United Slates
Environmental Protection
Agency
Air And
Radiation
(ANR-459)
EPA 520/1-90-014
September 1990
&EPA
Analysis And Evaluation
Of A Radioactive
Waste Package
Retrieved From
The Farallon Islands
900-Meter Disposal Site
Printed on Recycled Paper
-------
EPA 520/1-90-014
ANALYSIS AND EVALUATION
OF A RADIOACTIVE WASTE PACKAGE
RETRIEVED FROM
THE FARALLON ISLANDS 900-METER DISPOSAL SITE
September 1990
by
P. Colombo and M. W. Kendig
Waste Management R&D Group
Department of Nuclear Energy
Brookhaven National Laboratory
Associated Universities, Inc.
Upton, New York 11973
This study was conducted for the
U.S. Environmental Protection Agency
under Interagency Agreement No. EPA-IAG-D6-0166
PROJECT OFFICER
Robert S. Dyer
Office of Radiation Programs
U.S. Environmental Protection Agency
Washington, DC 20460
-------
FOREWORD
The Environmental Protection Agency (EPA) was given a Congressional
mandate to develop criteria and regulations governing the ocean disposal of all forms of
waste, pursuant to Public Law 92-532 (the Marine Protection, Research and Sanctuaries
Act of 1972) and, as amended, by Public Law 97-424. The EPA has taken an active role,
nationally and within the international nuclear regulatory community, to develop the
effective controls needed to protect the health and safety of man and to safeguard the
marine environment.
In 1974 the EPA Office of Radiation Programs (ORP) first initiated
feasibility studies to determine whether existing technologies could be applied toward
assessing the fate of radioactive wastes that had previously been disposed in the oceans.
After successfully locating waste packages in ocean sites formerly used by the U.S. to
dispose of radioactive waste materials, ORP developed a program of site characterization
studies to determine the biological, chemical, geological and physical characteristics of the
marine environment, in and near the disposal sites. These studies also included
evaluations of the concentration and distribution of radionuclides within and near the
disposal sites.
In addition, ORP has retrieved radioactive waste containers from three
deep-ocean disposal sites to evaluate the performance, with time, of past packaging
techniques. Under an interagency agreement with ORP, the Brookhaven National
Laboratory (BNL) has performed container corrosion and matrix analysis studies on the
recovered radioactive waste packages. The results of BNL's first analysis of a LLW
package recovered, in 1976, from the deep-ocean were published in EPA Technical Report
No. EPA 520/1-82-009, "Analysis and Evaluation of a Recovered Radioactive Waste Package
from the Atlantic 2800 Meter Disposal Site."
In 1977, ORP recovered another LLW package from the Faralion Islands
900-meter disposal site. This report details the BNL analysis of that waste package.
Readers of this report are invited to send comments or suggestions to Mr.
Martin P. Hal per, Director, Analysis and Support Division (ANR-461), Office of Radiation
Programs, U.S. Environmental Protection Agency, Washington, DC 20460.
Richard J. ttaimond, Director
Office of Radiation Programs
-------
Table of Contents
FOREWORD iii
List of Figures vii
List of Tables ix
1. INTRODUCTION 1
1.1. FARALLON ISLANDS RADIOACTIVE
WASTE DISPOSAL SITE 1
1.2 WASTE INVENTORY 1
13 WASTE TYPES 4
1.4 WASTE PACKAGING TECHNIQUES 4
2. SURVEY AND RECOVERY OF A WASTE PACKAGE
FROM THE 900-METER SITE 6
2.1 FARALLON ISLANDS 900-METER DISPOSAL
SITE SURVEY (1974) 6
2.2 RETRIEVAL OF A WASTE PACKAGE FROM THE
FARALLON ISLANDS 900-METER SUBSITE 8
2.2.1 Retrieval Operation 8
2.2.2 Radiation Surveillance 15
2.23 Shipboard Inspection 19
2.2.4 Storage and Transportation 21
3. ANALYSIS OF THE CONCRETE WASTE FORM 22
3.1 RADIOGRAPHY 22
3.2 CONTAINER REMOVAL 25
33 CONCRETE CORING 25
3.4 WASTE CONTENT 27
3.5 RADIOACTIVITY 30
3.6 WASTE FORM INTEGRITY 30
4. CORROSION ANALYSIS OF THE METAL CONTAINER 33
4.1 VISUAL INSPECTION AND SAMPLING 33
4.2 DIMENSIONAL ANALYSIS 34
43 PROTECTED REGION ON THE SEDIMENT SIDE 46
4.4 CORRODED REGION ON THE SEA SIDE 50
5. CONCLUSIONS 63
REFERENCES 64
-------
List of Figures
Figure 1.1 Farallon Islands Radioactive Waste Disposal Sites 2
Figure 1.2 Cut Away Isometric Views Showing Modifications of 55-Gallon
Drums for Packaging Radioactive Wastes Which Were Disposed
of In The Farallon Islands Disposal Site 5
Figure 2.1 Unmanned Cable Controlled Underwater Recovery Vehicle,
.CURV III 7
Figure 2.2 Manned Submersible PISCES VI 9
Figure 2.3 Canadian Research Vessel PANDORA II 10
Figure 2.4 University of Southern California Research Vessel VELERO IV 10
Figure 2.5 Waste Package Showing Severe Hydrostatic Implosion .... 12
Figure 2.6 Waste Package Showing Mild Hydrostatic Implosion 12
Figure 2.7 Waste Package Showing No Signs of Hydrostatic Implosion . . 13
Figure 2.8 Underwater Inspection of Recovered Waste Package 14
Figure 2.9 View of the Wire Rope Lifting Eye Imbedded into the Concrete
Waste Form 14
Figure 2.10 Submersible PISCES VI Being Launched For Waste Package
Recovery 16
Figure 2.11 Waste Package Being Hoisted Aboard the Research Vessel
VELERO IV 18
Figure 2.12 The Waste Package Being Prepared for Visual Inspection . . 18
Figure 2.13 A View of the Corrosion on the Sea Side of the Container . 20
Figure 2.14 Sea Worms Attached to the Concrete Surface of the Waste Form 20
Figure 3.1 Removal of the Waste Package From the Shipping Container . 23
Figure 3.2 View of the Sediment Side of the Waste Container 24
Figure 3.3 Concrete Waste Form After Removal of Container 26
Figure 3.4 View of the Position of the Cardboard Box Embedded in Cement 28
Figure 3.5 Closeup of Parted Waste Form 29
Figure 3.6 Legibility of Printed Matter on Cardboard Box 29
-vii-
-------
List of Figures (Continued)
Figure 4.1 Schematic of Coordinate System Used to Identify Locations
on the Container 35
Figure 4.2 View of Waste Package Showing the Sediment Side (0° - 90°) 36
Figure 4.3 View of Waste Package Showing 37
Figure 4.4 View of Waste Package Showing the Sea Side (180° - 270°) . 38
Figure 4.5 View of Waste Package Showing the Sea Side (270° - 360°) . 39
Figure 4.6 Nodule of Corrosion Product on Sea Side 40
Figure 4.7 Strips Extracted for Metal Loss Analysis 41
Figure 4.8 Perforations Observed Prior to Cutting Sheath 42
Figure 4.9 View of the Sheath from the Concrete Side 43
Figure 4.10 Closeup of a Chime Perforation 44
Figure 4.11 Metal Thickness as a Function of Position 48
Figure 4.12 Optical Micrograph of Well Protected Specimens Taken from
Positions (x - 9 inches, Q - 45), (x - 15 inches,
6 - 45) respectively 51
Figure 4.13 Initiation of Corrosion on the Sediment Side of the Container
(x-12 inches, 6 - 46") 52
Figure 4.14 Rapid General Corrosion on Sample Taken from X=27 inches,
6 - 225° 53
Figure 4.15 Metal to Scale Interface of Sample Taken From x-27 inches,
6 - 225° 54
Figure 4.16 Corrosion Scale of Sample (Concrete Side) Taken From
x=27 inches, 9 - 225° 55
Figure 4.17 Solid Phase Analysis of Concrete Side Lamina of Sample Taken
from x-27 inches, 6 - 225° 56
Figure 4.18 Schematic of Container with Scale Growth 57
-Vlll-
-------
List of Tables
TABLE 1.1 FARALLON ISLANDS RADIOACTIVE WASTE DISPOSAL SITE 3
TABLE 3.1 COMPRESSIVE STRENGTH OF CONCRETE CORES 32
TABLE 4.1 SCRAPINGS 45
TABLE 4.2 CALCULATED CORROSION RATES 49
TABLE 4.3 TRACE ELEMENT ANALYSES OF CONTAINER MATERIAL 60
TABLE 4.4 X-RAY FLUORESCENCE ANALYSIS OF TRACE COMPONENTS IN SCRAPINGS 61
TABLE 4.5 X-RAY DIFFRACTION ANALYSIS OF MAJOR IRON OXIDE COMPONENT . 62
-IX-
-------
1. INTRODUCTION
1.1 FARALLON ISLANDS RADIOACTIVE WASTE DISPOSAL SITE
Between 1946 and 1970, the United States disposed of radioactive wastes by
ocean disposal at designated sites licensed by the Atomic Energy Commission
(AEC).
