United States Office of Radiation Programs EPA/520/1-89-012
Environmental Protection (ANR-459) September!!
Agency
&EPA Characterization of
Contaminated Soil from
the Montclair/Glen Ridge,
New Jersey, Superfund
Sites
-------
Characterization of Contaminated Soil
from the Montclair/Glen Ridge, New Jersey
Superfund Sites
James Neiheisel
Office of Radiation Programs
Environmental Protection Agency
-------
FOREWORD
The Superfund Amendments and Reauthorization Act (SARA)
of 1986 and the revision of the National Contingency Plan (NCP)
placed new requirements on hazardous waste site cleanup actions.
Remedial action at Superfund sites must protect human health and
the environment and meet applicable or relevant and appropriate
requirements (ARARS) as established by Federal and State
standards. The selection of remedies must also be
cost-effective and use permanent solutions and treatment
technologies or resource recovery technologies to the maximum
extent practicable. Treatment methods which permanently and
significantly reduce the mobility, toxicity, or volume of
hazardous substance are preferred in this Superfund requirement.
Tne Office of Radiation Programs (ORP), in concert with
Superfund requirements, is evaluating the use of physical volume
reduction and chemical extraction (VORCE) to remediate radium
contaminated soils at the Montclair and Glen Ridge, New Jersey,
Superfund sites. The VORCE investigation at these New Jersey
sites consists of 1) soil characterization, 2) treatment
studies, and 3) technology implementation. The characterization
of the soil includes particle size distribution, mineralogical
identification, chemical measurements, and radioassay. The soil
characterization phase provides important data that paves the
way for the treatment and implementation paases that follow.
Knowledge of radioactive contaminant distribution as a function
of particle size and soil mineralogy is especially fundamental
in assessing remedial measures.
This report documents the soil characterization phase
at the EPA Eastern Environmental Radiation Facility
in Montgomery, Alabama. The identification of the
spatial distribution of the radium within the soil and its
dissociation with specific minerals or materials is shown to
relate to the later implementation phase. Procedures developed
in cms investigation have application to other radioactively
contaminated Superfund sites.
The Agency invites all readers of this report to send
any comments or suggestions to Mr. Martin P. Halper, Director,
Analysis and Support Division, Office of Radiation Programs
(ANR-461), U.S. Environmental Protection Agency, Washington, DC
^ U rr O 0 •
Ric
Office
Programs
111
-------
ACKNOWLEDGEMENTS
Tne author wishes to thank Mr. Robert S. Dyer and
Mr. Gary 3. Snodgrass o£ this Office for their critical
review and support in tha multiple phases of the soil
characterization study of the Volume Reduction and Chemical
Extraction (VORCE) investigation of radium contaminated
soil at the Montclair and Glen Ridge Superfund sites. I
also wish to thank Dr. William Richardson and Mr. Larry Coe
of Sanford Cohen and Associates, Incorporated, for their
comments and review of this report. Appreciation is
extended to Dr. Charles Porter and Mr/Charles Phillips of
the Eastern Environmental Radiation Facility, Montgomery,
Alaoama, for their indulgence and assistance in my use of
the EERF facility to perform mineral analysis and physical
testing of the contaminated soil.
I also wish to acknowledge the assistance of Dr. Fred Au
of the ORP Las Vegas Facility in supervising contractual
services from the University of Nevada and Huffman
Laboratories. I also wish to thank Mr. Ray Willingham of
the U.S. Army Corps of Engineers, Marietta, Georgia for
sedimentation and centrifuge analysis of select samples
which enabled completion of a grain size curve and the
preparation of silt/clay size fractions for further mineral
and radioassay analysis.
I am especially indebted to Mr. R. Jeff seme and
Dr. Robert Erikson of the Battelle Northwest Laboratories
for tneir study of the silt and'clay-size fractions with
tne scanning electron microscope and energy dispersive
spectrometer analysis technique and the x-ray diffraction
and radioassay of fractions of silt and clay separated by
tne linear density gradient method. I wish also to thank
Mss. Tonya Hudson and Deobie Whittaker of Sanford Cohen and
Associates, Incorporated, for their assistance in heavy
liquid separations at tha Eastern Environmental Radiation
Facilities Laboratory. Tnanks are also extended to Mss.
Virginia Stradford, Bonnie Wyvill, Eleanor Paige, and
Pnoeoa Suoer for typing this document.
v
-------
Table of Contents
Page
Foreword iii
Acknowledgements v
List of Figures ix
List of Tables x
Abstract xi
1. Introduction 1
2. Site Investigation and Geology 1
3. Characterization Procedure 3
3.1 Physcial Methods 4
3.1.1 Screening and Sieving 4
3.1.2 Sedimentation and Centrifuge
Separations 7
3.1.3 Density Separations 7
3.2 Mineral Identification Methods 8
3.2.1 Radioassay and Chemical Methods .. 8
3.2.2 Petrographic Polarizing Microscope
and Microscopic Methods .... 9
3.2.3 X-ray Diffraction 10
3.2.4 Scanning Electron Microscope/
Energy Dispersive Analyzer ... 10
3.2.5 Magnetic Separation 11
4. Radium 226 Activity Levels on
Soil Fractions 11
5. Mineral Composition of Contaminated Soil ... 13
5.1 Gravel-Size Material 13
5.1.1 Furnace-Fired Materials 15
5.1.1.1 Anthracite Coal 16
5.1.1.2 Coaly Slag 19
5.1.1.3 White Slag 20
5.1.1.4 Ferruginous Slag .... 20
5.1.2 Rock Particles 21
5.1.3 Glass and Trash 21
VII
-------
Table of Contents (Continued)
Pagjs
5.2 Sand-Size Materials 22
5.2.1 Quartz and Quartz Films 22
5.2.2 Feldspar 23
5.2.3 Heavy Minerals 23
5.2.3.1 Radioactive Minerals . . 26
5.2.3.2 Radiobarite 29
5.2.3.3 Natural Background
Minerals 30
5.2.3.4 Other Heavy Minerals. . . 30
5.3 Silt and Clay Size Material 31
8. Assessment of Percentage of Radium 226
Distribution on Soil Materials 36
6.1 Radium 226 in Secular Equilibrium
with Uranium Minerals 35
6.2 Radium 226 in Natural Background
Materials 37
6.3 Radium 226 Associated with Radiobarite. . 39
6.4 Radium 226 Associated with Amorphous
Silica 40
6.5 Adsorbed Radium 226 Associated with
Soil 40
7. Discussion 41
8. Summary 44
REFERENCES 46
VI11
-------
List of Figures
Page
1. Laboratory methods for characterization of
radium-contaminated soils 5
2. Weight percent of Montclair and Glen Ridge, NJ,
contaminated soil retained by selected sieves . . 6
3. Radium 226 activity on washed gravel-size
particles 17
4. Composition of gravel and sand-size materials in
Montclair and Glen Ridge contaminated soils ... 18
5. Comparison of heavy mineral suite in -50/+270
sieve size White Sands, Glen Ridge, and Montclair
materials 25
5. X-ray diffractograms of Glen Ridge 10 to 20
micron-size light and heavy density fractions . . 32
7. SEM photomicrograph and EDX spectrum for
carnotite/tyuyamunite in heavy mineral fraction
of the Glen Ridge soil 35
8. Relationship of particle size and mineral
composition to percent radium distribution in
Glen Ridge soil A3
IX
-------
List of Tables
1. Radium 226 activity and percentages
on gravel, sand, and silt/clay sizes 12
2. Mineral and material composition of
Montclair/Glen Ridge soil 14
3. Chemcial composition of Montclair/
Glen Ridge soils 15
4. Radiochemical analysis of Montclair/
Glen Ridge gravel particles 24
5. Radiochemical analysis of Montclair/
Glen Ridge sand-size material between
-50 and +270 sieve size 27
6. Mineral composition and radium 226
activity of silt-size (-45 micron, +2 micron)
soil from Glen Ridge site 34
7. Distribution of radium 226 on specific
materials of Montclair and Glen Ridge soil .... 38
-------
ABSTRACT
Volume reduction and treatment techniques are being
investigated for possible use in remediation of radium-
contaminated soil at the Montclair and Glen Ridge, New/ Jersey,
Superfund sites. The contamination occurs in residential areas
situated over former landfills that received various industrial
wastes including radium mill tailing wastes in the 1920's. One
of the first steps in the investigation was characterization of
the soil to determine the most efficient remediation methods.
The average radium activity of the contaminated soil is
estimated to be in the range of 60-200 pCi/g with higher
activity (1000-10,000 pCi/g} occurring in the fine materials.
The characterization procedure included detailed physical
sizing, special chemical tests, and radiochemical and
mineralogical analysis of 18 size fractions ranging from gravel
to clay size on each sample. Special magnetic, heavy liquid,
and linear density centrifugation separations were also made on
soil fractions to facilitate identification of minerals and
materials v/ith high radium activity.
About half of the radium activity is attributed to acid
leach radium milling products (radiobarite and amorphous
silica). The remainder of the radium is contained in uranium
ore minerals (carnotite and uraninite), incinerated radium
materials (ferruginous slag), and adsorbtion of radium from
solution onto native soil particles (illite, hematite, etc.).
An occasional radium vial is also a local anomaly at the sites.
Less than one third of the radium occurs in the soil
material greater than 50 sieve (0.15mm) size. This material is
comprised essentially of incineration products (coal, coaly
slag, white siliceous slag, and ferruginous slag) and native
rock particles (sandstone, siltstone, quartzite, etc.). Glass
and trash also occur in minor amounts. Approximately 10-15
percent of the radium activity in this fraction is associated
with magnetic ferruginous slag particles. The ferruginous slag
activity consists of uraninite from incinerated coal ash,
incinerated material containing radium, and adsorbed radium.
The magnetic properties and coarse particle size of tnese
materials lend themselves to physical separation procedures.
XL
-------
The contaminated soil less than 50 sieve size at both the
Glen Ridge and Montclair sites contains more than two thirds of
the radium activity which is predominately in the silt and clay
fractions comprising about one-fifth of the sample volume. The
radiobarite and amorphous silica contain the major portion of the
radium activity. Uranium ore minerals occur mainly in the sand
and upper silt size and co/nprise about 10 percent of the radium
activity. Background minerals (zircon, monazite, etc.) and
adsorbed radium on illite and other clay minerals comprise
another 5-15 percent of the radium activity.
XIL
-------
1. INTRODUCTION
The passage of the 1985 Superfund Amendments and
Reauthorization Act (SARA) and the revision of the National
Contingency Plan (NCP) placed new requirements on hazardous waste
site cleanup, emphasizing treatment over traditional remediation
approaches. In concert with these requirements for the cleanup
of sites on the National Priority List (NPL), the Environmental
Protection Agency's (EPA) Office of Radiation Programs (ORP) is
evaluating the use of physical volume reduction and chemical
extraction to remediate radium contaminated soils at the
Montclair and Glen Ridge, New Jersey, Superfund sites. These
residential sites are located over former landfill areas that
received radium contaminated tailings and other radium industrial
wastes in the 1320's.
The Volume Reduction/Chemical Extraction (VORCE) project
initiated as the remediation study of the Montclair and Glen
Ridge Superfund sites consists of 1) soil characterization,
2) treatment studies, and 3) technology implementation. The
characterization of the soil includes physical separations,
mineralogical identification, chemical measurements, and
radioassay. The soil characterization phase provides important
data that paves the way for the treatment and implementation
ptiases that follow. Knowledge of radioactive contaminant
distribution as a function of particle size and soil mineralogy
is especially fundamental in assessing remedial measures.
This report documents the soil characterization phase
conducted at the EPA Eastern Environmental Radiation Facility
(EERF) in Montgomery, Alabama; the Battelle Pacific Northwest
Laboratory (PNL), Richland, Washington; the EPA Las Vegas
Facility (LVF); the University of Nevada at Las Vegas, Nevada;
and the Army Corps of Engineers (COE), Marietta, Georgia.
2. SITS INVESTIGATIONS AND GEOLOGY
Residential areas of Montclair, Glen Ridge, and West Orange,
New Jersey, covering approximately 50, 45, and 9 acres,
respectively, contain an estimated 300,000 cubic yards of soil
contaminated with high levels of gamma radiation and radon gas.
