
and the addition of syntactic foam flotation material in conformancei with ALVIN's lifting
capacity.
>
Descent to the 3970-meter depth location of the selected waste package took
approximately two hours. On locating the package, ALVIN proceeded to capture it by
placing a metal harness around it using the ALVIN's mechanical arm (Figure 5). The
same harnessing procedure was used during the recovery of a waste package from
the Atlantic 2800-meter site in 1976 [1], A 100-meter tag line was attached between
the eye of the harness and an attachment point underneath the submersible using
explosive bolts for emergency release purposes [4]. The ascent to the surface took
three hours, with a total dive time of six and one-half hours. After ALVIN surfaced, a
lift line was extended from the stern-mounted A-frame of the R/V ADVANCE II to the
tag line attached to ALVIN's underside. The package was released from the ALVIN,
and recovery operations aboard the ADVANCE II began. After clearing the surface,
and before taking the waste package aboard, it was allowed to drain to minimize
shipboard contamination (Figure 6). During that time, it was subjected to a radiation
survey by health physicists aboard the ADVANCE II.
Immediately following the survey, the package was taken aboard, where it was
photographed and carefully examined for biological growth and corrosion products.
Within one-hour the package was transferred to a sealed storage container under a
positive pressure of argon gas to inhibit or retard further corrosion prior to analysis at
Brookhaven National Laboratory (Figure 7).
11
-------
Figure 6 After clearing the surface, the waste package was subjected to a radiation survey prior to be taken aboard the R/V
Advance II.
-------
Figure 7 Following visual examination the package was transferred to a sealed storage container to inhibit further corrosion prior
to laboratory analysis.
-------
-------
3. ANALYSIS OF THE CONCRETE WASTE FORM
3.1 Description of the Retrieved Waste Package
The waste package consisted of a 210-liter (55-gaIlon) container filled with
Portland cement concrete. The diameter of the concrete waste form was 57.8
centimeters; its length was 85.1 centimeters. It weighed 306.8 kilograms. Imbedded
into the center of the concrete waste form and running the entire length was a 7.6-
centimeter (3 inch) diameter steel pipe. One end of the pipe was exposed to the
concrete surface of the waste form. Common practice in preparing waste packages
for ocean disposal was to cap the containers with "clean" concrete for radiation
shielding and to minimize contamination during storage and transportation. The
exposed pipe and the unfilled portion of the container (Figures 6 and 8a) indicate that
the cap is missing. Judging by the weathering on the exposed concrete face and the
debris found inside the pipe (such as pebbles and sediment indigenous to the
environment at the package location) it is assumed that loss of the concrete cap may
have occurred during descent or upon impact of the waste package with the ocean
floor. Although there appears to be indications of erosion on the exposed concrete
surface (Figure 8a), it is minimal considering the turbidity and bottom current velocities
near the package. Figure 8b shows the condition of the bottom rim of the container
and part of the sea side to the left of the yellow markings.
3-2 Concrete Coring
After removal of the mild steel container for corrosion studies, cores were taken
of the concrete waste form to determine the types, quantities and distribution of
contained radioactivity. Cores were taken from three locations along each of the 90°
and 270° longitudinal axes extending to depths that approached the position of the
embedded steel pipe. Orientation of the waste form and positions of the cores taken
from it are shown in Figures 9a-9b.
-------
SIM#
Figure 8a Shows an exposed mild steel pipe which ran the entire length of the container and the weathered concrete surface.
-------
Figure 8b Overall view of the container showing the bottom side and the "markings" on the sea side.
-------
Sediment Line
Figure 9a Orientation of the waste package.
Core Locations
Figure 9b Position of cores taken from the concrete waste form for analysis.
18
-------
The concrete waste form was cored by a Target* concrete saw with a dual-
speed motor (500/1000 rmp) on a swivel base. Impregnated diamond core bits (2-1/4
inch diameter) were used to produce two-inch diameter cores. Although water is
normally used during concrete coring to both lubricate and to flush out debris, it was
not used during this coring in order to minimize spraying and accumulation of
contaminated liquid waste. Instead, a commercial teflon spray was periodically
applied to lubricate the outside of the coring bit. This equipment and procedure was
successfully used to obtain cement cores as previously reported [1,2]. However, due
to the nature of the concrete composition (Table 2) and the lack of integrity, it was not
possible to obtain satisfactory core samples for determining mechanical properties
such as compressive strength. Based on its behavior, the compressive strength of
the concrete waste form was estimated as <800 pounds per square-inch (5.515 MPa).
Weighed concrete samples removed from the waste form were digested in
aqua regia, as described in Section 3.3. During this process, the aggregate was
released from the cement matrix and the weight of the aggregate and (by difference)
the cement were determined. These results are shown on Table 2, along with
aggregate and cement percentages. The average contents of aggregate and cement
were 61.3 and 38.7 percent, respectively. The distribution of cement content ranged
from 22.4 to 66.2 percent, indicating that the waste form was not homogeneous when
prepared for disposal due to improper mixing of cement, aggregate and waste.
