CHAPTER 16
RANKING AND SELECTION OF CHEMICALS OF CONCERN
ABSTRACT
This chapter describes the ranking method used to determine which chemicals were to be
evaluated during Phase II of the Savannah River Site (SRS) dose reconstruction project. The
method involved comparing amounts of chemicals that could have been released to the
environment in a worse-case situation to those in air or water that would not be expected to cause
adverse health effects. The comparison resulted in a ratio that indicated whether there was a
sufficient quantity of a chemical onsite to have presented a hazard. From these ratios, we
concluded that is was important to derive a source term that described how much, when, where,
and in what form the following chemicals were released from the Site: arsenic, benzene,
cadmium, chromium, coal and coal ash, hydrogen sulfide, hydrazine, lead, manganese, mercury,
nickel, nitric acid, nitrogen dioxide, sulfur dioxide, and zinc. Chapters 17 and 18 describe the
estimates of the amount of chemicals released to the air and surface water and the information
used to develop them. Chapters 19 and 20 describe additional environmental and effluent
monitoring information used to develop or support chemical release estimates.
PURPOSE
Thousands of chemicals have been used at the SRS over the years., We used a ranking
method to determine which chemicals were important to evaluate for Phase II of the dose
reconstruction study because they may have been discharged to the air or water in quantities that
could have posed an offsite hazard.
We found documents and records that describe how chemicals were used, stored, purchased,
and disposed of as a part of specific operations. We used essential materials lists, purchasing
records, and inventories to try to determine what chemicals were onsite, when, and in what
quantities. From this information, we developed a list of potential chemicals of concern and
subjected the chemicals to a ranking exercise.
We determined a ratio for each chemical that compared an approximation of the amount of
chemical that may have been in the environment to the concentration in water and air that was not
expected to cause adverse health effects. To develop the ratios, we derived the amounts of
chemicals that may have been in the environment from inventory amounts using conservative
release fractions and air dispersion and water dilution factors. We used the term toxic levels of
concern or toxicity values to refer to the threshold concentrations in water and air that were not
expected to cause adverse health effects. The ratio indicates whether there was a sufficient
quantity of a chemical onsite to have presented a hazard. The more toxic or carcinogenic a
substance, the smaller the amount dispersed or released needed to exceed toxic levels of concern.
These ratios involved several very conservative assumptions and represent a worse-case or upper-
bound situation.
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-2 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
RANKING METHOD
Inventory Information
A list of potential chemicals of concern was compiled primarily from three sources: (1) a list
of 34 essential materials in a Survey of Effluent and Environmental Monitoring at the SRP, from
1973 (Reinig et al. 1973), (2) the Company Chemical Inventory (CCI) Report compiled by Du
Pont in the 1970s (Du Pont 1979), and (3) the 1994 Chemical Information and Inventory System
Database of materials onsite (Information Systems Engineering 1994). The latter two inventories
were developed to comply with U.S. Environmental Protection Agency (EPA) reporting and
listing requirements and are both described in more detail in the Phase I Task 3 Report (Meyer et
al. 1995) and below. In addition, we used lists of materials in safety analysis documents and other
documents that described processes and operations. Materials listed in the SRS Ground Water
Monitoring Program鈥檚 quarterly report (Westinghouse 1994) as being found above levels of
concern in groundwater were also included.
The May 1959 Monthly Progress Report from the Works Technical Department reported
that in response to a request by the Atomic Energy Commission Division of Biology and
Medicine, a survey of potentially hazardous materials used in bulk quantities at SRS was made.
The survey was said to identify a total of 48 potentially hazardous materials on the plant. The
monthly report stated that these survey results were issued as a report (Du Pont 1959). Tom
Cavenaugh and the SRS Central Records staff conducted a search of the central records and the
PINT database for the 1959 report during Phase I in 1994 when mention of the survey was first
found in a monthly report. They were unable to find a report of the survey
Radiological Assessments Corporation (RAC) team members interviewed many former and
current Site workers with knowledge of chemical usage, receipts, and records in Phases I and II.
They agreed that purchasing records would not be useful and purchase order and related records
would not have been retained far enough back in time to be useful. Some essential materials
ledgers were found and were used to develop the list of chemicals of potential concern. Inventory
amounts were all converted from pounds, tons, or grams per week or month to kilograms per year
using 2000 lb ton鈭?1, 0.4536 lb kg鈭?1, 1000 g kg鈭?1, 12 mo y鈭?1, and 52 wk y鈭?1 conversions.
The list of materials in the 1973 Survey of Effluent and Environmental Monitoring at the
SRP included estimates of the amount of material used each month (Reinig et al. 1973). These
estimates provided the largest amounts listed in any of the inventories found, for several
materials.
Du Pont developed an index of chemicals at the SRS, called the CCI Report (Du Pont 1979),
in the mid-1970s to comply with the Toxic Substances Control Act. The index was a list of about
400 chemicals that were present onsite. A printout of the index can be found in Appendix A of
the Task 3 Report for Phase I (Meyer et al. 1995). The inventory cataloged each chemical by
name; registration number; jurisdiction values (whether the material is in a published inventory,
reportable, user exempt, etc.); quantity; and date the material was added and/or deleted from the
inventory. Most of the chemicals were identified as a chemical synthesized at the Site, imported
onsite, a process chemical, a support substance that does not end up in the commercial product,
an intermediate chemical, a laboratory chemical, a nonwaste impurity, a waste by-product, or a
research chemical. Quantities were given by letter designations corresponding to ranges of less
than 454 kg; 454 to 4540 kg; 4540 to 45,400 kg; 45,400 to 454,000 kg; 454,000 to 4.54 million
Evaluation of Materials Released from SRS 16-3
Ranking and Selection of Chemicals of Concern
kg; 4.54 to 22.7 million kg; 22.7 to 45.4 million kg; 45.4 to 227 million kg; 227 to 454 million
kg; over 454 million kg; and quantity not reported or reported under another synonym. For the
ranking, the upper bound of the corresponding range was used as the inventory amounts for
materials. 鈥楺uantity not reported鈥? was the most common entry. The Du Pont inventory contained
quantity entries for only 65 chemicals, which accounted for about 15% of the chemicals listed. Of
these, seven chemicals (copper, dioctyl phthalate, gadolinium nitrate, lead, mercury, nickel and
nickel sulfate, and uranium) were listed as being present in quantities less than the quantities
reported in the Chemical Information and Inventory System (CIIS) database. Amounts for five
chemicals (cadmium, ceric ammonium nitrate, ferrous sulfamate, phosphoric acid, and potassium
permanganate) in the Du Pont inventory were consistent with the CIIS database. Quantities of
nine chemicals (hydrazine, hydrofluoric acid, hydrogen sulfide, hydroxylamine nitrate,
hydroxylamine sulfate, manganese nitrate, napthalene, oxalic acid, and sodium hydroxide) were
listed in the Du Pont database with quantities greater than CIIS database amounts. The 44
chemicals presented below were listed with quantities in the Du Pont database, but they were not
included on the list to be ranked because, for various reasons, they are very unlikely to present an
offsite hazard.
Acetylene Aluminum Aluminum nitrate
Aluminum sulfate Ammonium sulfamate Aquadag
Ascorbic acid Calcium Calcium fluoride
Several paints, thinners and primers Calgon Carbon
Calcium oxide Dodecane Ethylene glycol
Formic acid Gelatin Glucose
Helium Iron Methane
Nitrogen Oxygen Polyethyleneimine
Propane Sodium aluminum silicate Sodium carbonate
Sodium hypochloride Sodium nitrate Sodium nitrite
Sodium salicylate Sodium sulfite Steel
Sucrose Urea
The CIIS database, maintained by Westinghouse SRS, is a comprehensive listing of all
hazardous materials used or stored in the workplace. The database was designed to fulfill the
requirements of Worker-Right-To-Know legislation and to help organize inventory data for EPA
reporting requirements. It is an ORACLE relational database that runs on the VAX computer.
Information in the database is derived from material safety data sheets (MSDSs) that are also
stored in hard copy form in notebooks. The database is designed to be updated each year and to
provide information for annual reports. The SRS CIIS database is not historical but contains
materials currently inventoried onsite. It does not provide specific information on toxicity, use,
monitoring information, or time periods of use.
Two products were sought from the CIIS. Cheryl Hardy and John Harris, Westinghouse SRS
Environmental Protection Department personnel, selectively extracted this information from the
database. One product was a list of all of the chemicals in the database with a Chemical Abstract
Service (CAS) number. The material name, CAS number, average amount, and number of hits
(number of times the material is reported onsite) were printed for 1994. A list of about 4000
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-4 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
chemicals resulted. This list is printed in Phase I Task 3 report (Meyer et. al. 1995). The CIIS list
appeared to be the most complete listing of chemicals. It seemed prudent to start with a
comprehensive list and eliminate materials not of concern for dose reconstruction. Although some
of the chemicals used onsite at one time may not have been stored onsite in 1994, many
chemicals used historically were included in this large database. The second product was the
result of several data extractions. The resulting database of chemicals was transferred to
Radiological Assessments Corporation (RAC) on disk, converted to a FoxPro format, and
incorporated into the RAC SRSCHEM database developed for Phase I. This database included
information on the chemical name, manufacturer synonyms, symbols, quantities reported, and
uses onsite.