An ocean area near the Farallon Islands was one of the sites designated by the
AEC for ocean disposal of radioactive wastes. This site received the majority
'of all wastes deposited in the Pacific Ocean.
The Farallons are a chain of uninhabited islands extending in a northwest
direction approximately 70 kilometers (40 miles) west of San Francisco,
California. The site actually consists of three separate subsites, identified
as 1, 2 and 3, in Figure 1.1. Each subsite was officially used for varying
periods of time between 1946-1965.
The first area selected for receiving radioactive waste packages in 1946 was
subsite 1, having a depth of approximately 92 meters. During the latter part
of 1946, the disposal operations were transferred to subsite 3 which is
located 77 kilometers (45 miles) from land at a depth of 1700 meters.
Disposal operations continued at this site until 1951 when subsite 2, at a
depth of 900 meters, was designated as the only subsite for disposal of
radioactive wastes. The reason for this new subsite designation is not
entirely clear although it may be attributed to the fact that subsite 3 was
also used to dispose of chemical munitions. The greater distance from land
may also have been a consideration. In 1954, however, disposal was resumed at
subsite 3 and it continued until 1965 when land disposal sites were licensed
to receive radioactive waste [1].
1.2 WASTE INVENTORY
Table 1.1 gives the estimated number of waste packages and total activity
deposited at the three Farallon Islands subsites. During the time period when
ocean disposal was in effect, detailed recordkeeping was generally not
required since the disposal of radioactive waste was considered a "garbage
disposal type of operation" [2]. Consequently, a complete assessment cannot
be made of the total number of packages or curies of radioactivity of the
packages which were disposed of at the Pacific Ocean radioactive disposal
sites.
Most of the wastes disposed at the Farallon Islands site were generated by:
(1) the U.S. Naval Radiological Defense Laboratory, (2) the University of
California Lawrence Livermore Radiation Laboratory, and (3) the University of
California Radiation Laboratory at Berkeley. The radioactive waste disposal
operations were conducted by the U.S. Navy until July 1959 when private
companies assumed the responsibility under AEC license.
-------
123
38
o..
1-0
37«
San Jose
Figure 1.1 Farallon Islands Radioactive Waste Disposal Sites
-------
Subslte
1
2
3
Coordinates
37" 38'N
122° 58'W
37e 38'N
123° 08'W
37° 37'N
123° 17'W
TABLE 1.1
FARALLON ISLANDS RADIOACTIVE WASTE DISPOSAL SITE
Depth
(m)
90
(300 ft)
900
(3000 ft)
1700
(5300 ft)
Distance
From Land
(km)
45
(28 miles)
60
(40 miles)
77
(50 miles)
Years
Used
1946
1951-1953
1946-1950
1954-1964
Estimated No.
of 55-Gallon
Waste Packages
150
3,600
44,000
Estimated
Activity
(Ci)
unknown
1,100
13,400
-------
1.3 WASTE TYPES
Both defense waste and waste generated at commercial and medical facilities
consisted of materials having activity levels normally associated with
laboratory operations. The waste inventories represented an extremely
heterogeneous group of liquid and solid materials with physical and chemical
properties varying over a great compositional range. Typical solid waste
consisted of contaminated paper, metals, rubber, rags, and glass. The wet
waste included filter cartridges, aqueous solutions, evaporator concentrates,
solvents and other miscellaneous materials.
In general, radiation contamination was mainly by beta-gamma emitters evolving
from reactor experimental materials and byproducts of isotope production
having half-lives greater than one year. Many of the early waste packages
disposed in the Farallon Islands site contained alpha-emitting wastes generat-
ed by particle accelerators. It is estimated that approximately 30 curies of
alpha activity was packaged and disposed through 1953 [2].
1.4 WASTE PACKAGING TECHNIQUES
Most of the waste disposed at the Farallon Islands site was packaged in used
or reconditioned 55-gallon mild steel drums. The waste was usually mixed with
or encased in cement prior to or at the time of packaging so that the average
package density was sufficiently greater than seawater to ensure sinking to
the ocean floor after disposal. The drums served to contain the waste
mixtures, to minimize dispersion during handling and transportation and to
offer some radiation protection to personnel. No credit was given to the
container as a barrier to radionuclide migration in ocean disposal. It was
assumed that all the radioactive materials would eventually be released since
the packages were not designed or required to remain intact for sustained
periods of time after descent to the ocean floor. It was further assumed that
ocean currents would dilute and disperse the radioactivity to such low
concentrations that it would not constitute a significant hazard to man and
the environment.
Several of the techniques used for packaging radioactive wastes which were
deposited in the Farallon Islands disposal site are shown in Figure 1.2.
Beginning in 1951-52, all radioactive waste packages for ocean disposal at the
Pacific site incorporated a lifting eye which consisted predominantly of a
wire rope or a bent steel reinforcement bar with both ends embedded into an
exposed concrete cap. This information was useful in identifying and dating
radioactive waste packages during the EPA surveys at the 900-meter subsite in
1974-75 and at the 1700-meter subsite in 1977.
-------
SPACE
FOR
LIQUIDS
(a)
Weighted drum for
low level liquids
(b)
Carboy of low level
liquids
(c)
30 gallon drum of
solidified liquids
f
SPACE
FOR
MIXTURE
Mixture of solid waste
materials and concrete
Figure 1.2 Cut Away Isometric Views Showing Modifications of 55-Gallon Drums
for Packaging Radioactive Wastes Which Were Disposed of In The
Farallon Islands Disposal Site
-------
2. SURVEY AND RECOVERY OF A WASTE PACKAGE FROM THE 900-METER SITE
The framework for the eventual recovery and analysis of a waste package from
the Pacific Ocean began in 1974, when EPA successfully conducted an environ-
mental assessment survey of the radioactive waste disposal site near the
Farallon Islands [3].
The survey was conducted primarily to determine the fate of radioactive waste
packages disposed of at that site between 1946-65, and to make preliminary
determinations regarding the distribution of any released radioactivity.
2.1 FARALLON ISLANDS 900-METER DISPOSAL SITE SURVEY (1974)
The 900-meter site (Figure 1.1) was selected by EPA as the first in a planned
series of ocean disposal site surveys in the Pacific and Atlantic Oceans. The
site was singled out in preference to others because (1) it was the only site
used exclusively for disposal of radioactive waste, (2) it was used only from
1951-53, (thereby allowing estimates of the age of the waste package and the
rates of biofouling and corrosion), and (3) the precise coordinates of the
site were known.
The site survey was conducted through the use of the Cable-controlled Underwa-
ter Recovery Vehicle (CURV III), operated by the U.S. Navy Undersea Center
(NUC) in San Diego. The CURV III, shown in Figure 2.1, is a tethered,
unmanned submersible which is remotely controlled from shipboard and has
capabilities for both sediment sample collection and photographic documenta-
tion. Its equipment includes two movable television cameras, a 35mm color
camera with synchronized strobe, and a sonar system capable of scanning an
area of 120° and detecting waste packages at distances up to 400 meters [4].
On August 28, 1974, the first cluster of waste packages, consisting of 55-
gallon mild steel drums was located. The characteristic lifting eyes posi-
tively identified them as radioactive waste packages disposed of 21-23 years
earlier.
Although a pronounced number of packages appeared to have imploded due to
hydrostatic pressure, none of the observed packages showed signs of having
been breached solely from external corrosive forces, even though some surface
corrosion was evident on the containers. The hydrostatic implosions were the
consequence of air voids and/or the nonhomogeneity of the waste form within
the container when exposed to deepsea high pressure conditions. The condition
of some waste packages observed during this survey are evident in a series of
color photographs contained in an EPA operations report [1].