Radioactive contamination is associated with the natural soil
to depths up to 20 feet. The contaminated areas are former
landfills that received a variety of waste materials. The waste
material was randomly dumped and subsequently covered with a
layer of top soil up to 3-feet thick to provide land suitaole
-------
for housing construction. In 1979, an aerial gamma survey by
the New Jersey Department of Environmental Protection identified
the sites of elevated gamma radiation.
Studies suggested that the radium was discarded tailing
waste from extraction of radium from uranium ore (carnotite from
Colorado Plateau) by the acid leach process from mills located
in New Jersey. Field borings by Camp Dresser and McKee, Inc.
(CDM 385) found "hot spots" of these White Sands tailings as
thin layers in the reddish brown to grey colored admixture of
coal, ash, soil, construction material, and industrial wastes
found in various proportions in the former landfill areas that
constituted the contaminated soil sites. Olsson in a cursory
examination of the White Sands at West Orange, NJ, site, cited
tha presence of yellow carnotite and barite (coprecipitate in
acid leach process) in these sands (OL 86). The production of
ndium in New Jersey terminated in 1926 and after 1922, Belgian
Congo (Zaire) uranium ore (uraninite) replaced the Colorado
Plateau uranium ore (carnotite) as the world's primary source of
radium (La 84). While these early investigations did not
identify uraninite at these sites, this mineral is of common
occurrence in the landfill; however, the source of the uraninite
may not necessarily oe related to a foreign source, as will be
discussed in Section 5.2.3.1.
The only information from earlier investigations that
related radioactivity to particle size was on the thin bands
of tailing sands designated as hot spots. This information
provided focus on sand-sized and'finer materials as the most
troublesome source of radioactivity. If the radium milling
waste and raw ore constitute the major source of contamination,
it is not surprising that the sand size or smaller material
played the major role since the ore was crashed and passed
through number 16 sieve (1.18mm) prior to the acid leach
treatment (d'A 21). The lesser radioactivity of larger
materials is therefore related to accumulation of radium
residues in incinerated slags or from materials that adsorbed
radium.
In a reconnaissance visit to the Montclair/Glen Ridge, NJ,
sites, observations were made of the glacial deposits hosting
the contaminated landfill areas. Laboratory analysis of the
glacial material outside of the contaminated area in nearby
Nishuane Park in Montclair, NJ, revealed that the fines (fine
sand, silt, and clay) consist of approximately 70 percent
quartz, 15 percent feldspar, 5 percent mafic mineral
(hornblende, etc.), 5 percent mica/illite, 5 percent chlorite,
-------
and trace amounts of kaolinite and montmorillonite (Ne 87). The
cock particles in the glacial material arepredominantly smooth,
rounded quartzite and subangular, brownish-red sandstone and
siltstone; minor other varieties are graniter basalt, and
varieties of metamorphic rock.
The former landfills are contained in surficial host beds
of unconsolidated Pleistocene glacial deposits that range in
thickness from 28 to 84 feet above bedrock. The glacial
deposits at Montclair are poorly sorted glacial till (boulder
clay) while well sorted stratified drift of more uniform texture
occurs at the Glen Ridge site.
The bedrock underlying the glacial deposits consists of
alternating strata of sedimentary rock units of the Brunswick
formation of Triassic age. The sedimentary rock units are red
beds of sandstone and siltstone and local conglomerate that dip
gently to the northwest. Rounded to subangular gravel-size
particles of these sedimentary units are well represented in
the coarser materials of the glacial deposits that overlay the
bedrock.
Tha average activity of the Montclair and Glen Ridge
contaminated soils is estimated at 64.3 pCi/g (NJEPD 87). These
soils are located in the unsaturated zone above the water table
at all the sites. The radium 226 activity in the groundwater
beneath the contaminated soil is 2.3 pCi/L, at all but one
location. These are within the maximum limit of 5.0 pCi/L
dictated by the Safe Drinking Water (SOW) standard, however, the
11 pCi/L of radium 226 at one Montcliar site does exceed the SOW
standard.
3. CHARACTERIZATION PROCEDURE
The majority of the soil characterization testing phase was
conducted at the EPA's Eastern Environmental Radiation Facility
(EERF) Laboratory in Montgomery, Alabama, using representative
samples acquired and processed from the Montclair and Glen
Ridge sites. A small quantity of high radium activity tailings
sands from a location in the Montclair site was also tested.
This sample is referred to in this report as "White Sands."
The representative samples were oven dried at 60°C prior to
size separations into fractions by sieving, sedimentation,
-------
centrifugation, and heavy liquid methods. The upper limit of
60°c in the oven drying stage was fixed so as to maintain the
clay mineral structure which might otherwise be lost at higher
temperatures/ e.g., montmorillonite loses some interlayer water
at higher temperatures.
The testing methods used to characterize the contaminated
soil and the size fractions used in this initial phase of the
VORCE project are listed in Figure 1. Initial sizing and
radioassay of all size fractions were performed at the EERF
laboratory prior to sending representative fractions to other
laboratories for special tests. All heavy mineral and magnetic
sepacations and detailed radioassay and petrographic microscope
examinations of gravel and sand-size soil fractions were
conducted at the EERF laboratory; however, special testing of
silt and clay-size fractions as tfell as ancillary testing of
bulk samples (chemical and cursory X-ray diffraction analysis)
was conducted by other laboratories (identified in the
discussion of testing methods.)
3.1 PHYSICAL METHODS
Tne physical testing included screening, sieving,
sedimentation, and centrifugation techniques as well as special
density separations employing heavy liquids. Tne purpose was
to obtain a texture grain-size distribution curve and 18
gradational soil fractions on each sample for various testing
procedures designed to provide data for the treatment studies
and technology implementation phases.
3.1.1 Screening and Sieving
Dry screening, using a Gilson Screener (Model TM-4), was
performed at the EERF laboratory on approximately 2-kg samples
of Montclair and Glen Ridge soil. The nest of screens included
1 inch, 1/2 inch, 1/4 inch, and No. 4 sieve (4.75 mm). The
weight percent of gravel on each screen was determined for the
gravel-size (+ No. 4 sieve) material (Figure 2). The minus
number 4 size soil was saved for wet sieving.
Representative 100-gram portions of minus number 4 sieve
size samples were wet sieved in a Srinkman Table Top Vibrator
Screener using sieve numbers 10 (2 mm), 15 (1.18 mm), 50 (300
microns), 100 (150 microns, 140 (106 microns) and 200 (75
microns). The cumulative weight, in relation to total sample,
was determined for each sieve size and recorded (Figure 2).
-------
Ul
Sieve No.
4
10
16
50
60
100
140
200
270
400
Size(mm)
25.00
12.50
4.75
2.00
1.18
.30
.25
.15
.106
.075
.050
.038
.015
.005
.002
.0005
-.0005
Soil Size
Gravel
Sand
Silt
Clay
Sizing Method
Gilson
Mechanical
Screener
Brlnkman
Vibrator
Screener
Sedimentation
Centrifugation
Separation Method
Bromoform and
Tetrabromoethane
Sink Float Method
(heavy mineral
concentration)
Heavy Liquid
Linear Density Method
(high activity
separation)
Analytical Methods
Gamma Spectroscopy
Alpha Spectroscopy
Magnetic Properties
Gamma Spectroscopy
Alpha Spectroscopy
Petrographic Microscopy
Chemistry
Magnetic Properties
Gamma Spectroscopy
Alpha Spectroscopy
X-Ray Diffraction
Scanning/Transmission
Electron Microscope
w/X-Ray Analyzer
Figure 1. Laboratory methods for characterization of radium contaminated soils
-------
Grave/
Sand
Silt
Clay
Sieve Size
1 1nch
1/2 Inch
1/4 Inch
No. 4
No. 10
No. 16
No. 50
No. 60
No. 100
NO. 140
No. 200
Microns
4750
2000
1180
300
250
150
106
75
53
38
15
5
2
0.5
Pan
Weight Percent Retained
Montclalr
5
8
9
4
6
3
20
6
8
7
3
6
3
6
4
1
0.5
0.5
Glen Ridge
4
9
13
9
10
3
17
4
5
3
3
4
3
4
5
2
1
1
Figure 2. Weight percent of Montclalr and Glen Ridge contaminated soil
retained by selected sieves.
-------
The minus number 200 sieve size material was collected for
several runs and representative Montclair, Glen Ridge, and White
Sands samples were collected, radioassayed and delivered to the
Army of Corps of Engineers (COE), Marietta, Georgia, laboratory
for separation into 5 silt-size and 2 clay-size fractions.
3.1.2 Sedimentation and Centrifuge Separations
The representative silt and clay-size fractions of the
Montclair, Glen Ridge, and White Sands samples were separated
into the following fractions by sedimentation and centrifuge
techniques: silt-size into -75/+50 microns, -50/+38 microns,
-38/+15 microns, -15/+5 microns, and -S/+2 microns; clay-size
into -2/+0.5 microns and -0.5 microns. The cumulative fractions
from several runs were combined to obtain sufficient sample
material for the mineral identification tests and radioassay
analyses. The weight percent was obtained on the larger silt-
sizes by weighing on a top-loading balance. The percentage
distributions of the finest fractions were obtained by the pipet
method (Fo 74). The weight distribution is listed in Figure 2.
The COE prepared texture grain-size distribution curves
from pipet analysis (silt and clay) data and screen and sieve
(gravel and sand) data provided by EERF. These grain-size
curves and their construction provide the means of proportioning
weight distribution for any fractional size consideration.
3.1.3 Density Separations
Since high density uranium minerals and radium bearing
barite (radiobarite) could comprise a relatively high percentage
of the radioactive contamination, the sand and upper silt-size
soil was separated into a heavy (greater than 2.9 specific
gravity) and light fraction by simple sink-float techniques.
The heavy liquids used for the separations at the EERF
laboratory were bromoform and tetrabromoethene.
Density separation of silt and clay-size particles is
possible by linear density gradient separations using heavy
liquids and centrifugation techniques, such as described by
Mattigod and Ervin (Ma 83). The silt and clay-size fractions
-------
sent to the PNL, Richland, Washington, laboratory rfere prepared
for linear density gradient separations using continuous
mixtures of heavy liquid (tetrabromethane) and a less dense
liquid (absolute ethanol and polyvinylpyrolidonone). The
separation of the soil minerals was then achieved by
centrifugation until three isopycnic oands ware formed
in a series in the centrifuge tuoes (Er 38).
The sink-float density separation of sand-size soil
fractions provided positive identification of mineral
constituents using a Nikon Research Polarizing Microscope at
the EERF laboratory. Linear density separations of the silt
and clay-size particles at the PNL laboratory enabled x-ray
diffraction of the separate density bands and microprobe
fluoresence of particles to facilitate mineral identification.
3.2 MINERAL IDENTIFICATION METHODS
Tne goal of mineral identification is to provide data to
assess the association of radionuclides with specific minerals
or materials comprising the contaminated soil. Radionuclides
occur in the White Sands and in the complex admixture of native
soil, industrial waste, and unprocessed uranium ore as
components of mineral structures and adsoroates on particle
surfaces. Thus, in addition to isolation and concentration by
gradation fractions and density separations, a combination of
instrumentation and chemical methods for mineral identification
and soil characterization is required (Figure 1). These are
addressed in the following paragraphs.
3.2.1 Radioassay and Chemical Methods
The EERF laboratory radioassayed each size fraction of the
Montclair, Glen Ridge, and White Sands samples as an initial
stap in determining size ranges in which the radium 226 activity
was concentrated. The tradium 226 activity levels were obtained
by gamma-ray spectroscopy using nigh purity germanium detectors
(Li 84). Radium 226 was identified and measured using the 186
MeV photopeak. Since only very small quantities of uranium 235
were found in the samples ( 0.1 percent), interference by a 185
MeV photopeak was not a significant consideration (Ri 88). The
measured radium 226 activity will be discussed in Section 4.
-------
The uranium isotopes and their daughters were estimated
on the Glen Ridge and White Sands silt/clay size linear density
gradient bands by counting gamma and X-rays on a high-resolution
intrinsic germanium diode coupled to a Nuclear Data ND 6620
Nuclear Analyzer. Details of this procedure are contained in
the PNL report by Erikson and Serne (ER 89). The significance
of tnis data will be discussed in a later section.
The EERF laboratory also performed radiochemical analysis
of uranium 238, radium 226, thorium 230, and thorium 232 on
select gravel particles and the light and heavy density
sand-size fractions of the Montclair, Glen Ridge, and White
Sands samples. The radium 226 was measured by gamma-ray
spectroscopy and the other radioisotopes by radiochemical
methods (Li 84). The thorium was separated by ion-exchange
chromatography and counted by alpha spectroscopsy using
thorium 234 as a tracer to determine the chemical yield.
