I
CHEMICAL ENGINEERING BRANCH
MANUAL FOR THE PREPARATION
OF ENGINEERING ASSESSMENTS
by
IT Environmental Programs, Inc.
11499 Chester Road
Cincinnati, Ohio 45246
Contract No. 68-D8-Q112
Work Assignment No. P3-7
PN 3786-64
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF TOXIC SUBSTANCES
401 M STREET, S.W.
WASHINGTON, D.C. 24060
February 28, 1991
-------
PREPARATION OF ENGINEERING ASSESSMENTS
VOLUME I
Chemical Engineering Branch
Economics and Technology Division
Office of Toxic Substances
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
CONTENTS
Figures vii
Tables vii
Acronyms ix
I. Introduction 1-1
II, Review Processes 2-1
A, New chemical review 2-1
B. Existing chemicais review 2-7
C. Section 313 petitions 2-10
111. Approaches and Data Sources for Assessment 3-1
A. Approaches 3-1
B, Data sources 3-3
IV. Modeling Workplace Exposure 4-1
A. Estimating inhalation exposures 4-1
B. Estimating dermal exposure 4-33
C. Personal protective equipment 4-39
D, Engineering controls 4-49
V. Modeling Release to Water 5-1
A, Cleaning of equipment 5-1
B, Tank truck and tank car cleaning 5-5
C. Phase separation 5-8
D. Condensers and scrubbers 5-11
E, Poly-electrolytes 5-15
F. Metal working operations 5-17
G, Filtering solids from water 5-18
H, Spray-coating operations 5-19
-------
CONTENTS (continued)
I. Leather dyeing
J. Drilling operations
K. Recirculating water-cooling towers
VI, Modeling Release to Air 6-1
A. Process vents 6-1
B. Tank working and breathing tosses 6-3
C. Fugitive releases 6-4
D, Secondary sources 6-6
VII, Evaluating Release Controls 7-1
A. Water controls 7-1
B. Air controls 7-3
C. Liquid and solid waste controls 7-8
VIII, References 8-1
Appendices
A. Sample Initial Review Engineering Report A-1
B. Guidelines for Coordinated ETD PMN Standard Review B-1
C. Sample Production Exposure Profile (PEP) C-1
D. Sample TRI Data D-1
E. Industrial Process Profiles and Other Completed Studies E-1
F, Summary of Guidelines for Statistical Analysis of
Occupational Exposure Data F-1
G, Derivation of Formulas for Calculation of Workplace Airborne
Concentration G-1
H, Chart of Body Areas and Estimation of Skin Area H-1
I. Other Factors to be Considered in Respirator Selection 1-1
J. Standard Language for 5(e) Orders and SNURs j-1
K. Derivation of Equation for Evaporation From Open Surfaces (Revised) K-1
VI
-------
FIGURES
Number Page
3-1 Matrix of Data Contained in Volume II 3-8
3-2 Matrix of Data Bases Contained in Volume til 3-9
5-1 Schematic of Cooling Tower System 5-25
7-1 Venturi Scrubber Collection Efficiencies 7-9
VII
-------
TABLES
Number Page
4-1 Inhalation Rates 4-2
4-2 Estimated Airborne Concentrations of Total Mist for Spray Coating
Operations 4-8
4-3 Typical Parameters for Dye Weighing Operations 4-10
4-4 Estimated Airborne Concentrations for Metalworking 4-11
4-5 Typical Compound Composition for Tire Manufacturing 4-13
4-6 Summary of NIOSH Monitoring Data for Particulate Exposures in
Tire-Manufacturing Operations Based on Seven Plants 4-14
4-7 Default Distribution Factors 4-23
4-8 Saturation Factors for Bulk Loading Operation 4-26
4-9 Air Emission Factors for Loading 4-27
4-10 Typical Diameters and Areas 4-31
4-11 Summary of Concentration Calculations for Transfer Operations 4-32
4-12 Summary of Concentration Calculations for Open Surfaces 4-33
4-13 Typical Factors for Calculation of Dermal Exposure 4-36
5-1 Summary of Residue Quantities From Pilot-Scale Experimental Study 5-4
5-2 Typical Cooling Water Additive Concentrations 5-21
6-1 Average Fugitive Emission Factors for the Synthetic Organic Chemicals
Manufacturing Industry 6-5
7-1 Optimal Operating Conditions for F!ares 7-11
vi
-------
LIST OF ACRONYMS
ACGIH American Conference of Governmental Industrial Hygienists
AIHA American Industrial Hygiene Association
ANSi American National Standards Institute
APF Assigned Protection Factor
ASTM American Society for Testing and Materials
CCD Chemical Control Division
CEB Chemical Engineering Branch
CEPP Chemical Emergency Preparedness and Prevention Office
CFR Code of Federal Regulations
CHIP Chemical Hazard Information Profile
CPSC Consumer Products Safety Commission
CRSS Chemical Review Search Strategy
CSB Chemical Screening Branch
DD Division Director
ORE Destruction and Removal Efficiency
EAB Exposure Assessment Branch
ECAD Existing Chemical Assessment Division
EED Exposure Evaluation Division
EPCRA Emergency Planning and Community Right-to-Know Act
ESP Electrostatic Precipitator
ETD Economics and Technology Division
HEPA High Efficiency Paniculate Absolute
HERD Health and Environmental Review Division
HHE Health Hazard Evaluation
ICB Industrial Chemistry Branch
1DLH Immediately Dangerous to Life or Health
1H Industrial Hygienist
IPPEU Industrial Process Profiles for Environmental Use
IX
-------
IRER Initial Review Engineering Report
ITC tnteragency Testing Committee
IWS Industry-Wide Survey
LEL Lower Explosive Limit
LEV Local Exhaust Ventilation
LVE Low Volume Exemption
MCCEM Multi-Chamber Chemical Exposure Model
MSDS Material Safety Data Sheet
MSHA Mine Safety and Health Administration
NCB New Chemicals Branch
NiOSH National institute for Occupational Safety and Health
NOES National Occupational Exposure Survey
NSPS New Source Performance Standards
ORD Office of Research and Development
OSHA Occupational Safety and Health Administration
OSWER Office of Solid Waste and Emergency Response
OTS Office of Toxic Substances
PEL Permissible Exposure Limit
PEP Production/Exposure Profile
PM Program Manager
PMN Prernanufacture Notification
PQTW Publicly Owned Treatment Work
PPE Persona! Protective Equipment
RAB Risk Analysis Branch
RCRA Resource Conservation and Recovery Act
RIB Regulatory Impacts Branch
SAT Structure Activity Team
SIC Standard industrial Classification
SNUR Significant New Use Rule
STEL Short-Term Exposure Limit
-------
Ti Technical Integrator
TME Test Market Exemption
TRl Toxic Chemcial Release Inventory
TSCA Toxic Substances Control Act
TWA Time-Weighted Average
VOC . Volatile Organic Compound
XI
-------
Issued; February 28S 1991
Revised:
Page No.: 1-1
INTRODUCTION
Engineers in the Chemical Engineering Branch (CEB) evaluate occupational
exposures and environmental releases of new and existing chemicals under the Toxic
Substances Control Act (TSCA, PL94-469), CEB engineers also evaluate petitions
submitted under Section 313 of the Emergency Planning and Community
Right-to-Know Act (EPCRA, PL99-499), provide technical support to the Office of Solid
Waste and Emergency Response's (OSWER) Chemical Emergency Preparedness and
Prevention Office (CEPP), and participate in other Agency activities regarding the
regulation of chemicals, releases, and wastes. This manual describes the CEB
engineer's role in these activities. It also describes approaches and resources
available to conduct these activities in the absence of data on exposures or releases.
The duties of CEB engineers include:
Evaluating the methods used to manufacture, process, or use a specific
chemical substance in order to identify potential exposures and release
points.
Evaluating or estimating the extent of exposure or release (e.g., the airborne
concentration of a volatile liquid when it is drummed).
Evaluating the effectiveness of control alternatives, including personal
protective equipment and engineering controls, for reducing exposures or
releases.
Recommending appropriate controls for regulatory action.
To perform these duties, the CEB engineer collects information from many
sources. These include, but are not limited to, industry contacts, unpublished
contractor reports, journal articles, and scientific texts. The Environmental Protection
Agency (EPA), other government agencies, and industry increasingly are collecting
quantitative data on releases of chemicals from industrial facilities and exposures to
chemicals in the workplace. Databases and hard copy reports are routinely accessed
by CEB staff to obtain these data. Evaluating and accurately representing these data
are important functions of the CEB engineer. CEB also continues to develop
-------
issued; February 28, 1991
Revised:
Page No.: 1-2
databases, reports, and standardized approaches in order to use existing data more
efficiently.
The ideal exposure/release assessment would define the range of potential
releases and/or exposures, the central tendency of the estimates (mean or median),
the descriptive statistics of the exposure/releases (percentiles, standard deviation),
and characterize the uncertainty in the estimates. Sufficient data on releases and
exposures rareiy exist to permit full analysis. Frequently, outer bound or "reasonable
worst case" estimates are ail that can be made when tittle or no data exist,
In the absence of data, the CEB engineer must estimate releases and exposures.
Estimation methods are often used in the review of new chemicals that have yet to be
manufactured or used at the time of the review. Estimation methods are also used
during the assessment of existing chemicals under conditions that have not been
studied. To estimate releases or exposures, the CEB engineer must either use an
analogy (i.e., apply data on similar chemicals used in similar circumstances) or use
modeling techniques based on physical parameters such as vapor pressure or
solubifity in water. At times, the CEB engineer also must evaluate the reasonableness
of reported data.
Several standardized methods are available for estimating environmental releases
and workplace exposures under various conditions. Use of these methods provides
consistency in the review of a variety of chemical substances. This manual describes
these methods and provides values for key parameters necessary for their use. CEB
engineers and support contractors developed many of these estimating techniques
and parameters using the best available information, These methods are periodically
updated as new information becomes available.
The formulae presented in this manual are intended for use when better
information is not available. Conceptually, these formulae provide assessments of
"typical case" and "reasonable worst case" scenarios. In the absence of reliable
information, the "reasonable worst case* calculations should be used. If controls are
in use or will be used, the "reasonable worst case* estimates should be revised to
provide credit for the estimated effect of the controls.
-------
Issued: February 28, 1991
Revised:
Page No.: 1-3
A "reasonable worst case" represents conditions that may affect worker exposure
during the operation of a process or performance of a certain task by trained workers,
For example, it is reasonable to expect that a trained worker will occasionally spill a
chemical, or that limited ventilation conditions will exist during bagging of powders.
The intent of the formuiae, however, is not to represent such extraordinary
occurrences as a worker failing into a chemical tank, though such events can occur.
Although the actual exposures and releases may be considerably less than levels
derived from "reasonable worst case" scenarios, these estimates define an outer
bound to the potential for exposures and releases. This outer bound becomes an
important factor in the risk assessment of many chemicals. If the "reasonable worst
case" exposures lead to little or no concerns for risks to a substance, it may be
possible to drop the substance from further review. This will allow resources to be
focused on better characterization of other chemicals. If exposures estimated for the
worst case scenario are of concern, the engineer must reevaluate the parameters
used to develop the scenario and determine how representative they are of the
majority of situations.
The engineer should always attempt to obtain release and exposure monitoring
data for the substance under review or any substance with similar properties. If such
data more accurately represent the substance and the exposure and release scenarios
under review than "reasonable worst case" estimates, then they should be used in
preparing the assessment.
-------
Issued: February 28, 1991
Revised:
Page No,: 2-1
H. REVIEW PROCESSES
CEB engineers routinely review chemicals as part of three structured processes:
1} new chemical review, 2) existing chemica! review, and 3) Section 313 petitions.
Less-structured reviews conducted by CEB (and not further described in this manual)
include the review of regulatory proposals and supporting documentation developed
by other EPA program offices. CEB engineers also may evaluate consequences of
possible chemical accidents to decide whether characteristics such as flammability or
reactivity need be considered in listing chemicals for Section 302 of EPCRA.
A, New Chemical Review
The new chemical review process has two parts: 1) Premanufacture
Notification review and 2) fotlowup review.
Under Section 5 of TSCA, companies must submit a Premanufacture
Notification (PMN) at least 90 days before the commercial production (including
import) of any chemical that is not on the TSCA Inventory of chemicals in commerce
{"existing chemicals"). The PMN review focuses on the company's intent. The Agency
may act to restrict certain aspects of their activity by orders issued under Section 5(e)
and 5(f) of TSCA, "New Chemical Review: Process Manual" contains a full description
of the Office of Toxic Substances (OTS) process for evaluating new chemicals (USEPA
1986b).
The PMN rule provides exemption from the reporting requirements for
several chemical categories, including 1) chemicals being test marketed, 2} chemicals
used for research and development, 3) certain polymers, 4) chemicals manufactured
in quantities of less than 1000 kilograms per year, 5} certain non-TSCA uses, and
6) chemicals considered nonisolated intermediates.
Under 40 CFR 720.3, the PMN rule excludes from reporting chemicals
that are considered nonisolated intermediates. Based on this, OTS intends to exclude
from the PMN requirements any chemical substance that is manufactured and
consumed in the manufacture of another substance without intentional removal from
-------
Issued; February 28, 1991
Revised:
Page No.: 2-2
the process during its manufacture and use (Wong 1988). The chemical is considered
nonisolated if it is not intentionally removed from the equipment in which it is
manufactured, This includes the use of several reactors in a continuous process. The
definition does not apply if the chemical is transferred to tanks or other vessels in
which the chemical is stored after its manufacture. Mechanical or gravity transfer
through a closed system is not considered intentional removal if the chemical is not
transferred to storage or shipping containers. Volume II contains a more complete
discussion of nonisolated intermediates (Wong 1988).
Engineering analyses of new chemicals for which PMNs have been
received occur in two stages: 1) an initial (screening) review of all cases, and 2} a
detailed analysis for cases where it is decided in the Initial Review that the combination
of health/ecotoxicity concerns and exposure/release estimates may require regulatory
action,
a. Initial Review
[NOTE: At ihe present time, CEB is considering using a panel of engineers to prepare initial
reports and has begun a pilot effort.]
During the Initial Review, the CEB engineer prepares an Initial
Review Engineering Report (IRER) for presentation at the FOCUS meeting,
[At FOCUS, representatives of CEB, Industrial Chemistry Branch (ICB), and Regulatory impacts
Branch (RIB) present the results of initial assessments regarding the PMN chemical, FOCUS
occurs by day 21 of the review process (counting the date EPA receives the PMN) to cover
groups of cases at this step in the process. It is regularly scheduled on Mondays and Thursdays
at 1:00 p.m.]
In the screening stage, the CEB engineer drafts an IRER for all cases
not dropped at the CRSS meeting (see Appendix A for a sample IRER),
[CRSS stands for Chemical Review Search Strategy, This is a meeting of chemists from ICB to
determine chemical identity, physical properties, and other Information. CRSS can terminate the
review of polymers found to have high molecular weight without certain reactive end groups and
with negligible water solubility,]
The draft IRER forms the basis for CEB's presentation on exposure/release at
FOCUS. The CEB engineer has access to the Initial Review Chemistry Report (IRCR).
-------
Issued: February 28, 1991
Revised:
Page No.: 2-3
A brief description of health and ecotoxicity concerns identified by the Structure Activity
Team (SAT) is also available.
(SAT meets three days prior to FOCUS. It evaluates the potential hazard of the PMN based on
readily available data on analogues or information submitted with the PMN.]
The CEB engineer must use these resources to prepare the IRER, The CRSS
provides reaction pathways and estimates of properties. SAT results provide some
idea of the extent to which the CEB engineer should concentrate on exposure or
release (or on an impurity in the PMN) when doing the initial assessment, A complete
assessment is expected despite SAT outcome, however.