The following fields were entered into FoxPro for about 30,000 records:
MSDS number
CAS number
Material name
Manufacturer
Form (liquid, gas, solid, or mixture)
Maximum amount (onsite in 1994)
Average amount (onsite in 1994)
Synonyms
Formula
EPA storage code
A Above ground tank H Silo
B Below ground tank I Fiberdrum
C Tank in a building J Bag
D Steel drum K Box
E Plastic or nonmetallic drum L Cylinder
F Can M Glass bottle
G Carboy N Plastic bottle
Usage code
1 Import
2 Byproduct
3 Impurity
4 Reactant
5 Repackaged
6 Laboratory
7 Manufacturing aid
8 Processing aid
9 Other use.
The department, area, building, and room where the chemical was currently used or stored
could have been extracted from the CIIS. However, because many of the release point locations
were classified and the 鈥榓rea鈥? provided sufficient information about the location of the release,
this information was not included to expedite the security review.
Evaluation of Materials Released from SRS 16-5
Ranking and Selection of Chemicals of Concern
Several subsequent extractions of this database were conducted. One extraction targeted all
chemicals on EPA lists and other lists including, Superfund Amendments and Reauthorization
Act (SARA) Title III; Comprehensive Environmental Response, Compensation, and Liability Act
(CERCLA); Resource Conservation and Recovery Act (RCRA), Toxic Substances Control Act
(TSCA); and a list of reproductive hazards. The extractions excluded chemicals with no hazard
rating (no acute toxicity, chronic toxicity, flammability or reproductive hazard). We made a
second extraction of all chemicals in quantities greater than 50 lb. This, we felt, captured all of
the chemicals that might be of concern based on toxicity and quantity. We then conducted a line-
by-line review of the database to delete materials that were unlikely to present an offsite hazard,
such as concrete, plastic materials, and solid building materials. Records that involved the
following materials and/or synonyms for them were deleted:
Joy dishwashing Markers Crayons
Glass cleaner Copier Typewriter
Glue Analytical standard Deep Woods Off
Paint Correction fluid Ajax
Welding rod Battery Comet
Toner Air freshener Carpet
Saddle Soap Ink Crayons, pens
Mounting solutions GC column packing Mortar
Weatherstrip Cutting oils Enamels
Adhesives Polyurethane Shellacs
Resins Welding wire Salt solutions
Cements Sealants Buffer solutions
Lubricants Oil absorbants Gaskets
Insect repellents Spill absorbents Office supplies
Small volume cleaners Abrasives Laboratory supplies
Steel Pump oils Grease
Cast iron Art supplies Tape
Deodorizers Acetylsalicylic acid Solder and flux
Inert solids Sand and aggregate Scintillation fluids
Salts Insulation Lotions
We then reviewed the list of deleted items line-by-line and added back in items that could
not be identified by name (for example 鈥榝uel additive number 13,鈥? or 鈥榓ccelerator SFA342鈥?). A
total of 168 materials were listed in the database by trade name and with insufficient
manufacturer or synonym information to allow identification. A list of these was sent to Dick
Reynolds, Manager of the Chemical Commodity Group at SRS, for further identification. The
task of looking at these materials was assigned to Mark Lloyd, who provided synonym,
carcinogenicity, or reproductive hazard (yes or no) information and the hazardous constituents of
each material if known. Based on this information, trade name materials were deleted, retained as
discrete materials, or combined with and listed under their most hazardous component.
The following groups of materials were removed from the list of chemicals of potential
concern. The following materials contain noncarcinogenic components and relatively nontoxic
components or ingredients:
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-6 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
Mod U Formula Crimstar 30
CP 220 ChilWet Solucom Clear
Cronwear Eagle Electrostatic solution
Apiezon Q Compound Mojave
Inhibitor 526C Snapback
Magicfloc 985N Corrosion Inhibitor GC Formula
Sediperse All Drewguard 315
Polysperse Plus Monosodium titinate slurry
Bycothane 300 Formed Molecular Sieve
Durapox Primer CI Mastic
Industrial Grade PVC solvent Paraffin Wax
The following are relatively nontoxic and are used as food additives, medicines, anesthetics:
Sodium nitrate Sodium nitrite
Potassium nitrate Polyethylene glycols
Propylene glycol Phospholene
Sodium sulfite Fumaric acid
Hexane Boric acid
Trisodium phosphate Sodium sulfite
Sodium phosphate Nitrous oxide
Ascorbic acid
The following materials are used in office equipment, primarily copiers. Some of these materials
contain usually less than 1% of formaldehyde, which is a carcinogen:
Cronar reducer Itek activator
Cronolith liquid blender
The following materials contain components that are not chronic hazards, have relatively low
acute toxicity, and are not carcinogenic:
Evaluation of Materials Released from SRS 16-7
Ranking and Selection of Chemicals of Concern
Doubleteam Synthetic sludge
Alconox cleaner (sulfonate) Exotherm (calcium chloride)
Cryosan (bleach) Sulfamic acid
Chloroethane Biphenyl
Calcium Cerium
Barium Hydrogen peroxide
Graphite Oils and lubricants
Paraffin Aluminum
Chloride Iron
Phosphate Selenium
Silver Silver nitrate
Calcium hydroxide Butyl stearate
DPD free Cl reagent
(disodium phosphate)
The following materials contain components that are not carcinogenic but are sensitizers or
irritants:
Rynothane activator Concresive liquid hardener
(diisocyanate) (polyaminoamide)
Fortify (acrylic polymers 11鈥?25%) Virchem 931 hardener
HitCZO Part A (diglycidyl resorcinal ether) Scotchcast (vinylcyclohexanedioxide)
R process gum (<1% formaldehyde) Diphenyl methane diisocyante
diethyl hexyl phosphoric acid Cloro-m-cresol
Duraskid Epon
Phosphates Neodynium oxide
Lanthanum Polyphosphates
Terphenyls Erbium oxide
Polyethyleneamine Saframine
LTC accelerator CreteLease
ChemTreat
The database contained many paints, paint thinners, and epoxies that may be hazardous in a
liquid form but once dried would not be expected to pose an offsite hazard. These materials may
have been stored in large amounts, but they were likely to have been used in relatively small
amounts. Although resins and enamels may contain toxic components, they are generally used for
coating and painting structures and equipment and would be found in a polymerized state or
stored as active ingredients, separated in small quantities. Epoxy components known to be potent
sensitizers in occupational settings were listed above as sensitizers.
In another data extraction, substances with maximum annual inventory amounts of less than
0.25 lb were deleted. This extraction eliminated the thousands of hazardous materials that were
used in very small volumes, such as laboratory reagents, analytical standards, or samples. These
materials would have been present and used in such small amounts that they would not pose an
offsite hazard. The cut off of 0.25 lb was chosen from the results of an assessment of a
hypothetical release of TCDD, the most potent carcinogen and developmental toxin ever
evaluated by the EPA. The proposed cancer potency factor for TCDD yields a daily intake level
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-8 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
corresponding to a 10鈭?6 cancer risk level of 0.01 picograms per kilogram per day (pg/kg/day). In
this assessment, we attempted to back calculate an amount that if released over a very short time,
would be low enough not to be harmful. If 100 g (0.22 lb) of a chemical was released at one time
and a conservative dispersion factor (蠂/Q) of 10鈭?5 sec m鈭?3 was used, the resulting exposure would
correspond to a maximum daily intake of about of 0.00065 pg/kg/day. This value is about 15
times less than the daily intake level for a 10鈭?6 cancer risk. This justifies 0.25 lb as a very
conservative cut off amount. For similar exercises, others have chosen 1 or 0.5 lb, which are also
justifiable limits. Using 0.25 lb seemed to accomplish our goal of eliminating most of the
chemicals used in very small amounts.
Again, the list of deleted materials was reviewed line-by-line and exceptionally potent
carcinogens (such as some of the polyaromatic hydrocarbons and pesticides that were listed with
inventory amounts of less than zero but were likely to have been used onsite in the past) were
added back to the list. Many of the materials remaining on the list with inventory amounts of less
than 0.25 lb were not derived from the CIIS database. They were taken from essential material
lists or the list of groundwater contaminants of concern, so they were not eliminated as a result of
this data extraction.
Inventory amounts were not available for all materials. Many materials in the CIIS database
had a zero inventory amount. The staff at Westinghouse SRS who maintained the database in
1994 believed that the zero amounts on our ranking list were a result of less than 0.001 lb being
reported and rounded to zero. Benzo(a)anthracene, benzo(a)pyrene, benzofluoranthene,
benzoperylene, and chrysene were listed in the database with inventory quantities of zero or less
than 0.001. These were probably used as analytical standards and were evaluated using a quantity
of 0.001.
Gasoline and diesel fuels were added back to the list and were included in the ranking
because they may have leaked from tanks into soil and groundwater. A large amount of developer
was currently and has been historically used at the Site. Photographic developer was used in the
Site鈥檚 photo labs to develop photos and x-rays of welds. Developer and toner were also used in all
of the copy machines and in other equipment onsite and it may have been ordered in large
quantities by some departments. It is unlikely that developer would have caused an offsite hazard,
and entries totaling approximately 11,043 lb were deleted from the list. We also deleted many
materials that contain small amounts of carcinogens or are carcinogens but were used in small
quantities onsite. For example, we deleted Joy dishwashing liquid (which contains ethanol),
upholstery cleaners containing chlorinated solvents, nickel-containing lubricants, roofing tars,
chromium-containing corrosion preventatives used in small amounts (less than 2 lb), and
formaldehyde-containing solutions used in small quantities. We also deleted EDTA and similar
chelators from the list.