3.3 Radiochemical Analysis
The concrete cores taken to determine radionuclide distribution in the waste
form were dissolved in aqua regia prior to analysis. The aqua regia was made by
combining 3 parts of 12M hydrochloric acid, 1 part of 16M nitric acid, and 1 part of
distilled water by volume. Distilled water was added to inhibit the interaction of the
Robert G. Evans Company, Kansas City, Missouri 64130.
19
-------
TABLE 2
Cement and Aggregate Content of the Waste Form
Retrieved from the Atlantic 3800 Meter Site
Sample
Weight of
core (g)
Weight of
aggregate (g)
Weight of
cement (g)
%
Cement
%
Aggregate
91-1
197.74
128.77
68.97
34.88
65.12
91-2
122.93
80.53
42.40
34.49
65.51
91-3
85.52
66.34
19.18
22.43
77.57
92-1
172.02
108.58
63.44
36.88
63.12
92-2
102.74
67.85
34.89
33.96
66.04
92-3
68.14
41.57
26.57
38.99
61.01
93-1
141.09
85.16
55.93
39.64
60.36
93-2
109.70
72.98
36.72
33.47
66.53
93-3
103.17
50.80
52.37
50.76
49.24
270-1
64.33
34.62
29.71
46.18
53.82
270-2
119.97
66.87
53.10
44.26
55.74
270-3
72.42
41.17
31.25
43.15
56.85
271-1
70.85
23.93
46.92
66.22
33.78
271-2
91.75
65.59
26.16
28.51
71.49
271-3
125.81
80.12
45.69
36.32
63.68
273-1
75.77
51.92
23.85
31.48
68.52
273-2
73.49
50.51
22.98
31.27
68.73
273-3
247.63
141.13
106.50
43.01
56.99
-------
two acids during storage. Since the concrete waste form was composed of portland
cement, sand, and stone aggregate, only the cement phase dissolved In the aqua
regia. However, all the activity in the cores went into solution since it is associated
predominantly with the cement phase. No activity was noted in any aggregate
samples counted.
In the dissolution process, the concrete core to be dissolved was weighed and
then placed in a glass beaker to which aqua regia and a Teflon stirring bar were
added. The solution was stirred until the sample was dissolved and present in the
solution as a suspended floe, with the exception of the aggregate, which settled to the
bottom of the beaker. The solution containing the floe was then poured into a 500-ml
volumetric flask, which in all cases exceeded the volume of the dissolved sample. The
beaker and aggregate were washed with additional aqua regia to remove any residual
solution and/or activity. This rinse was also added to the volumetric flask along with
additional aqua regia to bring the liquid level up to the calibrated volume. The liquid in
the volumetric flash was mixed thoroughly, and a 250-milliliter sample was taken for
gamma-ray analysis.
The dissolved cement samples were analyzed using an Ortec" coaxial Ge (Li)
detector. The detector was horizontally mounted with an integral FET preamplifier, the
signal of which was fed into an Ortec 472A spectroscopy amplifier. The detector has
an efficiency of 20 percent with a resolution of 2 keV at 1.33 MeV. The energy
spectrum was analyzed during a Tracor Northern"* TN 1700 multi-channel analyzer
in the pulse height analysis mode. A hardwired peak search routine (ALI) was used
for peak identification and peak area determination.
"Ortec, Inc., 100 Midland Road, Oak Ridge, Tennessee 37830.
""Tracor Northern, 3551 W. Beltline Highway, Middleton, Wisconsin 53562.
-------
The gamma-ray analysis indicated the presence of a number of radionuclides.
Table 3 shows results for gamma emitting radionuclides commonly associated with
low-level radioactive waste. Very low activities of Cs-137 and Co-60 were observed in
the waste form. Three core sections (sections 91-1, 93-1 and 93-3) contained
detectable amounts of Cs-137. Three other sections (92-1,92-3 and 271-3) contained
observable quantities of Co-60. The inhomogeneous distribution of these
radionuclides make interpretation of release processes impossible.
Experimental studies investigating leaching of radionuclides from portland
cement waste forms have shown that Cs-137 will leach via diffusion through the
porous cement [5]. Based on curves from experimental studies of Cs-137 leaching
from cement waste forms in seawater, it was determined that a diffusion coefficient of
1.07 x 10* cnf/sec is a reasonable estimate for Cs-137 leaching [6], The ALT
computer code [7] was used to calculate the fraction of Cs-137 released from the
waste form over the 20 years (7300 days) that it rested on the ocean floor. The model
results indicate that 65.3 percent of the cesium would leach out in that amount of time.
The model curve is shown in Figure 10. Cobalt, however, behaves differently in the
high-pH conditions of a cement-based material and can be expected to remain in an
insoluble form and, therefore, resistant to leaching.