Uranium was extracted from the thorium analysis mixture into
triisoctylamine (TIOA), stripped from the extract with nitric
acid, coprecipitated with lanthanum fluoride carrier, and
counted by alpha spectroscopy using uranium 232 as a tracer
to determine the chemical yield (Li 84). These results and
additional chemical analysis of vanadium, iron, and barium,
oy Galbraith Laboratories, Inc., are reported in appropriate
sections of this report.
Specific chemical tests were performed on selected
fractions of the Montclair, Glen Ridge, and White Sands
samples. These tasts, conducted by Huffman Laboratories, Inc.
were for identification of chemical constituents that might
interfere with volume reduction and cnemical extraction
processes. Of special consideration are the values of barium
and vanadium which are used as chemical signatures in the
assessment of radiobarite and carnotite. The results of these
tests will oe discussed in a later section.
3.2.2 Petrographic Polarizing Microscope and Microscopic Methods
The gravel and sand-size materials were initially
identified by visual examination and a binocular microscope.
Trie Nikon Research Polarizing Microscope, Model OPT1PHOT-POL,
was used to examine the light and heavy minerals in the size
range from -50 sieve size (300 microns) to +270 sieve size (53
microns), using area statistical counts on 300 grains of each
sand fraction. Opaque minerals were examined using reflected
light. Select size fractions were also subjected to chemical
-------
assay for those elements diagnostic to specific minerals of
interest, e.g., Ba is diagnositic to radiobarite of the radium
extraction tailings sands and vanadium is a signature of
carnotite (hydrated potassium uranium vanadate) from the
Colorado Plateau uranium ore. X-ray diffraction and magnetic
separations were also used to support identification of opaque
minerals and some of the industrial gravel slag particles.
3.2.3 X-Ray Diffraction
Initial x-ray diffraction of select gravel particles and
all size fractions less than number 4 sieve size, was performed
at the University of Nevada (Las Vegas) on a Philips (Norelco)
x-ray diffraction unit, Model 12206/53. Sedimented powder
slides were subjected to a scan between 4 degrees and 50 degrees
2 theta to obtain qualitative cursory information on major
mineral phases. X-ray diffractograms of some of the fine sand
and silt fractions contained identifiable peaks of carnotite.
In the White Sands, barite was identified as the most abundant
of the heavy minerals (greater than 2,9 specific gravity) in
the sand-size material. These initial x-ray diffraction scans,
no^ever, did little more than show the complex nature of the
contaminated soils and the need for petrographic microscopic
analysis in the sand-size material and use of linear density
gradient separations for x-ray diffraction of silt and clay-size
materials.
The soil mineralogy of the linear density separated silt
and clay fractions prepared by the PNL Richland laboratory
v/as determined using a Philips ADP-3520 x-ray diffractometer
equipped with a graphite monochrometer. Each sample was scanned
between 4 and 65 degrees 2-thata using Cu-K alpha radiation at
40 kV, 20mA. All diffractograms were obtained on soil samples
oriented on glass slides using absolute ethanol.
3.2.4 Scanning Electron Microscopy/Energy Dispersive
Energy Analyzer" ~~
Silt and clay-size soil fractions having the highest
radionuclide concentrations were further characterized by the
PNL Richland laboratory to determine the associations of the
radionuclides with various soil minerals. Samples were mounted
on brass stubs and were carbon coated. A JEOL JSM-25S III
scanning electron microscope equipped with a Tracer Northern
TN-2000 energy dispersive x-ray spectrometer (EDS) was used at
an accelerating potential of 30 kV.
10
-------
3.2.5 Magnetic Separation
The sand-size heavy mineral fractions of the Montclair and
Glen Ridge contaminated soils were separated into magnetic and
nonmagnetic fractions by a standard laboratory hand-held magnet
with a steel needle attachment. The magnetic separates were
weighed on a table top balance and representative portions of
the magnetic and nonmagnetic fractions assayed for U-238,
Th-230, Th-232, Ra-226, Ba, V, and Fe.
4. RESULTS OF RADIUM 226 ANALYSIS ON SOIL FRACTIONS
The radium 226 activity level of the bulk soil samples was
814 and 182 pCi/g respectively for the Glen Ridge and Montclair
sites and 3400 pCi/g for the White Sands sample. This activity
level is significantly higher than the average reported for the
Montclair and Glen Ridge contaminated soils (NJEDP 87); however,
the more highly contaminated soils were selected intentionally
in order to more effectively evaluate contaminating source
materials.
The radium 226 activity levels for gravel sized materials
and the individual fractions of sand, silt, and clay-sized
materials are listed in Table 1. In all samples, the concen-
tratiaon of radium 226 is greatest in the fine fractions. The
Glen Ridge silt/clay fractions comprise approximately 20 percent
of the contaminated soil; however, they contain 57 percent of
the radium with highest values of greater than 3,000 pCi/g in
the fine silt and clay-size materials. The Montclair samples
contain less radium activity than the Glen Ridge samples, but
the percentage distribution of radium is generally similar.
The Montclair gravel size contains less slag and more rock
particles than the Glen Ridge gravel-size materials.
The White Sands are a recognized radioactive anomaly in the
contaminated soil and this material has been cited as probable
tailing waste from the extraction of radium by the acid leach
milling process (01 86). The radium 226 activity levels of this
material of predominantly sand-silt size is several times that
found in comparable Glen Ridge soil fractions and as much as an
order of magnitude greater than similar sizes of Montclair soil
fractions (Table 1). The highest radium 226 activity in the
sand, silt, and clay size fractions of the White Sand is
respectively 1,913, 17,620, and 21,800 pCi/g.
11
-------
Table 1. Radium 225 activity and percentages on gravel, sand,
and silt/clay sizes. Analysis by EERF.
Size
Sieve Microns White Sands
Glen Ridge
Montelair
pCi/g
Wgt % pCi/g %Ra
Wgt % pCi/g %Ra
-1 1/2/+4 4750
35
346
15
26
44
-4/+10
-10/+16
-16/+50
-50/+100
-100/+140
140/+200
2000
1180
300
150
106
75
50
38
15
5
2
296
641
993
980
1,036
1,913
8,656
14,490
9,525
15,170
17,620
10
3
17
9
3
3
4
3
4
5
2
307
268
500 27
472
498
677
1,047
953
1,164
3,034
3,039
6
3
20
14
7
3
6
3
6
4
1
26
39
100 33
113
138
170
304
240
360
430
644
57
59
0.5
-0.5
21,800
3,725
1
1
576
3,301
0.5
0.5
1,113
264
Notes:
1. The average Ra 226 activity on a representative White Sands, Glen Ridge,
and Montclair sample is respectively 3400, 814, and 182 pCi/g.
12
-------
The source of the radium activity in the soil fractions
and the association of radium 226 with specific minerals or
materials in the soil fractions is one of the major
considerations in the sections that follow.
5. MINERAL COMPOSITION OF CONTAMINATED SOIL
The contaminated soils at the Montclair and Glen Ridge
sites are heterogeneous mixtures of native soil, industrial
wastes, mining waste, and household waste materials. These
materials are characterized in Table 2 with tne average
composition computed as the sum of the weighted averages of
the 18 soil fractions tested for each sample. The average
mineral composition for each size class (gravel, sand, silt, and
clay-size) is also listed in Table 2. The chemical compositions
listed in Table 3 were used to help assess the percentages of
some of the minerals.
The composition of the coarser gravel-size materials stands
in sharp contrast to the more homogeneous fine sand, silt, and
clay-size particles that contain the highest activity levels of
radium 226. In order to better relate the activity levels of
Ra-226, U-238, and Th-230 to tne mineral composition, radio-
chemical and chemical data, each size class will be treated
separately.
5.1 GRAVEL-SIZED MATERIAL
The gravel sized materials are those retained between the
No. 4 sieve-size (4.75mm) and 3 inch sieve size as specified
in ASTM D643-78. The ASTM standard is one of the several
grain-size scales described that is an acceptable standard
currently in use in the United States (Di 82). Since the
boulders were removed from the samples to facilitate mixing, the
gravel-size material is the largest size class of the samples
tested. The gravel-sized materials, witn boulders removed,
comprise 35 percent of the Glen Ridge and 26 percent of the
Montclair contaminated soil. A typical gravel-size fraction is
shown for the Montclair sample in Plate 1. The radium 226
activity on particle types and the fractional percentage
distribution of particles within the gravel size material is
13
-------
TABLE 2. MINERAL AND MATERIAL COMPOSITION OF MONTCLAIR/GLEN RIDGE
SIZE CLASS
SfTE LOCATION
WEIGHT PERCENT
FURNACE-FIRED MATERIALS
ANTHRACITE COAL
COALY SLAG
WHITE SLAG
FERRUGINOUS SLAG
GLASS/TRASH ROCK PARTICLES
SANDSTONE/SILTSTONE
QUARTZITE
GRANITE/BASALT
MINERALS
QUARTZ
FELDSPAR
RADIOACTIVE1'
AMORPHOUS SILICA
RADIOBARITE
CARNOTITE
URANINITE
ZIRCON
MONAZITE
OTHER RADIOACTIVE*
CLAY MINERALS
ILUTE
KAOLINITE/CHLORITE
HEAVY MINERALS
MAGNETITE
HEMATITE
HORNBLENDE
OTHER3'
" Radioactive minerals percentage based on petrographte
GRAVEL
G/R Miclr
35
12
44
20
7
5
6
5
1
—
—
—
__.
—
—
—
—
—
microscope,
21 Other radioactive minerals include tyuyamunfte, autunite, thorite, and
* Other heauu minerals In nrHar nf Kot-raacinn aHiirwfancA- ilmenileflAi!
26
13
22
5
5
3
20
25
7
—
—
—
_
—
—
—
—
—
XRD, SEM,
gummite.
iRcmene rift
SAND
G/R
45
2
14
18
2
—
3
5
^^
51
2
0.1
0.3
0.7
0.8
0.1
0.1
0.1
0.1
0.1
0.4
0.3
0.1
0.4
EDX, and
rnc*t rutilc
Miclr
53
3
19
12
1
__
2
3
—
52
2
0.1
0.2
0.2
0.3
0.1
0.1
0.1
0.1
0.1
0.3
0.2
0.2
1.0
SILT
G/R
18
—
—
—
—
__
2
3
—
63
1
20
0.5
0.9
0.1
0.1
0.1
0.1
8
5
0.2
0.1
0.1
0.1
Mtclr
20
—
—
—
—
__
—
—
—
73
2
10
0.2
0.2
0.1
0.1
0.1
0.1
8
5
0.1
0.1
0.1
0.5
CLAY
G/R
2
—
_
_
—
__
—
—
—
15
1
20
0.5
0.9
0.1
0.1
0.1
0.1
40
22
0.1
0.1
0.1
0.1
radiochemical analysis of density separated
> inurmalir
>e pnlrlrtto
..^-Mft.,
mtiUHe still!
Mtclr
1
_
— rar_
___
^^^^
_ _
—
—
15
1
10
0.2
0.2
0.1
0.1
0.1
0.1
45
22
0.1
0.1
0.1
0.1
fractions.
nrtanfta ctai
AVERAGE %
G/R
5
22
15
3
2
4
4
1
35
1
4.1
0.2
0.5
0.2
0.1
0.1
0.1
1.2
0.7
0.2
0.2
0.1
0.2
i mitt a ar»*
Mtclr
5
16
8
2
1
6
8
2
43
2
2.2
0.1
0.1
0.1
0.1
0.1
0.1
1.4
1.1
0.2
0.1
0.1
0.6
\ WuanHa
-------
TABLE 3. CHEMICAL COMPOSITION OF MONTCLAIR AND GLEN RIDGE SOIL SAMPLES
CHEMICAL
COMPOSITION
SULFUR % —
CARBONATE C % —
TOTAL CARBON % —
ORGANIC C % —
ALUMINUM % —
BARIUM % —
CALCIUM % —
IRON % —
Fg2 IRON PPM
FB3 IRON % —
LEAD % —
MAGNESIUM % —
MANGANESE % —
POTASSIUM % —
SILICON % —
SODIUM % —
TITANIUM % —
VANADIUM % —
T-PHOS % —
SULFIDE-S % —
WHITE SANDS (TAILINGS)
UJ
_ a
"E
• 4£
Z *"
en ™
3 <
z t-
^™ ^)
EH
0.09
<0.02
0.95
0.95
1.36
0.30
0.07
0.45
8.4-
0.45
0.016-
0.08-
0.006-
0.22-
42.4
0.29-
0.14-
0.069-
0.04
<0.01
UJ
>
o m
in CO
CO ".