The CEB engineer should spend two to four hours reviewing the
PMN and preparing a draft IRER. Resources on hand should be used, including
reviews of past cases noted in the CRSS report, reference texts, standard scenarios
developed for specific end uses, Agency documents (e.g., Development Documents
for Effluent Guidelines), and National Institute for Occupational Safety and Health
(NIOSH) documents on workplace exposure. The engineer should contact the PMN
submitter to resolve any questions, Any such contact must be cleared with the
Program Manager (PM) and all information documented for the file,
[The Program Manager is a member of the New Chemicals Branch (NCB) of the Chemical Conlrol
Division (CCD) and is responsible for all policy and regulatory matters on a case. PMs are
assigned when the PMN is received.]
Generally, there is insufficient time before FOCUS to search databases, periodical
literature, or patents to obtain information to improve the assessment. CEB engineers
develop special expertise in one or more industrial categories and assignments are
made to take advantage of this expertise,
A panel of two or three CEB engineers and industrial hygienists
reviews the completed IRER for completeness, accuracy of estimates, and
reasonableness of assumptions. The release estimates are provided to the Exposure
Assessment Branch (EAB) for use in estimating environmental concentrations. One
panel member presents the engineering assessment to the decision makers at
FOCUS.
-------
Issued: February 28, 1991
Revised:
Page No.: 2-4
The FOCUS Representative informs the CEB engineer of FOCUS
results. If the case is dropped at this point, the CEB engineer completes the IRER.
This involves making any corrections or additions directed by the Focus
Representative and filing the report within two weeks. The IRER is completed in the
same manner when cases go to Follow-up for analysis of uses other than those
intended by the submitter (see Section H.B),
Cases may be subject to analysis based on either total exposure
(exposure-based cases) or risk (Standard Review), Exposure-based cases are
non-polymeric PMN chemicals with production volumes of 100,000 kg or greater,
These cases must meet at least one finding of worker exposure, environmental
release, or consumer exposure. PMN chemicals not meeting these findings are
dropped at FOCUS. Worker exposure findings are based on the overall exposure (the
total number of workers with routine dermal exposure or inhalation exposure during
manufacture, processing and use), EAB makes the findings on consumer exposure
and environmental releases.
b. Detailed Analysis
In risk-based cases, a risk assessment team is formed of the
appropriate disciplines. Teams usually consist of an engineer, a chemist, an
economist, 'lexicologists, environmental fate assessors, a Technical Integrator (T!), and
the PM. The CEB engineer generally must analyze exposures or releases in more
detail and respond to specific questions raised by the Tl. The Tl is responsible for
overseeing all technical aspects of the case to make a risk determination.
Three to four weeks are allotted for CEB's detailed analyses. The
CEB engineer prepares a draft report that is due before the Mid-Course Meeting, The
Section Chief reviews the report prior to submittal. CEB's report is consolidated into
an Economics and Technology Division (ETD) report. See Appendix B for a
description of the ETD report.
[Mid-Course Meeting is held by day 45 of the process. Its purpose is fo Inform revi&wers of each
others' findings and arrive a! a consensus on the overall assessment.]
-------
Issued: February 28, 1991
Revised:
Page No.: 2-5
cases may be dropped after this meeting. If the Tl recommends dropping the case, it
will be presented at CCD DISPO.
/CCD DISPO considers 1) cases that have been reviewed' in detail and are recommended to be
dropped, and 2) cases for which limited reviews have been conducted to decide whether more
detailed analyses are necessary. It is regularly scheduled for Monday and Thursday aterrtoons,
Again, chemist^1, health/ecotoxiciP/, and exposure/release estimates are presented to a
decision-maker from NCB,}
A subsequent draft of the engineering report, also reviewed by the
Section Chief, forms the basis for CEB's presentation at the ETD DtSPO, ETD DISPO
is regularly scheduled on Wednesdays at 3 p.m. At ETD DiSPO the CEB engineer
presents ail findings on exposure/release to ETD management and the ETD Division
Director, After ETD DISPO, the engineer files a completed report. The final step in the
review process is the Division Director's meeting.
[Division Directors meet each week. They review all cases for which regulatory actions are
recommended, with the Director of CCD determining ultimate recommendations for case
disposition. The engineer on the case should attend. Detailed analyses also will have been
prepared on health and ecotoxicity, environmental fate, consumer exposure, and economic
considerations, as needed, and are presented at this meeting.}
In completing a more thorough analysis, the CEB engineer should
research literature on specific end uses. This is often the most difficult area for which
to quantify exposure and release. The engineer also must search databases for
exposure data on analogues, it is especially important that the CEB engineer make
use of information gathered by the chemist and economist (also members of the
assessment team),
A written, fully-referenced report is prepared as support for
possible regulatory actions. In the event the Division Directors decide that regulatory
controls are needed, the CEB engineer identifies potential control alternatives and
participates as needed in drafting or reviewing orders under Section 5 of TSCA. Such
orders require the submitter to control exposure or release or to test the compound.
The CEB engineer also must participate in meetings with the submitter of a PMN, at
the request of the Program Manager.
-------
Issued: February 28, 1891
Revised:
Page No.: 2*6
It is the policy of CEB that engineers inform PIVIN submitters of
their findings regarding exposure or release and the basis for those findings. This
helps ensure that faulty assumptions are not used and that the submitter has an
opportunity to provide more data or information to clear up any discrepancies. During
detailed analysis of a case (and before ETD DtSPO), the CEB engineer should not
hesitate to check assumptions with the submitter.
2,
During the PMN review process, the Agency may identify concerns for
the chemical under conditions not covered by the PMN. The purpose of Foliow-up is
to decide whether Significant New Use Rules (SNURs) or Section 8 reporting rules are
needed to control exposure or release if 1) a chemical is manufactured rather than
imported; 2) a chemical is manufactured by another company, at another site, by
another process, or in larger volume; or 3) uses other that those intended by the
submitter could lead to increased exposure or release. PMNs reviewed in Follow-up
include those subject to regulatory controls for the intended use, and those not thus
controlled but presenting potential health/ecotoxicity hazards if used differently,
The Follow-up consists of an ETD Use Analysis, which occurs in three
steps:
Identification of possible uses for the PMN.
An assessment of promising uses, conditions of use, and the volume of
PMN that could be so used.
An evaluation of exposures and releases for these uses.
An ETD chemist, economist, and engineer are assigned to each case.
The chemist and economist are responsible for developing information on the potential
uses which are assessed in more detail by the economist.
The engineer prepares estimates of exposure or environmental release
for the new uses with the highest probability and completes a written report that is filed
after review by the Section Chief (usually within one week),
-------
Issued: February 28, 1991
Revised:
Page No.: 2-7
Other aspects of Follow-up include a Toxicity Validation review by
HERD (usually before a case is referred to ETD Use Analysis) and preparation of a
proposed rule (if needed) by CCD, The CEB engineer may be called on to contribute
to the proposed rule or to review public comments on the proposal.
Although there is little activity on individual Follow-up cases, many
SNURs are being prepared based on the TSCA Section 5 orders. These rules are
being promulgated under the "Generic SNUR," a lengthy rule with lists of requirements
such as control technologies, personal protective equipment, and disposal restrictions.
An individual case or rule would be promulgated under the Generic SNUR by listing
only the relevant requirements.
B. .Existing Chemical Review Process
Under Section 6 of TSCA, OTS selects chemicals and evaluates the need to
control aspects of their manufacture, processing, use, distribution in commerce, or
disposal, OTS may refer chemicals reviewed in the existing chemicals program to
other Federal agencies for action under TSCA Section 9. It also may regulate the
existing chemicals under TSCA Section 4 (testing for toxicity), TSCA Section 5 (SNUR
for other uses), or TSCA Section 8 (reporting rules), or refer them for action under
other EPA authorities.
Typically an existing chemical review has 3 phases: 1) risk characterization,
2) detailed risk assessment, and 3) risk management and implementation. Once
assigned to a particular chemical, the CEB engineer is typically responsible for work
on it throughout the process. This process may require several years to complete,
although there are significant variations with individual cases. The ETD existing
chemical coordinators, section chiefs from each branch, manage the existing chemical
activities.
Currently, the existing chemicals review process is being redefined. The
current process is based on the 3 phases mentioned above. The first phase is called
RM 1 short for Risk Management Phase 1. RM 1 is a 12 to 13 week review of a .
chemical based loosely on the PMN process. In this process there is a chemistry
-------
Issued: February 28, 1991
Revised:
Page No.: 2-8
review, a market study, an engineering assessment, a fate assessment, and a health
assessment all done in the first 7 weeks of the process. Approximately, 100 hours of
total OTS staff effort is expected to be expended on a typical RM 1 chemical case.
The CEB engineer will use readily availabte sources of information to determine the
populations exposed, the occupational exposures and the environmental releases,
The level of CEB effort is in the range of 10 -15 total hours per RM 1 case.
Over the next 3-4 weeks the lead ECAD person on the case would perform
a cursory risk assessment and recommend several options to control the risks. This
ECAD document is called a chemical's dossier. The engineer on the case will review
the dossier for technical accuracy and reasonableness of the proposed control
options. After the dossier has finished review the case is presented at a decision
meeting called the RM-1 meeting. At this meeting the case is discussed and final
recommendations are agreed to by the division directors. Decisions from this meeting
include: 1) drop the case as there ts no unreasonable risk or that the existing risks are
being dealt with by another agency (referral), 2) gathering more data under the RM 1
process, 3} putting the chemical into RM 2 or RM 3, or, 4) non-regulatory options such
as writing concern letters to the companies or an informal referral of the chemical case
to other agencies or organizations such as OSHA, NJDSH, CPSC, ACGH-J, and FDA.
The second phase of the existing chemical process is RM 2. This phase is
currently being redefined to better match with the division director's needs on pollution
prevention, toxic use reduction, and international activities. This phase is a type of a
risk assessment phase for cases which OTS may want to regulate under TCSA.
Although there is a risk assessment performed during the RM 1 process rt is very
cursory and it is not of sufficient quality to regulate under the "will present" finding
needed for section 6 regulations. However, the RM 1 risk assessment focuses on the
highest risks to see whether they are "unreasonable",
CEB's role during RM 2 is to evaluate the potential for worker exposures or
environmental releases associated with manufacture, processing, use, or disposal of
the chemical. Tasks to be performed by the engineer may include: 1) gathering and
reviewing al! available literature information on the chemical including TRI data, NIOSH
-------
Issued: February 28, 1991
Revised:
Page No.: 2-9
and OSHA data and data from other offices of EPA; 2} contacting manufacturers,
processors, or users and other sources for unpublished information; 3) characterizing
the releases and exposures expected in industry and the controls typically used; 4)
developing estimates of the populations exposed and the duration, frequency, and
levels of releases and exposures (where insufficient data exists). This assessment
should characterize the range and the central tendency of the available data or provide
estimates in the absence of data. The uncertainities in the presented information
should be discussed; 5) identifying monitoring techniques for EED to validate and
providing input on the data's quality; and 6} recommending exposure or release
scenarios that could be considered for additional monitoring studies. The actual tasks
the engineer preforms on a given case may vary a great deal from the list above.
The last phase of the existing chemicals review process is called RM 3. This
is the risk management and implementation phase. In RM 3 the regulatory and non-
regulatory options are selected to control unreasonable risks posed by a chemical and
they are implemented. The CCD lead person manages these activities drawing
support from the different branches in OTS as needed. CEB support includes refining
the exposure, release, and control technology assessments as needed to support the
different regulatory options and responding to comments generated by proposed rules
and advisories. Any changes or new information may impact on RIB's support
documents.
Bisk management requires the identification of feasible alternatives available
to reduce the risks associated with production, use, or disposal of a chemical. This
includes a consideration of substitutesmaterials and an economic assessment of the
control alternatives. The tasks assigned to CEB include: 1) identifying potential control
schemes to limit exposures and releases including engineering controls, administrative
controls, and personal protective equipment. The use of pollution prevention alter-
natives should be emphasized; 2) determining the effectiveness of the control alter-
natives and estimating the potential exposures after their application; 3) estimating the
operating and capital costs of each control alternative; and 4} assessing the impact of
substitutes' use on industrial processes.
-------
issued: February 28, 1991
Revised:
Page No.: 2-10
Section 313 Petitions
Section 313 of EPCRA requires annual reporting of releases for over 300
listed chemicals and chemical categories. The TRI database, which has all reported
information, is a valuable source of data for the engineer in existing chemical cases
and PMN reviews.
Under Section 313 of EPCRA, EPA may be petitioned to add chemicals to or
delist chemicals from the list of chemicals subject to TRI reporting. For chemicals
already on the list, the release information reported in the previous year is reviewed,
For chemicals to be added, data must be gathered from many disparate sources.
Anyone can petition EPA to add a chemical to or remove a chemical from the list.
EPA has 180 days either to deny a petition or take the initial rulemaking steps to
change the list
In the case of deleting a listed chemical, the engineer's role in petition review
is to review and evaluate all release data in the database for the previous reporting
years. The basis of the engineer's report is a printout of release data reported by
each facility, grouped by Standard Industrial Classification (SIC) Code (See Appendix
D for sample TRI data). The engineer must evaluate the reasonableness of reported
data, which are after all, estimates. Evaluation includes comparing the releases of
similar facilities, independently estimating a release based on such parameters as
known production volume of a facility and published emission factors, and contacting
the facility to discuss the data upon which the release estimate was based.
In the case of a petition to add a chemical to the list, the engineer must
develop an assessment based on published data, contact with manufacturers, and
various estimation methods. The economist's assessment of markets for the chemica!
is most important in beginning this work.
Petition review draws on a team representing all technical disciplines within
OTS. Other EPA program offices participate as well. If the Assistant Administrator
decides on a rutemaking, the engineer participates in preparing a proposed rule,
evaluating public comments on the proposal, and completing the rule.
-------
Issued: February 28, 1991
Revised:
Page No,: 3-1
ill. APPROACHES AND DATA SOURCES FQR ASSESSMENT
A. Aggma,chgs
Ideally, the CEB engineer would use information specific to the chemical
under consideration in all evaluations. This ideal may be possible in longer term
analyses of existing chemicals where existing data can be identified or monitoring can
be conducted. However, data on environmental releases and workplace exposures
almost never exist for PMN chemicals. These data also may not exist or be readily
accessible for many end uses of an existing chemical. The engineer therefore may
have to make preliminary estimates to decide whether further analysis is warranted.
When data are not available for the specific chemical (e.g., monitoring data
at onty one potential use site under one set of conditions), exposure or release may
be estimated based on analogy to similar chemicals with similar physical characteris-
tics that are similarly handled. Great care must be used in drawing such an analogy,
and the results should be presented recognizing the uncertainties in the data used and
any assumptions made.
The engineer must determine the process characteristics of operations
involving the chemical, as process details chosen by manufacturers or users of
chemicals directly affect exposure or release. For example, often the choice to use a
given reactor or a solvent cleanup is based solely on the existence of the necessary
equipment (i.e, the particular reactor is available or an existing solvent recovery system
at the site makes solvent use feasible). Engineering controls, personal protective
equipment, batch sizes, reactor types, and many other factors will vary even for
chemicals in the same class.
For PMN chemicals, where the single (usually) manufacturer is known, exact
information (e.g., number of sites, number of ¥/orkers, days per year, activities) may
be available and should be used. However, these plans may not be final and may
change. In addition, where there are multiple use sftes, conditions can be expected to
vary greatly and the identify of users is often not revealed (if known). The engineer
must try to accommodate these variations by estimating a range of exposures or
-------
Issued: February 28, 1991
Revised:
Page No,: 3-2
releases (e.g., reporting the range, not just the mean, of measured airborne
concentrations of a material).