This preliminary survey established the feasibility for conducting more
sophisticated studies at radioactive waste disposal sites and laid the
groundwork for subsequent surveys which were conducted in 1976-78 at other
subsites in the Pacific Ocean and at Atlantic Ocean disposal sites.
-6-
-------
Figure 2.1 Unmanned Cable Controlled Underwater Recovery Vehicle, CURV III
-------
2.2 RETRIEVAL OF A WASTE PACKAGE FROM THE FARALLON ISLANDS 900-METER SUBSITE
(1977)
In October 1977, a second survey of the Farallon Islands 900-meter disposal
site was undertaken by EPA for the specific purposes of: (1) retrieving a
radioactive waste package and (2) supplementing information acquired during
the 1974 survey regarding the geochemical, radiochemical, and biological
characteristics of the site.
2.2.1. Retrieval Operation
The retrieval of a waste package from the 900-meter site was a coordinated
effort involving the research vessel Pandora II. the manned submersible PISCES
VI. and the research vessel Velero IV. The 220-foot research vessel Pandora
II was the support ship for PISCES VI as shown in Figure 2.2 and is operated
by the Canadian Department of Fisheries and the Environment. On October 17,
1977, it arrived from Vancouver, British Columbia, to Fisherman's Wharf in San
Francisco, California, where it joined the 110-foot research vessel Velero IV
of the University of Southern California for a scheduled 10-day study at the
Farallon Islands disposal site. Figure 2.3 shows the highly specialized
Canadian research vessel Pandora II with its large articulated A-Frame for
raising and lowering the submersible. Figure 2.4 shows the research vessel
Velero IV. from which correlating biological, geological, and radiochemical
data were collected. The Velero IV was also designated for shipboard recovery
of a radioactive waste package.
On October 21, 1977, a pre-recovery survey was conducted at the 900-meter site
to select a waste package for recovery.
The criteria established for the selection and recovery of the waste package
included the following:
• It should have identifiable markings regarding its source, activity
level and date of disposal.
• It should be essentially intact with no signs of implosion or
breaching due to impact with the ocean floor; this would minimize
the possibility of shipboard and personnel contamination during
recovery.
• The exposure rates (as measured from the manned submersible prior to
recovery) should not be of a level which might affect the health and
safety of personnel during recovery, storage, and transportation.
The precise positioning of sighted radioactive waste packages and the accuracy
with which depths were recorded during the 1974 survey of the 900-meter site
identified the areas having appreciable numbers of waste packages. This
earlier information not only served to pinpoint locations but also provided
the opportunity to select a waste package which could best represent the
effects of the ocean environment during the time it sat on the ocean
floor.
-8-
-------
Figure 2.2 Manned Submersible PISCES VI
-------
Figure 2.3 Canadian Research Vessel PANDORA II
Figure 2.4 University of Southern California Research Vessel VELERO IV
10
-------
During the course of this survey, photographs were taken by PISCES VI for
documentation and for verification of the condition of the drums observed
during the 1974 survey of the 900-meter site.
An example of the implosion pattern characteristic of many of the waste
packages sighted is shown in Figure 2.5. Hydrostatic crushing at the center
of a drum is typical of radioactive waste packages prepared as shown in Figure
1.2, whereby the waste was essentially sandwiched between two plugs of
concrete within the container. Invariably the waste consisted of laboratory
trash which was insufficiently compacted to remove air voids or lacked the
physical strength to resist hydrostatic implosion. Although the center
portion of the drum in Figure 2.5 showed severe indentations, it did not
appear to be breached. Several thornyhead fish, either Sebastolobus alascanus
or Sebastolobus altivelis are seen taking shelter near the side of the drum.
The exposed concrete cap establishes a solid anchor site for a number of
marine invertebrates such as the white vasiform hexactinellid sponges attached
to the exposed concrete end of the package.
Figure 2.6 shows a waste package with a moderate amount of hydrostatic
crushing. A lifting eye, typical of waste packages disposed of at the site
between 1951-1953, is evident on the concrete end of the package. Also
evident are several sponges attached to the concrete.
By contrast, Figure 2.7 shows a barrel with no evidence of hydrostatic
crushing. A coating of fine, minimally disturbed sediment is prominent on the
exterior of the drum. Sediment taken by boxcore in the vicinity of this drum
is sandy silt having a mean grain size of approximately 5. 6 . At the surface
the sediment consists of 28.8 percent sand, 44.8 percent silt and 26.8 percent
clay [5]. The photo also indicates the small amount of drum penetration into
the sediment with little or no sediment scouring in the vicinity of the drum.
The waste package shown in Figures 2.8 and 2.9 was selected for recovery since
it closely complied with the aforementioned selection criteria and it appeared
to be in good enough condition to survive the trip to the surface and yet
provide meaningful information on past packaging technique. Figure 2.8 shows
the mechanical arm of the PISCES VI carefully rolling the drum to ensure that
it was not breached on its underside. The concrete cap and lifting eye of the
waste package, as seen in Figure 2.9, was inspected to ensure that they would
remain intact during the attachment of the lift line to the lifting eye and
recovery of the waste package from the ocean floor. In both Figures 2.7 and
2.8, some of the benthic and demersal fish population can be seen close to the
waste packages.
-11-
-------
Figure 2.5 Waste Package Showing Severe Hydrostatic Implosion
Figure 2.6 Waste Package Showing Mild Hydrostatic Implosion
-12-
-------
Figure 2.7 Waste Package Showing No Signs of Hydrostatic Implosion
-13-
-------
Figure 2.8 Underwater Inspection of Recovered Waste Package
Figure 2.9 View of the Wire Rope Lifting Eye Imbedded
into the Concrete Waste Form
14
-------
On October 22, 1977, at 0927 hours, the PISCES VI. Figure 2.10, was launched
to recover the waste package which had been selected by visual observation
from the submersible the previous day. The PISCES VI arrived at the site of
the selected package at 1150 hours. The depth of the water was 915 meters
(approximately 2,750 feet). The PISCES VI proceeded to capture the package by
placing a metal harness over the top of the package. The mechanical arm of
the submersible secured the harness around the package using a grip hoist
which cinched the wire rope of the harness around the package. Although the
same procedure was used for harnessing the package for recovery from the
Atlantic Ocean 2800-meter site in 1976 [6], the package was raised by connect-
ing it directly to a winch line from the recovery vessel. In the current
case, the package was lifted using two nylon ropes attached to the submers-
ible. At 1420 hours, the PISCES VI was off the bottom and pumping ballast to
ascend. The package was reported free from the bottom at 1445 hours and the
PISCES VI surfaced at 1536 hours. At 1600 hours, Velero IV closed for the
package recovery and by 1610 hours all lines were attached from the package to
the recovery ship. At 1613 hours, the package was released from the PISCES VI
and recovery operations aboard the Velero IV began. By 1700 hours, the waste
package broke the surface of the water. Before bringing it aboard, it was
allowed to drain while at the same time it was subjected to a radiation survey
by the health physicist aboard the Velero IV.
2.2.2. Radiation Surveillance
The necessary monitoring equipment for alpha-beta-gamma counting was assembled
and calibrated before taking it aboard the research vessel Pandora II in
preparation for the survey and recovery operations. Radioactive standard
sources also were taken along for periodic checking and recalibration of the
monitoring equipment.
Prior to the survey and recovery activities, a radiation safety lecture was
given to the crews of the PISCES VI and the research vessel Pandora II and the
research vessel Velero IV. This lecture included radiation exposure limits,
radiation contamination, biological effects of radiation, procedures for the
use of film badges, self-reading dosimeters and survey meter operations. The
operators of the PISCES VI carried a radiation survey instrument aboard during
each dive. After each dive, the health physicist performed a contamination
survey of the PISCES VI and the Pandora II to detect any radioactive contami-
nation in samples collected and sampling equipment used during the disposal
site surveys.
Prior to retrieval operations aboard the research vessel Velero IV. prepara-
tions were made to prevent radiation exposure or the spread of radioactive
contamination in the event of leakage or an accident during the recovery
operations.