3 0
Q. *L
0.07
<0.02
0.40
0.40
1.67
0.30
0.06
0.65
4.8
0.65
0.023
0.14
0.006
0.30
43.7
0.16
0.20
0.083
0.02
<0.01
UJ
^
o !ii
*~ CO
CO ->
3 °
•^ *?
o. ,L
0.04
<0.02
0.04
0.04
0.52
0.12
<0.01
0.07
1.2-
0.07
0.012
0.03
0.002
0.04
46.0
0.02
0.13
0.025
0.01
<0.01
ui
O °-
^
N 2)
„_ 1
en ™
3 <
Z H
E H
0.10
<0.02
0.67
0.67
1.39
0.38
0.05
0.30
3.4-
0.30
0.007
0.06
0.008
0.35
42.1
0.23
0.32
0.070
0.04
<0.01
GLEN RIDGE
Ul
w a.
• <
i
(0
2 H
— o
EH
0.11
<0.02
7.72
7.72
6.72
0.14
0.32
3.84
23.0-
3.84
0.028
0.30
0.034
0.96
31.4-
0.51
0.61
0.082
0.05
<0.01
ui
e "*
u> co
CO —
3 o
o. -L
0.013
<0.02
10.25
10.25
7.05
0.16
0.04
4.22
29.0-
4.22
0.056
0.36
0.036
1.13
30.1-
0.38
0.66
0.101
0.10
<0.01
ill
o ^
*- co
CO ~
3 0
o. «L
0.04
<0.02
1.87
1.87
4.47
0.08
0.15
2.06
13.0
2.06
0.056
0.24
0.024
0.66
39.3
0.31
0.39
0.061
0.04
<0.01
ui
Q.
E •
0 <
gco
CA — ^
3 <
Z H
— O
EH
0.08
<0.02
1.76
1.76
7.82
0.24
0.29
2.11
23.0-
2.11
0.048
0.36
0.023
1.26
32.1
0.47
0.86
0.105
0.10
<0.01
ui
Q.
S
X <
0«
••
CO ^
3 <
^ H
- o
EH
0.06
<0.02
4.70
4.70
5.44
0.09
0.36
3.49
22.0-
3.49
0.041
0.45
0.046
1.00
35.1
0.63
0.52
0.021
0.04
<0.01
MONTCLAIR
Ul
>
0^
in co
CO ~
3 0
_i 7
o. «L
0.06
<0.02
5.66
5.66
4.55
0.07
0.34
3.78
24.0-
3.78
0.066
0.28
0.051
0.87
36.3
0.39
0.39
0.021
0.03
<0.01
Ul
_ >
gui
»- co
CO ~
3 0
^ *?
Q. 2*
0.02
<0.02
0.77
0.77
2.94
0.05
0.10
1.69
12.0
1.69
0.031
0.18
0.022
0.63
39.7
0.41
0.32
0.011
0.02
<0.01
Ul
o.
s
0 <
Wl
^
3 <
Z H-
^~ ^D
EH
0.04
<0.02
1.36
1.36
6.77
0.11
0.31
2.58
21.0-
2.58
0.049
0.34
0.046
1.42
34.9-
0.79
0.82
0.02
0.05
<0.01
Analysis by Huffman Laboratories (Hu88).
-------
depicted graphically in Figures 3 and 4. Furnace fired coaly
material (coal, coaly slag, white slag, and ferruginous slag)
comprises most of the Glen Ridge (83 percent) and a considerable
portion of the Montclair (45 percent) gravel-sized materials.
Rock pacticles comprise more than half of the Montclair gravel
(52 percent) and a substantial portion of the Glen Ridge
(12 percent) material. Glass and other varieties of discarded
household waste (leather, ceramic material, etc.) comprise a
few percent of the gravel-sized material at both sites.
The radium 226 activity of the Glen Ridge gravel materials
averages 346 pCi/g and represents approximately 16 percent of
the radioactivity of the sample tested. The radium-226
activity at the Montclair site averages 44 pCi/g, approximating
8 percent of the radioactivity in the sample (Table 1).
Measurements of radium 226 activity on random gravel-size
particles by EERP found the highest readings in the slag
particles with ferruginous slag containing the highest radium
226 activity (Table 4 and Figure 3),
5'1'1 Furnace Fired Material
The furnace fired material, which comprises 83 percent
of the Glen Ridge and 43 percent of the Montclair gravel-sized
material, contains the highest radium 226 activity levels. The
furnace fired constituents consist of (a) unburned anthracite
coal, (b) interlayered coaly slag, (c) white slag and
(d) brownish-red ferruginous slag. (Figure 4).
5'1.1.1 Antracite Coal. Unburned, black subangular, anthracite
coal occurs in generally similar amounts at ,bo£h the Montclair
and Glen Ridge sites. Anthracite coal from'the Appalacian coal
fields of Pennsylvania deposits was in abundant supply in the
1920 's; however, this variety of coal has since been depleted.
The coal ranges from 5 to 68 pCi/g radium 226 (Figure 3) for.
several particles tested at EERF (Ri 89). Approximately 20
percent of the radium 226 is in secular equilibrium in the coal
particles assuming no process has separated the parent uranium
from the radium.
16
-------
Other Rock
Sandstone
Quartzite
Coal
Coaly Slag
White Slag
Ferruginous Slag
i
i
,
Eac
I
I
ti bar re
— 1
I
presents
one sarr
pie anal)
sis
•4 1 1 . t 1 1 I I '
25 50 75 100 125 150 175 200 225 250 275
pCI/gram
Figure 3. Radium 226 Activity on Washed Gravel-Size Particles
-------
Rock Partides
= Glass & Trash
o Ferruginous Slag
H
5
White Slag
Coaly Slag
Coal
Montclair (26 weight % gravel)
__
Rock Particles
Glass & Trash
Ferruginous Slag
White Slag
Coaly Slag
Coal
10 20 30 40 SO 60
Percentage of Gravel Size
Glen Ridge (35 weight % gravel)
Furnace Fired Malarial
B Basalt
0 Granite
Q Ouartzite
B Sandstona/Siltslone
10 20 30 40 50 60
-
I
-
B
j
Heavy Minerals
Feldspar
Quartz
Rock Particles
Slag
Montclair (53 weight % sand)
10 20 30 40 50 60
Heavy Minerals
Feldspar
Quartz
Rock Particles
Slag
Glen Ridge (45 weight % sand)
Percentage of Sand Size
Amorphous Coated Quartz
10 20 30 40 50 60
Figure 4. Composition of Gravel and Sand Size Materials in Montclair and Glen Ridge Contaminated Soils
-------
The National Coal Resource Data System of the United States
Geological Survey lists the relative amount of U and Th in
coal in the United States (Pi 88) as follows: (1 pCi/g U =
approximately 3 ppm U)
Geometric Maximum Number
Mean (PPM) Value (PPM) of Tests
ALL U.S. COAL
U 2.59 2674 6532
Th 3.12 92 7001
APPALACHIAN COAL
U 1.7 63.4 4213
Th 2.9 47.8 4361
ANTHRACITE COAL
U 1.4 25.2 47
Th 5.2 14.4 40
The uranium in the coal particle tested at EERF from the
Montclair and Glen Ridge sites exceeds the uranium in the
average coal in the three categories cited above. Extreme
uranium values occur in some of the lignite coal deposits of the
Dakotas and Texas. The lignite coal of the Dakotas, mined for
its uranium content, #as ashed to concentrate the uranium ore
prior to milling in order to reduce the hauling cost (Fi 88).
Apparently, burning the coal does not affect the uranium rfhich
in the ash is concentrated as much as 10 times or more. Thus/ a
considerable portion of radioactivity in the gravel is from the
uranium originally present in the coal from the Appalachian coal
fields. The radium activity of uranium mineral, however, is
probably not more than 4.5 percent of the total radium activity
in the samples.
5.1.1.2 Coaly Slag. The coaly slag comprises approximately 44
percent of the Glen Ridge gravel and 22 percent of the Montclair
gravel-sized materials. The coaly slag is mottled in appearance
and consists of partially burned coal, in various stages of
decomposition to slag. The radium 226 activity on several
-------
particles ranges from 12 to 110 pCi/g (Figure 3). Approximately
15 percent of the radium 226 is in secular equilibrium with
uranium.
5.1.1.3 White Slag. White flat-shaped siliceous slag averages
20 percent of the Glen Ridge and 5 percent of the Montclair
gravel-size materials. X-ray diffraction analysis of the white
slag reveals a composition of mullite, quartz, cristobalite, and
an amorphous silica-rich glass. The mullite (Als Si2 013)
is a refractory material resulting from heating of high
aluminous silicate minerals such as clay minerals. The
amorphous glassy material appears as a broad welt on the x-ray
diffractogram between 20 and 30 degrees 2 theta. The high
siliceous composition and fine grain size (large surface area)
are factors that would favor high adsorption of radium 226. The
initial radium 226 activity of 71.9 pCi/g in the white slags and
relatively low uranium (0.98 pCi/g) tend to support the view
that more than 90 percent of the radium 226 on this highly
siliceous material may be a result of adsorption or perhaps
radium paint residue from burned rags containing radium paint.
Nicdosh (1984) in studies of radium in Elliot Lake uranium
ore concludes that surface adsorption may play a major role in
retention of radium by solids when the radium content of the
solids is low (Ni 84). The Langmuir theory of surface
adsorption is also favored over leaching in very dilute
solutions and the adsorbed trace impurities are difficult
to wash off completely.
5.1.1.4 Ferruginous Slag. Reddish-brown ferruginous slag
comprises between 5 and 7 percent of the gravel and consists of
hematite and other metals that probably accumulate as a heavy
residue in a coal- fired furnace. Pyrite (FeS2) is a common
occurrence in coal and the oxidation of pyrite to hematite
(Fe203) or magnetite (Fe3 04) in the furnace probably
comprises the bulk of the accumulation of metals formed on the
slag at the bottom of the furnace. The relatively high magnetic
content of this material makes all of the particles weak to
strongly magnetic and much of this material can be removed with
a magnet.
The highest radium 226 activity occurs in the ferruginous
slag (Table 4). Secular equilibrium between radium and uranium,
however, is limited to about 15 percent of the radium activity.
20
-------
The remaining 85 percent of the radium 226 activity is believed
to be a result of radium painted residues from incinerated
clean-up rags or materials from the paint and adsorption of
the radium from solution.
A special test on representative soil samples from the
Montclair and Glen Ridge sites was conducted to determine the
feasibility of magnetic separation of gravel-sized materials
including also sand-size materials larger than the plus 50 sieve
size. The tests conducted at SERF found more than twice the
radium 226 associated with the magnetic fraction per unit
volume. It seems feasible to expect removal of approximately
10 percent of the radium activity by magnetic means on these
furnace fired ferruginous slag particles.
5.1.2 Rock Particles
Rock particles comprise approximately 52 percent of the
Montclair and 12 percent of the Glen Ridge gravel-size material
(Figure 4). The smooth, rounded, dense, tan, quartzite
particles constitute about 50 percent and 30 percent respec-
tively of the Montclair and Glen Ridge rock particles. The
quartzite is transported glacial drift from the Pleistocene
materials that blanket the bedrock. The red-brown, smooth-
to-rough surfaced, subangular, dense sandstone and siltstone
particles present in generally similar proportions as the
quartzite were probably derived from the underlying bedrock
(Brunswick Formation). Lesser amounts of granite, gneiss,
basalt and minor other rock varieties comprise the remainder
of the rock particles.
The radium 226 activity associated with the rock particles
ranges from 3.7 to 9.0 pCi/g, and approximately one-third of the
radium is in secular equilibrium. The other two-thirds of the
radium 226 is attributed to adsorption of the radium from
solution.
5.1.3 Glass and Trash
Glass and other debris typical of materials discarded
into dumpsites comprise about 5 percent of the Glen Ridge and
3 percent of the Montclair gravel-size materials. Radium 226
activity values on these materials may have local high anomalies
in organic-rich materials. Virtually all of the radium 226 is a
result of adsorption of the radium from solution.