Where information specific to the manufacturers, processors, or users of a
chemical is not available, the engineer should use a "reasonable worst case" analysis
to arrive at an estimate. For example, genera! ventilation rates in industry range from
a low of 500 ft3/min to over 10,000 ft3/min, with a typical value of 3000 fts/min
(Clement 1982). If no information is available on the general ventilation rates of the
users (individually or as a group), a "reasonable worst case" assumption is thai the
rate could be as low as 500 ft3/min.
The reasonable worst case approach is adopted for PMN chemicals in
particular to cover the range of possibilities for the future use of the chemical, If these
conditions do not pose an unreasonable risk, then concerns for the chemical may be
resolved. Careful consideration must be given to all aspects of the case to ensure
that the assessment is reasonable. For example, exposure or release may already be
controlled due to OSHA or EPA requirements for associated chemicals, or materials
may be corrosive so that workers wear gloves of necessity, or the industry in question
may be known to use local exhaust (rather than just general ventilation) to control
known hazards.
It is of fundamental importance to establish the circumstances of exposure
or release (e.g., daily handling of large quantities of a powder material). Estimated
levels of exposure can be misrepresented unless firmly based on an understanding of
the operations, work practices, controls, and other circumstances involved in the
manufacture or use of the chemical.
End use scenarios also must be developed through an understanding of the
operations involved in the use of a chemical. These scenarios are best constructed by
considering the principal business of the user. For example, based on the number
and range of automobiles built at an average facility and the amount of primer applied
per car, the engineer could estimate the amount of a pigment in the primer used per
site, the number of sites at which it could be used, and the number of workers
potentially involved (the business of the user is building cars, not using pigment).
-------
Issued: February 28, 1991
Revised:
Page No.: 3-3
Section IV presents models that should be used to estimate levels of worker
exposure in the absence of data on the chemical or on analogues. These models
require judgment by the engineer in choosing parameters to fit each case. Section V
presents standard approaches for estimating water release for basic operations often
found in manufacture. Section V! presents standard approaches for estimating
releases to air. Section VII presents information on controls that may be used to
reduce release to both water and air.
B, Data
There are many resources available to the engineer besides the information
provided by the PMN submitter or other direct industry sources. Appendix E identified
ORD, OTS, CEB, and OAQPS documents for different processes and chemicals. Six
standard sources should always be checked for relevant process and release infor-
mation.
Effluent Guideline Series. This source contains information on the release of
many pollutants to wastewater for industries that represent major sources of
wastewater release. See Volume II for a more complete list of the guidelines
(CEB n.d.).
.NewJjQurce Performance Standards (NSPS). This source contains air
release limitations that apply to new sources in 58 industries. Many
standards involve particulates or VOCs that may be useful in determining
reasonable releases for constituents of the release. Background documents
prepared in support of the New Source Performance Standards are most
helpful A listing of the New Source Performance Standards is contained in
Appendix E.
Industrial Process Profiles for Environmental Use QPPEU). This source
contains detailed process flow diagrams including input materials, process
conditions, and release estimates for 29 broad industries. A separate
volume addresses each industry, and some industries (e.g., plastics and
-------
Issued: February 28, 1991
Revised:
Page No.: 3-4
resins) contain over 20 specific process flow sheets and 60 process
descriptions, A listing of the IPPEU's is contained in Appendix E.
Kirk-Othmer Encyclopedia gf Chemical Technology. This source contains
detailed and well documented information on almost every type of chemical
process. The bibliography after each section represents an excellent source
of additional information. The author of each section represents an expert in
the fieid who is typically knowledgeable of current trends in the industry,
Compilation of Air Pollutant Emission Factors. AP-42. This source contains
process descriptions, emission factor estimates, and control information on
more than 120 processes. Emission factors are for the criteria pollutants.
Factors for total particulates or VOC can be used with composition
information to prepare rough estimates of releases from similar processes.
Sections on release from the storage and transfer of organic liquids are
particularly useful in calculating release for CEB assessments.
Past PMN cases. Many PMNs involve similar types of chemicals and similar
processes. Past PMNs therefore represent an excellent source of informa-
tion on methodologies that may be used to calculate worker exposure and
environmental release.
Title IJLSection J13JBMeasjJlejDortJn^ These reports
contain brief descriptions of the industry, identify potential release points,
and model calculations for estimating releases. A listing of these guidance
documents is contained in Appendix E,
In addition, there are standard sources that the engineer should always
check for occupational exposure information:
NIQS.H Health Hazard Evaluations (HHEsj and Industry-Wide Surveys
jQWSjj)- These sources contain well-documented occupational exposure
measurements. They generally also contain process and job descriptions,
OSHA NatimM^iiJIJl^irnpliDfl-Results. This source contains occupational
exposure measurements of approximately 600 chemicals from 1979 to.
present. Summaries are in CEB files.
-------
Issued: February 28, 1991
Revised:
Page No.: 3-5
NIOSH National Occupational Exposure ...Survey (NOES). This source
contains estimates of the number of workers exposed to many existing
chemicals. It also contains a product database giving the concentration of
the chemical in many product formulas.
Besides these sources, Appendix E contains a list of process-specific reports that
should be consulted when applicable.
Volume II of the manual contains copies of additional reports that may be
useful in the preparation of release or exposure assessments. These reports are
unpublished contractor reports done for CEB or published reports that were not widely
distributed. A description of the documents follows,
CEB Research Project: Effluent Guidelines Information (Parts A and Bj.
This document contains information that may be used to estimate release of
pollutants to wastewater from tank truck cleaning operations. It also
provides a description of the contents of development documents produced
by the Office of Water.
Carbon Adsorption Report. This document provides information on the use
of carbon adsorption systems, chemicals treated by carbon adsorption,
occupational exposure to spent carbon, and the environmental impact of
carbon adsorption.
Generic Engineering Assessment - Spray Coating: Occupational Exposure
an,d,.EnyirQnmienial. Release. This document contains generic exposure and
environmental release scenarios for various spray painting operations. It
contains specific information on automotive finishing and refinishing, metal
and wood furniture finishing, large appliance finishing, non-autornotive
transportation finishing, and heavy machinery finishing,
CEB Research Project:__Engjnggring Standards. This document contains a
listing of organizations that voluntarily develop standards that are of interest
to CEB. it also provides descriptions of the standards or recommended
practices suggested by each organization.
-------
Issued: February 28, 1991
Revised:
Page No.: 3-6
generic Engineering Assessment - LeatheLDyjng: Occupational
and Environmental Release. This document contains generic exposure and
release scenarios for leather dying operations. It also provides specific
information on the weighing and transfer of smal! quantities of dyes.
.CEB Research Project: Industrial Hazardous Wj§it. incineration. This
document describes methods of incineration, types of incineration devices
and expected efficiencies, air emission estimation, cost estimation, Federal
regulations, and occupational exposure.
PoJyetectrolytes. This document describes how polyelectrolytes are used
and how they are removed from wastewater. It also provides a method for
estimating release of polyelectrolytes to water.
Drilling _FMfe=:=&wirQjinigMal Release Analysis. This document describes
the types of waste fluids produced from off-shore and land-based drilling
operations. It also provides information on the types of chemicals that are
contained in drilling fluids and provides methods for estimating their release
to all media.
Information on the loading and Unloading of Chemicals under Nitrogen
Blanket, This document provides typical workplace concentration levels of
three chemicals during unloading and loading under nitrogen blanket. It
also describes general procedures for loading and unloading chemicals
under nitrogen blanket.
Particulates in the Workplace. This document compiles data on typical
paniculate concentration and size listed by industry segment and worker
activity.
Strategy for Recommending Respirators for CgntroljgfJExggguiggJo
.Substances Undergoing PremajTyfajctuj^J3evi.ew. The document provides
information to be used to identify types of alternative respiratory protection
when limited information is available on a PMN substance. It includes tables
listing the different types of respirators, their assigned protection factor
levels, and the capabilities and limitations associated with each.
-------
Issued; February 28, 1991
Revised:
Page No.: 3-7
Use of Oil Separators .in Drum Reconditioning and Transportation
Vessel-Cleaning Facilities. This document profiles the drum reconditioning
and transportation vessel-cleaning industry and characterizes the waste
streams generated and wastewater treatment devices employed in the
industry. It describes oil separation processes and presents estimates of oil
separator effectiveness,
Cost of Selected Engineering Controls. This document presents cost data
for engineering controls for five activities. These activities are quality control
sampling, drumming of a liquid, bagging of a solid, reactor process vent,
and open tank operations.
industrial Process Profiles to Support PM.N Review: Metal Treatment
Chemicals. This document provides descriptions of many types of metal
treatment operations and the chemicals used in each operation. It also
contains information on occupational exposure and environmental releases
associated with metal treatment operations. (A listing of other Industrial
Process Profiles to Support PMN Review is contained in Appendix E).
Figure 3-1 contains a matrix of the references contained in Volume II and the types of
information contained in each.
Volume 111 of this manual contains descriptions of the databases that may be
accessed for specific information. Figure 3-2 contains a matrix of these databases
and the types of information that may be accessed.
-------
Mrfflftfca
L-bintf !
, t :e • IH.II
r.-[t ian?.i
1 r.ib'-MH,
rrr.wr
r-P'Wd
j Ltb^SU
r~r.«iw
My»*". h d i
PEI 'In'il
PCI t9ft»h
Wilk in.d.)
Wong )»8
Plwslcatf
chcmlcsl
properties
K
X
X
Pr ««5«
1^
walw
y
X
X
X
X
X
X
X
" Ptvt-tsrse
Tri*Jl
abiiiiv
X
X
X
X
X
i»ll
mwjla}
j(
X
X
X
X
X
X
Ls« of
wqulamri
chcniiail'.
Technical •
«xp«its ^
X
-m.
Figure 3-1. Matrix of data contained in Volume II.
Q) CD CO
to < c
CD to CD
~ G> O,
CO
1
03
-n
CD
O"
c
Q)
ro
CD
-------
*™
CMKJ1 Of
CK*T*.
eus
D.PF.
IH Mr*-l!s*'i'ia
IN0K
1506
HOC*
OSHA
«,,*,
SPUD
Tridi .ft*sr3
KW*LK
sss
X
A
X
X
K
hoto^
X
>
0*J»t«5
X
»
X
S.C
3s
»
X
*r
K
X
K
If
^~
>
r,: ;:.
X
„,,„,.
X
.^n
^
_
*,.,.
je
'
T.«,
s
!(
,^ {
r;-'^ ,
K
T«7o7-
;
i
i
i
i
X
1
TJ 3D of
03 OJ CXI
Q.
CD
OJ CT
* 1
•2
ro
Figure 3-2, Matrix of data bases contained in Volume 111, to
-------
Issued: February 28, 1991
Revised:
Page No.: 4-1
IV. MODELING WORKPLACE EXPOSURE
CEB has developed standard approaches for estimating exposure levels to
workers via the inhalation and dermal routes and for assessing the effectiveness of
controls, including ventilation, respirators, and gloves. In the absence of represen-
tative data, CEB engineers should use these approaches to quantify worker expo-
sures.
A. Estimating Inhalation Exposures
The amount of substance inhaled by a worker is a function of many
variables including the airborne concentration of the substance, the amount of time
spent in an atmosphere containing the substance, the breathing rate of the worker, the
worker activity or job performed, the physical and chemical properties of the sub-
stance, the temperature changes, seasonal changes and the effectiveness of engi-
neering controls or personal protective equipment in protecting the worker. In
assessing the potential for worker exposures at multiple processing and use sites, the
controls used at any or all the sites may not be known. Thus, CEB engineers must
estimate reasonable worst case exposures representing sites where no controls are
used. If information is available about a particular site or general industry practices,
the engineer atso may estimate exposures assuming an effectiveness for the expected
control.
This section describes the assessment of occupational exposure in the
absence of respirators and engineering controls. Sections IV.C and tV.D discuss the
effects of respirators and engineering controls, respectively, in reducing worker
exposure.
1- Genera! Approach
As stated previously, inhalation exposure is a function of many factors
including airborne concentration, duration of exposure, inhalation rate, and effective-
ness of engineering controls and personal protective equipment. Neglecting any use
of controls or personal protective equipment to reduce the airborne concentration, the
-------
Issued: February 28S 1991
Revised:
Page No,: 4-2
amount of substance available for inhalation is calculated in units of mg/day as:
/ = Cn bh Equation 4-1
where: \ = Daily inhalation exposure, mg/day
Cm = Airborne concentration of substance, mg/rn3
b = Inhalation rate, rn3/hr
h = Duration, hr/day
Determination of inhalation rate and duration of exposure is generally straightforward,
and is discussed in the following paragraphs, Determination of airborne concentration
can be difficult, and is discussed in detail in Section IV.A.2.
With increased physical activity, inhalation rate increases. The typical worker
breathes about 10 m3 of air in 8 hours, or 1.25 m3/hr (NIOSH 1976). This value is
slightly above the volumetric flowrate for light work (NfOSH 1976). During the
workday, the volumetric flowrate at any given instant may vary widely as a function of
the type of work being performed. Inhalation rates range from about 0.56 m3/hr
during rest periods to 7.9 m3/hr during the maximum work, as shown in Table 4-1,
TABLE 4-1. INHALATION RATES
Activity
Rest
Light work
Medium work
Hed, heavy work
Heavy work
Maximum work
Minute volume
air flow
rates
myhr
0.56
1.18
1.75
2.63
3.6
7.9
Source: NIOSH 1976
The duration of exposure can only be estimated on a case-by-case basis through
knowledge of the activities that may lead to exposure. This information can be
obtained from persons knowledgeable about the process or from descriptions of
similar operations. Airborne concentrations (Permissible Exposure Levels [PELs]) or
-------
Issued: February 28, 1991
Revised:
Page No,: 4-3
measured data are often expressed as an 8-hr Time-Weighted Average (TWA), When
using an 8-hr TWA for purposes of estimating exposure, the actual duration of
exposure is not used to calculate the exposure, A TWA value implies that exposure
may be high for short periods of time as long as those periods are compensated by
periods of lower exposure and the average exposure over 8 hours does not exceed
the TWA. The maximum daily exposure is the total amount of substance to which the
worker can be exposed per 8-hr shift whether the exposures are for short duration at
high concentrations or long duration at low concentrations. It is calculated using a
duration of 8 hr/day.
If the inhalation rate on average is 1.25 m3/hr, the equation for calculating the
amount of substance available for inhalation becomes:
! = 1.25/1 Cm Equation 4-2
When Cm is estimated as an 8-hr TWA concentration, the duration of exposure is
assumed to be 8 hr/day and the equation reduces to:
I * 10Cm Equation 4-3
The remainder of this section discusses methods for estimating airborne
concentrations for use in Equations 4-1, 4-2, or 4-3.
2.
Worker inhalation exposure is best determined using personal
monitoring measurements for workers performing the job under study while being
exposed to "typical" pollutant levels. Section 111 discusses some sources of monitoring
data. Examples of chemicals for which the engineering assessments have been
based on industrial hygiene monitoring data include butadiene, acrylamide, and
chlorinated solvents.
Since monitoring data are seldom available for PMN chemicals, CEB
engineers usually rely on other methods to assess worker exposures. These methods
are also used in the analysis of existing chemicals for which no monitoring data exist
or when available monitoring data are not applicable to a particular exposure scenario.