Although they present little or no chronic health hazard, several caustics (such as nitric acid,
sulfuric acid, sodium hydroxide, and hydrochloric acid) were used and stored in very large
quantities; therefore, we retained them on the list for the ranking. Other materials that generally
present no chronic health hazard unless large amounts are ingested or they were used in quantities
or ways that releases or transport offsite seems unlikely include waxes, Freons, antifreezes,
diatomaceous earth, carbon monoxide, calcium carbonate, carbon, propane, and other inert and
asphixiant gases. Fly ash, a large volume of which exists onsite, may have been stored outdoors
and may have been subject to atmospheric dispersion or runoff into water; therefore, it was
Evaluation of Materials Released from SRS 16-9
Ranking and Selection of Chemicals of Concern
retained on the list. Because polyaromatic hydrocarbons and toxic metals may have been leached
or contained in rainwater runoff from coal piles, coal inventories were also retained.
Although we used the initial data extractions for the CIIS from Phase I and the FoxPro
database derived from the CIIS database was the starting point for the ranking, subsequent
treatment of the list of chemicals differed. Therefore, the list of potential chemicals of concern in
the Phase I Task 3 report (Meyer et al. 1995) is not the same as the list developed, with more
care, in this phase of the study.
Toxicity Values
Toxicity values were obtained or derived from the EPA鈥檚 1995 HEAST Tables; 1995
National Ambient Air Quality Standards (NAAQS) in the Code of Federal Regulations, Part 50;
the 1995 Integrated Risk and Information Systems (IRIS) Database; the Agency for Toxic
Substances and Disease Registry鈥檚 (ATSDR) Toxicity Profiles; and Workplace standards
published by the American Conference of Governmental Industrial Hygienists (ACGIH) in 1994
and by the Occupational Safety and Health Administration (OSHA) or National Institute of
Occupational Safety and Health (NIOSH) in 1994. The documents and databases used are listed
in the references for this chapter. The toxicity values are listed in Appendix C3, Tables C3-1 and
C3-2. We used the shaded values to calculate the ranking ratios.
The tables include a qualitative developmental effects designation (D) and reproductive
effects designation (R). The number of Ds and Rs corresponds to potency or certainty of the
information available. A chemical with a DDD or RRR designation is more potent or more
positive human data were available for the chemical than for chemicals with a DD, RR, D, or R
designations. The designations were determined from information obtained from a search of the
Reprotext铮? Database, information found in ATSDR Toxicity Profiles, and data reported on
developmental and reproductive hazards in the workplace in ACGIH (1994), NIOSH (1994), or
Lewis (1993).
A carcinogenicity designation was listed for all of the chemicals that might be carcinogenic. In an
effort to provide consistency, published EPA designations were used when available, followed by
International Agency for Research on Cancer (IARC) designations, ACGIH categories, and any
new information from the National Toxicology Program (NTP).
The EPA has evaluated many of the chemicals in Tables C3-1 and C3-2 for carcinogenicity
and has assigned the chemical a designation based on the weight of evidence. There are five
designations or groups:
Group A- Known human carcinogen. There is sufficient evidence from human
epidemiological studies to support a causal association between the substance and
cancer.
Group B - Probable human carcinogen. There is limited evidence of human carcinogenicity
based on epidemiological studies but sufficient evidence of carcinogenicity based
on animal studies.
Subgroups
B1 - Limited epidemiological but sufficient animal data
B2 - With sufficient animal data but inadequate or no data from human
epidemiological studies.
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-10 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
Group C - Possible human carcinogen. There is limited evidence in animals and no human
data.
Group D - Not classifiable, inadequate data or no data.
Group E - No evidence for carcinogenicity. Negative test results, usually in at least two
animal species, or adequate negative epidemiological data.
Many of these chemicals have not been evaluated by the EPA and are classified as Group D
because their carcinogenicity has not been adequately determined or are currently under
evaluation. Some of the chemicals have not been tested for carcinogenicity.
The IARC has a similar ranking scheme. The scheme defines sufficient evidence as when a
causal relationship has been demonstrated in humans; limited evidence as when there is causal
relationship but chance, bias, or confounding factors cannot be discounted; inadequate evidence
as when available studies cannot determine the carcinogenicity; and evidence suggesting a lack of
when there are adequate studies that are negative. IARC groups chemicals into four groups:
Group 1 - Compounds are carcinogenic to humans.
Group 2 - Compounds cause cancer in animals.
Subgroups
2A - Probable carcinogen in humans. There is limited evidence of
carcinogenicity in humans but sufficient evidence in animals and causal
relationship is clear in multiple species or strains in independent
experiments.
2B - Possible carcinogens in humans. There is insufficient evidence in
humans or animals.
Group 3 - Compounds are not classifiable. There is inadequate evidence and neither the
presence nor absence of a carcinogenic effect can be demonstrated.
Group 4 - Compounds are probably not carcinogenic. There are adequate studies involving
at least two species to suggest that the substance is not carcinogenic.
The ACGIH also has carcinogenicity designations for chemicals used in the workplace:
A1 - Confirmed human carcinogen
A2 - Suspected human carcinogen
A3 - Animal carcinogen
A4 - Not classifiable as a human carcinogen
A5 - Not suspected as a human carcinogen.
We also consulted reports published by the NTP of cancer bioassays that report 鈥榗lear鈥?
evidence, some evidence, equivocal evidence, no evidence, and inadequate evidence of
carcinogenicity. If available, an EPA designation was listed. If an EPA designation was not
available or was Group D, the IARC or ACGIH designation was entered. We listed several
chemicals that are not yet designated by the EPA, IARC, ACGIH, or examined by the NTP but
are mutagens or structurally resemble other carcinogens with the carcinogens in Table C3-2.
For noncarcinogens, we used a reference dose (RfD) for ingestion or a reference
concentration (RfC) for inhalation. The RfC or RfD is defined by the EPA as a provisional
estimate (with uncertainty spanning perhaps an order of magnitude) of the daily exposure to the
Evaluation of Materials Released from SRS 16-11
Ranking and Selection of Chemicals of Concern
human population (including sensitive subgroups) that is likely to be without appreciable risk of
deleterious effects during a portion of a lifetime in the case of subchronic RfD or RfC or during a
lifetime in the case of a chronic RfD or RfC (EPA 1995). When available, we used chronic RfCs
in units of milligrams per cubic meter air for continuous exposure and chronic RfD values in units
of milligrams per kilogram per day. The oral RfD was converted into a corresponding water
concentration in milligrams per liter by the equation:
脳 (16-1)
oral RfD in mg / kg / day 70 kg
mg / L in water =
2 L / day
which assumes a average human body weight of 70 kg and average water consumption of 2 L
day鈭?1. The RfC, in milligrams per cubic meter or micrograms per cubic meter, corresponds to an
ambient air concentration for continuous, 24 hr day鈭?1 exposure.
For carcinogens, cancer potency factors, called slope factors, have been determined by the
EPA for many environmental carcinogens. The slope factor is an upper-bound estimate. It is
estimated using mathematical extrapolation models, most commonly the linearized multistage
model that estimates the largest possible linear slope, within the 95% confidence limit. The EPA
believes true cancer risk to humans is not likely to exceed this upper limit and is likely to be
lower. The slope factor is expressed as risk per unit dose in units of risk per
milligram/kilogram/day (EPA 1995).
Another useful value is the unit risk value, which is the risk per unit concentration. The unit
risk for inhalation is the risk per concentration unit in air expressed as risk per micrograms per
cubic meter. The unit risk for oral exposure is the risk per concentration unit in water expressed
as risk per micrograms per liter. The unit risk is calculated by dividing the slope factor by the
body weight of 70 kg and multiplying by an average breathing rate of 20 m3 day鈭?1 for air or 2 L
day鈭?1 average consumption for water (EPA 1995).
The risk-specific air or water concentrations can be estimated using the unit risk value at a
given risk level. We choose to use a risk level of 1 in 100,000 or 10鈭?5. The concentration in air in
micrograms per cubic meter corresponding to
10-5
a lifetime cancer risk of 10-5 = (16-2)
.
unit risk per 碌g / m 3
Risk-specific concentrations in drinking water can be estimated from the oral slope factor.
The water concentration corresponding to an upper-bound increased lifetime cancer risk of
1 脳 10鈭?5 is calculated as
10 -5 脳 70 kg (16-3)
the risk per mg/L in water = .
slope factor (mg/kg/day) 脳 2 L/day
These values are published in units of micrograms per cubic meter, milligrams per cubic
meter, micrograms per liter, and milligrams per liter. There are 106 碌g g鈭?1, 109 碌g kg鈭?1, 103 mg
g鈭?1, 106 mg kg鈭?1, and 103 碌g mg鈭?1. Appropriate conversion factors were applied so that
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-12 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
concentrations corresponding to a 10鈭?5 risk level, drinking water standards, workplace standards,
and other toxicity values could be compared in Tables C3-1 and C3-2.
The NAAQS are primary standards designed to protect public health. Six have been
established to date. The NAAQS include an extensive database that has been rigorously reviewed.
The primary NAAQS and the inhalation RfC have essentially the same function, and the EPA has
stated that, except for lead, the NAAQS with annual averaging times can be used instead of the
RfC.
The time weighted average (TWA) threshold limit value (TLV) established by the ACGIH
or the permissible exposure limit (PEL) established by OSHA (in milligrams per cubic meter or
fiber per cubic centimeter) were used when EPA toxicity values were not available. Because the
TLVs are designed to protect healthy workers, these were divided by 10 to help account for the
fact that some members of the general public (especially young children, elderly people, or
people with preexisting health conditions) are more sensitive to the toxic effects of chemicals.
Drinking water standards (specifically the EPA maximum contaminant levels [MCLs]) were
used if the RfD, oral slope factor or drinking water unit risk value was not available. We
preferred to use the toxicity values in the order presented in Tables 16-1 and 16-2.