21
-------
5.2 SAND-SIZE MATERIALS
The sand-size materials are those particles retained
oetween the number 4 sieve (4.75mm) and number 200 sieve
(0.05mm) (ASTM D643-78). Seven fractions of the sand-size soil
were used to characterize the sand material; the sieve sizes
used included the number 10, 16, 50, 60, 100, 140, and 200. The
sand-size materials of the soil samples comprise 53 percent of
the Montclair and 45 percent of the Glen Ridge materials. White
Sands tailings from a near-surface strata at the Montclair site
constitute a high radium 226 activity anomaly for comparison
with similar materials dispersed in the radium contaminated
soils.
The sand-size materials listed in Table 2 average
35-38 percent slag particles, 5-8 percent rock particles, and
tne major portion consists of homogeneous minerals (quartz,
feldspar, and minor radioactive and other heavy minerals). The
slag and rock particles occur only in the plus 100 sieve size or
larger materials with virtually homogeneous minerals below that
size. The major composition change occurs at about the 50 sieve
size.
5.2.1 Quartz and Quartz Films
Quartz is the dominant material (80-95-percent) in the fine
sand-size fractions (100, 140, and 200 sieve sizes). Some of
the quartz grains contain an amorphous coating which is more
abundant on the quartz grains on the white sands (8 percent)
as compared to Glen Ridge (5 percent) and Montclair (3 percent)
samples (Figure 4). These coatings are believed to be (a)
predominantly white amorphous silica released as a precipitate
from the acid-leach process, and (b) minor amounts of soft,
yellow, radioactive carnotite adhering to quartz from
unprocessed sandstone uranium ore. Both types of films on
quartz were observed in reflected and transmitted light with the
petrographic microscope. The radium 226 activity on the light
less tnan 2.9 specific gravity) slag, feldspar, and quartz
(including quartz coatings) accounts for 36 to 69 percent of the
minus 50 and plus 270 sieve size material respectively at the
Montclair and Glen Ridge sites (Table 5).
The adsorptive properties of quartz for most radionuclides
are negligible as indicated by measured sorption distribution
coefficients (Kd). Measured Kd values for radium on quartz
reported by Fordham, however, are 1900 mL/g, suggesting that
22
-------
radium has a strong attraction for solids (Po 73). Nirdosh et
al (1987) in an extensive investigation of radium adsorption on
quartz found that radium, unlike other alkaline-earth cations/
adsorbs on quartz at pH 1 and the order of magnitude of
adsorption at pH 1 is comparable to that at pH 10 (Ni 87).
The nature of the radium adsorption is believed to be "specific"
adsorption and the radium ions are strongly bonded to the solid
quartz surface by chemical or Van der Waal forces and are
essentially nonhydrated. Other alkaline earth elements (such as
Ca) are bonded by weak electrostatic forces and are hydrated; as
such/ they can be easily desorbed when the electrostatic
attraction between the adsorbant and adsorbate is weakened.
Radium by contrast is strongly adsoroed but not easily desorbed
(Ni 87).
5.2.2 Feldspar
Feldspar comprises from 3 to 5 percent of the fine sand
fraction. Both K-feldspar and Na/ Ca/ feldspar are present in
generally similar proportions in the Montclair and Glen Ridge
samples but are negligible in the White Sands sample. The
radium 226 distrioution coefficient measured on albite {Na-rich
plagioclase feldspar) has been reported as 20,000 ml/g (Pi 86).
This high adsorption is probably due to cation exchange of Ra
for Na and/or Ca in this otherwise tightly bonded alumnosilicate
mineral structure.
5.2.3 Heavy Minerals
The heavy minerals (greater than 2.9 specific gravity)
comprise from less than 1 to 3 percent of the fine sand and
upper-silt size fractions. The heavy minerals of the White
Sands separated by bromoform contain barite (radiobarite) as
the orincipal constituent (Figure 5). The heavy minerals of
the White Sand fractions average 0.5 percent of the sand with
minerals composition of the heavies averaging 55 percent barite/
17 percent zircon, 16 percent black opaques (hematite, magnetite,
ilmenite and uraninite), 3 percent rutile, 2 percent carnotite,
and 7 percent minor others (tourmaline, monazite, garnet,
hornblende, pyroxene, etc.). The major portion of the heavy
minerals consists of residuals from the radium extraction acid
leach process (radiobarite) and uranium ore and associated
23
-------
TABLE 4. RADIOCHEMICAL ANALYSIS OF MONTCLAIR/GLEN RIDGE
GRAVEL PARTICLES. ANALYSIS BY EERF.
GRAVEL-SIZE
PARTICLES
FURNACE FIRED
COALY SLAG
FERRUGINOUS SLAG
WHITE SLAG
COAL
ROCK PARTICLES
SANDSTONE
QUARTZITE
OTHER ROCK
TRASH
WOOD (PLASTIC)
LEATHER
GLASS (POTTERY)
LOCATION
R
M
G
G
M
M
R
G
G
G
G
M
R
M
G
G
R
R
M
M
M
M
R
M
G
U-238
pCi/g
2.77
4.13
3.49
7.39
6.07
3.92
2.71
40.63
16.83
0.98
10.20
1.66
4.43
6.49
8.55
1.16
0.86
0.54
1.83
1.20
0.64
3.58
3.00
17.88
2.11
Ra-226
pCi/g
12.91
10.41
42.06
78.87
13.10
68.57
6.19
253.90
126.80
71.90
91.57
5.51
20.22
18.95
66.73
3.70
6.74
8.99
5.42
6.80
5.73
9.18
19.22
72.50
15.94
Th-230
pCi/g
4.40
19.18
2.84
31.89
2.75
3.38
3.32
63.59
50.14
14.95
13.72
0.70
1.70
1.76
7.30
1.11
0.76
2.41
0.82
1.28
0.51
3.80
6.09
21.28
7.63
Th-232
pCI/g
2.51
2.24
0.63
4.04
1.61
3.02
2.12
0.47
3.03
3.38
2.07
0.54
0.94
0.38
0.63
1.10
0.86
0.41
0.92
1.01
0.47
0.69
0.99
0.53
0.54
Fe
ppm
_
—
—
—
55,300
—
—
—
—
930
—
2,000
—
—
—
17,900
—
—
—
18,900
—
—
—
—
—
Note: 1. M = Montclalr soil; R = Representative soil;
G = Glen
2. Fe analysis t
Ridge soil.
iv Galbralth Labors
torlas
24
-------
White Sands
60
Glen Ridge
Montclair
.
i
K)
Ul
50-
40 •
30
20
0.3% Heavy Minerals
113,400 pCI/g Ra-226
o
1.6% Heavy Minerals
7082 pCi/g Ra-226
2.0% Heavy Miners
4878 pCI/g Ra-22(
Ilmenite/Leucoxene
Figure 5. Comparison of Heavy Mineral Suite in -50/+270 Sieve Size White Sands, Glen Ridge, and Montclair Material
-------
The average heavy mineral fractions of the Glen Ridge and
Montclair samples contain minor representation of the heavy
minerals common to the tailings as well as major minerals
common to the glacial fills and furnace-fired metallic opaque
materials (uraninite, magnetite, and hematite). The greater
radioactivity in the Glen Ridge sample as compared to the
Montclair sample is reflected in the heavy minerals. The
Glen Ridge heavy mineral fraction contains both greater
concentrations of uraninite from furnace fired materials and
somewhat larger proportions of radiobarite and carnotite from
the unprocessed ore and acid leach process.
5.2.3.1 Radioactive Minerals. The radioactive minerals in the
White Sands are restricted to carnotite and minor uraninite and
tyuyamunite. The latter was identified by x-ray fluorescence
probe of silt-sized material (Er 38). The carnotite is readily
observed under the petrographic microscope as yellow, smooth,
elongated, discrete particles and as yellowish films on quartz
grains. Since vanadium is abundant in carnotite and
tyuyamunite,xbut impoverished in other materials at the New
Jersey sites, this element is used as a chemical signature to
calculate the amount of the uranyl vanadate minerals in the
contaminated soils. The percent of vanadium for representative
material in the less than number 4 sieve size of the
contaminated soil samples from the Montclair and Glen Ridge
sites reported by Huffman Laboratory (Table 3) is listed below
with the approximate amount of carnotite it represents.
White Sands Glen Ridge Montclair
Vanadium % 0.069 0.082 0.021
Carnotite % 0. 59 0. 70 0. 18
The uraninite in furnace fired material associated with
magnetite was separated by a hand magnet and this fraction is
listed in Table 5. All of the furnace fired magnetic fraction
is brownish-red to black in color and generally in flat and
elongate particles of sheet-like structure, reflecting its
formation as furnace slag.
26
-------
TABLE 5. RADIOCHEMICAL ANALYSIS OF MONTCLAIR/GLEN RIDGE SAND-SIZE
MATERIAL BETWEEN -50 AND +270 SIEVE SIZE. ANALYIS BY EERF.
SAND
(-50/+270 SIZE)
MTCLR LIGHT (-2.9)
MTCLR HEAVY (+2.9)
MAGNETIC
MTCLR HEAVY (+2.9)
NON-MAGNETIC
G/R LIGHT (-2.9)
G/R HEAVY (+2.9)
MAGNETIC
G/R HEAVY (+2.9)
NON-MAGNETIC
W/S LIGHT (-2.9)
W/S HEAVY (+2.9)
Wt%
98.0
0.4
1.6
98.4
0.3
1.3
99.7
0.3
Ra-226
pCi/g
43.03
190.70
4,687.00
181.10
1184.00
5898.00
806.5
204,700
%
Ra
36
1
63
69
1
30
57
43
U-238
pCi/g
8.75
19.28
54.51
28.44
190.40
728.49
2.49
9.62
%
U
90
1
9
22
4
74
99
1
Th230
pCi/g
39.89
74.89
470.02
194.80
432.25
862.15
12.02
47.01
TH232
pCi/g
0.60
1.86
16.08
1.95
2.84
23.01
0.02
0.43
Fe
ppm
476
6,800
7,700
365
5,200
7,000
24
398
G/R = GLEN RIDGE
MTCLR = MONTCLAIR
W/S a WHITE SANDS
27
-------
A larger fraction of the uraninite occurs in the
nonmagnetic heavy mineral fraction with about half of the
uraninite similar in appearance to the furnace material and the
rest clearly naturally occurring uraninite. The latter consists
of pitch black, equidimensioned, high lustered, botryoidal or
"buobly11 textured particles. The highest thorium 232 and
uranium 238 also occurs in the nonmagnetic heavy mineral
fraction (Table 5). While thorium 232 occurs in monazite, which
is present in similar amounts in all the heavy mineral fractions
from the sites, an anamolous amount occurs in the Glen Ridge
heavy mineral fraction. Uraninite from high temperature vein
deposits is known to contain thorium 232 in variable amounts
in solid solution with uranium 238. Thus the natural high
temperature uraninite occurring in the Glen Ridge nonmagnetic
heavy mineral fraction appears to correlate with the anomalous
thorium 232 (23.01 pCi/g) reported in Table 5.
The Belgian Congo uranium ore, reportedly used for most
of the world uranium supply after 1923 (La 84), is a high
temperature uraninite. However, the high temperature uraninite
in the Glen Ridge sample lacks some of the characteristics of
that deposit and could have originated in the Colorado Plateau
deposits. Garrels and Larsen (1959) in a geochemical study of
uranium ores of the Colorado Plateau, reported that the
classic low temperature, oxidized, sedimentary, carnotite
deposits sometime occur in proximity to veins or "pipes"
that are of hydrothermal origin (Ga 59). According to Frondel
(1958), oxidized deposits of uranium and vanadium have been
known in the Colorado Plateau since 1898, but uraninite was not
recognized as a valuable uranium ore mineral in the Colorado
Plateau area until 1943-49. Thus, in the 1920's, when the
carnotite was used for radium production in New Jersey, any
uraninite that occurred with the carnotite was probaoly treated
as a gatigue mineral rather tnan an ore mineral.
Radiochemical studies of minus 4 sieve size materials at
EERF and special studies on silt and clay fractions by Sattelle
Northwest Laboratory indicate that the uranium 233 occurring in
the contaminated soils would account for only 10 percent of the
progenies. Thus, approximately 10 percent of tha radioactivity
at the sites is a result of the uranium minerals carnotite and
uraninite and very minor other uranium bearing minerals. The
relative proportion of canotite may be approximated from
vanadium assay and the uraninite from uranium assay of tne
magnetic and nonmagnetic heavy fractions.