-------
Issued: February 28, 1991
Revised:
Page No.: 4-4
Predictions of the expected airborne concentration of a substance may be based on
monitoring data for analogous substances, OSHA PEL'S for substances present in the
workplace, and mass balance models.
a. Using Monitoring Data
When monitoring data on occupational exposure are available, the
utility of the data should be evaluated following the process established for CEB that is
described in "Guidelines for Statistical Analysis of Occupational Exposure Data" (PEI
1989}, The quality of the exposure data that CEB assesses varies from poorly
characterized, summary data to well-characterized sets of individual data points. In
almost all instances, the quantity of available data is limited. The guidelines describe
the treatment of uncertainties, assumptions, and biases in the data. With the assis-
tance of an industrial hygienist and a statistician, the CEB engineer can use the
guidelines to categorize the data and perform the statistical analysis. Appendix F
briefly describes these guidelines.
b. Using Analogous Data
(1)
The airborne concentration of vapors and particulates may be
estimated using personal monitoring measurements for analogous chemicals or
processes, In each case, similarities must exist in physical/chemical properties of the
chemicals, nature of workplace environment, quantites of material handled, and worker
activities associated with use of the chemical. Chapter HI discusses typical sources of
chemical-specific data, such as the OSHA Compliance Database, NIOSH industry-wide
surveys, and NIOSH Health Hazard Evaluations (HHE's). In addition, CEB has
developed descriptive generic scenarios that include summaries of the available
monitoring data for industries that are frequently assessed. Currently, these scenarios
cover metalworking, textile dyeing, and spray coating.
Although estimates of airborne concentrations may be based
on analogous chemicals or processes, caution should be used when making the
-------
Issued: February 28, 1991
Revised;
Page No.: 4-5
analogy. This is a type of reality check, If the exposure calculations exceed the PEL,
Short-Term Exposure Level (STEL), or Immediately Dangerous to Ufe or Health (IDLH)
levels, then the assumptions or calculation methods should be checked for their
validity.
To estimate airborne concentrations for vapors from
analogous data, the following simple relationship has been derived:
P X
C = C , s Equation 4-4
V.. V,
where: Cvc = Estimated airborne concentration of the PMN chemical, pprn
C, , - Measured airborne concentration of the known chemical, ppm
Ps, = Vapor pressure of PMN chemical, torr
Pk = Vapor pressure of known chemical, torr
Xs = Mole fraction of PMN chemical in mixture
Xk = Mole fraction of known chemical in mixture
The derivation of this relationship assumes that;
Vapor generation is driven by either evaporation from an open surface or
the displacement of saturated vapors from a container.
The liquid temperatures (T) and the mass transfer coefficients (K) of the
\
PMN chemical and the known substance are simiiar.
The workplace environments are similar: the ventilation rates (0) and mixing
factors (k) for the workplaces in which the PMN and the known chemical are
handled are essentially the same; the quantities of materials handled are
similar (for example, in an activity such as drumming, the volume of the
container and the fill rate for the PMN and the known are equal).
Raou!t's Law is valid.
Assuming the ideal gas law, the airborne concentration
expressed on a volume basis in units of ppm [C ] can be converted to airborne
-------
Issued: February 28, 1991
Revised:
Page No,: 4-6
concentration expressed on a mass in units of mg/m3 [Cm] using the following
equation:
Cm = Cv — Equation 4-5
where: Cm = Airborne concentration, mg/m3
Cv = Airborne concentration, ppm
M = Molecular weight of the chemical of interest, g/g-mole
V = Molar volume, liter/mole (use 24,45 liter/mole at 25* C and 760
mm Hg)
For participates, the analogous data may be expressed as
the airborne concentration of a particular substance or as total solids. To calculate the
airborne concentration, assume that the composition of the airborne parttculates is the
same as the composition of the bulk material. A ratio of the weight fractions of each
substance is used to calculate the airborne concentration of the PMN from
concentration of the known chemical,
Cm = Cmk -4 Equation4-6
where: C^,,, - Estimated airborne concentration of the PMN chemical, mg/m3
Cm> = Measured airborne concentration of the known chemical, mg/m3
Y5 = Weight fraction of PMN chemical in mixture
Yk = Weight fraction of known chemical in mixture
If the airborne concentration is measured as total solids, this equation becomes:
-------
Issued: February 28, 1991
Revised:
Page No.: 4-7
Cm.s = C^ >y Equation 4-7
where; C,Tj, = Estimated airborne concentration of the PMN chemical, mg/m3
Cn, k = Measured airborne concentration of total dust, mg/rrr
Y5" = Weight fraction of PMN chemical in the solids
When using an analogy for particulates, the following
parameters should be judged to be similar: hygroscopicity, moisture content, density,
particle shape, particle size and distribution, and static buildup potential Whenever
available, the engineer should obtain particle size information for the PMN chemical.
This information may be presented in several different ways, from a complete particle
size distribution to the limited identification of average size or percent of particulates
above or below certain cutoff sizes. Depending on the type of data submitted, the
p article size information may be used to more completely characterize the exposure to
the worker. The data should represent the PMN particle size distribution at the
potential exposure points.
If properly collected and analyzed, particle size information
may be used to identify the percent of particles in the respirabie range, Respirable
particulates are defined as those with an aerodynamic diameter of 3.5 ^m or less,
These particles are expected to reach the alveolated gas exchange portions of the
human respiratory system where they may be absorbed. Almost at! particulates that
are inhaled and are larger than the respirabie size are deposited in the upper respira-
tory tract. Those which deposit in the nasopharanx behind the nasal hairs tend to be
carried downward to the throat. Those which deposit in traecNobronchiai system are
carried upward to epiglottis. Those particles that are larger than the respirabie
particles tend to be ingested. A CEB industrial hygienist should be consulted to
determine the appropriate type of testing protocol for particle size testing, and to
provide guidance to the submitter.
(2) Spray Coating
One use scenario which CEB frequently assesses is industrial
spray application of coatings. To standardize and facilitate the engineering analysis,
-------
Issued: February 28, 1991
Revised:
Page No.: 4-8
CEB developed a report entitled "Generic Engineering Assessment Spray Coating"
(CEB 19878) that describes several generic spray coating scenarios. The report,
which is included in Volume II, is organized according to the industry of application:
automotive finishing and refinishing, wood and metal furniture finishing, large appliance
finishing, railroad car finishing, light aircraft finishing, and heavy machinery finishing.
For each industry, a matrix provides information that can be used to estimate typical
usage rate of coating, number of use sites, numbers of workers exposed by activity,
duration and frequency of exposure, types of protective equipment likely to be used,
and inhalation exposures to total mist and organic solvents,
Of the PMN chemicals that are intended for use in coatings,
most are nonvolatile substances used as resins, pigments, or other additives in the
coating. Therefore, the PMN chemical is usua!ly part of the solids, and inhalation
exposures can be estimated based on expected airborne concentrations of total mist
using Equation 4-7. Table 4-2 lists airborne concentrations of total mist categorized by
scenario.
TABLE 4-2. ESTIMATED AIRBORNE CONCENTRATIONS
OF TOTAL MIST FOR SPRAY COATING OPERATIONS
Industry Estimated 8-hr TWA (mg/m3) Total
Hist
Automotive
- Finishing Not expected to exceed levels for
refinishing operations.
- Refinishing 5
Furniture
- Wood 0,1 to 2,5
- Metal 0,1 to 23.5
Large appliance 35
Railroad car
Light aircraft
Heavy machinery 1 to 18
Source: O'Brien 1981, CEB 1987a
-------
Issued: February 28, 1991
Revised:
Page No,: 4-9
When no particular scenario represents the PMN use, the
OSHA PEL for particulates, not otherwise regulated, should be used to represent the
upper bound for potential exposure. Estimation of exposure based on PELs is
discussed later.
If the PMN is volatile, the potential airborne concentration can
be based on the airborne concentration of a known similar volatile compound.
Airborne concentration levels for a variety of known solvents are listed in the NIOSH
technical report entitled "Evaluation of Engineering Control Technology for Spray
Painting" (O'Brien 1981), If most of the volatile portion of the coating becomes
airborne during spray application, the airborne concentration of the PMN is estimated
to be a function of the ratio of moie fractions:
CViS - CVih -1 Equation 4-8
Mi
where: C¥,t = Estimated airborne concentration of the PMN chemical, ppm
Cy_ = Measured airborne concentration of the known chemical, ppm
Ys = Weight fraction of PMN chemical in mixture
Yi = Weight fraction of known chemical in mixture
(3) Textile Dye Weighers
A description of textile dyeing operations is found in the
in-house report entitled 'The Dyeing and Printing of Textile Fibers" (Heath 1984). This
comprehensive overview includes information on types of dyes, typical formulations of
dyes, days of operation, numbers of workers, throughput rates, and releases for both
batch and continuous dyeing operations. Based on the report, dyes are used that
range from less than 0.001 percent to greater than 4 percent fiber active coloring
material based on the weight of the fiber (wof). The level of dye as formulated will
depend on the depth of color needed. For example, the active coloring agent may be
1 percent of the dye solution if used as a moderate-deep base color and 0,1 percent if
-------
Issued; February 28, 1991
Revised:
Page No.: 4-10
used as a shading agent. Darker shades such as black, navy blue, and orange may
require 5 percent or more dye as formulated.
A typical dyehouse employs one dye weigher per shift and
may operate 3 shifts per day over 5 to 7 days per week. Inhalation exposures to
particulates during dye weighing are expected.
To better assess occupational exposures to powder dyes,
EPA has conducted a study of dye weighers at 24 textile dye/print houses. Quantities
of individual formulated dyes handled by workers in the study using a "scoop/shake/
pour" method ranged from 0.001 to 54 kg/worker/shift. On an interim basis, CEB
estimates the worst case and average dye dust concentration per weighing ratio as a
basis for assessing workplace inhalation dust exposures. The assessment of worker
exposure to dust via the inhalation route is calculated from the following formula:
Typical Case:
/ = 0.0314 x % concentration, x no, weighings/day Equation 4-9
based on solids
Worst Case;
/ = 0,170 x % concentration x no. weighings/day Equation 4-10
based on solids
The percent concentration is either provided in the PMN or
other source or may be estimated (see Appendix C of Heath 1984). The number of
weighings of dyestuff is an engineering judgement based on mass used per site and
conditions of use. Table 4-3 provides typical values.
TABLE 4-3. TYPICAL PARAMETERS FOR DYE WEIGHING OPERATIONS
Type of No. of dye units No. of dyelots No. of dyeings
operation per site-day per shift/unit per dyelot
Batch 3 1,33 1-3
Continuous 1 1 3-4
JPrint 1 _ _ _1 2-4/screen
Source: Seath 19B4
-------
Issued: February 28, 1991
Revised:
Page No.: 4-11
(4) Weighing and Transfers of Small Quantities ofJSojids
Workers who weigh or transfer chemicals similarly to those in
dye weinghing activities (i.e., similar quantities of less than 54 kg/worker/shift and
similar handling techniques of "scoop/shake/pour") are expected to be exposed to
similar levels of dust. For example, airborne concentration levels of paper and leather
dyes have been found to be similar to levels of textile dyes (Heath 1988). Thus, in
these instances, the equations for dye weighers can be used to estimate exposures.
(5) Metal Working Operations
Information on the operations, numbers of workers and
activities, and occupational exposures at machine shops and selected metalworking
operations can be found in "Exposure to N-Nitrosodiethanolamine in Machine Shops'*
(CEB 1984b) and "Exposure to N-Nitrosodiethanolamine in Selected Metalworking
Operations" {CEB 1984c). Table 4-4 presents estimated inhalation exposures based
on NIOSH field studies and OSHA compliance monitoring data collected for n-nitro-
sodiethanolamine (NDELA) and oil mist during the period of 1972 to 1984,
TABLE 4-4, ESTIHATEO AIRBORNE CONCENTRATIONS FOR HETALHQRtCING
Type of
facil ity
Machine shops
Machine shops
Rolling mill
Contaminant
NDELA (vapor)1
Oil mist
Oil mist
Airborne concentration
arithmetic average
(range)
0.04 M9/m3
(rtondetectable to 0,08 jug/itt }
1.2 mg/m3
(0.001 to 5 mg/m3r
0.24 ing/m3
(0.18 to 0.3 mg/m5)
* N-Nltrosodiethanolamine (NOELA)
CAS: 1116-54-7
HW: 110
VP: not available
The upoer end of this rsnqe based On OSHA monitoring data Is
actually 8.3 ing An' which exceeds the OSHA PEL for oil wist. Since
exeeedance of the PEL occurred only in a small number of Inspec-
tions and the DSKA data ray be biased by the fact that Inspections
are often conducted based on employees complaints aboyt the work-
place, the DSHA PEL fcr oil trust has been used to represent the
upper bound.
Source; CEB 1983b, CEB !9B4c
-------
Issued: February 28, 1991
Revised:
Page No.: 4-12
The OSHA PEL for oil mist should be used to represent worst
case exposures for metalworking operations, Use of OSHA PEL'S as surrogates for
exposure or as upper bounds is discussed later.
(6) Pjjntjng
In assessing inhalation exposures in the printing industry,
CEB relies on a very limited data set. High-speed letterpress and lithographic printing
operations are expected to generate ink mist. High-speed processes include printing
of newspaper (300 to 400 m/min) and publications (300 to 370 m/min). Based on
two studies, occupational exposures to ink mist during printing operations range from
0,3 to 6.2 mg/mj (Gikis 1983), Occupational exposures to solvents during printing
operations are also expected.
(7) lire Manufacturing
Air monitoring data for occupational exposures to total
particulates are available for tire manufacturing (SIC code 3011). From 1979 to 1980,
NIOSH conducted detailed control technology assessments of seven tire
manufacturing sites. The individual plant reports contain air sampling data, identify
emission sources, describe work practices, and evaluate engineering controls for
various operations such as weighing, mixing, milling, and curing. The results of these
field studies are summarized in the NIOSH report entitled "Control of Air Contaminants
in Tire Manufacture" (NIOSH 1984).
The NIOSH studies found that workers may be exposed to
particulates from rubber, carbon black, process oils, vulcanizing agents, accelerators,
activators pigments, softeners, and plasticizers. Table 4-5 gives a typical compound
for tire manufacture.
-------
Issued: February 23, 1991
Revised: November 8, 1991
Pae No.: 4-13
TABIE 4-5. TYP1C3UL OMPOOND OMPOSEElai
TjDRE
wt %
Rubber 71
Carbon black is
Zinc oxide 2
Stearic acid 1.4
Softener 3 , 4
Antioxidant o . 7
Sulfur 2 . 0
Primary accelerator 0.5
j>econdary accelerator o.i
Source: N10S» T9S4
Workers may be exposed to dusts during transfers, weighing,
and mixing of raw materials. During milling, extrusion, calendering, and curing, workers
may be exposed to fumes that are emitted from the hot milled rubber. During tire
repair, workers are potentially exposed to rubber panicles from grinding. Exposures are
generally higher in the earlier stages of the manufacturing process when dry ingredients
are handled than in the latter stages that involve the building and handling of cured
product. The NIOSH report describes the operations and worker activities in detail.
PMN chemicals are frequently nonvolatile substances used as
resins, pigments, or other additives in rubber compounds. Therefore, the PMN chemical
is usually pan of the solids and inhalation exposures can be estimated based on expected
airborne concentrations of total solids using Equation 4-7.
Table 4-6 lists airborne concentrations of total paniculates
categorized by worker activity based on the NIOSH field studies. In engineering
assessments for lire manufacturing, note that carbon black has an OSHA PEL of 3.5
mg/rrr. 8-hr TWA.
An earlier study found that paniculate levels in compounding
and mixing are about 1 to 3 mg/nr', while those in milling, curing, and finishing are
generally less than 1 mg/nv7 (Williams 1980). These results agree with the NIOSH data
presented in Table 4-6.