Gamma spectroscopy also showed the presence of Pb-212, Pb-214, Bi-214, Ac-
228, K-40 and, intermittently, Th-234 and Ra-226/U-235. All of these radionuclides
occur naturally, although they and/or their parent radionuclides may have been placed
in the waste form. Table 4 shows the activities observed for Pb-212 and Bi-214. The
activity of Pb-212 ranges from 0.014 to 0.222 pCi/g of cement. Lead-212 Is one of the
daughter radionuclides in the Th-232 decay chain. It was observed at 238.6 KeV (
43.1 percent abundance). It has a short half-life (10.6 hours) and Is maintained by the
decay of the parent Th-232.
22
-------
TABLE 3
Activities of Cs-137 and Co-60 In Cores
Taken From The Waste Form Recovered From the Atlantic 3800 m Site
SAMPLE
WEIGHT (g)
Cs-137 (pCi/g)
Co-60 (pCi/g)
91-1
68.87
0.15
BDL
91-2
42.40
BDL
BDL
91-3
19.18
BDL
BDL
92-1
63.44
BDL
0.18
92-2
34.89
BDL
0.37
92-3
26.57
BDL
BDL
93-1
55.93
0.88
BDL
93-2
36.72
BDL
BDL
93-3
52.37
2.04
BDL
270-1
29.71
BDL
BDL
270-2
53.10
BDL
BDL
270-3
31.25
BDL
BDL
271-1
46.92
BDL
BDL
271-2
26.16
BDL
BDL
271-3
45.69
BDL
4.88
273-1
23.85
BDL
BDL
273-2
22.98
BDL
BDL
273-3
106.5
BDL
BDL
The Detection Limit varies from about 0.10 to 0.20 pCi/g depending on the
mass of concrete that was digested.
-------
Cumulative Fraction Leached
0.8
0.6
0.4
0.2
0
2
3
4
5
6
7
1
Time (Days in Thousands)
Leached in Sea Water
De=1.07 x 10'7 cm2/sec
Figure 10 Modeled leaching of Cs-137 from a cement waste form; release after 20 years
in water is 65.3%.
24
-------
TABLE 4
Activities of Pb-212 and Bi-214 In Cores
Taken From the Waste Form Recovered From the Atlantic 3800 m Site
Sample
Weight (g)
Pb-212
(CPS)
Bi-214
(CPS)
Pb-212
(pCi/g)
Bi-214
(pCi/g)
91-1
68.87
0.0073
0.0018
0.088
0.056
91-2
42.4
0.0057
0.0019
0.112
0.096
91-3
19.18
0.0051
BDL
0.222
BDL
92-1
63.44
0.0075
0.0011
0.098
0.036
92-2
34.89
0.0043
0.0024
0.104
0.015
92-3
26.57
0.0023
0.0030
0.072
0.240
93-1
55.93
0.0024
0.0031
0.036
0.118
93-2
36.72
0.0062
0.0019
0.142
0.110
93-3
52.37
0.0027
0.0040
0.044
0.162
270-1
29.71
0.0029
BDL
0.082
BDL
270-2
53.1
0.0066
0.0027
0.104
0.108
270-3
31.25
0.0043
BDL
0.116
BDL
271-1
46.92
0.0039
BDL
0.070
BDL
271-2
26.16
0.0046
BDL
0.148
BDL
271-3
45.69
0.0052
0.0029
0.096
0.134
273-1
23.85
0.0033
TXTIT
nlJLi
0.116
BDL
273-2
22.98
0.0020
BDL
0.072
BDL
273-3
106.5
0.0018
0.0061
0.014
0.122
BDL, Below Detection Limits.
-------
Bismuth-214 ranges in activity from 0.036 to 0.240 pCi/g. It is a daughter in the
uranium-238 decay series, with a half-life of 19.7 minutes and is therefore maintained
by the decay of U-238. It was observed at the 609.3 KeV line (46.1 percent
abundance) and the 1120.3 KeV line (15 percent abundance). These radionuclides
(Pb-212 and Bi-214) can be used in gamma spectroscopy as indicators of the
presence of their parent radionuclides. The parents have only very low energy and/or
low abundance gamma rays that are difficult to detect.
The presence of U-238 and Th-232 could be due to one of several possibilities:
1. These elements are naturally present in the raw materials from which the
cement was produced;
2. U/Th or Ra waste was mixed into the waste form; or
3. The chemical nature of the cement concentrated some of these
radionuclides as seawater moved into the waste form. This last hypothesis
is the least likely.
To determine if the U-238 and Th-232 concentrations supporting the Bi-214 and
Pb-212 are at environmental levels or higher (due to their presence in the waste), the
Origen 2 computer code (version 2.1) was used. This code is typically used to
calculate activities and masses of radionuclides generated in nuclear processes, such
as those that take place in nuclear reactors. It can also be used to determine the
quantities of daughter radionuclides resulting from the decay of natural radionuclides.