Table 16-1. Order of Preference for Carcinogens
Air Water
鈭?5 unit risk 鈭?5 unit risk
10 10
RfC RfD converted to a water
concentration
NAAQS MCL
TLV 梅 10 or PEL 梅 10
Table 16-2. Order of Preference for Noncarcinogens
Air Water
RfC RfD converted to water
concentration
NAAQS MCL
TLV 梅 10 or PEL 梅 10
Because they are more relevant to environmental contaminants, we first chose to use the
EPA鈥檚 RfD or concentration, followed by unit risk values, NAAQS values, MCLs, then TLVs. In
general, the concentrations corresponding to a risk level of 10鈭?5 are the lowest concentrations of
concern. However, in a case where the MCL, TLV/10, RfD, or RfC was lower, we used the
lower, more stringent level to calculate the ranking ratios.
In some cases, a single toxicity value and combined inventory quantity was given to a group
or class of compounds that contain a common element or component. Although different valence
states of metals may vary tremendously in their toxicity, for the purposes of screening, the
toxicity value we used was based on the toxicity or carcinogenicity of the most toxic or elemental
form depending on the data available. For some materials, like manganese, sodium, and
potassium compounds, more toxicity information is available on specific compounds, so we used
it. We combined compounds of arsenic, cadmium, chromium, mercury, nickel, and tin for
screening. For example, the chromium compounds include chromium metal; chromium oxide;
Evaluation of Materials Released from SRS 16-13
Ranking and Selection of Chemicals of Concern
dichromic acid; lead chromate; potassium chromate; potassium dichromate; sodium chromate;
sodium dichromate; chromium azo dye; chromic acids and salts (acetate, chloride, nitrates,
sulfates); and 51Cr. We used the toxicity values for carcinogen hexavalent chromium.
If both the anion and the cation of a compound are responsible for the toxicity, then we
considered both. For example, zinc cyanide was included in both the zinc and cyanide inventory
amounts, and we included both compounds in the ranking. Some metal salts (like mercuric
nitrate) were used in large quantities, so we considered this compound separately from elemental
mercury.
Many radionuclides, such as 51Cr, are hazardous because of their radioactivity and chemical
toxicity. Uranium, for example, is toxic to the kidney. Although uranium will be evaluated with
the radionuclides of concern, for the purposes of this ranking we treated uranium as a toxic metal
and assessed it using the RfD for kidney toxicity.
Isomers of tetrachloroethane, dichlorobenzene, dichloroethane, dichloroethylene,
dimethylphenol, dinitrotoluene, dichlorophenol, chlorophenol, nitrophenol, trichlorobenzene,
trichloroethane, tetrachloroethane, and trichlorophenol were summed, and we used the toxicity
value for the most toxic or carcinogenic isomer. All of the inventories for various arochlors and
polychlorinated biphenyls (PCBs) were summed and the toxicity values for arochlor 1260 were
used for the PCBs.
We included chemicals used at the Naval Fuel Facility in the ranking. The names of these
materials are classified; however, the materials of concern were not unique to the Naval Fuels
Facility and were already included on one or more of the Site inventories.
Ranking Ratio Calculation
Using inventory amounts and toxicity values described above, we calculated ranking ratios
for 170 chemicals in Tables C3-1 and C3-2. Initially, we did not calculate ranking ratios for 12
chemicals, primarily pesticides, for which there was no inventory amount or for which there was
an inventory amount that seemed far too small. An effort was made in April 1996 to actively
solicit information on inventory amounts from former and current Site employees who might
have knowledge of chemical usage, purchase, storage, disposal, or other relevant operations
onsite. A letter was sent to 118 individuals by post or electronic mail, and 18 individuals
responded. Several respondents agreed that some of the pesticides listed may have been used
onsite in the 1960s and 1970s; however, we did not obtain new or different inventories for any of
the chemicals.
No published toxicity values could be obtained for 11 chemicals. We looked at the MSDSs
for these chemicals in the SRS files and sought toxicity testing data and structure activity
information for these compounds to determine air and water concentrations of potential concern.
Because better toxicity values have not been determined, we used LD50 divided by 100,000
(Layton et al. 1987) as a toxicity value.
Some chemicals (for example, dichlorodifluoromethane and sulfur oxides) are only a
concern if released to the air and inhaled. We did not calculated ratios for water for these types of
chemicals. We did not calculate air ratios for chemicals, such as dichlorobenzidine and
chlorophenol, because toxicity values have only been determined for consumption of water.
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-14 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
We calculated the ranking ratio for materials released to air using the following equation:
Q a ( kg / s) 脳 蠂 / Q (s / m 3 ) 脳 10 9 ( 碌g / kg)
air ratio = (16-4)
Tox a ( 碌g / m 3 )
where
Toxa= air concentration that corresponds to a level of concern. The preferred toxicity values
used are described on page 14.
蠂/Q = atmospheric dispersion factor. The average 蠂/Q, based on tritium monitoring done in
1990, for onsite locations of 5.4 脳 10鈭?8 s m鈭?3 (Cummins et al. 1991), was used
Qa = the release rate
I (kg) 脳 RFa
Qa (kg / s) =
3.16 脳 10 7 s / year
where I = annual inventory amount in kg and Rfa = the release fraction.
Release fractions account for the fact that some materials may be released in total (like volatile
solvents or gases that are put out the stacks) and others (like solids) will not all be released. We
chose to conservatively assume, for this first screening, that 100% of the inventory was used.
Although more material is usually stored than is used, it is reasonable to assume that it all was
used and the entire annual inventory was subject to release each year.
We assigned release fractions according to the volatility of the compound (that is, how
readily it evaporates). For materials released to the air, we chose a release fraction (RFa) of 1.0
for volatile liquids and 0.005 for nonvolatile liquids and solids. It is likely that workers would
have made some effort to conserve reagents, keep lids on solvent baths, and prevent spills and
leaks of volatile substances. However, for the purposes of this ranking, we assumed that all of the
volatile material inventoried was eventually released to the atmosphere and we used the
conservative release fraction of 1 (signifying 100% of the inventory quality was released).
The ranking ratio for materials released to water was calculated using the following
equation:
Q w (kg/s) 脳 10 6 mg/kg (16-5)
ratio w =
DF (L/s) 脳 Tox w (mg/L)
where
Toxw = water concentration that corresponds to a level of concern. The preferred toxicity
values used are described on page 14.
DF = the dilution due to flow of the river. We assumed that the Savannah River is the
surface water of concern and that it was used for drinking water (untreated). Other
activities, such as fishing, boating, swimming, etc., will be considered in later phases
of this work. The 1990 SRS Environmental Report includes a plot of the flow rate of
the Savannah River for 1980鈥?1990. The median value corresponds to a DF of 2.55 脳
105 L s鈭?1 (Cummins et al. 1991).
Qw = the release rate.
Evaluation of Materials Released from SRS 16-15
Ranking and Selection of Chemicals of Concern
I (kg) 脳 RFw
Q w (kg / s) =
3.16 脳 10 7 s / year
We used the release fractions (RFw) 0.02 for volatile liquids, which would tend to evaporate out
of the water, and 0.1 for nonvolatile liquids and solids.
To compare among values, we converted all of the concentrations corresponding to 10鈭?5 risk
levels, MCLs, TLVs, and NAAQS to units of milligrams per cubic meter or milligrams per liter
in the tables. In summary,
I (kg) 脳 RF 脳 5.4 脳 10 -8 s/m 3 脳 10 6 mg/kg 梅 toxicity value mg/m 3
air ratio = . (16-6)
3.16 脳 10 7
If all the toxicity values are in units of milligrams, then the two ranking ratio equations can
be summarized as
inventory amount in kg 脳 release fraction 脳 1.7 脳 10 -9
ratio air = (16-7)
Tox a
inventory amount in kg 脳 release fraction 脳 1.24 脳 10 -7
ratio water = . (16-8)
Tox w
Ranking Results
Tables C3-1 and C3-2 in Appendix C3 list the resulting ranking ratios for 166 chemicals.
Ratios could not be determined for methyl bromide and methyl chloride (two chemicals detected
in groundwater but not used onsite and for which there was no inventory) and ozone, an air
pollutant produced as a byproduct for which there is no inventory. Materials with the highest
ratios are listed in Tables 16-3 through 16-6.
Table 16-3. Chemicals with Ranking Ratios > 1.0
Chemical Ranking ratio Ratio
Coal 254 Water
Hydrazine 36.7 Air
Uranium 1.87 Water
Gasoline 1.77 Water
Hydrazine mononitrate 1.26 Air
Table 16-4. Chemicals with Ranking Ratios < 1.0 but > 0.1
Chemical Ranking ratio Ratio
鈭?1
Tetrachloroethylene 2.29 脳 10 Water
1.10 脳 10鈭?1
Hydroxylamine sulfate Water
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-16 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
Table 16-5. Chemicals with Ranking Ratios < 0.1 but > 0.01
Chemical Ranking ratio Ratio
鈭?2
Benzene 1.04 脳 10 Air
1.20 脳 10鈭?2
Nickel compounds Water
鈭?2
Mercury 1.21 脳 10 Water
1.32 脳 10鈭?2
Sulfuric acid Air
鈭?2
Trichloroethane 1.54 脳 10 Air
2.32 脳 10鈭?2
Chromium compounds Water
鈭?2
Lead compounds 3.25 脳 10 Water
3.53 脳 10鈭?2
Mercuric nitrate Water
鈭?2
Asbestos 9.40 脳 10 Air
Table 16-6. Chemicals with Ranking Ratios < 0.01 but > 0.001
Chemical Ranking ratio Ratio
1.03 脳 10鈭?3
Zinc compounds Water
1.28 脳 10鈭?3
Manganous nitrate Air
1.35 脳 10鈭?3
Copper compounds Water
鈭?3
Nitric acid 1.48 脳 10 Air
2.48 脳 10鈭?3
Aldrin Water
鈭?3
Cadmium 2.52 脳 10 Water
3.44 脳 10鈭?3
Hexanol Air
鈭?3
Chlorine 3.95 脳 10 Air
4.27 脳 10鈭?3
Gadolinium nitrate Air
鈭?3
Coal tar 4.45 脳 10 Water
In addition, we found no inventory amount or the inventory amount seemed far too small, for
19 chemicals:
Aldrin Carbon tetrachloride Chloroethane
Chloromethane DDT Dieldrin
Endosulfan Endrin Chlordane
Heptachlor Hydroxyquinoline Lindane
Toxaphene Trichlorophenol (2,4,5-T) Velpar-L
Chloroethane DDT Chloromethane
PCBs
Because many hazardous, persistent pesticides (like DDT) have been banned for some time,
we might expect little or no inventory amounts although these materials may have been used in
the past. We found little mention of these materials in the periodic reports. We compiled recent
monitoring study results to see if they could help us quantify some of the pesticides of concern.