-------
Issued: February 28. 1991
Revised: Novembers, 199
Page No.: 4-14
TZSJ3EE 4-6. SU-WKRY OF NIOSI M3OTORING DKEA FOR
EXPOSURES IN Tim~wtiUFMmmm QPERKTICNS BSSED cw
SEVEN
Total partlculates
range of exposures
_ Activity _ (mg/m ) _
Bin-filling (emptying bags into bins) 0,6 to 4,3
Manual weighing with LEV 1.5 to 2.2
Manual weighing without LEV 2.1 to 2.5
Mixing 0.08 to 1.54
Milling with LEV 0.2 to 1.22
Calendaring (under canopy hood) 0.07 to 0.4
Curing (automatic press/general ventila- 0.1 to 0.22
lion) 0,17 to 0.26
Tire repair with LEV _ ^^
Source; MIOSH *984
If the PMN is volatile, the potential airborne concentration can
he based on the airborne concentration of a known similar volatile compound using
Equation 4-4.
Worker exposure to pentane, hexane, heptane, benzene, and
toluene during tire manufacture has been studied (Van En 1980), Since PMN chemicals
typically are non-volatile, the solvent exposure data are not presented here.
,*. Use of OSHA PEL'S
When neither representative monitoring data nor a generic scenario
are available to describe potential exposures to the chemical of interest, the upper boimJ
"o the airborne concentration may be estimated using OSHA PELs for analogous
substances or substances present in the same workplace as the PMN chemical. To
estimate the airborne concentration, the PEL is used in the same equations as those u-<>J
to calculate concentrations from measured data,
The most frequent use of PELs is in the estimation of airborne
concentrations of solids for operations such as spray coating or solids handling. The
OSHA PEL for Paniculate, Not Otherwise Regulated, is IS mg/nrl 8-hr TWA (29 CTR
1910.1000, Table Z-l-A). Data collected by OSHA and NIOSH for total dust exposure-.
have been tabulated by SIC code (Mure 1985). A mean exposure value was
-------
Issued: February 28, 1991
Revised:
Page No.: 4-15
calculated for each SIC code. The mean exposures support the use of 15 mg/m1 to
represent airborne concentrations of total dust in the worker's breathing zone as a
substitute for specific data. This study has been included in Volume II.
The OSHA PEL is frequently used to describe exposures to oil
mists generated in operations such as metalworking and printing presses. The 8-hr
TWA PEL for oil mists is 5 mg/m3 (29 CFR 1910.100, Table 2-1-A).
To calculate the airborne concentration of PMN from either of
these PEL's, the following variant of Equation 4-7 is used:
Cn... = KC, y.' Equation 4-11
where: Cfns = Estimated airborne concentration of PMN substance
KCK = 8-hr TWA airborne concentration of known (either paniculate, not
elsewhere regulated or oil), mg/m3
Ys' = Weight fraction of PMN chemical in the solids or oil
To calculate the airborne concentration for vapors based on the
PEL, use Equation 4-4 substituting the PEL concentration (in pprn) for the measured
concentration.
d, US£-QLMajsJM3.nce Models
When information on analogous chemicals is not available, CEB
engineers use mass-balance or "box* models to predict airborne concentrations.
These models are most applicable for vapor and gaseous emissions because vapors
follow currents freely and are not influenced by gravity,
Worker exposures can be influenced by many variables: 1)
degree of automation, 2) employee work practices, 3) equipment design, age, and
frequency of maintenance, 4) container and closure design, 5) ventilation type and
rate, 6) employee use of protective equipment, 7) effectiveness of emission control
devices, and 8) product flammability. These factors must be considered before
estimating worker exposure using these models. This is especially true I the default
parameters are used to represent worst and typical conditions in the workplace?.
-------
issued; February 28, 1991
Revised:
Page No,; 4-16
One check of the models is that good design of equipment keeps
concentrations below 25 percent of the Lower Explosive Limit (LEI). A reasonable
worst case assumption should not exceed good design practice. For PMN chemicals,
however, the LEL or operating practices may not be known,
(1) Dimple Mathematical Model
The model frequently used to estimate worker exposures is
based on the mass balance of a substance in an enclosed space. It assumes the
fol towing:
There is perfect and instantaneous mixing of the contaminant with incoming
general dilution air,
The airborne concentration of the contaminant in the exfiltration air is the
same as in the room.
Only one source within the work area emits the contaminant,
The basic mass balance is expressed as:
dC
V—! = QC, + S -
eft
0 * el VC
Equation 4-12
, , c _ Contaminant concentration in workplace, [volume chemical/volume
air]
Q = Volumetric ventilation rate [volume/time]
~ _ Contaminant concentration in incoming dilution {infiltration) air
\j —
[volume/volume]
S = Source generation rate (volumetric) [volume/time]
e = Extinction rate [time"1]
V = Room volume [volume]
The following assumptions are made to simplify the model:
Extinction of the chemical (adsorption, absorption, or chemical
transformation) resulting from deposition on walls and equipment,
-------
Issued: February 28, 1991
Revised; November 8, 1991
Page No,: 4-17
condensation of hot vapors, and photodegradation of chemicals is
negligible.
The incoming air is contaminant-free,
The concentration of contaminant at initial time (t=0) is negligible.
The generation and ventilation rales are constant over time.
Room air and ventilation air mix ideally,
The concentration approaches the equilibrium concentration,
In its most simplified form, the model is expressed as:
C,. = £ Equation 4-13
where: C,_, = Contaminant concentration in workplace, ppm
S = Source generation rate (volumetric), cmj/h
Q = Volumetric ventilation rate, cnrVb
If available, a known ventilation rate for the workplace under
study is used. Volumetric ventilation rates are conventionally represented by units of
tV/min or cfm. General ventilation rates in industry range from a tow of 500 frVmin to
over 10,000 ftj/min; a typical value is 3,000 frVmin (Clement 1982). If the ventilation
rate is not known, a rate of 3,000 cfm (85 rrr/min) is considered typical and 500 cfm
(14.2 rrr/min) represents worst case. For outdoor operations with only minimal structure,
the ventilation rate in cfm is estimated as 26,400 v where v is the wind speed in mph
(Clement 1982). The average wind velocity is assumed to be 9 rnph (Clement 1982).
In reality, general dilution ventilation air does not always mix
perfectly and instantaneously with contaminated room air. Pockets of poorly mixed air
may be found in the room. Thus, a dimensionless mixing factor (k) is introduced to
describe the degree of mixing of the displaced ventilated air. Ideal mixing is represented
by a mixing factor (k) of L This mixing factor is a function of room size and locations of
the air iniet and exhaust. The ACGIH Ventilation Handbook suggests the following
-------
Issued: February 28, 1991
Revised: November 8. 1991
Page No.: 4-18
factors (ACG1H 1988): best (0,67 to 1), good (0.5 to 0.67), fair (0.2 to 0,5), and poor
(0. i to 0.2).
In using this model, a mixing factor of 0.5 represents a typical
case and 0.! represents a worst case. Incorporating the mixing factor and assuming the
ideal gas law, the model is expressed as:
(1.7 x I05) 7 G i--,,i
Equation 4-14
.If Q k
where: Cv, = Contaminant concentration in workplace, ppm
Ta = Ambient temperature of the air, K
G = Vapor generation rate, g/sec
M = Molecular weight, g/g-mole
Q = Ventilation rate, tV/min or cfm (Note that "cfm" is a conventional unit
for volumetric ventilation rates)
k = Mixing factor, dimensionless
Note: the factor 1.7 x itf in Equation 4-14 accounts for units con%'ersion and
is expressed in units of:
i »* i
sec atm cm' fr
g-rnole K min cm3
Equation 4-14 is the model which CEB engineers frequently use to
estimate the airborne concentrations of contaminants in the workplace. Use of this
model for specific scenarios will be discussed later.
(2) Comp'ex Mass Balance Models
The basic mass balance approach can be used to derive
models for other scenarios (e.g.. when the infiltration air is not contaminant-free, the con-
centration is not constant, or the initial concentration of contaminant is known). There
are three complex mass balance models that have previously been used by CEB to
-------
Issued: February 28, 1991
Revised: Novembers, 1991
Page No.: 4-19
estimate room concentrations. These are the Multi-Chamber Chemical Exposure Model
(MCCEM), the Consumer Products Safety Commission (CPSC) model, and the Monte
Carlo Simulation model. They have similar mass balance approaches to model exposures
to chemicals,
The MCCEM was developed for EPA, Office of Research and
Development (ORD) in Las Vegas, Nevada. MCCEM is available on diskette and
estimates airborne concentrations in the home based on user-supplied generation rates
(Geomet 1989). The model provides default values for room size, house size, infiltration
rales, and interzonal air flows. CEB has used this model to estimate occupational
exposures to house painters when using latex paints containing formaldehyde-releasing
biocides. Further information about the model is available in Volume III.
The CPSC model calculates exposure concentrations resulting
from a continuous release of a chemical substance in an enclosed space. This model has
been used by CEB to estimate exposure to chlorinated solvents. If the release of the
PMN chemical occurs over a relatively short time but the worker remains in the area for
other activities throughout the period, the mass balance equation must be solved for two
distinct periods. The first period is the time when the PMN chemical is released (t0)
until the time the PMN chemical has ceased being released (tL). The second time period
is after the PMN chemical has ceased being released (tj) until the time the worker leaves
the room (t2). Other assumptions for the model include that the pollutant concentration
at the start of the release, the outdoor pollutant concentration, and the rate of removal
by extinction are all zero. The model also assumes that the source strength and air
changes per hour are constant over the time of interest. The model also assumes ideal
mixing. The general solution for estimating room concentrations during the two time
periods is presented in Equation 4-15,
-------
Issued; February 28, 1991
Revised: Novembers, 1991
Page No.: 4-20
M =
a V
£- [1 - e^'r'o^-oWi^ Equation 4- 15
l
where: C(t) = Contaminant concentration in the workplace, ppm
S = Source strength, crrr/hr
Number of air exchanges per hour for the ventilation system, hr"1'
Room volume, trr1
Time when the pollutant is released, hr
Time when the pollutant release ceases, hr
Time when the worker leaves the room, hr
The Monte Carlo Simulation Model is a combination of a
complex mass balance mode] to estimate exposure concentrations of the chemical
substance in an enclosed space and a Monte Carlo analysis. The continuous release
model estimates the concentration of a chemical substance that is released at a constant
rate over a period of time until the exposure duration ends or the source ceases to emit
a substance, whichever comes first. This mode! has been computerized by EPA, ORD.
This model has been used to estimate room concentrations after releases of enlorofluoro-
carhons in various scenarios such as mobile air conditioning and refrigeration. Assump-
tions for this model include ihat the pollutant concentration at the start of the release,
the outdoor pollutant concentration, and the rate of removal by extinction are zero. The
model contains a factor that accounts for nonideal mixing, The solution of the mass
balance mode! for this model is presented in Equation 4-16.
-------
Issued: February 28, 1991
Revised: November 8, 1991
Page No,: 4-21
C =
Sr.
l-e
Equation 4-16
m =
V =
C0 =
where: C = Contaminant concentration in the workplace, g/m3
S = Source generation rate, g/hr
Residence time of the indoor air, hr
Mixing factor
Room volume, m3
Outdoor concentration, g/m3
Cjn, = Concentration in the room at the beginning, g/m3
t = Time, hr
The Monte Carlo analysis portion of this model consists of
three major steps. First, the CEB engineer selects a mean, standard deviation, minimum,
maximum, and distribution type for each of the input variables in Equation 4-16. Second,
the model is executed numerous times (usually between 500 and 1000 times), each time
wnh a unique combination of values for the input factors. The unique sets of inpui
factors are generated by the Monte Carlo model randomly sampling from the distribution
assigned to the factors, Third, the numerous values of exposure generated from this
iteration technique are then analyzed statistically to estimate a mean, standard deviation,
minimum and maximum, and 95th percentile. The primary statistic is the mean value for
each scenario. This value is more appropriately referred to as the estimated likely
exposure because it is derived using a variation in input values. In most cases, the
predicted 95th percentile value may be used to represent the reasonable worst-case
exposure. The validity of this model is highly dependent on how well the distribution of
each input parameter is understood. Interpretations of the results must be made very
cautiously if uncertainty exists in the understanding of the input distributions,
-------
Issued: February 28, 1991
Revised: November 8, 1993
Pas,; No.: 4-22
Other approaches
Other approaches that are not commonly used are included
here for general information. The mass balance modeling approach does no! account for
dispersion patterns, room size, or location of the worker with respect to the source. To
determine the effect of worker location on inhalation exposures, data were collected
during laboratory pilot studies and an empirical model for predicting airborne concen-
trations during drumming was developed (MRI 1986). This model incorporates a factor
to account for the positioning of a motionless worker in relation to the source and the
ventilation flow direction. This distribution factor (d) represents the amount of time a
worker .spends in from of the fill station. The mode! does not incorporate the effect of
worker movement because standardization of worker movement is nearly impossible.
Slight movement may reduce the breathing height that may affect concentrations. The
model is expressed as:
Cm =
Q
- (100fW)S G Equation 4-P
where: Cm - Airborne concentration, mg/mj
Q = Ventilation rate, frVmin
d = Distribution factor, sec/or'
G = Vapor generation rate, g/sec
The final MRI report provides the following table of default distribution factors for drum
filling operations of liquids. One conclusion drawn from the study is that cross ventila-
tion ;iir flow produced lower exposures than rear ventilation air flow. CEB has not used
this model to dale in preparing assessments.
-------
Issued: February 28, 1991
Revised:
Page No.; 4-23
TABLE 4-7. DEFAULT DISTRIBUTION FACTORS
Filling operations
Automated
Manual
Airflow
Side
Rear
Side
Rear
d factors (sec/m3)
MW < 100 MW > 100
0.2 0.1
0,6 0.6
0.3 0,1
3 3
Source; MR! 1986.
(4) Modeling the Generation Rates
To estimate the airborne concentration of the contaminant,
the generation rate of the contaminant must be estimated. Several approaches to
calculating the generation rate have been developed.
Many models which CEB routinely uses to estimate the
generation rate are functions of vapor pressure of the substance or mixture. The
vapor pressure of a pure substance may be available in the literature, ICB usually
provides estimates of vapor pressure for PMN chemicals based on the chemical
structure.
if two boiling points at two different temperatures (T1t T2)
and pressures (P,,Pj) are known, the vapor pressure at 25" C (P25) may be calculated
as shown in equation 4-18;
-------
Issued: February 28, 1991
Revised;
Page No.: 4-24
25
ito
[P] = aim
where:
b =
(298 - 7t)
Equation 4-18
2QB(T2 - 7,)
[TJ = Kelvin - K
Equation 4-16 is derived from the Clausius-Clapeyron Equation, and assumes that the
heat of vaporization, 'H^, is constant. If the heat of vaporization and one boiling point
(T1 at P,,) are known, the Clausius-Clapeyron equation may be used to calculate the
vapor pressure. This is presented in Equation 4-19,
ln\
25
•H.
m
1
298
Equation 4-19
v/here: R - Universal gas constant, 8,314 J/mol "K
° Hm = Heat of vaporization, J/mol
For mixtures, the vapor pressure of a component {Pa in
atmospheres) may be calculated using Raoult's Law as shown in Equation 4-20:
-------
Issued: February 28, 1991
Revised:
Page No.: 4-25
Pa = P° Xa Equation 4-20
where:
M.
M
8
P * = Vapor pressure of pure substance, atm
X = Mole fraction of component
W- = Weight percent of component
M. = Molecular weight of component, g/g-mole
(3,0,1 denote components)
Raoult's Law may be too simplistic in certain circumstances because vapor/liquid
equilibria data do exist for certain binary systems. Where data exist, this should be
used instead of calculated values. For chemicals that are solid at ambient temperature
and volatilize by sublimation, standard chemical engineering techniques for estimating
physical properties are not adequate for predicting vapor pressure.