During the modeling, an assumption of 1 gram of Th-232 or U-238 was made, and
each was allowed to decay until it was at secular equilibrium with the daughters. The
ratio of daughter to parent was then adjusted with the activity of the daughters
observed in the waste form to give the activity and mass of the parents.
26
-------
Lead-212 cam© into secular equilibrium with Th-232 rather rapidly, requiring
between 50 and 100 years. Based on the measured quantity of Pb-212, it is estimated
that the mass of Th-232 ranges from 1.3 x 10'7 g/g of cement to 2 x 10"" g/g of
cement
The activity of Bi-214 comes into secular equilibrium with U-238 after
approximately 1x10® years. Assuming that there was no separation of uranium and
bismuth, the activity of Bi-214 indicates that the mass of uranium ranges from 1.2 x 10*
7 g/g of cement to 8 x 10 7 g/g of cement. It is reasonable to assume that the uranium
is natural and has not been chemically separated, as one might suspect in a waste.
This is because of the very slow in-growth of Bi-214. The activity of Bi-214 observed
in the waste form would require a large mass of uranium to support it, if it had been
separated recently. For example, If it had been effectively separated 20 years prior to
analysis, the observed maximum Bi-214 activity would require 4.9 x 104 g/g of cement.
This is obviously impossible. The conclusion is that the uranium and thorium
concentrations observed in the waste form are typical of those expected in naturally
occurring materials.
27
-------
-------
4. CORROSION ANALYSIS OF THE METAL CONTAINER
As stated in previous reports [1,2], the corrosion analysis assesses the effect
that the disposal site environment has upon the carbon steel sheathing material.
Although one assumes total loss of this sheath, experience has shown that the sheath
remains intact for very long periods of time and retards the leach rate by an
observable amount. This aspect provides further motivation for studying the corrosion
of the sheath material.
The method of corrosion analysis of this waste package follows the procedures
described in previous reports [1,2] and includes the following tasks:
• Visual Inspection
Dimensional Analysis
• Micro-Analysis
Chemical and Metallurgical Analysis
4.1 Visual Inspection
As a result of visual inspection, one could see evidence of greater attack on the
mud- buried portion than the side exposed only to the water. Figure 11 shows a
schematic of the container and indicates how the container rested in the bottom
sediment. As illustrated in Figure 11, the sediment intersects the package at the line
ABCD corresponding to respective coordinates: (x,r,6)=(0,28.6 cm,25"), (82.6 cm,28.6
0X1,56"), (82.6 cm,28.6 cm, 190°), (0,28.6 cm.200.4'). The arc's B, 90°, A and C,
90°, and D represent the portion of the container exposed to the sediment. Figure 8b
shows a clear transition between the attached sediment side at angles less than 190 °
and the protected region corresponding to angles greater than 190°. Figure 12
shows a closer view of the difference between the sediment side, which exhibits
severe general corrosive attack, and the more protected sea side. The corrosive
29
-------
c
U)
•o
90
\
\
,10
A\€>°
Figure 11 Schematic of mud-line on the container.
-------
Figure 12 Close view of severe general attack on sediment side compared to the more protected sea side.
-------
attack goes from a pitting mode slightly above the sediment line (-190") to a more
generalized attack below the line. The attack below the sediment line shown in Figure
13 takes the form of a severe general pitting, while that on the sea side, Figure 14,
appears more or less protected, possibly due to a coating. Figure 14 shows about a
0.1-millimeter diameter blistering of an apparent "coating." An instance of initiation of
corrosion appears where a scratch is present.
Although the sea side exhibited less general attack, there were locations of
specific high corrosion rates. For example, Figure 15 shows the chime near the top
end of the container between 270* and 0°. This falls on the sea side. The chime
exhibits an accelerated attack with a buildup of orange product, appearing to the left
of a number "953," originally painted as an identification mark on the container. Figure
16 gives a closer view of this chime attack. The pH of the underlying material
indicated that small regions less than 0.5 centimeters in diameter indicated slight
acidity as a result of active pits.
Figure 15 also shows the perforation that occurred at the end of the container,
which exposed the concrete. A closer view of this attack is seen in Figure 17. This
rapid attack on the rim perforated the container at the sea and sediment-buried sides
as well (Figure 18).
After extracting two strips, one from the sea side at 270* and one from the
sediment side at 90°, two shells of the carbon steel sheath remained. Photographs of
the inside of the shells appear in Figures 19a through 19c. Instances of perforation
appear at the top of the container near the rim at the chimes and at the mud line.
However, the perforation is less than 1 percent of the total sheath area, a somewhat
better protection than was observed for the containers retrieved from the Atlantic
2800-meter site and the 900-meter Farallon Islands site [1,2] with respect to high
localized attack.