Although we will could not determine an inventory amount for many of these pesticides and
many could not be adequately evaluated in the ranking, it seemed inappropriate to simply dismiss
these pesticides at this early stage of the ranking simply because of a lack of inventory.
Evaluation of Materials Released from SRS 16-17
Ranking and Selection of Chemicals of Concern
Limitations of the Ranking
This approach to ranking chemicals of concern has several limitations that are important to
recognize. The amount of material listed on an inventory is not necessarily related to the amount
that was discharged out a stack as a result of a process. For example, nitrogen dioxide stored in
tanks or cylinders onsite is in no way related to the amount of nitrogen dioxide produced by a
process and discharged to the environment. A careful examination of processes and materials that
may have been produced and released at certain points of an operation will be another important
part of this work.
Another problem is that the CIIS database contains the inventory for 1994. The amounts
onsite now may be more or less than when facilities at the plant were in full operation. We
compared the amounts in the 1994 CIIS database and the Du Pont Index from the 1970s, and
noted differences. We used the largest amounts (usually those in the listing from 1973 or the
upper value from the range in the Du Pont database). It is unfortunate that we could not locate
additional inventories from the 1950s, 1960s, and early 1970s.
One of the most significant problems with this approach is a lack of information.
Information on inventory amounts and toxicity is missing for many of the materials. Often,
toxicity values are available for inhalation or ingestion but not both pathways. Information on
reproductive and developmental effects is not available for most of the materials. Eleven
materials have no published toxicity values, so we used a very conservative and uncertain value
of the lowest LD50 value reported (usually an oral LD50 in rodents or rabbits) divided by
100,000 (Layton et al. 1987). For several materials on which no toxicity testing has been done,
we used values derived for similar compounds.
Second Stage of the Ranking
After the ranking ratios were calculated, it was obvious that the ratios alone could not
adequately be used to prioritize chemicals of concern. The conservative assumptions used created
unrealistic scenarios for many of the chemicals that had ranking ratios greater than 10鈭?3.
The next step was to further evaluate the chemicals identified in the ranking by considering
environmental fate and transport characteristics, information on chemical use at the plant; and
release potential. We also assessed the chemicals based on the physical and chemical properties
relevant to their behavior in the environment, such as water solubility, volatility, susceptibility to
biodegradation and chemical breakdown, and mobility. These properties can be very important to
environmental behavior. For example, highly water-soluble compounds are generally less likely
to adsorb to soils and sediments, more likely to remain in the water, more biodegradable, and less
likely to volatilize from water. The octanol-water partition coefficient, Kow, is a measure of the
degree to which an organic material will preferentially dissolve in octanol compared to water.
The greater the Kow, the greater the tendency for the material to partition from water to a more
organic phase.
Some chemicals, for example titanium tetrachloride, break down so rapidly that
environmental exposure for people living offsite from this chemical is unlikely. Other chemicals,
for example asbestos, were used in building materials and were not subject to storage in large
amounts, leakage, spillage, or routine release.
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-18 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
Chemicals with ranking ratios greater than 10鈭?3 are evaluated further in the following
paragraphs.
Alcohols
Large amounts of ethanol and hexanol and smaller amounts of propanol, ethoxyethanol, and
other alcohols were used for cleaning and as part of processes at the SRS. The ranking ratios for
these compounds were very low. The highest ratio was for hexanol, with an air ratio of
3.4 脳 10鈭?3. Because toxicity information on hexyl alcohol was limited, it was ranked using the
very conservative LD50/100,000 value. Hexanol is a volatile skin and eye irritant that appears to
cause liver toxicity and other toxic effects similar to those caused by other alcohols. The
availability of documentation specific for this alcohol limited our ability to develop a source term.
It is likely subject to biodegradation and has fate and transport characteristics similar to other
alcohols. Emission estimates in the 1985, 1987, and 1990 Air Emissions Inventory are low, and
significant offsite concentrations would not be expected. Therefore, we did not develop a source
term for this chemical.
Aldrin
Aldrin is a particularly toxic chemical for which we have no reasonable SRS inventory and
usage information. In general, in the U.S., it was used as an agricultural pesticide until 1970 and
for termites until 1987. Aldrin and its toxic breakdown product, dieldrin, are very persistent, not
soluble in water, and tend to bioaccumulate. Further evaluation of this chemical was not possible
because documentation on its use was not found.
Asbestos
Asbestos is a known human carcinogen. It causes lung cancer and lung disease. Asbestos
insulation was used extensively in the 1950s, when many facilities at the SRS were initially
constructed. Asbestos materials can be found in roofing, pipe, and vessel insulation; building
insulation; gaskets; packing; siding; and other building materials. The materials have been
removed as they deteriorate or as renovations, maintenance, and repairs have been done. Worker
protection has necessitated that asbestos controls be in place during all removal operations (DOE
1987).
More than 80,000 linear feet of asbestos-containing materials were reported to have been
removed in 1984. Nonradioactive asbestos has been disposed of in a separate trench in the
sanitary landfill. Radioactive asbestos was buried in the Solid Waste Disposal Facility (formerly
referred to as the Radioactive Waste Burial Grounds). An asbestos disposal site in C-Area was
closed and capped in 1984 (DOE 1987).
Although a concern for workers onsite, asbestos has not been considered an environmental
contaminant and has not been a concern for offsite exposure. In 1967, asbestos concentrations in
air at several plant locations were measured. The maximum concentration was 鈥?1 million
particles per cubic foot,鈥? which was said to have been one-fifth of the TLV for workers at that
time (Du Pont 1967). No facility has listed asbestos as an emission for the Air Emissions
Inventory (Faugl 1996). We did not evaluated asbestos further because it is primarily found in
Evaluation of Materials Released from SRS 16-19
Ranking and Selection of Chemicals of Concern
building materials that are resistant to environmental transport unless seriously degraded or the
building is demolished. If a building is imploded, exploded, or demolished all at once, the
asbestos would be released to the air and would be subject to resuspension with time unless
precautions were taken to contain it. It is likely that most of the buildings containing asbestos
materials have been remediated or destroyed in the last decade, when awareness of asbestos
hazardous was high and precautions were taken to protect workers and limit dispersion of
asbestos fibers. For the purposes of the conservative stage one ranking, if all of the asbestos
reported onsite was released to the air, concentrations about 100 times the level of concern could
have been reached. This is a very unreasonable scenario, however, because most of the asbestos
onsite is in insulation and building materials. Asbestos has not been subject to storage in large
amounts, leakage, spillage, or routine release to the environment; therefore, we did not estimate a
source term for asbestos.
Benzene
Benzene is a very volatile, slightly water-soluble chemical used in the past as a solvent. It is
a contaminant of several other materials used at the SRS, including gasoline. Chemical
degradation, primarily reaction with hydroxyl radicals, limits the persistence of benzene in air to a
few days or even a few hours. Benzene released to soil and water is subject to biodegradation,
photooxidation, and volatilization. The half-life in surface water has been estimated to be about
17 days for photolysis. Biodegradation half-lives are estimated to be about 8 to 16 days in surface
water and about 30 days in groundwater. Although a large volume of benzene is released to the
environment from a variety of sources, environmental levels are generally low because of rapid
removal and degradation. The Log Kow is 2.15. Benzene is considered highly mobile in soil and
water, but it will adsorb to organic matter in soil. Benzene does not bioconcentrate, and it is a
known human carcinogen. Because large amounts of petroleum products were used at the SRS,
we investigated environmental monitoring information and releases of benzene further.
Chlorine
Chlorine is a severely irritating gas that is also more of a hazard in the workplace than an
environmental contaminant of concern. Chlorine is reactive and would not be expected to persist
in the environment as an irritating gas. It is not a carcinogen or chronic toxicant and would not be
likely to have caused health effects offsite unless released in very large quantities at one time.
Therefore, chlorine was not included in the chemicals to be considered further.
Coal
Coal is primarily carbon containing varying amounts of toxic or carcinogenic metals, sulfur,
and other contaminants. Coal is regulated in the workplace as a nuisance dust (Lewis 1993).