28
-------
5.2.3.2 Radiobarite. Radiobarite (radium barium sulfate)
comprises the major portion of the White Sands heavy mineral
fraction which averages approximately 0.3 percent of the sand
fraction. When viewed under the petrographic microscope, the
barite grains are white to yellowish color, anhedral in shape,
and have surfaces ranging from smooth and clear to rough,
mottled, and pitted. The radiobarite grains were probably
smooth and unblemished when they first formed as a coprecipitate
during the acid-leach stage of the radium milling process. The
radiobarite subsequently incorporated into mill tailings and
removed to burial in the Montclair and Glen Ridge landfills may
have encountered local anaerobic microenvironments resulting
from contact with organics or other redox governing factors.
Investigations of uranium mill tailings have disclosed possible
reducing conditions in a large uranium pile of the Western
United States that is in proximity to uranium mills,
Sulfate-reducing bacteria acting on sulfate compounds in these
uranium mill tailing piles may have occurred in localized
anaerobic microenvironments and may have posed possible
localized sulfide-forming conditions (La 86 and Sh 84). The
pitted surfaces of the radiobacite in the Montclair and
Glen Ridge samples may similiarly be a result of such localized
anaerobic microenvironment activity which resulted in pyrite
(FeS2) being formed on radiobarite surfaces. The pyrite would
have since been oxidized and removed resulting in the pitted
surfaces of the radiobarite. There is probably a correlation
between radium 226 activity and degree of surface alteration of
the radiobarite. Since some of the radium 226 was probably
released in the alteration process, the unblemished radiobarite
would contain higher radium 226 activity levels.
The radiobarite is more dispersed in the average Glen Ridge
and Montclair contaminated soil than in the White Sands. Since
barium is negligible in the glacial till, barium may be used
as a signature for the abundance of radiobarite formed as a
coprecipitate in the radium extraction acid leach process.
The percent of oarium in the minus number 4 sieve size material
(Table 3) is listed below with the amount of radiobarite it
represents.
Whits Sands GJ.en Ridge Montclair
Barium % 0.38 0.24 0.11
Radiobarite % 0.76 0.48 0.22
29
-------
The 0.08 and 0.04 percent barium reported respectively for
the average Glen Ridge and Montclair sites by the New Jersey
Environmental Protection Department (NJEPD 87) would appear to
correlate with the samples used in this investigation, since
higher activity samples were used in this characterization study,
5.2.3.3 Natural Background Minerals. zircon and monazite, two
slightly radioactive heavy minerals that contain minor amounts
of uranium, radium, and thorium, are common background minerals
in most natural soils. Both of these minerals are highly
resistant to weathering and hence occur in many deposits.
Zircon (ZrSi04) comprises 17 percent of the White Sands heavy
mineral fraction and 5 percent of the Montclair and Glen Ridge
sand-size heavy mineral fractions whereas monazite (CePC>4)
comprises generally less than 1 percent of the heavy mineral
fractions. As much as 3 percent uranium and 13 percent thorium
have been reported in some zircons (He 58). Monazite contains
variable amounts of Th02 (up to 30 percent) and occasionally
contains some uranium. Both monazite and zircon are the chief
contributors to the uranium and thorium content of the natural
background radiation.
5.2.3.4 Other Heavy Minerals. Other heavy minerals, exclusive
of the radioactive minerals previously cited, that occur in both
the Glen Ridge and Montclair soils include: magnetite,
hematite, hornblende, illmenite/ leucoxene, garnet, rutile,
tourmaline, epidote, staurolite, mullite, sillimanite,
staurolite and kyanite (Table 2). Magnetite, hematite, and
mullite are heavy minerals formed from furnace-fired materials.
The remaining heavy minerals are" rock forming minerals that most
abundantly originate from the host glacial tills. A few highly
resistant minerals (zircon, monazite and rutite) occur with the
tailing sands.
The most significant mineral as regards retention of
radium 226 is probably hematite. Some of the hematite is an
oxidation product of magnetite from furnace-fired material and
other waste debris while a lesser amount is from glacial till
(red sandstone and siltstone, etc). This mineral is highly
adsorptive of radium in the process of formation in the fresh
state but less adsorptive when fully formed. According to
Krishnaswami et al (1982), freshly precipitated ferric hydroxide
had a measured radium Kd of 28,000 mL/g which is the most
effective adsorbent known for radium (Ki 82). The highest
activity for radium 226 found in the ferruginous slag (524
pCi/g) appears to corroborate this view (Ri 89).
30
-------
5.3 SILT AND CLAY-SIZE MATERIAL
The silt size-material comprises the soil particles between
0.074mm and 0.002 mm size. Clay-size materials are all
particles less than 0.002 mm. The boundary between the silt and
clay-size is in accordance with the soil scientist classi-
fication (USDA 75). The weight percents of the silt (18-20
percent) and clay-size (1-2 percent) fractions are listed in
Table 2. The estimates of mineral composition of the silt and
clay-size materials (Table 2) are based on: 1) x-ray diffraction
(XRD) analysis (bulk soil fractions and linear density
fractions), and 2) chemical composition. Two of the x-ray
diffractograms of the light (2.10-2.25 g/cm;?), medium
(2.25-2.71 g/cm3) and heavy (2.71-2.96 g/cm3), density
fractions of the 10 to 20 micron silt fraction of the Glen Ridge
contaminated soil are depicted in Figure 6.
The most abundant mineral in the silt-size material is
quartz, comprising approximately two thirds of the sample.
Amorphous silica and clay minerals (illite, chlorite, and
kaolinite) comprise respectively 10-20 percent and 5-8 percent
of the silt-size material (Table 2). Chemical analysis of
barium and vanadium in the bulk silt and clay-size fractions
were used to compute barite and carnotite for the Montclair and
Glen Ridge fines. The results were: 0.2 percent barite and 0.2
percent carnotite for Montclair; 0.5 percent barite and 0.9_ ,
percent carnotite for Glen Ridge (Table 2). Minor amounts of
hematite, feldspar, cristobalite and mullite ara also disclosed
in the X-ray diffractograms in the density fractions of the Glen
Ridge silt (Figure 6).
The clay-size materials comprise one to two percent of t;he
samples and are comprised of amorphous silica, illite, chlorite,
kaolinite, and quartz (Table 2).
Of special interest is the percentage of radium associated
with the linear density layers of the 20 to 45 micron, 10 to 20
micron, and 2 to 10 micron-size fractions of the Glen Ridge soil
samples (Table 6). In all three of these size fractions, the
amorphous silica predominates in the light density band (2.0
to 2.35 specific gravity). The amorphous silica in the light
density bands contains between 25 to 31 percent of the
radium-226 activity in the silt/clay fractions. The
identification of the x-ray amorphous material as amorphous
silica was provided by the scanning electron microscope and
x-ray fluoresence probe (Er 88).
31
-------
GLEN RIDGE 20MIN (1020)
I
0.9
0.8
0.7
0.5
0.4
0.3
0.2
0.1
1
0.9
0.8
0.7
8 0.5
0.4
0.3
0.2
0.1
20 40
2-THETA DEGREES
FLOAT
HEAVY LIQUID SEPARATION
60
»
0 20 40 60
2-THETA DEGREES
BOTTOMS
Figure 6. X-ray diffractograms of Glen Ridge 10 to 20 micron-size
light and heavy density fractions
32
-------
The amorphous silica concentration decreases in the middle
density fraction of tne Gien Ridge silt fractions where qaartz,
kaolinite, illite/muscovite, feldspar, and barite are identified.
Mullite, hematite, and cristabolite from the furnace-fired slag
material also occur in this density fraction.
Bacite (tadiobacite) clearly dominates the heavy density
fraction (2.62 to 2.69 specific gravity) in the silt samples
(Table 6). Tha highest concentration of hematite also occurs in
this fraction. Minor quantities of quartz, illite, kaolinite,
feldspar, mullite, and cristabolite are also present. Although
no specific uranium mineral peak was observed by XRD in any of
tne density bands because concentrations //ere not sufficient to
exceed background scatter, positive analysis was made of a
uranium mineral with the x-ray fluorescence probe (SEM/EDS).
The mineral was an uranyl vanadate (carnotite or tyuyamunite).
The use of autoradiography techniques provides positive _
information on the association of specific radionuclides with
specific minerals. The alpha tracks emitted by the radionuclide
enable focus on specific target minerals ls in the gravel and sand-sized soil fractions (Tables 4 and
5!) " However, in the silt and clay-sized fractions, PNL measured
tnorium at higher activity levels than radium 226 (Sr 83). Tnis
33
-------
TABLE 6. MINERAL COMPOSITION AND RADIUM 226 ACTIVITY OF SILT-SIZE (-45 MICRON +2 MICRON) SOIL FROM
GLEN RIDGE SITE. ANALYSIS BY PNL.
Wt% DENSITY
Light
2.10-2,35
Medium
7.2 2.35-2.71
Heavy
2.71-2.96
Light
2.10-2.25
Medium
11.7 2.25-2.71
Heavy
2.71 - 2.96
% FRACTION
29.22
48.26
22.52
32.30
55.69
12.01
Ra-226
ACTIVITY
GLEN RIDGE
2,620 pCi/g
1,400 pCi/g
4,590 pCi/g
GLEN RIDGE
1,640 pCi/g
1,040pCi/g
8,270 pCi/g
Ra-226
CONTENT
20-45 MICRON SIZE
766 pCi
676 pCf
1,034pCi
10-20 MICRON SIZE
530 pCi
579 pCi
993 pCi
%Ra
30.94
27.30
41.76
25.21
27.55
47.24
GLEN RIDGE 2-10 MICRON SIZE
Light
2.10-2.27
Medium
3.6 2.27-2.62
Heavy
2.62-2.96
31.68
45.01
23.31
2,010 pCi/g
1,450pCI/g
5,050 pCi/g
637 pCI
653 pCf
1,770pCi
25.62
26.47
47.71
MINERAL
COMPOSITION1-2
Major: AS;Q
Minor: M/l, K, M, F
Major: Q, 1, K, F, C, M
Minor: AS, B, H
Major: B, H, M, Q
Minor: K, M/l, F,C
Major: AS, Q, M/l
Minor: K,H
Major: Q, M/l, K, F, C
Minor: M, B, H, AS
Major: B, H, M
Minor: Q, M/l, K, F, C
Major: AS, Q
Minor: M/l, K, M, H
Major: AS,Q,K,M/I
F.M.C
Minor: H, B
Major: B, H, M
Minor: Q, M/l, F.C
Notes: 1. Mineral composition based on x-ray diffraction (XRD) analysis linear density gradient bands and x-ray
fluorescence of x-ray amorphous material.
2. Amorphous Silica = AS; Quartz = Q; Mica/lllite = M/l; Kaolinite = K; Feldspar = F, Mullite = M;
Barite (Radiobarite) = B; CristaboIHe = C; Hematite = H.
-------
LT= 100 SECS
HEAVY MINERAL FRAC
5000
4000
CO
O
LJ
3000
2000
1000
U
0
0.000
i
5.
10.000
ENERGY
U
U U
15.000
20.000
keV
Figure 7. SEM photomicrograph and EDX spectrum for carnotite/tyuyamunite in the heavy
mineral fraction of the Glen Ridge soil. The length of the bar is 100 microns.
-------
relative increase of thorium 230 associated with the finer
material could possibly be related to the greater abundance of
amorphous silica that occurs in the fine fractions. In studies
of migration of radium and thorium from natural uranium deposits,
Airey (1982) reports that radium tends to be associated with
clay/quartz, whereas uranium and thorium tend to be associated
with the iron phase of the weathering profile (Ai 82). Hence,
it is possible that the highest activity levels of thorium 230
to radium 226 in the light density fractions of the silt is a
consequence of thorium attachment to amorphous silica during the
acid leach process. The generally high percentage of radium
226 in the higher density fractions may be the result of
adsorption to hematite particles.
6. ASSESSMENT OP PERCENTAGE OF RADIUM 226 DISTRIBUTION
ON SOIL MATERIAL?~
The percentage of radium 226 associated with specific soil
materials was determined by fractional separations of minerals
and materials by sieving, heavy liquid separations, magnetic
separations, radiochemical analyses, and chemical analyses. The
uranium minerals were identified by petrographic analysis and
chemical assay for uranium. The radium 226 occurs both as in_
growth progeny of uranium 238 in uranium minerals and as radium
separated from uranium. The radium separated from uranium by
the acid leach process occurs as (a) radiobarite (3a,Ra 304}
and (b) as radium fixed to amorphous silica. Some materials,
sucn as the furnace fired slags, have adsorbed radium from
solution on their surfaces or by cation exchange.