32
-------
Figure 13 Severe general pitting below the sediment line.
-------
Figure 14 Blistering of an apparent "coating" on the sea side of the container.
-------
Figure 15 Accelerated attack near chime on sea side of container.
-------
Figure 16 Close view of chime attack with buildup of orange product.
-------
u>
-J
Figure 17
Close view of rim perforations
-------
Figure 18 Upper rim perforations at the sea and sediment buried sides of the container.
-------
-------
Figure 19b Inner side of sheath showing perforations at the top of the container, at the chimes and at the mudline.
-------
Figure 19c Inside bottom of container showing perforations below the sediment line,
41
-------
4.2 Dimensional Analysis
Strips cut from the respective sea and sediment sides of the carbon steel
sheath provided specimens for dimensional analysis and metallographic examination.
Specimens were extracted at 7.6-cm (3-inch) intervals along each strip, mounted in
epoxy and ground past the saw cut damage with 320 sandpaper. The thickness from
the photographed cross-section was measured at each 0.01-inch (0.025-cm) interval
and averaged to obtain a dimension specific to each point taken at 7.6-cm intervals
along the x direction. The standard deviation of the determination depends upon the
localized nature of the corrosion. Large deviation indicates pitting.
Figure 19d shows the results plotted as metal thickness vs., x, the dimension
down the container axis, in quantitative agreement with the qualitative visual
observation, the sediment side shows a greater amount of metal loss than the sea
side. The sediment side has an average metal thickness of 0.095 ± 0.015 cm and the
sea side shows an average thickness of 0.116 ± 0.0095 cm. The majority of the metal
loss on the sea side appears at either end of the package. The bottom at x=24
inches (60.6 cm) shows pitting. The middle of the sea side is protected, as was the
case for previously-retrieved waste packages [1,2].
The initial dimension is unknown since the rim fold was compressed so that its
dimensions fell under the maximum dimension of the thickest exposed portion.
Hence, the initial dimension of 0.125 cm, the maximum observed thickness in the
protection region, served as an initial dimension d0, used to obtain the calculated
corrosion rate, r:
r _ d» " d
T ~ T
42
-------
0.10
E
o
(/) 0.05
in
UJ
•2.
M
O
LU
0.10
0.05
1 1
1
SEA
1 1 w
o
o
o
o
o
o
h-CH
SIDE
° o °
SEDIMENT
o
SIDE
o
o
1—OH
O
i
1 1
i
10
20
X, inches
30
Figure 19d Plot of thickness of sheathing material vs. longitude.
43
-------
where T is the time in the environment, taken as 20 years for this specimen. This
gives a calculated corrosion rate of 0.015 ± 0.002 mm/year or 0.0006 ± 0.0001
in/year. The sea side exhibits an average corrosion rate too small to be determined
by this macroscopic technique. One must keep in mind that these corrosion rates are
based on the assumption of a constant rate over the period of time involved and
should only be used as a means of comparing the different samples. Table 5 shows
a comparison of the corrosion rates calculated for this container with the previous two
retrieved from other Atlantic and Pacific radioactive waste disposal sites.
The calculated corrosion rate for the sediment side of this container is less than
that observed for the container extracted from the Atlantic 2800-meter site and lies
close to the zero-oxygen limit described by the empirical formula determined for
specimens having well defined initial conditions [8]. To summarize the results of the
dimensional analysis, the following statements can be made:
1. The sediment side showed greater general corrosion than the sea side,
with a rate of 0.015 mm/yr vs. <.005 mm/yr for the sea side. This
corresponds to a time of 42 years to cause 50 percent loss at the
sediment side.
2. Nevertheless, the corrosion rate of the sediment side was close to that of
empirically determined zero-oxygen limit.
3. The general corrosion rate of the sediment side fell under that for the
package retrieved from the 2800-m Atlantic site.
As mentioned before, both the sea side and the sediment side experienced
perforation. These corrosion rates apply only to general attack, not to the high rate
localized attack that produces perforation.
44
-------
TABLE 5
Corrosion Rate Data for Waste Packages Retrieved from Deep Sea Disposal Sites
Waste Package
Corrosion Rates
inches yr"1
Corrosion Rates
mm yr1
Reference
Atlantic 2800-meter Site
Gen. Attack
Sea Side
Sediment Side
Local Attack
0.0013 + .0002
0.0019 ± .0002
> 0.0026
0.033
0.048
> 0.066
(1)
1 wmmmmMMMmmmmmsmm?
II
Pacific 900-meter Site
Gen. Attack
Sea Side
Sediment Side*
Local Attack
0.00075 ± .00015
0
> 0.0021
0.019
0
> 0.054
(2)
Atlantic 3800-meter Site
Gen. Attack
Sea Side
Sediment Side
< 0.0002
0.0006 ± 0.0001
< 0.0045
0.015
This report
'Microscopic examination suggests a shallow pitting and scale formation resulting
in a calculated 0.0025 mm/yr average.