-15-
-------
Figure 2.10 Submersible PISCES VI Being Launched For Waste Package Recovery
16
-------
A collision mat was secured to the ship's transom with a double protective
cover consisting of a heavy duty nylon reinforced polyethylene tarpaulin on
the bottom and a canvas tarpaulin on the top, as shown in Figure 2.11. The
tarpaulins served to reinforce the collision mat and to confine "run-off" in
the event that water was discharged from the waste package during pressure
equalization or from void pockets within the package. It should be noted that
the pressure exerted on the surface of the waste package at a depth of 900
meters is approximately 1500 psi. The tarpaulins were extended from the
transom inboard to cover an area beyond that required to accommodate the waste
package. A 2-inch thick open cell polyurethane slab was then secured in
position to receive the waste package as shown in Figure 2.12. The purpose of
the open cell polyurethane slab was to: (1) act as a cushion for minimizing
damage while lowering the waste package from the winch to the deck in rolling
seas, (2) absorb any aqueous run-off from the waste package, and (3) facili-
tate containment of any radioactivity by rolling and properly wrapping the
flexible polyurethane slab.
Before the retrieval operations began, the health physicist and the monitoring
equipment were transferred from the Pandora II to the Velero IV. The ship-
board area designated to receive the waste package was restricted to the
health physicist and to personnel specifically assigned to assist in the
retrieval operation. This step was taken to minimize potential contamination
and personnel radiation exposure during shipboard operations. During the
recovery operation, a preliminary radiation survey of the package was per-
formed before taking it aboard and while it was suspended over the stern of
the Velero IV. This was followed by a complete radiation and contamination
survey after it was taken aboard.
The recovered package showed no external radiation or contamination. Smear
samples of the research vessel Pandora II. Velero IV. and the submersible
PISCES VI showed no contamination during or at the conclusion of operations.
The film badges and self-reading dosimeters worn by personnel involved in the
survey and recovery operations showed no radiation exposures above background.
-17-
-------
Figure 2.11
"T
Waste Package Being Hoisted Aboard the Research Vessel VELERO IV
•••Bi
Figure 2.12 The Waste Package Being Prepared for Visual Inspection
18
-------
2.2.3. Shipboard Inspection
Immediately after being hoisted on board the waste package was secured to the
deck of the ship where it was photographed and sampled for corrosion products
and biological growth. Sediment adhering to the bottom side of the package
was also sampled.
The mud, which clung strongly to the container during retrieval, appeared
greenish-black in color and had a faint odor of hydrogen sulfide, indicating
an anoxic environment. The pH of the sediment, measured with indicator paper,
was between 8-10. The mud line can be seen on the upper side of the concrete
face in Figure 2.11.
The sea-exposed surface of the container contained corrosion products and
scale ranging in color from black to reddish orange, as seen in Figure 2.13.
The whitish areas around the rim of the container in Figure 2.13, resemble
polychaete (sea worm) tubes. Sea worms appear to be prominent as observed on
the concrete surface in Figure 2.14. In addition to the worms, several small
or broken sponges can be seen attached to the concrete surface.
In general, the condition of the container appeared to be good considering the
duration of exposure in the deep ocean (estimated 21-23 years). The area of
heavy corrosion appeared to be on the outer rim of the sediment side of the
container. A more detailed description of the types and effects of corrosion
on the metal container will be given later in this report.
-19-
-------
Figure 2.13 A View of the Corrosion on the Sea Side of the Container
Figure 2.14 Sea Worms Attached to the Concrete Surface of the Waste Form
-------
2.2.4. Storage and Transportation
The on-board Inspection of the waste package was limited to approximately one
hour to minimize any chemical changes which might occur in an oxygenated
environment. Following inspection, the metal capture harness was removed and
the waste package was sealed into a cylindrical steel storage container, which
was especially designed to accommodate a standard 55-gallon drum. The
container conformed with all Department of Transportation (DOT) criteria for
Type A packages except for pressure requirements. Modifications were made to
provide a gasket seal between the cover flange and the flange of the container
which complied with the DOT requirements for maintaining the prescribed over-
pressure .
The cover of the storage container was further modified to provide the
necessary hardware for introducing an inert gas (argon) to replace the air in
the container. The use of inert gas was to minimize further corrosion of the
metal container during storage and transportation. Air was removed from the
storage container by continuously feeding argon gas into the container until
the oxygen concentration in the exit gas was reduced to 0.5 percent. The gas
inlet and outlet valves were then sealed to maintain a pressure of argon gas
slightly above atmosphere.
At the point of debarkation, (San Francisco, California), the storage contain-
er accommodating the waste package, was placed into an overpack for transpor-
tation to Brookhaven National Laboratory. The overpack used is a DOT approved
metal container suitable for the transportation of B type (transuranic) waste.
Because the radioactive content of the waste package was not known following
recovery, the above special transportation options were used to assure
compliance with regulations regarding the handling and transportation of such
wastes.
-21-
-------
3. ANALYSIS OF THE CONCRETE WASTE FORM
The recovered waste package consisted of a conventional 55-gallon mild steel
drum filled to within 1.5-2.0 inches from the top with a concrete mix.
Protruding through the exposed concrete end was a wire rope lifting eye which
was firmly embedded into the concrete. Although the overall appearance of the
metal container was good, the corrosion on the sea side appeared to be highly
localized. The nonuniform corrosion was surprising in that one would expect a
more even distribution. However, since most of the drums were used or
reconditioned prior to packaging and disposal, a large degree of uncertainty
exists regarding the initial conditions of any individual container and the
effects of the ocean environment on the corrosivity of the waste containers.
The blackish mud, indicative of anaerobic bottom conditions, remained strongly
bonded to the container when it was removed from the sealed storage container,
as shown in Figure 3.1. A faint odor of hydrogen sulfide could be detected,
typical of anaerobic bottom sediments where sulfate-reducing bacteria are
present. This condition is initiated through the formation of ferrous
sulphide (FeS) scale on the steel, which is cathodic to the bare metal surface
[7,8].
When the waste package was exposed to atmospheric conditions, the blackish
coloration immediately began to change to greenish-brown, as shown in Figure
3.2. This can be attributed to oxidation processes, with some metals such as
iron, during the transition from anaerobic to aerobic conditions.
A pool of liquid was observed at the bottom of the shipping container upon
removal of the waste package. This liquid, consisting of 13 liters, was
immediately collected for radiochemical analysis and pH determination. It was
presumed that the liquid had drained from the waste form since the shipping
container was hermetically sealed during storage and transportation. Such a
volume of liquid would require an appreciable void space to accommodate it
within the waste form. From this it was hypothesized that a packaging
technique along the lines of that demonstrated in Figure 1.2 (b) had been used
to package the waste.
3.1 RADIOGRAPHY
During the analysis of the waste package which was recovered from the Atlantic
2800-meter disposal site in 1976 [6] it became apparent that radiographs can
provide the necessary information to determine the presence and location of
objects which were incorporated into the cement waste form for disposal.
Radiographs for this study were produced using a 45-curie cobalt-60 source
supplied by the Consolidated Testing Laboratories, Inc., New Hyde Park, New
York. Since the waste form was suspect, a total of four sides of radiographs
were taken with the source located on the 0°, 90°, 180°, and 270" longitudinal
axis of the drum. Fiducial markers were placed on the drum to allow the
subsequent positioning of one radiograph relative to another.
-22-
-------
Figure 3.1 Removal of the Waste Package From the Shipping Container
23
-------
Figure 3.2 View of the Sediment Side of the Waste Container
24
-------
Interpretation of the radiographs indicated a rectangular-shaped cavity
situated off-center and adjacent to the 270° line. The size of the void was
estimated as 10 inches (25.4 cm) wide x 10 inches (25.4 cm) high x 12 inches
(30.5 cm) long. Positioning of this cavity permitted cores to be taken at
various locations and depths without impacting the void area. The radiographs
also indicated the precise location of the deeply embedded wire rope ends.
3.2 CONTAINER REMOVAL
The steel drum was carefully removed from the waste form using a pneumatic
chisel. For maintaining continuity, the same orientation system was used for
both the metal container and the concrete waste form. Longitudinal sections,
4 inches (10.2 cm) wide, were cut from the drum above and below the sediment
line for corrosion analysis. More detailed descriptions of the metal sampl-
ings and results are given in Section 4 of this report.