Although the ranking ratio is high, coal should not be the first priority in evaluating source terms
for chemicals because of the ultraconservative ranking assumptions applied. For the purposes of
consistency in the ranking exercise, it was assumed that all the coal piles were discharged to the
Savannah River and that the toxicity value for the entire pile was that of benzo(a) pyrene, with an
MCL of 0.002 mg L鈭?1. This led to a ranking ratio of 254. If the entire store was evaluated using
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-20 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
the MCL of 0.005 mg L鈭?1 for benzene, another carcinogenic component, the ranking ratio would
be 101. It is unreasonable to treat the entire coal inventory as if it were carcinogenic aromatic
hydrocarbon or benzene. We could conservatively estimate that coal tar and coal tar creosote
material could contain as much as 3% aromatic hydrocarbons (Gosselin et al. 1984) and as much
as 10% benzene (ATSDR 1995). If the inventory amount is reduced by these percentages, the
resulting ranking ratios are 7.6 for benzo(a)pyrene and 10.1 for benzene. It is also unreasonable to
assume that all of the coal inventory was discharged into the river. We know the piles remain
onsite, but we do not have an estimate of the amount of material that left the pile in runoff or
leachate. A preliminary evaluation of the runoff history, means taken to prevent runoff, and the
results of environmental sampling around the coal piles was worthwhile for this phase of the
study. We also evaluated coal ash.
Coal Tar
Coal tar and coal pitch are volatiles formed during the distillation of coal, and they contain
carcinogenic polyaromatic hydrocarbons like pyrene, benzo(a)pyrene and chrysene, benzene, and
phenol. If stores of coal tar were not contained, they could be subject to rainwater runoff and
leaching, and they could contaminate surrounding soil, surface waters, and underlying
groundwater. We investigated this material as a part of the qualitative evaluation of coal stores.
Freons
Chlorinated, fluorinated hydrocarbons, also called Freon (a registered trademark of Du Pont),
are clear, colorless, noncombustible liquids. Perhaps best known for their use as refrigerants, they
were also used at the Site for cleaning, degreasing, and as decontamination solutions. Freons were
used in solvent degreasers, especially after 1988. Freon was also used as a coolant during
operation of saws and lathes in M-Area. About 530 gal y鈭?1 was used during the peak production
times of 1985鈥?1988. All of this was probably released to the atmosphere (Radian 1992). Freon
was also released from heating, ventilation, and air-conditioning equipment.
Freon vapors are four to five times heavier than air and tend to accumulate in tanks, pits,
sumps, and other low places; therefore, inhalation of concentrated vapors can be an occupational
hazard (Haynes and Stoddard 1984). In a 1975 industrial hygiene summary report, mention is
made of 800 gal of Genetron 11 being removed from a refrigeration unit and disposed of. The Du
Pont Freon Products division memo informed SRS staff that the company had 鈥渘o method or
requirement for chemically changing these materials into other substances that would be more
acceptable for release to the environment the recommendation (and practice) is to remove the
material to a remote area and allow it to evaporate鈥? (Harper and Croley 1976). Although release
of these compounds to the environment is recognized to contribute to ozone depletion, it is very
unlikely these materials would have presented a health hazard offsite; therefore, we did not
determine releases of Freon.
Gadolinium Nitrate
Because toxicity information on this chemical is limited, we ranked gadolinium nitrate using
the very conservative LD50/100,000 value reported in the MSDS. Gadolinium nitrate is described
Evaluation of Materials Released from SRS 16-21
Ranking and Selection of Chemicals of Concern
as a severe irritant and no chronic health effects are known. It is a relatively stable material, but
environmental fate and transport of this material has not been studied. It is likely that precautions
were taken to conserve this material and large releases were probably uncommon. The
unavailability of documentation specific for this material would have seriously limited our ability
to develop a source term; therefore, we did not evaluate gadolinium nitrate further.
Gasoline and Other Fuels
Gasoline is a complex mixture of hydrocarbons containing small amounts of benzene,
toluene, xylene, 1,3-butadiene, and sometimes lead. Gasoline may have been released to the air or
water and it may have seeped into groundwater from leaking storage tanks, pipelines, or as a
result of spills. For the purposes of the ranking, we assumed all gasoline stores spilled into the
river or discharged into the air, and the entire amount was ranked using the drinking water
standard for a carcinogenic component, benzene.
The composition of gasoline varies tremendously. Compositional data studies sponsored by
the American Petroleum Institute found the benzene content of gasoline mixtures were as high as
2% (ATSDR 1994). The IARC estimated that gasoline typically contains 0.5 to 2.5% by volume
benzene (ATSDR 1994) or as high as 3.2% by weight. Benzene, ethylbenzene, and other
aromatics related to benzene (like toluene and xylene) may make up as much as 30.5% by weight
of a gasoline mixture (ATSDR 1994). Measurements of the amount of benzene versus total
hydrocarbons gasoline released to the atmosphere vary greatly when tanks are vented, gasoline is
pumped, or storage tanks are excavated. However, measurements do suggest benzene vapors
could account for nearly 10% of the gasoline vapors under certain conditions of discharge.
Studies have shown that benzene concentrations in air are highest during refueling operations.
Studies on service station attendants suggest that the gasoline in the air they breathed averaged
about 0.25% benzene. Taken together, these studies suggest an upper-bound estimate of 30% for
the percentage of benzene in gasoline or gasoline vapors. Using the toxicity values for benzene
and 30%, the inventory amount results in a ranking ratio of 0.531 for gasoline.
Gasoline is very volatile and does not dissolve readily in water. Most of the hazardous
components of gasoline are broken down in a number of hours to weeks after their release. The
Log Kow for gasoline ranges from 2.13鈥?4.87. Most chemicals in gasoline do not bioaccumulate.
After it is released to the environment, gasoline is not transported as a mixture. The components
of the mixture selectively partition into different environmental media according to their
individual chemical and physical properties. The compounds of greatest health concern, like
benzene, are water-soluble and are transported in groundwater, surface water, and through soils.
These compounds are also subject to photochemical oxidation in air and have half-lives on the
order of one or several days. They are also subject to biodegradation in water and volatilization,
photooxidation, and biodegradation in soil. Although gasoline has caused liver and kidney tumors
in animal studies, there is no evidence that gasoline causes cancer in humans. We did not find
documentation, interview notes, or other evidence of leaks, spills, and large releases of gasoline
that may have traveled offsite; therefore, we did not determine a source term estimate for gasoline
released to surface water.
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-22 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
Hydrazine
Hydrazine is a reactive, flammable liquid used as a reducing agent for nuclear fuel
reprocessing. It has also been used as an intermediate in the production of agricultural and
industrial chemicals, as rocket fuel, and as a medication for sickle cell disease and cancer.
Hydrazine has a relatively low vapor pressure and is soluble in water. Hydrazine could have been
released to air, water, or soil. This chemical rapidly degrades in most environmental media by
oxidation and biodegradation. Hydrazine in air is quickly destroyed by chemical reaction within
minutes or hours, depending on the concentration of ozone and hydroxyl radicals in the air. Most
hydrazine in air would be expected to have degraded within several hours of its release.
Hydrazine released to water and soil can become dissolved in water or bind to soil. Hydrazine can
sorb onto clay soils. The Log Kow is 鈭?3.08. Hydrazine is subject to oxidation and biodegradation,
and most of the hydrazine in water and soil would be expected to be gone within a few weeks.
Hydrazine does not tend to biomagnify up the food chain. The potential for hydrazine to have
been transported offsite is limited by its rapid degradation. Hydrazine causes several types of
tumors in animals and is classified as a probable human carcinogen by the EPA (ATSDR 1994).
We reviewed the use, storage, release history, and environmental monitoring results for hydrazine
and hydrazine mononitrate.
Hydroxylamine Sulfate
Hydroxylamine sulfate is a white crystalline material that is very corrosive and can cause
severe burns, ulceration, and sensitization reactions. There is a lack of toxicity information for
this material, which is listed as being used in relatively large amounts in the 1970s Du Pont
Index. The ranking ratio for hydroxylamine sulfate is comparatively large because we used the
upper value of the range reported in the Du Pont inventory and the high LD50/100,000 value to
calculate it. The LD50 was derived from data on mice given hydroxylamine sulfate i.p. and
reflects the corrosive, irritant effects rather than any chronic effects. Although specific
environmental fate data on this chemical are also lacking, it is a reactive chemical and would not
be expected to persist in the environment. Based on this and the extreme conservatism of the
toxicity values, we did not develop a source term for hydroxylamine sulfate.
Manganese
Manganese was evaluated using the EPA鈥檚 RfC value, which is quite conservative, with an
uncertainty factor of 900 to account for a lack of toxicity data. Available monitoring data and
source term information were examined, but the lack of documentation specific to this material
limited our ability to develop a source term. We reviewed and evaluated monitoring and usage
data on manganese compounds to the extent possible.
Mercury
The physical and chemical characteristics of metals that influence their behavior in the
environment include solubility, oxidation state, and tendency toward sorption on materials in soils
and sediments. Mercury exists in the elemental form (a volatile liquid), in the +1 and +2
Evaluation of Materials Released from SRS 16-23
Ranking and Selection of Chemicals of Concern
oxidation states, and as organic mercury. Mercury sorbs strongly to organic material and oxides
and tends to accumulate in sediments. Microorganisms in sediments can convert mercury to
methymercury, which bioaccumulates. Mercury and methylmercury are very potent neurological
toxins. The inventory amounts used for the ranking may not include mercury in pumps that was
disposed of by burial through the years. We further evaluated discharges of mercury to the air and
water and evaluated mercuric nitrate along with elemental mercury.
Other Metals
Chromium, lead, and nickel had ranking ratios for water that were greater than 10鈭?2. Zinc,
copper, and cadmium compounds had ranking ratios for water that were greater than 10鈭?3. These
metals are toxic, and chromium, nickel, and cadmium are carcinogenic. Lead is a potent
developmental neurological toxin.