The majority of the soil particles contain very minor
amounts of radium 226. The radium 226 is associated with
specific materials that include (a) uranium minerals ; (b)
acid-leach precipitates or coprecipitatas, and (c) highly
adsorbent materials. Reasonable approximations of the
radium 226 distribution will be made from measured radium 226
activity associated with specific fractions of known composition.
6.1 RADIUM 226 IM SECULAR EQUILIBRIUM WITH URANIUM MINERALS
Radium 225 nas no specific mineral of its own that occurs
in nature. As a daughter product of uranium 238, it has the
same specific activity as uranium tfhen in secular equilibrium
with the parent uranium. Assuming the uranium minerals
(carnotite and uraninite) ara relatively intact, the uranium and
radium radiochemical analyses of the minus Wo. 4 sieve material
indicate that approximately 10 percent of the radium 226 is in
36
-------
secular equilibrium with the uranium 238 (Ri 88). Measurements
of the silt/clay fractions also indicate a similar percent of
secular equilibrium materials (Er 89). The gravel-size
material, however, contains 15 percent to 33 percent of the
radium in secular equilibrium with uranium. This difference is
not surprising since the gravel is comprised largely of rock
particles and furnace-fired slag and is free of radium solids of
sand-size and smaller size formed from the acid leach process or
which occur as uranium ore minerals.
The uranium minerals, for calculation purposes, used to
assess the radium 226 activity distribution include (a) the
furnace fired slag uranium minerals from the Appalachian coal
fields and (b) the uranium ore minerals associated in the
tailings sands from the radium processing plant. The former is
uraninite associated with the larger soil particles and the
latter is carnotite and uraninite in essentially equal amounts
that occur in the finer soil materials.
A major composition change occurs at the plus 50 sieve
size. Practically all the furnace-fired slag and rock particles
occur above this size (Table 2). The percentage of radium 226
activity in the plus 50 sieve size for the Glen Ridge soil
approximates 30 percent. Special tests conducted at EERF on
plus 50 sieve materials on Montclair and Glen Ridge soils found
that magnetic particles have more than twice the radium activity
per unit volume. It is estimated that about 10 percent of the
radium activity on the plus 50 sieve size particles could be
removed by simple magnetic separation.
In the assessment of the percentage of radium 226 activity
associated with uranium ore minerals, the carnotite and
uraninite are found to be essentially equal. The percentage of
radium 226 activity on the minus 50 sieve size soil fraction
from Montclair and Glen Ridge is respectively 83 and 70 per-
cent. Thus, if the carnotite and uraninite in this material is
considered equal in distribution for the 10 percent of the
radium 226 in secular equilibrium, the percent of radium 226 for
each of these minerals in the total Glen Ridge and Montclair
samples is 4 percent (Table 7).
6.2 RADIUM 225 IN NATURAL BACKGROUND MINERALS
Tne radioactive background minerals that are ubiquitous in
the Colorado Plateau ore minerals and the glacial tills of
New Jersey include zircon and monazite. These two minerals and
several very minor others (gummite, autunite, thorite, etc.) are
estimated to account for one percent of the radium 226 activity
in the total sample (Table 7).
37
-------
Table 7. Distribution of radium 226 on specific materials
of Montelair and Glen Ridge
Materials
Percent Radium
Glen Ridge
Distribution
Montclair(l)
Weight Percent
Glen Ridge Montclair
Radium in Secular Equilibrium in Uranium Minerals
Uraninite (ashed)
Uraninite (ore)
Carnotite (ore)
Zi rcon/Monazite
5
4
4
1
4
4
4
1
0.2
0.2
0.5
0.1
Acid-Leach Radium in Tailings Precipitates and Coprecipitates
Radiobarite
Amorphous Silica
36
26
45
35
Adsorption of Radium on particles (2)
Magnetic Ferruginous Slag
Other Furnace-Fired Slag
Illite/Mica
Feldspar 24
Hematite
Kaolinite/Chlorite
Rock Particles
Quartz
Other Materials
0.2
4.1
3,
42,
1,
1.0
0.2
0.7
9.0
35.0
2.6
0.1
0.1
0.1
0.1
0.1
2.2
2.0
29.0
1.4
2.0
0.1
1.1
16.0
43.0
1.7
1. Estimated values of Montclair samples for silt/clay fractions (lack
linear density gradient measurements).
2. Materials are listed in order of highest degree of adsorption
to lowest degree of adsorption from top to bottom.
3. Magnetic ferruginous slag and other furnace slag contains the major
occurrence of radium 226 in the gravel and upper sand-size fractions
as uraninite ash and as high adsorption on iron and other slag surface
materials.
38
-------
6.3 RADIUM 226 ASSOCIATED WITH RADIOBARITE
Radiobarite in the contaminated soil is a product of the
acid leaching process which removed uranium and thorium from the
ore mineral to isolate radium for industry. It was precipitated
with barium from solution as barium (radium) sulfate and the
uranium and thorium remained in solution for removal by
decantation. The radiobarite is restricted to the sand and
silt-size materials in the contaminated soil. The barite is
readily concentrated by heavy liquids and is associated with
the nonmagnetic heavy mineral fraction (Table 5).
The barite in the nonmagnetic heavy liquid fraction is
associated with uranium minerals and other heavy minerals,
generally low in radium 226 activity. The radium not in
secular equilibrium with uranium is assessed as radium affixed
to radiobarite for calculation purposes. Since hematite and
some background radioactive minerals (zircon and monazite) may
account for a small fraction of the radium 226 activity, the
radium 226 activity assigned to radiobarite is a maximum figure.
In the Glen Ridge sample, 30 percent of the radium 226
activity in the sand-size material is affiliated with the
non-magnetic heavy mineral fraction. Since 12 percent is
required for secular equilibrium, 88 percent is used in the
calculation that finds 26.4 percent radium 226 assigned to the
radiobarite in the sand-size fraction. Since the sand-size
fraction is 30 percent of total radium activity in the sample,
the radiobarite from the sand-size fraction is 8.7 percent of
the total radium 226 in the sample.
The radium 226 activity in the radiobarite of the Glen
Ridge silt/clay size linear density bands is 47 percent Er 89).
Since the radium 226 activity of the silt/clay size fraction
comprises 59 percent of the total sample, the maximum radium 226
associated with the radiobarite is 59 x 47 or 28 percent. The
sum of the percentage of radium 226 activity for radiobarite in
both the sand and silt/clay size fraction of the Glen Ridge
sample is 36 percent (80 + 28 percent) (Table 7).
The Montclair sample lacks the linear density gradient
measurements; however, if it is assumed to be generally similar
in proportion to the Glen Ridge sample, the radiobarite possibly
contains 46 percent of the radium 226 activity in the Montclair
sample.
39
-------
Th
hus, while radiobarite comprises bat 0.2 weight percent of
the Glen Ridge sample (Table 2), it comprises more than a third
of the radium 226 activity. The radiobarite in the Montclair
site comprises 0.1 weight percent of the sample but could
possibly contain nearly half of the radium 226 activity.
6.4 RADIUM 226 ASSOCIATED WITH AMORPHOUS SILICA
Amorphous silica was observed on the quartz grains of the
sand fractions with more representation in the White Sands. The
Glen Ridge sand-size light fractions, free of the heavy minerals
containing radiobarite and uranium minerals, contains 69 percent
of the radium 226 activity (Table 5). The only other material
in competition for this relatively high radium activity is
feldspar and quartz with moderate adsorption capabilities and
some carnotite adhering to some of the quartz grains. The
latter may be determined by assigning the amount necessary for
secular equilibrium; i.e., 28.44 pCi/g Ra 226 to match the
23.44 pCi/g U 238. Thus, 16 percent of the radium 226 is
attributed to carnotite/ leaving 84 percent for amorphous silica
and adsorbed radium. The percent of radium activity in the sand
approximates 9 percent for the total sample.
In the silt/clay fraction, the linear density band of light
material contains 30 percent of the radium activity (Table 5).
Since the silt/clay Glen Ridge fractions comprise 57 percent of
the radium 226 activity, the amount of radium 226 on the
amorphous. silica is calculated as 17.1 percent. Thus, the sum
of Ra 226 percentages for the sand and silt/clay size of the
Glen Ridge sample represented by the amorphous silica is
26 percent (Table 7) .
Similar calculations for the Montclair sample, assuming
parallel conditions for the silt/clay fractions at Glen Ridge,
indicate that 35 percent of the radium 226 may be a result of
the amorphous silica and radium adsorbed onto quartz and
feldspar. Since the activity level of Ra 226 is 4 to 5 times
greater on the Glen Ridge sample, the Montclair sample may have
a higher percent of the radium as adsorption onto particle
surfaces than occurs on the Glen Ridge sample. Thus/ the
amorphous silica values are maximum values of radium 226
association since they include adsorbed radium.
5.5 ADSORBED RADIUM 226 ASSOCIATED WITH SOIL
There are no sits specific measurements of the radium
equilibrium distribution coefficient (Kd) . The Kd is commonly
used to determine the degree of adsorption of an ion by an
40
-------
adsorbent from a solution. The few measurements in the
literature range from 1700 mL/g for quartz to 28,000 mL/g for
ferric hydroxide; some intermediate values include 20,000 mL/g
foe muscovite, 20,000 mL/g for albite, 6,500 mL/g for
montmorillonite, and 1900 mL/g for kaolinite (NJEPD 87), Since
iliice, the most abundant clay mineral at the Montclair and
Glen Ridge sites, has a mineral structure similar to muscovite,
the adsorption in the silt/clay fraction could be significant.
On the other hand, considerable adsorption is also apparent
from measured values of radium 226 on the furnace-fired slag
materials (Table 4). This assumption is depicted graphically
in Figure 8.
An approximation of the percentage of radium activity
resulting from adsorption is calculated by difference from the
sum of the Ra 225 percentages assessed for the uranium minerals
and acid leach products. The result is 24 percent for the
Glen Ridge sample and 6 percent for Montclair (Table 7).
The assessment of the percent radium distribution for the
Glen Ridge sample is depicted in Figure 8.
7. DISCUSSION
Tne methodology used in this investigation has provided an
assessment of the percentage of radium 226 activity associated
witn specific materials of known size and distribution. The
magnetic ferruginous slag and other furnace-fired slag material
comprises the major distribution of radium activity in the
gravel and larger sand-sized materials. The major radium
activity in the Glen Ridge sample occurs in the fine sand and
silt/clay size fractions as acid leach-derived radiobarite and
amorphous silica. Soma of the radium 226 activity of the
Glen Ridge sample is assgciated with uranium ore minerals and
uraninite coal ash. The remainder is attributed to furnace
fired radium paint residue and adsorption of radium on surfaces
of materials. This methodology has application to other
radium-contaminated soils which exist at several Superfund sites,
Magnetic ferruginous slag contains furnace-firad uraninite,
radium paint residue, and adsorbed Ra 226. It is expected that
as much as 25 percent (the +50 sieve size) of the radium 226
activity might be removed from material by a simple magnet
applied to the washed material.
In any chemical treatment to remove radium, it is well to
consider the following:
41
-------
o Amorphous silica probably contains both Ra 226 and
Th 230 in a weak bond. This material contains
approximately 26 percent of the radium 226 activity on
the Glen Ridge sample in the sand-size (as coatings on
quartz) and in the silt-size as discrete particles.
o Carnotite occurs as coatings on quartz sand and as
discrete particles in the sand and silt size. This
soft, yellow mineral is readily soluble in weak HC1
and comprises about 4 percent of the radium activity
in both samples.
o Uraninite occurs as coal-fired slag and could be
removed with the magnetic ferruginous slag with a
magnet. As much as 4.5 percent of the radium activity
occurs in furnace-fired slag with most associated with
the magnetic slag. A black, equidimensional shaped,
hard, dense, uraninite occurs in the fine sand and
upper silt size of both the Montclair and Glen Ridge
samples as discrete particles, with a density of
approximately 10, these heavy projectile-like
particles will with agitation sink through layers of
soil to bottom positions in separation pans for
removal. This same mineral is, however, insoluble in
weak HC1.
o Radiobarite occurs as dense particles of sand and
silt size and is relatively insoluble. It comprises
more than a third of the radium activity of the
Glen Ridge soil.
o Radium 226 adsorbed on solid surfaces is probably
difficult to remove without strong agitation^or
solvent extraction. This is based gn relatively high
radium equilibrium distribution coefficient (Kd)
values reported in the literature.
o Extraction tests conducted on contaminated soils of
average 75 pCi/g from the Montclair and Glen Ridge
sites (using standard RCRA method) found radium 226
releases of 10 pCi/L (NJEPD 87). Ground water beneath
these sites contains 2.3 pCi/L Ra 226. This suggests
that relatively weak acid solutions might be effective
in some of the radium removal.