The homogeneity of the concrete, along its entire length, indicated that a
continuous pouring was made in preparing the waste form. This is shown in
Figure 3.3. Since the radiographs indicated that the upper side of the cavity
was approximately 8.5 inches (21.6 cni) from the exposed concrete surface, the
wire rope was cut and an attempt was made to reach the cavity by probing the
concrete surface. The hardness of the concrete, however, made this task
difficult and it was decided to use coring equipment.
3.3 CONCRETE CORING
The cores were obtained using a concrete hole saw with a dual speed motor
(500/100 rpm) on a swivel base [6]. Two-inch cores were obtained using
impregnated diamond core bits. Water was sparingly used during the drilling
operations as a lubricant and to minimize dispersion of fines. With this
method, cores were obtained having 2 inch (5.1 cm) diameters and 4 inch (10.2
cm) lengths. Cores were taken along the 0°, 90°, 180° and 270° longitudinal
axes at distances of 5.25 inches (12.7 cm), 16 inches (40.6 cm) and 26.5
inches (67.3 cm) from the top of the concrete waste form. The concrete cores
were taken to determine the compression strength of the concrete and for
radiochemical analysis.
Prior to obtaining the cores, exploratory drillings were conducted to substan-
tiate the location of the cavity as indicated by the radiographs and to
determine the contents of the cavity.
The cavity was located at a depth of 8.5 inches from the top of the waste
form, as indicated by the radiograph. Using a steel-hooked rod, pieces of
cardboard were extracted from the cavity. The same material was extracted
from drill holes made along the periphery of the cavity. During this opera-
tion, all extracted materials were continuously surveyed for radioactivity.
25
-------
...*•,
Figure 3.3 Concrete Waste Form After Removal of Container
26
-------
Based on the estimated location of the cavity, it was decided to attempt
cutting the waste form along the upper chime line as demarcated by the 55-
gallon metal drum which according to the radiograph, coincided with the
estimated center of the cavity. The waste form was finally parted using a
pneumatic chisel.
3.4 WASTE CONTENT
Figure 3.4 shows the parted waste form and the cavity, which housed a thick,
empty cardboard container. Figures 3.5 and 3.6 show a closeup of the parted
waste form. Note the integrity of the cardboard and the legibility of the
print in Figure 3.6.
The cardboard box was carefully removed from the cavity to determine if it
contained any information which might shed some light on its original con-
tents. The following markings were identified on the side and bottom of the
box.
SIDE OF BOX
TOP
HANDLE WITH CARE
GLASS
NET 35 LBS
GR 54 LBS
NO. 4052
BOTTOM OF BOX
3 GALS
1
1006
1048
0
The certification seal, also located on the bottom of the box read:
GAIR BOGATA CORP.
CERTIFICATE OF BOX MAKER
THIS BOX CONFORMS TO ALL CONSTRUCTION REQUIREMENTS
OF CONSOLIDATED FREIGHT CLASSIFICATION
BURSTING TEST 200 LBS PER SQ. INCH
SIZE LIMIT 5 INCHES
GROSS WT. 65 LBS
BOGATA N
27
-------
Figure 3.4 View of the Position of the Cardboard Box Embedded in Cement
28
-------
Figure 3.5 Closeup of Parted Waste Form
Figure 3.6 Legibility of Printed Matter on Cardboard Box
-------
3.5 RADIOACTIVITY
No radioactivity was detected in samples taken during each phase of this
operation. This included shipboard smears of the waste package and analysis
of the discharged liquid, the cardboard box, and the concrete cores. Yet, the
evidence is there to indicate that historically the recovered waste package
was disposed of at sea as radioactive waste. In view of this, the following
can be postulated:
(a) The cardboard box was discarded as "suspect" waste and was not
radioactive at the time of packaging. The practice of combining
"suspect" waste with radioactive contaminated waste was prevalent in
most laboratories since it eliminated the task of performing radio-
logical assays and segregation of the waste.
(b) The cardboard box was contaminated with short half-lived radionuc-
lides. It seems unlikely that a concrete casting was made specifi-
cally to house an empty cardboard box unless the box was heavily
contaminated with radioactivity. In the absence of any detectable
radioactivity it can be presumed that the contaminants consisted of
one or more short-lived radionuclides.
(c) All of the radioactivity leached out of the waste form. Judging by
the integrity of the cardboard box, the concrete waste form and the
container, it is difficult to visualize that all of the activity
leached out of the waste form. This is unlikely based on several
facts. The cardboard box, the concrete waste form and the contain-
ers had maintained their integrity during 22 years of disposal.
This would impede leaching. Other waste forms recovered from ocean
disposal sites had significant quantities of radionuclides present
after periods of up to 16 years. A diffusion coefficient of 10"7
cm2/sec is a typical, high release rate observed for concrete waste
forms in seawater. With this rate, assuming diffusion as the
release mechanism, a homogeneous waste form having the dimensions of
the recovered package would still retain 34 percent of its activity
after 22 years. If it can be assumed that the activity were origi-
nally contained in the vicinity of the cardboard box, then signifi-
cantly more than 34 percent of the activity would be retained.
3.6 WASTE FORM INTEGRITY
The durability of the waste form in the ocean environment was determined by
measuring the compressive strength of concrete cores taken along its longitu-
dinal axes, as described in 3.3 of this report. The tests were conducted
using ASTM Standard C 39-72, "Method of Test for Compressive Strength of
Cylindrical Concrete Specimens." Table 3.1 lists the average compressive
strengths of the tested cores.
-30-
-------
Although the original concrete formulation is not known, dissolved core
samples were analyzed to determine the mix proportions used to produced the
concrete. The results indicated that the mix consisted of 49.2 wt. percent
aggregate (crushed rocks), 30.5 wt. percent sand and 13.3 wt. percent portland
cement. From this, it was estimated that 6.8 wt. percent was water, having a
water to cement ratio of 0.5.
It is difficult to predict the effects of the ocean environment on compressive
strength since the initial values are not known. However, the waste form did
not exhibit attack due to the chemical action of the dissolved salts in
seawater, to mechanical attribution, to the corrosion of the metal container,
or to bacterial actions. This is surprising since sulfates in seawater attack
cements very markedly. The conversion of calcium hydroxide (CA(OH)2) to
gypsum (CaSOA»2H20) through sodium or calcium sulfate attack more than doubles
the solid volume, resulting in the expansion and deterioration of the cement.
Magnesium sulfate, however, forms a hard dense skin on concrete surfaces and
tends to hinder penetration of sulfate solutions by depositing magnesium
hydroxide in the cement pores.
Although these compounds ((CaOH2), CaS04»2H20, and Mg(OH)2) were detected in
varying concentrations both on the surface of the concrete and within the
concrete form, it is difficult to predict the reaction mechanisms or the rates
of formation under the conditions to which the waste form was exposed to over
the 21-25 year period.
-31-
-------
TABLE 3.1
COMPRESSIVE STRENGTH OF CONCRETE CORES
Longitudinal
Axis
0°
90°
180°
270°
Distance from top
of Waste Form. in.
(5.25, 16.0, 26.5)
(5.25, 16.0, 26.5)
(16.0, 26.5)
(26.5)
Core Size,
in. (d x H )
1.73 x 3.85
1.73 x 3.96
1.73 x 3.92
1.73 x 3.87
Compression
Strength
psi (average)
2,780
2,850
2,790
2,760
32
-------
4. CORROSION ANALYSIS OF THE METAL CONTAINER
From the previous corrosion analysis of the waste package retrieved from the
2800-meter disposal site in the Atlantic Ocean [6], a well-defined procedure
has been developed. This procedure includes, when possible, those tasks
outlined in the National Association of Corrosion Engineers (NACE) Standard
RP-01-73 describing recommended practices for collection and identification of
corrosion products [9],
The package was kept in an inert argon atmosphere from the recovery to
commencement of analysis. However, once analysis commenced, the trepanned
samples and surface scrapings were kept in a desiccator prior to physical
analysis.
The task sequence for corrosion analysis is as follows:
Task I: Visual Inspection. Visual inspection of the overall container,
sampling of scale, pH, and sediment provides an early qualitative record
of the container and samples for further analysis. Also, coordinates to
which samples are indexed are defined.
Task II: Dimensional Analysis. The inspection resulting from Task I
allows selection of areas on the sea side and sediment side from which
strips of material are cut. Sections from these strips are mounted in
epoxy for cross-section dimensional analysis. In addition, samples are
trepanned from sites of specific attack or protection. Estimates of
corrosion rates as a function of position result from this analysis.