Transfer ....... Operations. When liquids are transferred,
displacement of saturated vapor from the container must be considered. If
evaporation rate is negligible in comparison to the displacement rate, the generation
rate is expressed as:
-------
Issued: February 28, 1991
Revised:
Page No.: 4-26
« / M V r P* !-.•„„<
G = _ Equation 4-21
3600 Ft TL *
where: G = Vapor generation rate, g/sec
I = Saturation factor, dimensioniess
M = Molecular weight, g/g-mot
V = Volume of container, cm3
r = Fill rate, units/hr
P ° = Vapor pressure of pure substance, aim
R = Universal gas constant, 82.05 atm cm3/g-mol ° K
TL = Liquid temperature, ° K
Equation 4-21 is frequently used to estimate worker
exposures during tank truck or tank car loading and drumming operations. Summary
Table 4-11 at the end of Section 4A presents the default values generally assumed by
CEB for container volume and fil! rate. If complete saturation of the vapor space within
the vessel is assumed, the saturation factor is equal to 1.
This model is used to estimate losses due to vapors
generated from bulk loading of petroleum products as described in AP-42 (USEPA
1985b), Saturation factors for tank truck loading of petroleum liquids are expected to
range from 0,5 to 1.45 (USEPA 1985b). Table 4-8 lists typical saturation factors by
mode of loading for tank trucks and tank cars,
TABLE 4-8. SATURATION FACTORS FOR BULK LOADING OPERATION
Mode of operation Saturation Factor (f),
dimensJonless
Submerged loading;
Clean cargo vessel 0.50
Normal dedicated service 0,60
Dedicated vapor balance service 1.00
Splash loading
Clean cargo vessel 1.45
Normal dedicated service 1.45
Dedicated_vagor_ba1ance service 1.00
Source; U.S. EPA 19B5b.
-------
Issued: February 28, 1991
Revised:
Page No.: 4-27
CEB engineers typically assume complete saturation of the
vapor space within the vessel (f = l) for both tank truck and tank car loading, Typical
and worst case parameters for tank truck and tank car loading are- presented in
summary Table 4-11 at the end of Section 4A.
An alternative method for estimating the vapor generation
rate for bulk loading of tank trucks and cars is the use of emission factors. OAQPS
has collected information on air emissions of volatile chemicals from different industrial
processes. These data are compiled into a report entitled Toxic Air Pollutant
Emission Factors - A Compilation for Selected Air Toxic Compounds and Sources"
(USEPA I988a). Emission factors from tank truck and car loading expressed as
amount of chemical released per tank truck/car loaded are presented for several
chemicals. Table 4-9 presents estimated generation rates assuming that two tank
trucks or one tank car can be loaded in an hour.
TABLE 4-9. AIR EMISSION FACTORS FOR LOADING
Generation rate
Operation/chemical (9/sec)
Tank truck loading
ethylene dibromide (product) 1 x 10"6
ethylene dibromide (gasoline) 1 x 1fJ3
ethylene dichloride (product) 1 x 10"5
ethylene dichloride (gasoline) 2,6 x 10"3
benzene 0,06
carbon tetrachloride, controlled 0.4
carbon tetrachloride, uncontrolled 2.7
chloroform 3.9
Tank car loading
acrylonitrile, wet scrubber 4.72 x 1CT6
acrylonitriie 4.6 x 1CT4
Source: USEPA 1988a.
These generation rates were compared to the model
(Equation 4-21). The model was found generally to overestimate the generation rate
by several orders of magnitude. This overestimation is likely because the model does
-------
Issued: February 28, 1991
Revised:
Page No.: 4-28
not consider the effects of engineering controls used to recover vapor losses as do
the actual data.
Two modes of drum-filling were studied under two simulated
drum-filling conditions: 1) splash filling in which the liquid dispenser remains at the top
of the container and the liquid splashes freely, and 2) bottom filling in which the liquid
dispenser remains at the bottom of the drum to minimize volatilization (MRI 1986).
During splash filling, the saturation concentration was reached or exceeded by misting
(MRI 1986). The generation rate for bottom filling was one-half that for splash filling.
Thus, for bottom filling of drums, the saturation factor is expected to be about 0.5.
.Open Surfaces. Open surface operations include work
related to open vats or tanks, solvent dip tanks, open roller coating, and cleaning or
maintenance activities. Although this model is frequently used to estimate air
emissions from spills, CEB typically estimates exposures only during routine
operations.
Under contract to EPA, the evaporation rates of different pure
compounds in a test chamber were measured to determine an empirical model to
describe the relationship between evaporation rate and physical chemical properties.
The experiment is described in "Evaporation Rate of Volatile Liquids" (Pace Laborato-
ries 1989). The data are presented in this reference for 16 compounds studied at
different air velocities and temperatures, The data were curve-fitted. CEB is especially
concerned about low volatility chemicals and low air flow rates.
Based on mass balance of a differential element above a
liquid pool, the evaporation rate was derived (see Appendix K),
-------
Issued: February 28, 1991
Revised: November 8, 1991
Page No.: 4-29
G =
13,3167 M P" A
f
MX5
Equation 4-22
where: G = Generation rate, Ib/hr
M = Molecular weight, Ib/lb-mole
P8 — Vapor pressure, in. Hg
A = Area, ft2
D^ = Diffusion coefficient, ft2/hr of a through b (in this case b is air)
v2 — Air velocity, ft/hr
T = Temperature, °K
Az = Pool length along flow direction, ft
Gas diffusivities of volatile compounds in air are available for several existing chemicals.
However, the diffusion coefficient often will not be known. An equation to estimate
diffusion coefficients has been developed (see Appendix K). The expression for the
diffusion coefficient is expressed as:
0.5
4.09 x ltrW[JL + Ij (M)-*33 Equation 4-23
P,
where: Dab - Diffusion coefficient, crrr/sec
T = Temperature, °K
M = Molecular weight, g/g-mole
Pf = Pressure, atm
-------
Substituting into evap:
where: G
M
P* =
v-i =
A =
T =
A 2 =
yO.OSAr0.5 p».5
Generation rate, g/sec
Molecular weight, g/g-mole
Vapor pressure, mm Hg
Air velocity, ft/min
Area, cm2
Temperature, CK
Pool length along flow direction, cm
Overall pressure, atm
Issued: February 28, 1991
Revised; Novembers, 1991
Page No.: 4-30
8.24 x 1')(v°'625)
where:
G = Generation rate, lb/hr ft2
M = Molecular weigh! of evaporating substance, Ib/Ib mole
Pe = Vapor pressure at liquid temperature, in.Hg
v = Air velocity, ft/min
Equation 4-25
-------
Issued: February 28, 1991
Revised: November 8, 1991
Page No.: 4-31
Because open surface operations are highly process-specific, the engineer must make a
judgment about the surface area from which volatilization is occurring. Table 4-10 gives
conversions for typical diameter openings and surface areas:
TABLE 4-10. TYPICAL DIAMETERS AND AREAS
Diameter
3 ft
2ft
1 ft
6 in
3 in
2 in
1 in
z, cm
91.5
61
30.5
15.25
7.6
5.08
2.54
Surface area, cm2
6,500
3,000
700
180
42
20
5
Sampling. Sampling methods vary and include dipping, open
loop and closed loop sample bombs. For worst case assumptions, sampling from a tap
into an open container should be assumed. Equation 4-24 is used assuming that the
surface area is equivalent to the sample bottle opening. When no other information is
available, estimate the surface area to be 80 cm2 and 40 cm2 for the worst case and
typical case, respectively.
Fugitive Releases. Fugitive releases from valves, pump seals,
and connections can lead to significant occupational exposures, especially in enclosed
areas. Potential exposures are expected to be much less outdoors. To estimate occupa-
tional exposure from fugitive releases of gas or vapor for a new chemical from equipment
used in its manufacture, the number of fugitive sources and the duration of worker
exposure must be known. This information is typically not available at the PMN review
stage. OTS rareiy estimates occupational exposure from fugitive emissions of PMN
chemicals during the review process since it is assumed that fugitive releases are not the
most significant sources of occupational exposure.
The amount of contaminant leaking into the atmosphere can
be estimated based on emission factors for the different equipment components (see
-------
Issued; February 28. 1991
Revised: Novembers. 199!
Page No,: 4-32
Table 6-1). The vapor generation rate is calculated as a function of the number of
components, the appropriate emission factor for the component, and the weight fraction
of the component in the process stream. These emission factors were developed for air
pollution purposes and are discussed in Section VI of this manual.
(5) Summary Tables
The appropriate vapor generation rate is substituted into the
mass balance model to calculate the airborne concentration of the pollutant. The worst
case workplace conditions are represented by low general ventilation rates; typical
conditions are represented by greater ventilation rates. If the workplace is known to
have different ventilation rates than those generally assumed, the documented values
should be used as the basis for estimation.
To facilitate calculations of exposure, Tables 4-1] and 4-12 list
the typical and worst case assumptions for each potential source of exposure and present
the simplified concentration model The generation rates are "based on equation 4-21 for
transfers and 4-24 for open surfaces. The engineer should consider the applicability of
the parameters before using ihese models.
TABLE 4-11. SUMMARY OF CONCENTRATION CALCULATIONS FOR
TRANSFER OPERATIONS
Drumming (55 gal ) .
worst case
Typical case
Can5/boc:'.cs (5
93 1 . )
Worgt CflSC
typical case
Tanlr truck (5,000
gat . )
Uorst case
Typical, case
*ank ear (20,000
gal.)
worst case
"VD-. cal case
f CI«1
1 2.1 x 10*
0.5 2.1 x 105
1 '.9 x tot
„ 0,5 t.9 x 10"
1 1.9 * 10^
1 1.9 x 10'
1 7.6 x 1fl£
1 ?.S v, 10
r^
30
20
30
20
2
2
1
1
Q
ft /win
500
3,000
500
1,000
b
26,400 v
237,600
26,400 v
237,600
C
k
0.1
0.5
0.1
0.5
0.1
0.5
0,1
0.5
= 0.00075^ vPfr/Qk
ppm*
95 P
o
8,6 P
0.1 P
11 P/v
0. 24 P
22 P/v
0.48 P
The units for P are in run Hg.
Worst case out*»r ventilation (lowest yfnd sisced, v in nph) should be estimated by CEB engineer.
-------
Issued: February 28, 1991
Revised: November 8. 1991
Page No.: 4-33
TABLE 4-12. SUMMARY OF CONCENTRATION CALCULATIONS FOR
OPEN SURFACES
* z a k Cv = 31.4 P (1/29 * t/M)°-25 A/M °-165z°'5CSc
cm era ft /win
Sampt ing
Worst case
Typical cas*
Open surface
Worst ease
Typicttl c»se
80 10 500 0.1 16 P (1/2-?+ 1/M)°-2S/M °"1d5
40 7 3,500 O.S 0.3 P (1/29 + 1/H)°'Z5/M 0>1i5
b b 500 0.1 0.628 P (1/29 + l/M>°-25 A/H D-1651°-
b b 3000 Oj 0.021 ? (1/29 + i/«)°-25 A/H °-165Z:0-5
The units for P are in am Hg.
See Table 4-10.
B. Estimating Dermal Exposure
1. General Information
In comparison to inhalation exposure, the assessment methodology for
predicting dermal exposures is relatively simplistic. To assess the potential for derinal
contact with a chemical, it is necessary to identify the activity where potential contact
may occur, the likelihood of contact, the frequency of contact, the potential surface area
•.'-{ contact, the physical state of the contacted substance, associated chemicals, and the
likelihood and effectiveness of the use of protective equipment. With this information.
appropriate assumptions can be made to complete the assessment.
For liquids and many solids such as powders, granules, or flakes,, a
quantitative estimate of contact should be made. For materials such as gases, cast solids,
or corrosives, a quantitative estimate should not be made. For these materials a
qualitative estimate should be made using the following guidelines:
Corrosives - Express contact as negligible due to the corrosive nature of the
substance or associated compounds. Use for PMNs determined to be corro-
sive by SAT/HERD or for PMNs/mixtures with pH greater than 1.2 or less than
2. Consider contact points at which conrosivity may not apply, as in dilute
solutions, and quantify for them as needed.
-------
Issued; February 28, :
Revised: November 8, 1991
Page No.; 4-34
High temperatures - Express contact as negligible for materials that are at
temperatures above 140°F. Consider contact points at which temperature
would not be a factor.
Cast sQlids/PMN in .matrices - Do not quantify contact. If the material is
.manually transferred, acknowledge that some surface contact may occur,
"Dry" surface coatings (e.g.. fiber spin finishes) - If manual handling is neces-
sary and there is an indication that the material may abrade from the surface,
quantify contact with fingers/palms as appropriate.
Ssses^agors - Do not quantify contact, but acknowledge thai some dermal
contact will occur in the absence of protective clothing.
Qualitative estimates should describe dermal exposure using the following
exposure categories:
- This is used to describe workers who have no chance of dermal contact
during normal job activities.
Very low - This is used to describe workers who during typical job activities
would have no dermal exposure but who, on occasion, may have short periods
of exposure after which all contact surfaces would be washed.
Incidental contact - This is used to describe workers who during typical job
activities have occasional dermal contact of a minor nature.
Intermittent contact - This is used to describe workers who during typical job
activities have dermal contact such as splashes, wiping with contaminated rags,
contact with contaminated tools, or surfaces, Contact may be with either
liquids or soiids.
Routine contact - This is used to describe workers who during typical job
activities routinely have dermal contact not including immersion in a liquid.
Routine immersion - This is used to describe workers who typically immerse
their hand(s) into a liquid,
Worker practices and the use of dermal personal protective equipment
are strongly dependent on the industry under study. Therefore, when making
-------
Issued: February 28, 1991
Revised:
Page No.: 4-35
qualitative estimates it is important to obtain as much current information on work
practices in the industry as possible.
2. Quantifying Dermal Contact
Dermal contact is best quantified with measurements made in specific
operations in the industry under study. Techniques used to quantify dermal exposure
include a patch type of derma! dosimeter made of gauze or charcoal cloth, a skin
wash technique (most appropriate for chemicals with low rates of dermal absorption),
urinary excretion of the chemical or metabolites of the chemical, and fluorescence of
selective chemicals. It should be noted that these methods are commonly used to
quantify exposure, but are very difficult to interpret (this science is still in formative
stages). Since many of these techniques may be applied when the worker is wearing
personal protective equipment it is important to find out the exact circumstances when
the measurements were made.
Once it has been determined that dermal contact is likely and no
monitoring data are available, the contact may be quantified by using Equation 4-26.
D = SOC Equation 4-26
where: D = Dermal exposure, mg
S = Surface area of contact, cm2
Q = Quantity typically remaining on skin, mg/cnf
C = Concentration of chemical of concern, percent
The time of exposure is estimated qualitatively and the dermal exposure calculated
using Equation 4-26 is then expressed as mg/day. It is very important to note the
methodology used to determine the amount of dermal exposure.
Table 4-13 provides typical factors for estimating the amount of dermal
contact that may occur in particular situations, if such contact is not ruled out by
factors such as temperature or corrosivity. When using the typical values In
-------
Issued: February 28, 1991
Revised:
Page No.: 4-36
TABLE 4-13. TYPICAL FACTORS FOR CALCULATION OF DERMAL EXPOSURE
Activity
Routine immersion,
2 hands
Routine contact,
2 hands
Typical examples
Cm"
mg/cm
' Handling wet surfaces 1300
" Filling/dumping con-
tainers of powders,
flakes, granules
* Spray painting
1 Maintenance/manual 1300
cleaning of equipment
* Unloading filler cake
* Changing filter
* Pi King drums with
liquid
1-3
Resulting
typical
contact,0
mg
6-14 6500 to 18,200
1300 to 39GQ
Routine contaci,
1 hand
IneidersSaf contact,
2 hands
Incidental contact,
i hand
660 1-3
• Connecting transfer 1300 1-3
line
* Weighing powder/scoop-
ing/mixing (i.e., dye
weighing!