We used combined inventories of small metal stores all over the Site and numerous metal-
containing compounds like rust and corrosion inhibitors that were used and stored in relatively
small amounts to obtain the ratio. Many of these materials were primarily found in solid form,
like sheeting, bricks, pellets and pipes. Especially large amounts of lead shielding, and lead
pellets were used at the SRS. These materials were relatively resistant to environmental transport
and were not likely to have been subject to resuspension in air or leaching into surface or
groundwater. For the purposes of ranking, we assumed that all stores of these materials were
placed in the Savannah River. This is especially unreasonable because these materials were not
stored in large piles or tanks but were used all over the Site in many different facilities, in small
and large amounts, and in many forms.
The mobility of all of the metals depends on soil and water chemistry and pH. For example,
cadmium usually occurs as the Cd+2 ion at pH levels less than 8 or quite commonly as cadmium
sulfate. Cadmium will adsorb to soils and sediments by cation exchange. Chromium occurs in the
+3 or +6 oxidation state in water. Cr(III) is insoluble and readily absorbs to metal oxides in soils.
Cr(VI) is soluble and is more mobile in the environment.
We compiled environmental monitoring data and information useful for source term
determination for arsenic, cadmium, chromium, lead, manganese, mercury, nickel, and zinc.
These metals were released into surface water from their use as cooling water treatment
chemicals, during surface water runoff from coal and ash piles, and as a component of process
waste in M-Area, F-Area, and H-Area.
Nitric Acid
Nitric acid is a caustic, severely irritating compound that is more of a workplace hazard than
an environmental contaminant of concern. It is subject to rapid degradation. It is not a carcinogen
or chronic toxicant and would not likely have caused health effects offsite unless released in very
large quantities at one time. We compiled releases of nitric acid fumes and further evaluated
nitrates and nitrogen dioxide, which are chemicals associated with the use of nitric acid.
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-24 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
Sulfuric Acid
Large amounts of sulfuric acid were used onsite, and several spills of sulfuric acid to Site
streams occurred. Sulfuric acid is a caustic, corrosive material that would not be expected to
persist in the environment. We noted documentation of sulfuric acid releases, but it did not seem
necessary to develop a source term for this material because transport of hazardous amounts
offsite would have been limited by its rapid degradation or conversion into other materials.
Tetrachloroethylene
Tetrachloroethylene is a volatile liquid used as a solvent, cleaner, and vapor degreaser.
Tetrachloroethylene is slightly water-soluble, and tetrachloroethylene released to surface water
would be expected to rapidly evaporate into the air. Tetrachloroethylene also evaporates from
soil, but it can easily travel through soil into groundwater. The Log Kow is 3.40.
Tetrachloroethylene in air is photochemically degraded. The half-life in air has been estimated to
be 3 to 4 months, while degradation in water is much slower. Tetrachloroethylene is subject to
biodegradation, but it can persist in soils and groundwater for decades. Tetrachloroethylene has a
very low tendency to bioaccumulate. Tetrachloroethylene has caused cancer in animal studies, but
it has not been shown to cause cancer in humans. The ranking ratio for tetrachloroethylene in
water was quite high because the drinking water standard for this chemical is quite protective. It
is likely that most of the tetrachloroethylene used onsite was released into the atmosphere. We
evaluated the discharge of this solvent and others to M-Area sewers, into groundwater, and into
the air; however, the priority assigned to tetrachloroethylene was not as high as the water ranking
ratio might suggest.
Trichloroethylene
Like tetrachloroethylene, trichloroethylene was used as a solvent and cleaner.
Trichloroethylene released to surface water would be expected to rapidly evaporate into the air.
Trichloroethylene also evaporates from soil, but it can easily travel through soil into groundwater.
Large amounts of trichloroethylene were used in M-Area, and we further evaluated discharges to
M-Area sewers, into groundwater, and into the air.
Trichloroethane
Trichloroethane is a volatile cleaner and solvent that was released to the air and to M-Area
process sewers. It dissolves slightly in water and would be expected to evaporate rapidly from
soil and water. Once in the air, it is estimated to persist for about 6 years. Trichloroethane is
thought to be important in reducing the stratospheric ozone layer. Trichloroethane in soil and
water is also subject to biodegradation. An estimated half-life for degradation in groundwater is
about 10 months. Trichloroethane does not bioaccumulate. 1,1,1-Trichloroethane is not
carcinogenic, and the likelihood that environmental exposure would cause significant health
effects is low. 1,1,2-Trichloroethane has caused cancer in some animal studies, but it has not been
shown to cause cancer in humans. For the purposes of the screening, we combined different
isomers of trichloroethane and calculated a ranking ratio using the more conservative toxicity
Evaluation of Materials Released from SRS 16-25
Ranking and Selection of Chemicals of Concern
values for the carcinogenic isomer. We further evaluated discharge of this solvent and others to
M-Area sewers and into groundwater and the air.
Uranium
Uranium is a radioactive metal that can cause cancer and kidney disease. In general,
uranium releases were reported in curies rather than kilograms or pounds of material. Uranium is
one of the radioactive materials being evaluated and we developed a source term for it.
Understanding the kidney toxicity of uranium may be an important step in later phases of the
project when health risks will be determined.
Other Chemicals of Potential Interest
There are several chemicals that, as a result of the screening, were not evaluated further but
are of public interest. A discussion of these chemicals follows.
Anthracene. Anthracene is a polycyclic aromatic hydrocarbon. Unlike similar aromatic
hydrocarbons, anthracene is not carcinogenic. Benzanthracene and other similar carcinogenic
compounds have been listed in essential materials ledgers, but how they were used was unclear.
Benzo(a) anthracene, benzo(a)pyrene, benzofluoranthene, benzoperylene, and chrysene were
listed in the 1994 SRS CIIS database with inventory quantities of zero or less than 0.001 lb,
which is consistent with use as an analytical standard or laboratory reagent. A September 1953
monthly progress report for the Works Technical Department mentioned that assistance had been
given to the instrument department concerning the use of anthracene in relatively large quantities
to coat tubes in building 773-A. The Works Technical Department recommended that this be
done in a hood, using protective clothing and a respirator, and special attention be given to
personal clean up after each job (Du Pont 1953).
Carbon Tetrachloride. An inventory amount for carbon tetrachloride was not given in the
CIIS database. The inventory amount in the CIIS database would not have been particularly
useful for estimating quantities of carbon tetrachloride used in the past because it is likely that use
of this chemical as a solvent was phased out. As at other U.S. Department of Energy facilities, it
is likely carbon tetrachloride was replaced with less toxic solvents, such as tetrachloroethylene,
trichloroethylene, and trichloroethane. Very large amounts of carbon tetrachloride were used at
the Rocky Flats Plants in Colorado. However, no documentation has been found to support the
use of large amounts of carbon tetrachloride at the SRS. It seems that tetrachloroethylene,
trichloroethylene, and trichloroethane were the solvents used in M-Area to clean materials such as
targets and cans. Recent RCRA and CERCLA monitoring data have detected carbon tetrachloride
in groundwater but not in quantities suggesting a significant use in the past. Carbon tetrachloride
was not reported in the 1974 inventory. Based on an apparent lack of inventory, we did not
evaluate carbon tetrachloride further.
Fluoride. Hydrogen fluoride was released from the JB-Line stack in F-Area (Reinig et al.
1973). Fluoride was also a component of liquid waste from the separations area processes.
Potassium fluoride was used in the frames process to isolate 238Pu. A 1988 summary of
separations activities reported that 59 lb of fluoride was discharged to the seepage basins without
evaporating because of corrosion problems with evaporators. This amount was said to represent
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-26 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
the maximum annual amount that would have been discharged during 1 year from one separations
plant (Du Pont 1988).
Fluorides are ubiquitous in food and water. Human exposure to toxic amounts of fluorine
and hydrogen fluoride is unlikely outside of an occupational setting. Fluoride is highly
electronegative and reacts vigorously with other compounds. Fluoride, fluorine, and hydrofluoric
acid released into soil or surface waters would rapidly react to form fluoride salts. Fluorides
discharged to the seepage basins have been retained by the soil.
Hydrofluoric acid was spilled on the grounds west of central shops sometime before 1970.
Monitoring wells have been installed in the area. Contaminated groundwater has not moved
offsite (Christensen and Gordon 1983), and it is not likely to move offsite in the near future
because the central shops are centrally located. A hydrofluoric acid solution has been used in the
773-A glass shop, which makes glassware used at the Savannah River Technology Center. The
1996 Operating Permit Application described the releases as being exhausted from two small
stacks. The application considered the releases to air to be very small (Westinghouse 1996). A
1987 annual environmental report describing ambient air quality monitoring onsite and offsite
said that gaseous fluorides were not monitored because the potential release was insignificant
compared to the standard (Mikol et al. 1988). We did not evaluate fluorides further.
Scintillation Fluids. Scintillation fluids are photofluoric compounds and surfactants in
organic solvents, such as xylene, toluene, dioxane, napthalene, or trimethylbenzene solutions. An
annual waste generation rate for scintillation fluids used to analyze samples for radioactivity was
estimated to be about 200 gal y鈭?1 (Smithwick 1984). Liquid scintillation solutions have been
buried at the Radioactive Waste Burial Grounds since 1965. As of 1984, the amount sent to the
burial grounds was estimated to total 10,000 gal, most of which had been used for tritium
analysis. The solutions were buried in plastic or glass vials in containers with absorbent material.
After 1987, all liquid scintillation solutions were supposed to be incinerated (DOE 1987). There
is no evidence that these fluids traveled offsite; therefore, we did not evaluate them further.