Task III: Micro-Analysis. The analyses made in Task I and II pinpoint
specific sites of either high protection or high corrosion rates for
further microscopic scrutiny. Optical and scanning electron microscopic
techniques provide metallographic evaluation of cross sections and
surface morphology at points of either high local failure or enhanced
protection.
Task IV: Chemical Analysis. In parallel with Task III, X-ray diffrac-
tion (XRD), X-ray fluorescence spectroscopy (XFS), and bulk chemical
analysis provide information on the local and bulk chemistries involved
in the corrosion protection mechanisms.
The corrosion analysis takes as its objective an assessment of the
effect of the disposal site environment upon the carbon steel sheathing
material. In addition, the results of this analysis will determine
specific instances of protection or failure thereby providing input for
future practices.
4.1 VISUAL INSPECTION AND SAMPLING
Figure 4.1 is a schematic of the coordinate system (x, r, 6) used to specify
locations on the container. The coordinate x specifies the distance parallel
to the cylindrical axis from the end containing the exposed waste form. The r
coordinate is the radial distance from the cylindrical axis; the surface of
-33-
-------
the carbon steel sheath, therefore, is at r =» 11.25 inches (28.6 cm). Looking
down the cylindrical axis from the open end, the angle, 8, is in degrees
clockwise from an initially marked 0° position. The cylindrical container,
Figures 4.2-4.5, has an average diameter of 22.5 inches (57.2 cm) and an
average length of 34.9 inches (68.6 cm).
As seen from Figure 4.2, the portion of the container buried in black mud
comprises those coordinates between ~Q° and =135° and the sea side comprises
regions between 135" and 360°. The sea-exposed surface contains black to
orange scale and corrosion products. In addition, many nodular conglomera-
tions of orange to black corrosion products cover the sea side of the con-
tainer. Figure 4.6 shows some of these regions. Where a nodule was removed
an adherent black underscale was observed, characteristic of much of the other
regions of the sea side (for example, Figure 4.5). Measurements using pH
paper in areas where a nodule was removed indicated values between 8 and 10,
corresponding to mild alkalinity. The mud also showed an alkaline pH between
8 to 10.
Black mud caked the entire portion of the mud-buried side of the container,
Figure 4.2. Large spots of silvery material lie on the surface of the mud
layer. Scrapings listed in Table 4.1 were taken using a porcelain spatula and
stored in a desiccator.
The sheath was cut and 4 inch strips were extracted from the mud side at 45.8"
and the sea side at 255.8° (Figure 4.7). Samples from these strips allowed
dimensional analysis. Before cutting the container for samples, perforation
of the carbon steel sheath at the rim was noted (Figure 4.8) at x-0 and 0-0.
An inside view of the container exhibited further perforation initiated along
the chimes and radial corrugations (Figure 4.9), where cold work probably
occurred during the forming of the drum. Figure 4.10 provides a closer view
of the specific chime attack and the red to orange scale formation on the
inside of the vessel. Perforation occurred on both the sea side and the
sediment side and consumed 1 percent of the total container surface area.
4.2 DIMENSIONAL ANALYSIS
The strips cut from the respective sea and sediment sides of the carbon steel
sheath provided specimens for dimensional analysis and metallographic examina-
tion. Specimens , 1 inch x 0.5 inch (2.54 cm x 1.27 cm) were extracted at 3-
inch (7.6 cm) intervals along each strip. The samples were mounted in epoxy
and ground past the saw cut damage with 320 paper. Photographs of these
cross-sections (50 X magnification) provide a view of 0.1 inch (0.25 cm)
lengths of the cross section at each 3-inch (7.6 cm) interval. The thickness
from the photographed cross sections was measured at each 0.01 inch (0.025 cm)
interval and averaged for a dimension determination for each 3-inch interval
of x. The standard deviation of the determination depends upon the localized
nature of the corrosion.
•34-
-------
Sea Side
270
90*
Sediment Side
Figure 4.1 Schematic of Coordinate System Used to Identify Locations
on the Container
35
-------
Figure 4.2 View of Waste Package Showing the Sediment Side (0° - 90°)
36
-------
Figure 4.3 View of Waste Package Showing Sediment Sea Side Interface
(90° - 180°)
37
-------
Figure 4.4 View of Waste Package Showing the Sea Side (180° - 270°)
38
-------
:
C
Figure 4.5 View of Waste Package Showing the Sea Side (270° - 360°)
39
-------
Figure 4.6 Nodule of Corrosion Product on Sea Side
40
-------
Figure 4.7 Strips Extracted for Metal Loss Analysis
-------
-------
SEDIMENT
SEDIMENT
SEA
Figure 4.9 View of the Sheath from the Concrete Side
43
-------
Figure 4.10 Closeup of a Chime Perforation
-------
TABLE 4.1
SCRAPINGS
Scraping #
II-l
II-2
II-3
II-4
II-5
II-6
II-7
II-8
II-9
Position
X (cm.)
0 to .88
0
0
23
32
86
41
16
16
8 (degrees)
0 - 270
0
0 - 360
163
180
0-270
25
180
180
Description
"Silver" scale
Metal from rim
Chips
"Alkaline" nodule
Rib nodule
Scale and nodule
Mud
Metal near perforation
Crystalline material
from perforation
45
-------
Figure 4.11 shows the results plotted as metal thickness vs. x, the dimension
down the container axis. As can be seen, the sediment apparently inhibits
the general corrosion of the container material. The sediment side shows
virtually no metal loss while the sea side shows approximately 0.045 cm
average metal loss. A precise calculation of the metal loss is made difficult
by insufficient knowledge of the initial thickness of the container. A fold
in the rim which was presumably not attacked had a metal thickness of 0.12 cm,
which was slightly less than the average thickness of the sediment side of the
container. This part of the rim, however, may have been reduced somewhat by
the forming process. Assuming the sediment side to have not had a significant
metal loss, its average dimension may serve as the initial dimension. This
average "initial dimension" is 0.130 ± .005 cm. The average dimension loss
for the sea side is, therefore, 0.045 ± 0.01 cm.
Table 4.2 lists calculated corrosion rates based on the assumptions of (1)
constant rate with (2) no induction time [10,11,12). For this container, a
rate of 0.00075 in/yr (0.019 mm/yr) for the sea side falls under those rates
observed for a previously analyzed container retrieved from the 2800-meter
Atlantic site. Indeed, the uniform corrosion rate of 0.00075 in/yr relates
closely to the zero oxygen limit estimated from the empirical relationship of
Reinhart (12). An oxygen minimum of 0.5 cc/1 exists at 900 meters off the
Pacific coast near where the package was recovered. This concentration is an
order of magnitude less than typical dissolved oxygen concentrations found
near the Atlantic 2800 meter disposal site where a waste package was recovered
in 1976. The corrosion rate may result from dissolved oxygen levels at this
disposal site, but the nearly negligible corrosion rate experienced by the
sediment side of the container suggests that perhaps the alkaline sediment
also plays a role in suppressing general corrosion. While these measurements
provide information on the general mode of corrosive attack and indicate some
specific forms of attack, microscopic evaluation provides more information on
mechanisms of protection or local failure.
4.3 PROTECTED REGION ON THE SEDIMENT SIDE
Figure 4.12 shows micrographs of a well-protected surface at coordinates along
the 45° axis of the container sheath. The microstructure of the metal shows a
grain size of between 10-25pm. A compact scale typically 40-50pm thick with
pits in the metal containing up to 250pm of scale characterizes the attack on
the well-protected sediment side of the container. While the macroscopic
corrosion rate measurement could not precisely detect this low rate, the scale
thickness can give an estimate of the corrosion rate under the assumptions
that the scale represents the entire corrosion product and its density is half
of that of the metal. A laminar material on the left of the photo at AB of
Figure 4.12 (a) suggests an initial surface scale consistent with the assump-
tion that the entire product formation has remained as scale. Using this
argument, an estimated corrosion rate for the sediment buried side falls near
0.0025 mm/yr, assuming a linear scale growth with no induction time. This
probably represents an upper limit for the general corrosion on the sediment
side since a non-linear parabolic law would seem more realistic because the
film appears to be protective. However, at this time it is premature to
postulate a corrosion mechanism and corrosion rate. Of course, as described
in previous sections, high localized corrosion completely penetrated the
-46-
-------
sediment side as well as the sea side showing no apparent preference. This
mode may in fact dominate at longer times thereby reducing the effectiveness
of the container as a barrier to fission product leaching. Assuming the scale
growth mode to dominate, a rather bold calculation based on a linear growth of
this scale yields a time of 260 years for 50 percent loss in thickness by
general corrosion, of a 1.3mm sheath buried in the sediment.