* Sampling 650 1-3
* Ladling liquid/bench
scale liquid transfer
650 to 1950
1300to 1900
65010 1950
Popendort and Leffingwetl 1982.
3 Versar 1984.
" These estimates also must be adjusted by the conoentratson of the chemical in the mixture and
the percent of the hand exposed if this is less than- what wouW be typical, Concentrations
that change over time due to evaporation of other factors tlsc should be aeeownlecf for.
Table 4-13, the exposure estimate also should be adjusted by the following factors
when applicable:
The concentration of the chemical in a mixture;
The percent of the hand exposed if less than what would be typically
expected for the activity;
Rapid evaporation of the chemical (lessening exposure time); and
-------
Issued: February 28, 1991
Revised:
Page No.: 4-37
The effect of an industrial hygiene program,
The likelihood of the use of protective clothing and its effectiveness
should be evaluated as described in Section IV.C. If gloves are worn for reasons
other than strictly to protect against corrosivity, it must be assumed that there is some
potential they will not always be used. In instances where gloves are expected, the
engineer should dearly state the reasons for their use and estimate contact if they are
not worn. The engineer must always clearly state the reasons for the conclusions,
The surface areas used in Table 4-13 are derived from Poppendorf and
Leff ing well 1982, Additional values for other parts of the body may be found in
Appendix H. These surface areas may be used in some special circumstances,
In Table 4-13, the amount per cm2 of skin contacted is derived from
Versar 1984. In that study, immersion of the hands in liquids of varying viscosity was
found to result in retention of 5 to 14 mg liquid per cm* surface area (increasing with
increasing viscosity). This range is appropriate for activities where there is frequent,
required contact with the material. Versar also reported results generally ranging from
1 to 3 mg/cm2 for activities such as wiping up a spill or wiping the hands with a
contaminated cloth, and for the amount remaining when the hands were "partially"
wiped with a dean cloth after immersion. This range is appropriate for situations
where there may be contact, but immersion is not expected,
Versar evaluated immersion by immersing a hand in a jar of liquid,
allowing excess liquid to drip into the jar, and weighing the jar to determine how much
liquid was removed. Six liquids of various viscosities were used. The data for
immersion of the hand range from 4,99 to 13.85 mg/cm? (for water and mineral oil,
respectively). The concentration was a direct function of the kinematic viscosity of the
liquid being tested. After partial wiping, the range was 1.3 to 2.72 mg/cm2, and was
not related to viscosity. The lowest quantities remaining after partial wipe were for the
more viscous liquids tested.
The data for handling a rag ranged from 1.37 to 3,88 mg/cm2. The
least viscous materials had the highest data points. Data for cleaning a spill ranged
from 0.67 to 1,11 mg/cm3 and were not related to viscosity. Partial wiping of the hand
-------
Issued: February 28, 1991
Revised:
Page No,: 4-38
afterward resulted in the remaining liquid ranging from 0.31 to 3,44 and 0.48 to 1,27
g/'cm?, respectively. Full wiping after the handling of a rag reduced the amounts
remaining to 0.01 to 3.3 mg/cm2,
The tests for initial and secondary uptake gave results of 1 .2 to 3.0
rng/crrf for the initial quantity on the hand. Again, the highest values were for the
least viscous materials. Partial and full wipes reduced the amounts to 0.1 to 2.7
mg/cm7,
These data suggest that for contact with rags and for liquid
remaining after partial wiping when the hands have been immersed, the amount of
material on hands ranges from 1 to 3 mg/cm2 in nearly ail cases, as concluded by
Versar, No relationship to viscosity is obvious. For immersion (without wiping) values
ranged from about 5 to 14 rng/cm?.
The Versar data are the most complete known for liquids. A PMN
submitter measured immersion data and found a similar relationship to viscosity when
gloved hands were used.
Kin Wong evaluated wiping both hands with glycerin, paraffin, oil,
and water. He obtained values of 0.5 to 1.8 g on the hand, equivalent to 0,4 to 1.4
mg/cm'2, assuming 1300 cm2 surface area for two hands.
3.
The inclusion of duration of exposure must acknowledge, If not
quantify, several mechanisms affecting dermal absorption. First is the issue of
volatilization of some (if not most) of the initial deposit from the skin before complete
absorption can take place, Second are workplace factors affecting the dermal
absorption rate. These include humidity (skin hydration), dermatitis, abrasion, and
clothing practices. Some consideration also should be given to the presence or
absence of chemical warning properties such as visual changes to the skin, irritation,
or corrosive properties. The engineer must check that an estimated dermal exposure
does not exceed the amount of material that could be present on the worker's skin.
-------
Issued: February 28, 1991
Revised:
Page No.: 4-39
The amount of time for which the contact can occur should generally
be estimated as 1 to 4 hours, based on the expectation that the worker will, at a
minimum, thoroughly wash the hands at lunch or end of the day. However, volatile
materials may evaporate rapidly from the hand. The actual duration of dermal contact
therefore may not exceed the duration of the activity leading to the contact. For
example, the volatilization model used to estimate airborne concentrations of materials
can be used to show that a thin layer of toluene will evaporate in several minutes. If
an activity such as sampling were estimated to require half an hour, this duration can
be used as the maximum exposure time. The number of contacts per day can be
estimated as one or more, depending on whether the worker is expected to handle
the chemical throughout the day or the chemical is rapidly absorbed but replenished
by additional contacts. This is based on expecting the worker to wash up at meal
breaks, When two distinct periods of contact are expected, the duration, but not the
quantity, of potential contact should be doubled. That is, if a worker's activities
involving a particular chemical last throughout the day, the engineer should report
contact as potentially lasting 8 hours, but totalling only the estimate for a single
contact.
This method provides an estimate of the amount of material on the
skin. The Health and Environmental Review Division (HERD) usually estimates
absorption of the material through the skin as a percent of the amount of contact.
Sometimes the flux of a material through the skin may be known or estimated in terms
of mg absorbed per cm2 per unit of time. In such cases, the amount absorbed could
be calculated based on CEB's estimate of area contacted and duration of contact, not
the amount on the skin.
C. Personal Protective Equipment
The primary types of personal protective equipment used to reduce worker
exposure are gloves and respirators. Aprons, coveralls, goggles, and face shields
also may be used v/here necessary.
-------
Issued: February 28, 1991
Revised:
Page No,: 4-40
1, Gloves
The ability to predict glove use practices, especially for end-users of a
chemical, is poorly established. Glove protection depends both on glove selection and
work practices. Therefore, it is difficult to define the level of protection resulting from
the general use of gloves. There are no gloves available that are totally impervious to
chemicals. Also, there are no existing models to predict the degree of protection
offered by a gtove of a particular material when used in contact with a specific
chemical. However, if gloves are properly selected and appropriately used
(considering available information on permeation, penetration, degradation, and
frequency of replacement), their use will significantly reduce exposure. The
effectiveness of gloves is a function of characteristics of the glove, the type of
chemical to which exposure may occur, the conditions of exposure, and the activities
of the worker.
The permeation rate is the rate at which a chemical moves through a
glove material. It is normally measured by testing a small piece of glove material using
a standard method such as ASTM F-739-81, Resistance of Protective Clothing
Materials to Permeation by Hazardous Liquid Chemicals. Breakthrough times for
many chemicals are listed in A.D. Little 1985. Where many breakthrough times are
consistent for a chemical or chemicals, this may be referred to as a guide for
breakthrough times for that chemical.
Glove materials (plastics or rubber) are not impermeable. Permeation
tests of intact glove material usua !y result in the determination of a breakthrough time
(i.e., the time between start of the test and the first detection of the test chemical in the
test chamber). Only after breakthrough has occurred can steady-state permeation
rates be established. The ASTM method aims to establish this steady-state rate.
Gloves or glove materials are a!so subject to penetration (ASTM
definition: "flow-through zippers, seams, pinholes, or other imperfections") and to
degradation by chemical or mechanical means. Permeation rates do not refer to these
phenomena, although a materiai undergoing a permeation test may degrade to the
point that the test is invalid.
-------
Issued; February 28, 1991
Revised:
Page No.: 4-41
Requirements for gloves aiso must address the conditions of exposure
{e.g., one hour with the glove immersed in the chemical). A sufficiently long
breakthrough time relative to the duration of exposure may mean that a glove prevents
exposure as long as it is intact and undegraded. The glove may be determined to be
"impervious" because exposure is too short for the chemical to achieve breakthrough.
The ASTM method states that resistance is determined by measuring
the breakthrough time and monitoring the subsequent permeation rate. A "resistant"
material is not defined in the standard. Gtove makers do rate the resistance of gloves
subjectively as "excellent, good," etc. Unless "resistance" is very strictly defined, this is
a less stringent requirement than that the material be "impervious." A material that
resists breakthrough for 30 minutes may be rated resistant though exposure may
occur for hours.
"Guidelines for the Selection of Chemical Protective Clothing" contains a
matrix of glove materials rated for many chemicals (A.D. Little 1985). The guideline
provides qualitative recommendations for twelve common glove materials based on
both computer predictions and test data. The recommendations are presented by
chemica! and chemical class. This matrix should be used only as guidance.
A computer model that predicts the amount of a chemical that will
permeate glove materials has been developed by A.D, Little (A. D, Little 1989), It is
designed to predict the amount of permeation through five different glove materials for
some organic chemicals. The glove materials are natural rubber, nitrile rubber,
neoprene, tow density polyethylene, and butyl rubber. Material thickness can be
varied. Besides gloves, the model can predict permeation rates for aprons, coveralls,
and other protective clothing that are constructed of one of the five materials.
The model requires chemical-specific input: molecular weight, density,
and vapor pressure or molecular structure of the chemical. The mode! also
incorporates specialized versions for certain chemical groups for which rt requires
more detailed input resulting in a more accurate prediction. The output is a cumulative
amount of chemical permeating through the material over time, presented as mass per
surface area for a designated period. The model also provides breakthrough times.
-------
Issued: February 28, 1991
Revised:
Page No.: 4-42
The following language has been adopted as standard for requiring the
use of gloves in orders written under Section 5 of TSCA:
Workers who may be dermaliy exposed to the PMN substance shall wearjgloves determined by
the Company to be impervious to the PMN substance under the conditions of exposure, including
the duration of exposure. The Company may decide this either by testing the gloves under the
conditions of exposure or by evaluating the specifications provided by manufacturers of the
gloves. Testing or evaluation of specifications should include consideration of permeability,
penetration, and potential chemical and mechanical degradation by She PMN or associated
materials.
A requirement for gloves usually should not specify their length (e.g.
elbow length). For most exposure scenarios, it seems reasonable to require
protection of the hands when there is potential that they would be immersed in a PMN
chemical or otherwise contact bulk amounts of the chemical Longer gloves should
only be needed where there is considerable potential for splashing of the material. In
such cases, aprons or other body covering also may be in order. The engineer on
the case may be in the best position to make recommendations in this area, based on
familiarity with the processes that may lead to extensive exposure. It should be noted
that CEB does not consider the physical environment in which the gloves will be worn,
but considers only the potential for chemical permeation, penetration, and degrada-
tion. The physical environment is an important consideration as gloves are often
composed of polymeric materials that may be stressed, punctured, or otherwise
damaged in actual used.
2. Respirators
Respiratory protection for new and existing chemicals presents some
unique and complicated problems for the assessment and control of occupational
exposure. Often, the potential for inhalation exposure is a primary concern for the
workers who handle new or existing chemicals.
a. OS HA Reguirernents
OSHA regulations in 29 CFR 1910.134 cover the proper use,
selection, care, and maintenance of respiratory protection. OSHA Permissible Practice
-------
Issued; February 28, 1991
Revised;
Page No,: 4-43
requires that engineering controls are first considered for reducing worker exposure
levels. However, if engineering controls are not feasible, need to be supplemented, or
are in the process of being instituted, appropriate respiratory protection should be
used to reduce worker exposure. Use of even one respirator by one employee
requires the implementation of a respirator program that meets the minimum program
elements specified by OSHA, Besides the use of an appropriate NIOSH or Mine
Safety and Health Administration (MSHA) approved respirator, the minimum
requirements for this respirator program are summarized as follows:
Written standard operating procedures.
Selection based on the hazards to which the worker is exposed.
Training and instruction.
Provisions for cleaning, disinfection, storage, and maintenance,
Respiratory Program inspection and evaluation.
Determination that employees are physically abie to perform the work and
use the respiratory protective equipment.
Further respirator use, selection, and program details are available
in the American National Standards Institute (ANSI) Standard Z88.2 1990 (in publica-
tion). Details on respirator construction and performance testing are covered in the
Bureau of Mines regulations at 30 CFR 11 (latest revision 7/26/89). tt should be
noted that the OSHA standard (29 CFR 1910.134) and the NIOSH certification
standards (30 CFR 11) are both being revised at this time.
b. Selection of a Respirator
(1) Background
industrial hygienists go through a series of decisions to select
the appropriate respirator. These include:
The desired exposure level (e.g., TLV);
The ambient concentration of the chemical in the work environment; and
The specific work conditions under which the respirator will be worn.
These and other human factors are discussed more fully in Appendix I.
-------
Issued; February 28, 1991
Revised:
Page No,: 4-44
NIOSH has developed a document entitled "A Guide to
Industrial Respiratory Protection" which contains a decision logic for assessing the
respiratory protection requirements under many conditions (NIOSH 1976). This
decision logic was updated in 1987 (NIOSH 1987). This reference should be
consulted when selecting respirator needs.
TLVs are rarely established for new chemical substances and
no information about absorption of the chemical by carbon in organic vapor cartridges
or odor threshold is typically available. Selection of an appropriate respirator is difficult
without this information. Therefore, the rationale of selection should be clearly
presented. It also must be remembered that wearing a respirator is often a burden to
a worker performing tasks. Not all worker activities or types of respirators are
amenable to use of respirators over extended periods of time.
(2) Sejeclion o.f EespJr.atQ.ry/Protection by CEB
Selection of respiratory protective equipment by CEB is
constrained by many factors as described in Appendix I. For CEB, selection is limited
to recommending different classes or categories of respirators based on state-of-the-
art knowledge of respiratory protection. The document "Strategy for Recommending
Respirators for Control of Exposure to Substances Undergoing Premanufacturing
Notice (PMN) Review" (Myers nd), which is contained in Voiume II, should be used to
select the appropriate class of respiratory protection for a PMN chemical The
selection of a particular device within a recommended category must be made by a
knowledgeable person at the workplace who is familiar with the actual conditions of
use. When respiratory protection is required, industrial hygienists typically weigh many
factors into the consideration of alternative respirator devices including;
The permissible exposure limit (PEL) specified by OSHA or threshold limit
value (TLV) specified by the American Conference of Governmental
Industrial Hygienists (ACGIH) or other allowable exposure level established.
The warning properties of the contaminant
-------
Issued: February 28, 1991
Revised:
Page No.: 4-45
The concentration of the contaminant (and other contaminants) in the work
area.
"The type of hazard (gas, fume, mist, particulate, etc.).
The degree of protection afforded by various respirators.
The conditions of use of the respirator including other hazards present, such
as lack of oxygen or confined space.
The workers' ability to wear a particular device.
Worker comfort, degree of wear time, need for communication or specific
duties that may be affected by the respiratory device.
For PMN substances, there is generally no established PEL
or TLV and it is generally unknown whether there are adequate warning properties.