-------
4.3 Micro-Analysis
Microscopic examination revealed specific attack or mechanisms of protection.
4.3.1 Protected Regions
The concrete/metal interface and the side not buried in the sediment
experienced the greatest protection. Hence, these regions were closely examined to
determine possible modes of protection. Figures 20a and 20b show the well-
protected region of the container sheath at the concrete/metal interface from the
portion of the drum not buried in the sediment. This particular sample has the
following coordinates: (r=11.25 cm, x=21 cm, 0=270°). An apparent coating with a
thickness of 5 to 7 microns (/*) covers the metal. Energy dispersive x-ray analysis
(EDAX) revealed that the layer or "coating" directly covering the metal contained no
heavy metals other than iron. However, deposits on this inner coating (Figure 20c)
contained calcium, aluminum and chlorine deposits attributed to sea and concrete.
At the concrete/metal interface, the 5 to l-\x film or "coating" remains continuous
and adheres to the metal. However, on the environment side of this particular sample
the 5 to 7-#i film or "coating" also remains continuous even over regions where a thick
scale growth has produced a pit, Figure 21a. The scale growth on the pit exhibits
cracks and voids, but the 5 to 7-/x initial film or coating remains, with the exception
that a break appears above the center of the pit. In addition, a deposit lies over this 5
to 7-fi film or "coating" and contains compounds of aluminum and calcium, as shown
in Figure 21b.
In a case where the thick, predominately calcium deposit remains, little
corrosive attack has occurred. Also observed for this case, magnesium has
concentrated in a layer closest to the 5 to 7-fi film under the outer calcium-rich deposit
(Figures 22a and 22b). As shown by a considerable body of experimental data,
46
-------
0.002. em.
Figure 20a Optical photomicrograph of the metal/concrete interface from the well-protected
portion of the container.
Figure 20b SEM corresponding to the same area as 19a.
47
-------
+
pHfi 12 ADO , 028KEV/CH- 30. 5ECS
Figure 20c Energy Dispersive Spectroscopy (EDS) of deposits (Ca, Al, CI, etc.)
on the inner coating of the container sheath.
48
-------
Figure 21a SEM photograph of a pit initiated on the sea side of the container at R= 11.25,
x=21 cm, 0=270°.
TASK:
*
PHR12 ADD . 828KEV/CH-
30. SECS
n
y :[ T ^ A. ^
~
. 080
10. 220
30,
Figure 21b EDS of scale above 5-7/i film or "coating" showing A1 and Ca compounds.
49
-------
Figure 22a SEM showing the distribution of elements in the scale formed on the sea exposed
side of the well-protected region of the container.
Figure 22b X-ray DOT mapping of magnesium layer.
50
-------
alkaline earth precipitates can protect metal in the marine environment [9] because of
diminished oxygen depolarization of the cathodic sites [10]. Conceivably, the ocean
side of the container could be predominantly a cathode where the following reaction
dominates, 2e" + Oz + 2HzO ¦* 40H, and the anodic dissolution, H20 + Fe -+ FeOH+
+H+ +2e", would dominate the sediment side. Initially, this differential aeration
between the sea and sediment sides will encourage these reactions. Precipitation of
alkaline earth scale would slow the entire process, probably due to diminished mass
transfer of the 02 depolarization [11]. Although the ocean side shows better
protection against general corrosion than the sediment side, high localized attack
penetrated several places on this side. Specific penetration occurred at the rim of the
end with exposed concrete and at several points on or near the chimes. Severe, but
localized, pitting also occurred on the sea-facing side. Figure 23a shows an oxide-
filled pit found on the side of the sheath facing the sea. The oxide is 0.02 cm thick at
the maximum and appears to have grown in layers. It exhibits cracks, extending to
the base of the pit. The base of the oxide contains "ghosts" of the metal, seen as light
spots in the micrograph (Figure 23b). Scrutiny of the initial-etched sample by a
scanning electron micrograph, Figure 23c, showed inclusions having the same
morphology as the embedded "ghosts." EDS analysis, Figure 23d, indicates that
these "ghosts" originate from the metal as grain boundary precipitated carbides.
4.3.2 Rapidly Corroded Region on the Sediment Side
While the ocean-exposed surface of the container provides regions of high
protection, its sediment side showed the greatest metal loss resulting from general
attack. This attack takes the form of a general pitting. The base of a typical pitted
region in a sample taken from (x=38.1 cm, 0=90°) shows the morphology of the
attack (Figure 24). In contrast to the side above the sediment, this surface has
porous, loosely adherent scale. The scale also contains some of the carbide "ghosts."
The concrete/metal interface of this specimen exhibited virtually no attack and
contained the 5 to 7-fi film of "coating" (Figure 25). Thus, the concrete/metal interface
51
-------
1 I
0,&02<} err*
Figure 23a Optical photomicrograph of an oxide-filled pit on the sea side of the container.