47
-------
03
O.I
e
«* °-°5
CO
LU
2
*:
o
X
>- c
_l
<
u O.I
S
0.05
1 1 5
SEA
~ 5 a o a SIDE -
^ i . . s ? .-_
)Q2oxo2 _
"• u x
SEDIMENT
SIDE
45°
*••• M_w
1 I 1
10 20 30
X, inches
Figure 4.11 Metal Thickness as a Function of Position
-------
TABLE 4.2
CALCULATED CORROSION RATES
Clear Surface Waters Off
Coast of Japan
Projected from Five Year
Tests in Surface Waters
Empirical Formula
Zero 02 Limit
General Attack Sample I
Sea Side
Sediment Side
Local Attack
This Container, Sample II
Sea Side
Sediment Side*
Local Attack
Corrosion Rate
in/yr
0.002
0.0023
0.00084
0.0013±.0002
0.00191.0002
>0.0026
0.000751.00015
0
>0.0021
mm/yr
0.051
0.058
0.021
0.033
0.048
>0.066
0.019
0
>0 . 054
Ref.
(10)
(11)
(12)
(6)
(6)
(6)
*Microscopic Examination suggests a shallow pitting and scale formation
resulting in a calculated 0.0025 mm/yr average.
49
-------
Figure 4.12 (b) shows a micrograph of the most severe pitting, other than the
previously mentioned chime perforation, within the region of the container
protected by the sediment. There is a scale growth of about 250pm on the side
of the container facing the seawater environment. Scanning Electron Microsco-
py (SEM) analysis of this particular sample, and the use of X-ray fluores-
cence showed a significant proportion of silicon in the large scale growth.
On the side of this sample facing the concrete, Figure 4.12 (c), a 20pm scale
growth between the metal and dense "initial" »5pm scale growth can be seen.
Figure 4.13 gives some insight into the initiation of this secondary scale
growth below the initial 5pm scale. To the left, the initial scale can be
seen breaking away from the metal. It appears that stresses on the protective
initial scale can accelerate high local corrosion.
The above discussion outlines the behavior of the region on the sediment side,
which appears relatively protected. Scale formation between 20 and 50pm with
some 250pm pits characterize this region and correspond to an average 0.0025
mm/yr corrosion rate. The protection probably relates to the alkaline
sediment and lower oxygen activity. The overall metal loss is much less than
that seen on the sea side. However, the sediment side showed the same
susceptibility to high rates of localized corrosion even though the more rapid
general rates appeared on the sea side of the container.
4.4 CORRODED REGION ON THE SEA SIDE
A sample taken at coordinates (x-27", 5-225°) represents the "worst case"
general corrosion (here general corrosion is distinguished from the high local
attack experienced by the rim and chimes). Figure 4.14 shows the scale formed
on the sea side of this sample. In contrast to that of the more protected
region, Figure 4.15 shows that this surface exhibits a very porous structure
with many cracks in the scale extending entirely through the metal surface.
As shown in Figure 4.16, the concrete side of this sample shows a more compact
scale formation. Significantly a thick 20pra to 50pm oxide (EF) forms under
the initially existent lamina, oxide (CD) and "coating" (BC). The identifica-
tion of these layers are based on X-ray fluorescence (XRF) evidence provided
in Figure 4.17. Here the scales (CD and EF) 'are shown to contain iron and are
assumed to be oxides of iron. Area DE represents a void resulting from
delamination. Area AB contains aluminum and is, therefore, part of the
concrete waste form, while region BC shows neither iron nor aluminum. Section
BC probably represents the remnants of a coating, although no evidence of this
"coating" has been found on the remainder of the vessel even though repeated
searches were made with the aid of the Brookhaven National Laboratory metal-
lography laboratory. The 5pm thick scale, CD, remains on much of the well-
protected portions of the container.
-50-
-------
a.
Typical Outer Scale
Worst Case Outer Scale
c.
H
Concrete Side Scale
Figure 4.12 Optical Micrograph of Well Protected Specimens Taken from
Positions (x = 9 inches, e = 45), (x = 15 inches, 6 = 45)
respectively
51
-------
to
Figure 4.13 Initiation of Corrosion on the Sediment Side of the Container
(x-12 inches, 6 = 46°)
-------
00
Figure 4.14 Rapid General Corrosion on Sample Taken from X-27 inches, 6 - 225'
-------
Ln
k
I
.
Figure 4.15 Metal to Scale Interface of Sample Taken From x=27 inches, 9 - 225'
-------
A
CONCRETE
METAL
B C
Figure 4.16 Corrosion Scale of Sample (Concrete Side) Taken From x-27 inches, 9 - 225'
-------
'
Scanning electron
micrograph
(SEM)
b. x-ray fluorescence
map of iron
Figure 4.17
c.
x-ray fluorescence
map of aluminum
Solid Phase Analysis of Concrete Side Lamina of Sample Taken
from x=27 inches, 9 •• 225°
56
-------
The inner scale represents a metal corrosion rate of about 0.001 mm/yr, for
the concrete side of the sample. This is an order of magnitude less than the
measured average of the total corrosion rate on the exposed sea side (.019
mm/yr). However, any detrimental effects of corrosion on the concrete side of
the container should be considered since it impacts on the longevity of the
container. For this reason, the following calculations have been made to
assess the significance of the growth of the oxide film on the mechanical
integrity of the package.
Referring to Figure A. 18, an oxide scale of thickness 5x will produce a
fractional radial strain 5r/R, expressed as:
e- - (i- P oxide) 5x
R p metal R
where: p oxide - oxide density R - radius of the container
p metal - metal density £r - change in radius caused by corrosion
Using a density ratio of 0.5, € - 8.7xlO"5 for 100/im oxide growth. This is
equal to the circumferential strain on the metal. An assumed elastic modulus
of 30xl06 psi [13] for the material results in a calculated stress on the
metal of 2.6xl03 psi corresponding to an excess pressure of 24 psi on the
concrete. These values correspond to nonsusceptible regions for either stress
corrosion cracking (SCC) of the carbon steel [14,15) or breakdown in the
concrete [6] and hence are not significant.
Table 4.3 shows an analysis of the trace elements present in the carbon steel
sheath of the waste package. The composition or structure of the material
does not differ significantly from the previous container [6]. Differences in
apparent corrosion rates between the two containers must, therefore, be due to
the respective environments.
The scrapings primarily contain iron compounds. Traces of other heavy metals,
as shown in Table 4.4, have also been found and probably originate from the
environment.
Table 4.5 shows that the scrapings are comprised primarily of the a and 7
hydrated Fe203 species. The most prevalent form, a - hydrated ferric oxide
(7-FeOOH) results from the aging of the 7 form in the neutral to alkaline
water [16]. The predominance of the a form of the hydrated iron oxide prod-
uct suggests a slowing of the corrosion process from the initial rates which
would result in the predominance of the 7-FeOOH product.
In general, little knowledge exists about precise mechanisms governing
corrosion in deepsea environments for long time periods. Typically, metal
loss will be a nonlinear function of time, temperature, and activity of
certain reactive, catalytic or inhibiting elements in the environment and
material. Earlier work has shown metal corrosion of well characterized
samples to be linear in time, temperature and oxygen concentration for low
-57-
-------
temperatures and time <5 years [6]. This must be considered to be a first ap-
proximation. Longer term analysis requires a more detailed description of the
mechanisms contributing to the corrosion of the sheath in order to obtain the
higher order time and temperature terms for the rate equation required for the
time periods >5 years.
Furthermore, use of accelerated tests to specify mechanisms or extrapolate
rates can produce misleading results for metal-concrete systems [17]. With
these considerations in mind, the "corrosion rates" specified throughout this
report are calculated rates based on assumptions of constant rate with no
induction time. These are questionable assumptions, but their use for
estimating rates are justified on the grounds that the estimated rates provide
a means of comparison. For example, the general rate of corrosion on the sea
side of this container exceeds that of the previously examined sample taken
from the Atlantic Ocean 2800 meter site [6].