42
-------
to
Radium In Secular Equilibrium
(16%)
100
10
1-
IE -a
I S
n
.01 -
.001
Caznofclta
Gkunita
XeM LMCft /todAim Materials
(62%)
Adsorbed Radkim
(22%)
OnuuLnit*
fin coal
and
Zircon
Jfonazlta
0 10
JUaorphoa* Silica
Hadiobarlfc*
7
7
r Furnace
Slug*
Book
Particl*s
Quartz
Hematite
Jtaollnlt*
^Chlorite
16
1
a
a
1
57
t
Figure 8. Relationship of Particle Size and Mineral Composition to Percent Radium Distribution In Glen Ridge Soil
-------
8. SUMMARY
The radium contaminated soils at the Montclair and
Glen Ridge, Naw Jersey, Superfund sites were characterized as a
first step in a remedial investigation to determine the physical
sizing, mineral composition, and radium distribution in the
minerals and materials comprising the soil. A grain size
distribution curve was constructed on each soil and 18 size
fractions of each soil were obtained for complete mineral and
radiochemical analysis. In addition, soil fractions were
further segregated by magnetic separation with a hand-held
magnet and by heavy mineral separations using sink float methods
for sand-size materials and linear density centrifugation
methods for the silt and clay size fractions. Chemical analysis
was also performed on aggregate and fractionated samples to
provide chemical signatures on diagnostic minerals and to
correlate with mineral determinations.
The characterization process found that uranium ore
minerals and the precipitates and coprecipitates from the acid
leach process used to obtain radium were the chief contaminants
in the soil with most of this high radium contaminated material
confined to the fine particles (fine-sand to clay-size
particles). Incinerated slag particles in the coarse particles
(gravel and coarse sand-size) were also contaminated to a lesser
degree with radium. Some of the radium of the furnace fired
slag was a result of ashing of coal containing uraninite;
however, a larger more significant amount was probably from (a)
burning of radium contaminated materials that also became part
of the slag and (o) adsorption of radium on slag surfaces after
being placed in the landfill, some of the radium also occurs in
natural background minerals (zircon, monazite, etc.) and
adsorbed on host mineral surfaces in the landfill.
The acid-leach radium in tailings materials was found
associated with radiobarite (Ba Ra 804) and amorphous silica.
The radiobarite comprises 36 percent on the Glen Ridge and
46 percent of the Montclair soil. Amorphous silica, occurring
as amorphous coatings on quartz grains in sand-size and as
discrete particles in siLt and clay-siza, comprises 26 percent
of the Glen Ridge and 35 percent of the Montclair soil, whereas
the radiobarite is relatively insoluble, the radium and thorium
associated with amorphous silica is mora readily solubilized.
The uranium ore minerals were identifed by mineral analysis
and correlated with secular equilibrium (balance Ra activity
with U values). The uraninite from coal ash, in slag particles,
44
-------
averages 5 percent radium activity in the Glen Ridge and
4 percent radium activity in the Montclair samples. Uraninite
and carnotite occurring as discrete ore minerals average
4 percent radium activity for both the Glen Ridge and Montclair
samples. Zircon and monazite, limited to microscopic identi-
fication, approximate 1 percent of the radium activity in both
the Glen Ridge and Montclair samples.
The magnetic ferruginous slag in the coarse particle size
fractions contain the ashed uraninite from coal and any radium
introduced by incineration of radium contaminated materials.
This material also contains some adsorbed radium from solution.
It is estimated from radiochemical measurements of magnetic
separations of the plus number 50 sieve size (0.15 mm) that
10 percent of the radium 226 activity is associated with this
fraction. This information is especially significant in volume
reduction remedial measures for the coarse particles in the soil,
The remainder of the radium activity in the soil is
associated with radium adsorbing to particle surfaces. The
chief adsorbates are illite, chlorite, kaolinite, and hematite;
however, radium also adsorbs to feldspar and quartz. The amount
of adsorbed radium probably comprises from 5 to 10 percent of
the radium activity.
The information made available by the soil characterization
has application to volume reduction and treatment processes in
remedial investigation procedures.
45
-------
REFERENCES
Ai 83 Airey, P.L., "Radionuclide Migration Around Uranium
Ore Bodies - Analogue of Radioactive Waste
Repositories," U.S. Nuclear Regulatory Commission,
NUREG/CR 3941, 1983.
COM 85 Camp Dresser and McKee, Inc., Roy F. Weston, Inc.,
Clement Assoc., Inc., "Remedial Investigation Study
for the Montclair/West Orange and Glen Ridge/
New Jersey Radium Sites," Vol. I, and II and
appendices, EPA Contract No. 68-6939, U.S. Environ-
mental Protection Agency, New York, September 13, 1985.
d'a 21 d'Aguiar, H.D., "Radium Production in America I,
"Chemical and Metallurgical Engineering," 25,
p. 825-828, 1921.
Di 82 Dietrich R.V., et al, American Geophysical Institute
Data Sheets, AGI Sheets 17.1-3, 1982.
Do 30 Doerner, H.A., 'Processes for Extracting Radium from
Carnotite," Bureau of Mines Report 3057, 1930.
Er 89 Erikson, R., and Serne, R.J., "Mineralogy and Radium
Distribution by Linear Density Gradient Methods on
Glen.Ridge, NJ Soil," Battelle Pacific Northwest
Laboratory Report under EPA Contract No. DW 96931162,
April 1989.
Fi 88 Finkelman, R.3., U.S. Geological Survey, personal
communication, 1988.
Fo 73 Fordham, A.W., "The Location of Iron-55, Strontium-85,
and Iodine-125 Sorbed by Kaolinite and Dickite
Particles," Clays and Clay Minerals, 21, pp. 175-184,
1973.
Fo 74 Folk, R.L., "The Petrology of Sedimentary Rocks,"
Hemphill Publishing Company, Austin, Texas, 182 p.,
1974.
Fr 58 Frondel, C., "Systematic Mineralogy of Uranium and
Thorium," U.S. Geological Survey Bulletin 1064, 1958.
46
-------
Ga 59 Garrels, R.M. and Larsen, E.S., "Geochemistry and
Mineralogy of tne Colorado Plateau Uranium Ores," CJ.S
Geological Survey Professional Paper 320, p. 54, 1959,
He 58 Heinrich, "Mineralogy and Geology of Radioactive Raw
Materials, McGraw-Hill, Inc., New York, 1958.
Hu 88 Huffman Laboratory, "Report on Chemical Analysis of
Soil Samples from New Jersey Superfund Sites," Lab
Report No. 054238 fro EPA, 4 April 1988.
Ka 86 Katz, J.J., Seaberg, G.T., and Morse, L.R., "The
Chemistry of the Actinide Elements," 2nd Ed., Chapman
and Hall Ltd, New York, 1986.
Ki 82 Krishnaswami, S., W.C. Graustein and K.K. Turekian,
"Radium, Thorium and Radioactive Lead Isotopes in
Groundwaters: Application to the In-Situ
Determination of Adsorption-Desorption Rate Constants
and Retardation Factors," Water Resources Research,
18, No 6, pp. 1663,1675, 1982.
La 84 Landa, E.R., "Geochemical and Radiological
Characterization of Soils from Former Radium
Processing Sites." Health Physics, Volume 46,
No. 2, p. 385-394, 1984.
La 86 Landa, S.R., Miller, C.L., and Updegraff, D.M.,
"Leaching of Radium 226 from Uranium Mill Tailings by
Sulfate-Reducing Bacteria," Health Physics, v. 51,
n. 4, pp. 509-518, 1986.
Li 84 Lieberman, R., ed., "Eastern Environmental Radiation
Facility Radiochemistry Procedures Manual," U.S.
Environmental Protection Agency Report 520/5-84-006,
June 1984.
Ma 83 Mattigod S.V. and Ervin, J.O., "Scheme for Density
Separation and Identification of Compound Forms in
Size-Fractionated Ash," Fuel, v. 62, p. 927-931, 1983.
Ne 83 Neiheisel, J. "Quantitative Mineral Assessment and
Radionuclide Retention Potential of Atlantic 3800
Meter Nuclear Waste Dumpsite Sediments,"
EPA 520/1-83-003, 1983.
47
-------
Ne 87 Neiheisel, J., "Characterization of Radium
Contaminated Soil at Montclair/Glen Ridge, New Jersey,
for Remedial Planning," EPA Inhouse Report, Office of
Kadiation Programs, Washington, DC, 1987.
Ni 84 Nirdosh, I., Muthuswami, S.V., and Baird, M.H I.,
Radium in Uranium Mill Tailings" - Some Observations
on Retention and Removal," Hydrometallurgy, 12,
-1-31 — 176, 1984.
Ni 87 Nirdosh, I., Trembley, W.3., Muthuswami, S v and
R2S?S2n^?:2"o"A?sor?t:ion"De8cci^tion Studies on the
Radium-Silica System," Scnool of Engineering, Lakehead
University, Thunder Bay, Ontario, Report Number
P7B-5E1, 1987.
NJEPD 87 New Jersey Department of Environmental Protection
Technical Background Information Report for Soil
Blending Program, Baker, Inc. Report, 1987.
01 86 Olsson, R.K., "Geological.Analysis of and Source of
the Radium Contamination at the Montclair, West
Orange, and Glen Ridge Radium Contaminated Sites."
Department of Geological Sciences, Rutgers University-
New Brunswick, New Jersey, 1986. "ivetait^
On 81 Onishi, Y., Serne, R.J., Arnold, E.M., Cowan, C E
and Thompson, F.L., "Critical Review: Radionuciide
Transport, Sediment Transport, and Water Quality
Mathematical Modeling; and Radionuciide
Adsorption/Desorption Mechanisms," Battelle Pacific
Northwest Laboratory, PNL-2901, p. 339, 1981.
Pi 86 Pingitore, N.E. and M.P. Eastman, "The Coprecipitation
of SR2+ with Calcite at 25°c and 1 atm,"
Geochimica et Cosmochimica Acta, 50, pp. 2195-2203,
1986.
Ri 88 Richardson, W.S., "An Interim Report for VORCE-
Particle Size Distribution, Radiochemical
Distribution, and Chemical Wash Studies on the
Contaminated Soils from Montclair and Glen Ridge New
Jersey," December 15, 1988.
Ri 89 Richardson, W.S., "An Interim Report for VORCE: Acid
Extraction Studies on the Contaminated Soils from
Montclair and Glen Ridge, New Jersey, March 15, 1989.
48
-------
Se 38 Serne, R.J., Erikson, R.L. and Young, J.F., "Studies
at Battelle Northwest Laboratory on Mineralogy of
Radioactive Soil from Montclair/Glen Ridge, New Jersey
Sites/" personal communication, 1988.
Sh 84 Shoesmith, D.W., "The Behavior of Radium in Soil and
in Uranium Mine Tailings," Atomic Energy of Canada
Limited, AECL-7818, 1984.
Si 87
•LJl.'tt.LW'M-XAf m_t*wU I \J Jm ^ / .L .X V ^ *
Simon and Schuster's Guide to Rocks and Minerals,
Simon and Schuster, Inc., New York, 1987.
Ta 47 Tanner, C.B., and Jackson, A.L., "Nomographs of
Sedimentation Times for Soil Particles Under Gravity
or Centrifugal Acceleration," Soil Science Providings,
pp. 60-65, 1947.
Th 88 Thomas, E.F., "X-ray Diffraction Analysis of Silt/Clay
Fractions from Montclair/Glen Ridge Sites," University
of Nevada (Las Vegas), Personal Communication, 1988.
USDA 75 U.S. Department of Agriculture, Handbook No. 436, 1975,
49
S. Government Printing Office! 1989-718-528
------- |