Figure 23b SEM photograph of the same area showing metallic "Ghosts."
52
-------
Figure 23c
SEM photograph of inclusions and precipitates in the metal.
TASK:
PHfl 12
820KEV/CH.-
5 fin Ft Fk.
Figure 23d EDS analysis of inclusions and precipitates of the metal and the MnS inclusion.
53
-------
Figure 24 Optical photograph of severe attack on a sample taken from the sediment side at
x=38.1 cm, 0=90°
Figure 25 Optical photomicrograph of the concrete/metal interface at x=38.1 cm, 0=90°,
54.
-------
over the entire package showed no attack except at the end of the container where
the concrete was exposed.
4.4 Chemistry and Metallurgy
Immediately after removal of the waste package from the sealed storage
container, scrapings and samples were taken from regions indicating high corrosion
rates (Table 6) to determine product composition. Table 7 shows the samples taken
for x-ray diffraction (XRD) analysis. The samples consisted primarily of y-Fe203 and y-
Fe203H20 which can form from a rapid oxidation of ferrous ions in neutral solution
[12]. a-Ferric oxides also contributed to the composition of these scrapings.
The analyses of trace elements present in the carbon steel sheath (Table 8) did
not differ significantly from that of a previous container retrieved from the 2800-meter
site in the Atlantic Ocean [1]. Differences in corrosion rates can therefore be
attributed to the respective environments. The metallurgy of the drum did not differ
significantly from that of the two previously-retrieved containers (Table 8).
4.5 Discussion
The calculated rates provided by the dimensional analysis allow some
interesting comparisons among the three samples. As stated in previous reports, the
calculated rate results from the assumption of a constant rate over the entire time of
storage of the package. This represents a rather severe assumption, but must be
used to compare the different samples quantitatively. Under this assumption, the
metal loss as a function of time would appear as the schematic, Figure 26a, the slope
being determined by a chemical variable, 02 concentration for instance. With similar
composite systems [13], initiation and inhibition play important roles in determining the
metal loss (Figure 26b). In general, there is an initiation time, a propagation period,
and a possible inhibition at longer times.
55
-------
TABLE 6
Position and Description of Container Scrapings and Samples
Scraping No.
Position
Comments
R, (cm.)
X, (cm.)
0 (degrees)
III-l
28.7
34.2
305
Corrosion product near
chime
III-2
28.7
11.43
255
Adherent materials near rim
III-4
17.2
83
45
Corrosion product from
container end
III-5
28.7
66
30
Deposit over perforation
III-6
28.7
15
170
Pits-orange molecule
-------
L/1
-4
TABLE 7
t
X-ray Diffraction Identification of Surface Scrapings
III-l
7-Fe203 some a-FeO(OH)
III-2
and some other oxides
III-4
a-Fe203 H20 a-Fe304
y-Fe203 H20 y-Fe203
III-5
y-FeA , FeO(OH)
a-FeOOH
III-6
y-Fe203, y-Fe203 H20
-------
TABLE 8
Metallurgical Analysis of Waste Container
Element
Weight Percent
C
0.10
Mn
0.38
P
0.005
S
0.030
Si
< 0.02
Ni
< 0.02
Cr
< 0.02
Mo
< 0.02
Cu
0.1
Fe
Balance
-------
100%
I02l
Metal Loss
Constant
Rate
[02]2
[o2]3
time
Figure 26a Schematic of metal loss vs. time assuming a constant corrosion rate.
100%
Metal Loss
d
©
•p*
*3
c
o
•mm
$
W>
m
€k
£
P-i
P
0
•im
Xi
•mm
A
fi
time
Figure 26b Schematic of generalized corrosion kinetics.
59
-------
Keeping in mind these above points, the "calculated rate" for this sample can be
compared to data for other containers and data from the literature [1,2]. The
summary of the general observations follow:
1. The sediment-facing surface showed greater metal loss than the sea side for
this package. This is the reverse for the losses observed in the container taken
from a 900-meter Pacific site.
2. The calculated rate for the sediment side is 0.015 ± 0.002 mm/yr (0.0006 ±
0.0001 in/yr) and corresponds to a 50 percent reduction in 42 years.
3. The rate observed for general attack of the sediment buried side is close to that
for the Pacific container and to the zero-oxygen limit (see Table 5).
4. The rate observed for general attack of the sediment-buried side is significantly
less than that for the greatest general attack experienced by the Atlantic
container retrieved from 2800 meters.
Clearly, there are many factors involved that protect the container sheath from
corrosion, since metal loss probably follows a law similar to the schematic in Figure
26b.
Coatings, cathodic depolarizers such as dissolved oxygen, inhibiting materials
such as Caz+ and Mgz+ precipitates, pH and ocean current all have effects. Part of the
Atlantic 2800-meter container and the entire Pacific 900-meter container appear to
have a coating that would contribute to the time to initiate corrosion.