58
-------
Figure 4.18 Schematic of Container with Scale Growth
-------
TABLE 4.3
TRACE ELEMENT ANALYSES OF CONTAINER MATERIAL
ELEMENT
C
Mn
P
S
Si
Ni
Cr
Mo
Cu
WEIGHT PERCENT
0.10
0.36
0.007
0.030
ND <0.
ND <0.
ND <0.
ND <0.
02
02
02
02
0.1
-60-
-------
TABLE 4.4
X-RAY FLUORESCENCE ANALYSIS OF TRACE COMPONENTS IN SCRAPINGS
Sample
II-l
II-4
II-5
II-7
Position
X(cm.) e(degrees)
88.5
23
31
41
0 - 270
163
180
25
Trace Elements
Cu, Ti, V, Mn ~30 ppra
Zn, Pb, ~10ppm
Ti ~30 ppm >Zn
Small Cu, Zn
Si, Ci, K, Ca ~50 ppra,
Ti, Cr, Mn 10-30 ppm
-61-
-------
TABLE 4.5
X-RAY DIFFRACTION ANALYSIS OF MAJOR IRON OXIDE COMPONENT
Sample
II-l
II-4
II-5
II-6
Position
X(cm.), G(degrees)
88.5
23
32
86
0 -270
165
180
0 - 270
Corrosion Products
a - FeOOH, 7 - FeOOH
a - FeOOH
a - FeOOH
a - FeOOH
62
-------
5. CONCLUSIONS
The following summarizes the conclusions of this study:
(a) The concrete waste form maintained a high degree of integrity during the
time (21-23 years) that it was exposed to the ocean environment. This
can be attested to by the condition of the cardboard box within the
waste form and the resistance of the concrete to hydrostatic implosion,
considering a void cavity of approximately 1,200 cubic inches.
(b) The measured compressive strength of the concrete cores is in a much
higher range than that found at the Atlantic 2800-meter disposal site
[6]. The degree of uniformity in the strength values obtained from
cores taken throughout the waste form are indicative of its durability
in the ocean environment.
(c) The sediment apparently inhibits general corrosion. The scale thick-
ness, observed for the sediment side of the container, corresponds to an
estimated 0.0025 mm/yr constant corrosion rate with no induction time or
a 260-year time for 50 percent reduction in thickness of a 1.3mm sheath.
(d) The metal loss on the sea side of the container corresponds to a
constant corrosion rate of 0.019 mm/yr or a 34-year time required for 50
percent reduction in thickness of a 1.3mm sheath.
(e) High rates of localized attack perforated regions of cold work, e.g.
rims and chimes. More perforation occurred on the sea side. This
effect could result from (1) higher elementary processes on cold work
metal surfaces, (2) higher stresses produced in protective scales or
coatings as a result of underside film growth, or (3) a combination of
these effects. Further investigation is recommended in order that the
best alemiorative design can be made for future use.
(f) The metal loss results primarily from the sea side formation of a porous
loosely adhering scale and not from concrete side corrosion.
(g) The corrosion on the concrete side typically formed 50/im to 100/m scale
with some 250 pm pits in the 24 years. The calculated mechanical effect
this has on the concrete or metal is negligible even when possible
alkaline stress crack corrosion is considered.
-63-
-------
REFERENCES
1. "A Survey of the Farallon Islands 500-Fathom Radioactive Waste Disposal
Site-Operations Report," U.S. Environmental Protection Agency, Report
No. ORP-75-1, Washington, DC (1975).
2. Joseph, A.B., "A Summary to December 1956 of the U.S. Sea Disposal
Operations," WASH 734, U.S. Atomic Energy Commission, Washington, DC
(1957).
3. Dyer, R.S., Environmental Surveys of Two Deep Sea Radioactive Waste
Disposal Sites Using Submersibles," In: Management of Radioactive
Wastes from the Nuclear Cycle. Vol. II, IAEA, Vienna, Austria (1976).
4. Interstate Electronics Corporation, "A Survey of the Farallon Islands
500-Fathom Radioactive Waste Disposal Site," IEC Report 4460C1648,
prepared for the U.S. Environmental Protection Agency Office of Radia-
tion Programs and Office of Water Program Operations by IEC, Anaheim,
CA, December 1975.
5. Dayal, R., I.W. Duedall, M. Fuhrmann, and M.G. Heaton, "Sediment and
Water Column Properties at the Farallon Islands Radioactive Waste
Dumpsite," Draft Report, April 1979, Submitted to the Office of Radia-
tion Programs, U.S. Environmental Protection Agency, Washington, DC.
6. Colombo, P., R. Neilson and M. Kendig, "Analysis and Evaluation of a
Radioactive Waste Package Retrieved from the Atlantic 2800 Meter
Disposal Site," EPA 520/1-82-009, BNL 51102, Office of Radiation
Programs, U.S. Environmental Protection Agency, Washington, DC, May
1982.
7. Uhlig, H.H., Corrosion and Corrosion Control, second edition, John
Wiley, P. 92, (1971).
8. Miller, J.D.A., Microbial Aspects of Metallurgy. Editor, American
Elsevier, (1970).
9. National Association of Corrosion Engineers, Technical Practices
Committee, "Recommended Practice, Collection and Identification of
Corrosion Products," NACE Standard RP-01-73, February 1973, Houston,
Texas.
10. Kakehi, T. and H. Yoshino, "Corrosion of Steel in Polluted Sea Waters,"
Proc. 5th Int. Cong, on Met Corrosion. 1972, NACE (1974), p. 76.
11. Tamada, A., M. Tammura and G. Tennyo, "Corrosion Behavior of Low Alloy
Steels in Sea Water," Ibid, p. 786.
64
-------
12. Reinhart, F.M. and J. R. Jenkins, "The Relationship Between the Concen-
tration of Oxygen in Sea Water and Corrosion of Metals," Proc. 3rd
Congr. on Marine Corrosion and Fouling, p. 562 et. eq. 1972.
13. American Society for Metals, Metals Handbook, p. 427, Vol. 1, 8th
Edition, Metal's Park, Ohio (1975).
14. Reinoehl, J.E. andW.E. Berry, Corrosion. 28 (4), 151 (1972).
15. Humphries, M.J. and R.N. Parkins, Corrosion Sciences. 7, 474 (1967).
16. Misawa, T., K. Hashimoto and S. Shumodiara, "The Mechanism of Formation
of Iron Oxide and Oxyhydroxides in Aqueous Solutions at Room Tempera-
ture," Corrosion Science. 14, 131 (1974).
17. Tuutti, K., "Cracks and Corrosion. The Corrosion of steel in Concrete--
the Effects of Cracks in the Concrete Cover." CBI Forsk. 1978 (6),
referenced in Chem. Abstracts.
65
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
. REPORT NO.
EPA 520/1-90-014
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Analysis and Evaluation of a Radioactive Waste Package
Retrieved from the Farallon Islands 900-meter Disposal
5. REPORT DATE
September 1990
Sit€ 6- PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
P. Colombo and M.W. Kendig, Brookhaven National Laboratory
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Brookhaven National Laboratory
Nuclear Waste Research Group
Department of Nuclear Energy
Ilnt-on. M«avr Vor-V 1107"3
10. PROGRAM ELEMENT NO.
1 1. CONTRACT/GRANT NO.
Interagency Agreement No.
EPA-IAG-D6-0166
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Radiation Programs
401 M Street, SW
Washington, DC 20460
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
ANR-461
15. SUPPLEMENTARY NOTES
16. ABSTRACT
In October of 1977, a 55-gallon low-level radioactive waste (LLW) package was
retrieved from the Farallon Islands 900-meter disposal site located 40 miles west of San
Francisco, California, at coordinates 37 38'N and 123°08'W. This was the second recovery
of a LLW package from an ocean disposal site and was conducted by the EPA Office of
Radiation Programs. The waste package was transported 'to Brookhaven National Laboratory
where container corrosion and matrix analysis studies were performed. This report presents
the final results of the laboratory analyses and contains detailed photographic
documentation of the recovery operation and condition of the waste package.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS 0. COSATI Field.GfOUp
1. radioactive waste disposal
2. radioactive waste packaging
3. marine corrosion
4. ocean dumping/sea disposal
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS iTIns Rtportl
Unclassified
21 NO. OF PAGES
78
20. SECURITY CLASS (TMlp
-------
oEPA
United States
Environmental Protection
Agency
(ANR-459)
Washington, DC 20460
Official Business
Penally for Private Use
$300
------- |