CONCLUSIONS
Based on the ranking results and the discussion above, we concluded at the beginning of
Phase II of the project that a source term should be developed for the following chemicals:
Benzene Mercury and mercuric nitrate
Coal Nitric acid
Coal Ash Trichloroethylene
Hydrazine Tetrachloroethylene
Gasoline Trichloroethane
We concluded the following metals should also be evaluated:
Arsenic Cadmium
Chromium Lead
Manganese Uranium
Nickel Zinc
Evaluation of Materials Released from SRS 16-27
Ranking and Selection of Chemicals of Concern
A source term was estimated for chromium, cadmium, and lead releases to air and nickel releases
to Site streams. We did not find enough information to estimate a source term with reasonable
certainty for arsenic in coal and ash pile runoff; chromium releases to surface water; or nickel,
zinc, and arsenic releases to air. We compiled monitoring data for these metals and described
releases using all available information. Information on gasoline, coal, and ash storage, use,
transport, and disposal was reviewed and summarized. We made release estimates for benzene
and toxic components of metals in coal and ash and releases of toxic metals and other pollutants
from coal burning. In addition, we developed source terms for nitrogen dioxide, sulfur dioxide,
ash particulates, and hydrogen sulfide. The SRS operations had the potential to release large
amounts of these pollutants into the air.
The following chemicals may have been released, but additional analyses were impossible
because of a lack of available inventory or toxicity information:
Aldrin/dieldrin Chloroethane
Chloromethane DDT
Endosulfan Endrin
Heptachlor Hydroxyquinoline
Lindane Toxaphene
Trichlorophenol (2,4,5-T) PCBs (polychlorinated biphenyls,
arochlors)
An inventory for these chemicals was not reported in 1974 or 1994. Chloroethane and
chloromethane have been detected in onsite groundwater. There is no amount listed in the
inventory for PCBs, but PCBs have undoubtedly been used in electrical equipment onsite, and
some of this equipment was likely to have been buried in onsite waste areas. The RAC researchers
reviewing monthly reports and other documents for Phase II of the dose reconstruction study
were given a list of these compounds and asked to flag any information regarding their use,
release, monitoring. or disposal. All of this documentation was reviewed and used to develop a
source term estimate for as many of these chemicals as possible. The lack of documentation for
some of the chemicals seriously limited our ability to develop a source term for them. However,
where a release estimate could not be calculated, a qualitative evaluation of the use and potential
release of the compound was conducted and the resulting characterization is provided in Chapters
17 and 18. Environmental monitoring for chemicals is addressed in Chapters 19 and 20.
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-28 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
REFERENCES
ACGIH (American Conference of Governmental Industrial Hygienists). 1994. 1993鈥?1994
Threshold Limit Values for Chemical Substances and Physical Agents and Biological
Exposure Indices. ACGIH, Cincinnati, Ohio. OpenLit.
Anderson, M.E., H. Clewell, III., and K. Krishnan. 1995. 鈥淭issue Dosimetry, Pharmacokinetic
Modeling and Interspecies Scaling Factors.鈥? Risk Analysis 15 (4). OpenLit.
ASTDR (Agency for Toxic Substances and Disease Registry). 1992鈥?1997. Toxicological
Profiles. Atlanta, Georgia. OpenLit.
Christensen, E.J. and D.E. Gordon, eds. 1983. Technical Summary of Groundwater Quality
Protection Program at SRS, Vol. 1. Site Geohydrology and Quality Protection Program at
Savannah River Plant. DPST-83-829. Du Pont, Savannah River Laboratory. SRS Phase II
Database SKR1993101316. December.
Cummins, C.L., D.K. Martin, and J.L. Todd. 1991. 1990 Savannah River Site Environmental
Report. WSRC-IM-91-28. SRS Phase II Database MOL1993120613. May.
DOE (U.S. Department of Energy). 1987 Environmental Survey Preliminary Report Savannah
River Plant, Aiken, South Carolina. DOE/EH/OEV-10-P. Environment, Safety and Health
Office of Environmental Audit. SRS Phase II Database EIW199309238. August.
Du Pont. 1953. Explosives Department Atomic Energy Division Monthly Report for the SRP.
DPSP-53-14-3. August, Georgia. SRS Phase II Database EAS199407251. March.
Du Pont. 1959. Monthly Progress Report Works Technical Department. DPSP-59-1-5. SRS
Phase II Database PDM1994071261. May.
Du Pont. 1967. Savannah River Laboratory Radiological and Environmental Sciences Monthly
Report for July 1967. DPST-67-1-7. SRS Phase II Database LWB1994011926.
Du Pont. 1979. CCI - Index of Chemicals Onsite at the SRS in the 1970s. Wilmington, Deleware.
Phase II Database KRM199812151.
Du Pont. 1988. SRP Plant History Separations Areas January 1973 through December 1987.
DPSP-74-454-2. SRS Phase II Database EIW199308188.
EPA (U.S. Environmental Protection Agency). 1995. Health Effects Assessment Summary
Tables. EPA 540/R-95-036. Office of Research and Deveopment. Washington, DC. May.
Faugl, T. 1996. Environmental Protection Department, Westinghouse Savannah River Company.
Communication with P. McGavran, Consultant to Radiological Assessments Corporation.
Subject: Air Emission Inventory. SRS Phase II Database Interview # 5.
Evaluation of Materials Released from SRS 16-29
Ranking and Selection of Chemicals of Concern
Gosselin, R.E., R.P. Smith, and H.C. Hodge. 1984. Clinical Toxicology of Commercial Products,
5th edition. Baltimore, Maryland: Williams and Wilkins. OpenLit.
Harper, J.A. and J.J. Croley. 1976. Summary of SRP Industrial Hygeien Activities.
DPSPU-76-276. Du Pont. SRS Phase II Database LWB1994021823. March.
Haynes, H.R. and D.H. Stoddard. 1984. Revision of 1977 Report: Systems Analysis - 200 Area
SRP Building 234-H Tritium Processing Facilities. DPSTSY-200-3-WD-Rev 1.
Environmental Anlaysis Division, Savannah River Laboratory. SRS Phase II Database
SVK1994082425.
Information Systems Engineering. 1994. Chemical Information and Inventory System Full User鈥檚
Guide, CIIS. SRS Phase II Database LWB199407051.
Layton, D.W., B.J. Mallon, D.H. Rosenblatt, and M.J. Small. 1987. 鈥淒eriving Allowable Daily
Intakes for Systemic Toxicants Lacking Chronic Toxicity Data.鈥? Regulatory Toxicology and
Pharmacology 7: 96鈥?112. Open Lit.
Lewis, R.J., Sr. 1993. Update, Sax鈥檚 Dangerous Properties of Industrial Materials, 8th ed. New
York: Van Nostrand Reinhold Company. OpenLit.
Meyer, K.R., P.D. McGavran, P.G., Voillequ茅, H. A. Grogan, L.W. Bell, H. R. Meyer, S.K.
Rope, and J.E. Till. 1995. Savannah River Site Dose Reconstruction Project, , Phase I. Task
3. Evaluation of Materials Released from the SRS. RAC report # 1 CDC-SRS-95.
Radiolgical Assessments Corporation, Neeses, South Carolina. June.
Mikol, S.C., L.T. Burckhalter, J.L. Todd, and D.K. Martin. 1988. SRS Environmental Annual
Report for 1987. DPSU-88-30-1. SRS Phase II Database EIW199310211.
Morgan, J. 1996. Chemical Commodies Management Center, Westinghouse Savannah River
Company. Communication with P.D. McGavran, Consultant to Radiological Assessments
Corporation. Subject: Chemical inventories and purchases and receipt records. SRS Phase
II Database Interview 15.
NIOSH (National Institute for Occupational Safety and Health. 1994. Pocket Guide of Chemical
Hazards. 94-116. June. OpenLit.
OSHA. 1990. US Department of Labor, OSHA Regulated Hazardous Substances, Health
Toxicity, Economic and Technological Data. Noyes Data Corporation, New Jersey. OpenLit.
Radian (Radian Corporation). 1992a. SRS M-Area Air Emission Inventory Calculations
Procedures Notebook. Prepared for Westinghouse. SRS Phase II Database PDM199710013.
November.
Risk Assessment Corporation
鈥淪etting the standard in environmental health鈥?
16-30 The Savannah River Site Dose Reconstruction Project
Phase II: Source Term Calculation and Ingestion Pathway Data Retrieval
Reinig, W.C., J.E. Hoy, W.T. Jacobsen, H.A. McClearen, and I.S. St. John. 1973. Survey of
Effluent and Environmental Monitoring at the Savannah River Plant. DPST-73-293. SRS
Phase II Database MJC199312148. June.
Sax, N.I. 1975. Dangerous Properties of Industrial Materials, 4th ed. New York, New York: Van
Nostrand Reinhold Company. OpenLit.
Sax, N.I. and R.J. Lewis. 1992. Dangerous Properties of Industrial Materials. New York, New
York: Van Nostrand Reinhold. OpenLit.
Smithwick, G.A. 1984. Information on Department of Energy Mixed Wastes, Mixed Hazardous
Waste Stream Generated at the SRP. DOE-03330. SRS Phase II Database SVK1994091423.
August 23.
U.S. Department of Labor. 1990. OSHA Regulated Hazardous Substances, Health Toxicity,
Economic and Technological Data. Noyes Data Corporation, New Jersey. OpenLit.
Westinghouse (Westinghouse Savannah River Company). 1994. The Savannah River Site鈥檚
Ground Water Monitoring Program Fourth Quarter 1994 (U) Report. ESH-EMS-940517.
Environmental Protection Department, Environmental Monitoring Section. SRS Phase II
Database KRM199810225.
Westinghouse. 1996. Part 70 Operating Permit Application. Submitted to the South Carolina
Department of Health and Environmental Control, Bureau of Air Quality. Volumes I-XIX.
SRS Phase II Database PDM199710023.
|