The differences in oxygen concentrations appear to play an important role also.
The Pacific 900-meter container, an apparently uncoated specimen, shows less loss
than the uncoated portion of the Atlantic 2800-meter container. The sample comes
60
-------
from a 900-meter Pacific site having a lower 02 concentration than the 2800-meter
Atlantic site from which a sample was obtained [2].
In addition, the mass flow of the cathodic depolarizer, Oz, to the metal surface and
the precipitation of calcium and magnesium precipitates probably inhibit the cathodic
process over the entire container. If the cathodic process is the rate-determining step,
then this will control the corrosion rate. The effect of the Mg and Ca precipitates upon
the cathodic process has been well established [9,10] and even provides the basis for
a patent application [14], Feigenbaum has used the high-frequency (>1KHZ)
impedance across protective scale to assess the corrosivity of fresh waters [10].
Perhaps, this applies to this system and environment. If this were to be the case, a
rapid in-situ probe could be easily developed.
4.6 Summary and Conclusion
The following summary and conclusions concerning the third container can be
made at this time.
1. The loss of metal as a result of corrosion is greater on the sediment side than
on the sea side. Assuming a constant rate (Figure 26a), the calculated rate is
0.015 ± 0.002 mm/yr. This corresponds to a 50 percent reduction of wall in 42
years. The general loss of the sea side is within the error of the measurement
of average wall thickness.
2. An apparent coating and Ca2+ or Mg2* precipitates appear to inhibit the
respective initiation and propagation of corrosion for this material in this
environment.
3. High local attack appears at chimes and at the rim of the concrete-exposed
end, where loss of concrete to metal adhesion occurs.
61
-------
-------
5. REFERENCES
1. Colombo, Peter, R.M. Neilson, Jr. and M.W. Kendig, "Analysis of a Radioactive
Waste Package from the Atlantic 2800 Meter Disppsal Site," EPA 520/1-82-009,
Office of Radiation Programs, U.S. Environmental Protection Agency, Washington,
DC. May 1982.
2. Colombo, Peter and M.W. Kendig, "Analysis and Evaluation of a Radioactive
Waste Package Retrieved from the Farallon Islands 900-Meter Disposal Site," EPA
520/1-90-014, Office of Radiation Programs, U.S. Environmental Protection
Agency, Washington, DC, September 1990.
3. Hanselman, David S. and W.B.F. Ryan, "1978 Atlantic 3800-Meter Radioactive
Disposal Site Survey -- Sedimentary, Micromorphologic and Geophysical
Analyses," EPA 520/9-79-001, Office of Radiation Programs, U.S. Environmental
Protection Agency, Washington, DC, June 1983.
4. Walden, Barrie B.f "Recovery of Low-Level Radioactive Waste Packages from
Deep-Ocean Disposal Sites," EPA 520/1-90-027, Office of Radiation Programs,
U.S. Environmental Protection Agency, Washington, DC, September 1990.
5. Fuhrmann, Mark, R.F. Pietrzak, J. Heiser, E-M. Franz, and P. Colombo,
"Accelerated Leach Test Development Program," BNL-52270, Brookhaven National
Laboratory, Upton, NY, November 1990.
6. Fuhrmann, Mark and P. Colombo, "Radionuclide Releases from Cement Waste
Forms in Seawater," Radioactive Waste Management and the Nuclear Fuel Cvcle.
Vol. II, No. 4, pp. 365-380, 1989.
63
-------
7. Fuhrmann, Mark, J. Heiser, R. Pietrzak, E-M. Franz, and P. Colombo, "User's
Guide for the Accelerated Leach Test Computer Program," BNL-52267,
Brookhaven National Laboratory, Upton, NY, 1990.
8. Dyer, Robert S., "Environmental Surveys of Two Deep Sea Radioactive Waste
Disposal Sites using Submersibles," Proc. Int. Svmp.. IAEA, Vienna (1976), 317-
338.
9. LaQue, F.S., "Marine Corrosion," and references therein.
10. Feigenbaum, C., et. al., Corrosion Science. 34(4), 133 1978.
11. Reinhart, F.M., et. al., "The Relationship Between the Concentration of Oxygen and
Seawater Corrosion of Metals," Proc. 3rd Cona. on Marine Corrosion. 1972, 562.
12. Misawa, T.K., et. al., Corrosion Science. 14, 131, 1974.
13. Tutti, K.f "Cracks and Corrosion," Swedish Cement and Concrete Research
Institute, Institute of Technology, Stockholm, ISSN 0346-6906, 1978.
14. Discussions with Drs. Hermann and Clayton, State University of Stony Brook; Dr.
Hermann informed me that a Dr. Wolf Hibberts of University of Texas, Austin, has
applied for such as patent.
64
------- |