According to 30 CFR 11, air-purifying respirators are prohibited for protection against
organic vapors with poor warning properties unless there is an OSHA or other Federal
standard that permits their use. OSHA has allowed the use of air-purifying devices for
substances with poor warning properties (e.g., acrylonitrile, benzene, and vinyl
chloride), if cartridges or canisters are changed prior to the end of the service life
(before breakthrough) or before the beginning of the next shift, whichever comes first
(29 CFR 1910,1017, 1910.1028, and 1910.1045).
The selection of an air-purifying, organic vapor respirator
requires a complex analysis of the anticipated performance of the chemical cartridge
or canister in atmospheres containing the substance of concern, CEB must consider
the following factors in this analysis:
Human levels of detection (odor threshold) or properties that make
recognition easy (irritation, lacrimation).
How long it takes the PMN substance, at the concentration in the workplace,
to break through the sorbent (cartridge service life),
The expected duration of exposure compared to the cartridge service life.
How often and when cartridges will be replaced.
Reaction products or amount of heat generated during sorption,
-------
Issued: February 28, 1991
Revised:
Page No,: 4-46
The consideration of an organic vapor respirator for a PMH
substance should be based on the following:
A decision by Division Directors that a protection factor of 50 or less will
afford adequate protection for the PMN substance.
An initial screening by CEB that determines that the PMN is a good
candidate for an air-purifying organic gas and vapor respirator based on
consideration of:
Warning properties.
Possibility for heat generation or generation of toxic chemicals upon
sorption,
Results from models that predict cartridge service life.
Other sources of information, such as ACGIH, NIOSH, American
Industrial Hygiene Association (AIHA recommendations for analogues,
etc.
If CEB determines that a PMN substance is not a good
candidate, the 5(e) Order would specify a supplied-air respirator, but the company
could petition EPA and provide the necessary documentation proving that an
air-purifying respirator is acceptable and that the organic vapor cartridge provides
acceptable performance.
In either case, the submitter would be required to document
the effectiveness of an organic vapor cartridge respirator by submitting the following
information:
The service life of the cartridge.
Information showing that high temperatures (>40°C) or toxic chemicals
are not generated upon sorption of the PMN substance.
A respirator cartridge change-out schedule established as part of the
respirator program.
Any known information on the warning properties of the PMN
substance.
Any administrative controls that will be used by the submitter.
-------
Issued: February 28, 1991
Revised:
Page No.; 4-47
NIOSH is preparing guidance to be used for testing organic
gas/vapor cartridges for service life. However, until this guidance is available, CEB
recommends that organic gas/vapor cartridges be tested in accordance with the
proposed NIOSH 42 CFR 84, Tests and Requirements for Certification of Permissibility
of Respiratory Protective Devices Used in Mines and Mining, published in the Federal
Register on August 27, 1987, A CEB industrial hygienist should be consulted to
determine the appropriate test concentrations for the PMN substance, and answer any
questions the submitter may have.
This information should be submitted for CEB review before
OTS approves use of an air-purifying respirator. This occurs either before an order is
written or upon a company's request to modify an existing order. Submitted
information will be reviewed by a CEB industrial hygienist on a case-by-case basis.
(3) Protection Factors
Protection factors for various classifications of respirators
have been assigned by NIOSH (see Appendix J). CEB uses these values will be used
to decide the appropriate type(s) of devices necessary to achieve the degree of
protection determined by Division Directors,
The protection factor of a respirator is an expression of the
performance based on the ratio of the concentration of contaminant measured outside
the facepiece cavity to the concentration of contaminant measured inside the
facepiece cavity. The Assigned Protection Factor (APF) is defined by NIOSH as a
measure of the minimum anticipated workplace level of respiratory protection that
would be provided by a properly functioning respirator or class of respirators to a
percentage of properly-fitted and trained users.
The table of protection factors developed by NIOSH may be
simplified since most classes of respirators have Assigned Protection Factors that fall
into one of four categories: APF > = 2000; APF > = 50; APF > = 25; and APF > =
10, Thus, respirators will be selected based on the need for a 100Q-, 50-, or 10-fold
reduction in the estimated potential inhalation exposure.
-------
Issued: February 28, 1991
Revised:
Page No.: 4-48
APF of at Least 2000 is Considered. Devices which have
been assigned a protection factor of 2000 or greater are suppfied-air respirators
equipped with a full facepiece and operated in positive pressure mode. Since the
selection of a supplied-air respirator is independent of the physical and chemical
properties of the contaminant, no further information is required to make this choice.
APF of at Least 50 is Considered. Supplied-air devices that
have been assigned a protection factor of at least 50 are equipped with a tight-fitting
facepiece and operated in continuous flow mode, or are equipped with a full facepiece
and operated in demand (negative pressure) mode. However, QSHA does not allow
the use of supplied-air respirators operated in demand mode. Supplied-air respirators
operated in positive pressure mode have much better performance as facepiece
leakage is minimal with positive pressure. Thus, EPA will allow the use of positive
pressure, supplied-air respirators only. Air-purifying respirators equipped with a full
facepiece and powdered air-purifying respirators equipped with a tight fitting facepiece
have been assigned protection factors of 50.
APF of at Least 25 is Considered. Powered air-purifying
respirators equipped with a loose fitting hood or helmet have been assigned a
protection factor of 25.
APF of at Least 10 is Considered. Most approved respirators
have a protection factor of at least 10. For a paniculate exposure, a high efficiency
particulate filter is required. To recommend the use of an organic vapor respirator, the
effectiveness of the cartridge or canister must be evaluated as described above.
-------
Issued: February 28, 1991
Revised:
Page No.: 4-49
(4) StandatilLanguage for 5(ej Orders
The standard language as presented in Appendix J reflects
the use of the NIOSH APF values. It describes the general classes of respiratory
protection that should be recommended in 5(e) or other orders. Any questions
regarding respiratory protection for 5(e) Orders should be referred to a CEB industrial
hygienist
D. Engineering Controls
1. Local Exhaust Ventilation
Local exhaust ventilation (LEV), in addition to general ventilation, is
the primary control used to reduce worker exposure to chemicals. If LEV is used for a
specific activity, the specifications of the control should be compared to the recom-
mendations made by ACGIH in "Industrial Ventilation: A Manual of Recommended
Practice" (ACGIH 1986). Although a qualitative determination of control use can be
made, it will usually not be possible to account quantitatively for local exhaust.
The actual reduction in worker exposure from the use of LEV is
not a simple relationship. It is dependent on several factors:
The design capture efficiency of the LEV system;
The work practices of the employee;
Air currents in the work area;
Other process-specific factors such as dragout or sudden releases;
and
Actual maintenance of the LEV system over time,
The design capture efficiency is primarily dependent on hood
design and exhaust volume. The best hood design is one that encloses or confines
the process. This is not always possible when worker access to the process is
considered. Exhaust volumes (or face velocities) presented in the ACGIH ventilation
manual for a similar process should be compared with information supplied by the
submitter to ensure that adequate face velocity will be used in the design.
-------
Issued; February 28, 1991
Revised:
Page No.: 4-50
The addition of an LEV system can affect the work practices of the
employee. Enclosure of the process can cause changes in work practices that may
make the LEV less effective. The LEV design should never allow the breathing zone
of the worker to pass between the pollutant source and the exhaust.
Air currents in the work place from other processes, natural
ventilation, general ventilation, or movement in the area can reduce the effectiveness
of an LEV system. All possible cross currents should be minimized.
Process-specific factors such as dragout can cause the chemical
to be carried out of the capture zone of the LEV system, thus decreasing its efficiency.
Processes that release sudden surges of hot gases or vapors must design the LEV
system to account for these releases.
Finally most LEV systems are not maintained at peak efficiency
throughout their life. If a system is older, the questions of design parameters such as
face velocity versus actual parameters should be addressed,
One process where standard LEV design can obtain very high
efficiencies is the use of lay-on or slot LEV to control worker exposure during drum
filling. In a series of tests on LEV use in drum filling performed for OTS, capture
efficiencies were between 99 to 100 percent when the ACGIH design flow rate was
used (PEI 1987). The efficiency of the lay-on and slot LEV systems in drum filling
operations generally was independent of the concentration of emissions leaving the
drum. The efficiency was affected by the fill rate that determines the emission velocity
at the drum bung hole.
2. JjitrQgen_BJ.a.n.Keting
Chemicals that are extremely volatile or that react with water or air
on contact are often loaded into shipping containers under a nitrogen blanket. This
handling procedure (known as nitrogen blanketing, padding, or purging) displaces air
and moisture and therefore prevents degradation of the chemical and decreases the
chance of explosion. Volume il contains a description of the process and work
-------
Issued: February 28, 1991
Revised:
Page No.: 4-51
activities associated with the procedure for aeetaldehyde, hydrazine, and toluene
diisocyanate (Mitre 1984).
The only release expected from nitrogen blanket transfer is a small
spilt from the end of the transfer line of the liquid remaining between the valve and the
end of the coupling collar. Assumptions to be made in the absence of other
information are described In the report and include:
Nominal inside diameter of transfer line equals 3 inches.
Distance between valve body and end of collar equals 3 Inches,
Area of spill is assumed to be 730 cm2.
In the original document describing nitrogen blanketing (Mitre 1984), a
model was assumed for the generation rate and an empirical equation for the mass
transfer coefficient. In this revision of the engineering manual, CEB has revised the
mode! for generation rate from an open surface. Using the equation for estimating
airborne concentration from an open surface (Equation 4-25) and substituting the
surface area of the spill:
Worst case:
Cv = 1679 P Equation 4-27
Typical case:
Cv = 54,8 P Equation 4-28
where: Ctf = Airborne concentration, ppm
P = Vapor pressure, mm Hg
3. Transfers
Emissions from controlled loading operations can be calculated by
multiplying the uncontrolled emission rate (Equation 4-21) by the control efficiency
term:
-------
where
G
E
I
M
V
R
T,
issued: February 28, 1991
Revised:
Page No.: 4-52
G -1 -
f M V r P*
100 J 3600 fl IL
Vapor generation rate, g/sec
Control efficiency, percent
Saturation factor, dimensionless
Molecular weight, g/g-mol
Volume of container, cm3
Fill rate, units/hr
Vapor pressure of pure substance, aim
Universal gas constant, 82,05 atrn cm3/g-rnol * K
Liquid temperature, • K
Equation 4-29
Measures to reduce loading emissions include use of vapor recovery equipment to
capture the vapors displaced during loading, recovery by refrigeration, absorption,
adsorption or compression, and piping back to storage. Vapors also can be
controlled through combustion in a thermal oxidation unit with no product recovery.
Control efficiencies of modern units range from 90 percent to over 99 percent depend-
ing on the nature of the vapors and the type of controls used (USEPA 1985b).
-------
issued: February 28, 1991
Revised:
Page No.: 5-1
V. MODELING RELEASE TO WATER
This section covers standard procedures to be used when estimating release to
water during manufacture or processing of chemical substances. It is primarily
intended for batch operations, so the basis for most estimates should be per batch.
In completing a scenario the engineer should obtain as much information as possible
from the submitter of a PMN or manufacturer of an existing chemical. The
assumptions presented here can be used in the absence of data from the submitter,
or when evaluating the reasonableness of data, or for preliminary evaluation of the
potential for release of existing chemicals. To reflect the uncertainty in the assessment
methodology, the engineer should report releases as a range and identify the basis for
the range.
The areas covered are cleaning of equipment, tank truck/car cleaning, phase
separation, condenser/scrubber operation, and polyelectrolyte wastewater treatment.
These operations are commonly encountered in the PMN review process and are
often the most important sources of water releases. Little data are available on actual
losses of chemicals to water from these simple batch operations.
A. Cleaning of Equipment
To estimate the amount of material that may be lost to water in cleaning of
equipment, the engineer should consider the equipment cleaning process, the
equipment to be cleaned, the cleaning schedule, the residual quantity of the PMN
chemical (or other material of concern) in the equipment, the type and amount of
solvent used (water or organic), the solubility/miscibility of the material in water, and
any treatment of the wastewater.
Equipment is either rinsed or flushed with water or an organic solvent,
depending on the solubility of the chemical in various solvents. If water is used, the
waste will typically be sent to wastewater treatment or reworked into a new batch. If a
solvent is used, it will typically either be recycled or incinerated (for estimation of
releases from incineration see Section VII).
-------
Issued: February 28, 1991
Revised:
Page No.: 5-2
The first step in the analysis is to determine the type and volumetric capacity
of major equipment that is cleaned (or volume of chemical per batch processed
through that equipment). Once all equipment and piping to be cleaned are identified,
the amount of chemical in the equipment during operation can be determined. This
will be the batch volume. Adjustments should be made for the concentration of the
chemical in the mass contained in the equipment
When size of equipment is unknown, the engineer must assume values
based on experience in the industry in question or information from similar cases or
the literature.
The frequency of cleaning of the equipment must be determined. Clean-out
after every batch may occur if quality of product demands it. Other reasons for
frequent clean-out are changes in the type of batch being run (e.g., color change in ,
paint mixing), possible solidification of product within a reactor, or proper operation of
mechanical equipment (e.g., a plate and frame filter may not close if not cleaned after
each use). Clean-out after every batch should only be assumed if a specific reason
for such cleaning can be identified. Otherwise, the engineer should assume cleaning
only after one week's run or at the completion of a campaign, whichever comes first.
The frequency of cleaning must be clearly stated, and release per day from cleaning
reported.
Another factor to be considered is the possible recycle of defining effluent
back to the process. Although such flushing may occur after every batch, it may not
result in a release (Le., the residue may be added to the product stream or used in the
next batch). This can occur when mixing vessels are rinsed with water that is sub-
sequently reworked into batches of similar product, or when product is to be
subsequently isolated from the cleaning solvent by distillation. Cleaning that results in
a release may be very infrequent in these cases.
The amount of the PMN chemical or other material of concern remaining in
equipment prior to cleaning is the amount available for loss. Many parameters affect
this quantity, including the design configuration of the equipment, the method of
removing or unloading the chemical from the equipment, the viscosity of the chemical,
-------
' ' . ' Issued: February 28, 1991
Revised:
Page No.: 5-3
and the material of construction or lining of the equipment. Typically, the amount of
chemical available for loss is calculated as a percent of the total amount of the
material in the equipment during normal operation of a batch.
Table 5-1 presents factors for estimating percent chemical remaining in
drums and tanks after unloading. These factors were derived from a pilot scale
research project investigating the effect of the four parameters listed in the preceding
paragraph on residue quantities (PEI 1986a). It was concluded that the amount of
residue is generally influenced most by the method of unloading. The viscosity ..of the
chemical and the design configuration of the equipment will affect residue quantities to
a lesser degree. Material of construction or lining of the equipment has little effect on
residue quantities. The values listed in the table represent residue quantities as a
weight percent of vessel capacity (pound chemical residue per 100 pounds of
chemical). The mean values presented in Table 5-1 may be used to represent typical
residues, while the upper end of the range may be used to represent reasonable worst
case. The values in Table 5-1 should only be applied to similar vessel types, unload-
ing methods and bulk fluid materials. The research was performed with materials with
viscosities below 100 cp. For materials with significantly higher viscosities (>200 cp),
estimates of percent residue were made based on engineering judgment.
Dow has developed a new "drainable" drum design which they claim has
residual losses of about 0.02 percent. Dow compares this to losses from conventional
drums of 0.5 to 0.68 percent for inversion draining and 0.68 to 2 percent for dumping.
Engineers should be aware of these new drums and consider contacting Dow for
individual PMN cases to estimate the likelihood that the new drums will be used.
Besides drums and tanks, losses from process piping should be considered.
The amount of chemical available for loss from process piping is calculated as a
percent of the total volumetric capacity of the piping being cleaned. A residue quantity
of 1 percent of piping volume may be assumed.
A final factor in estimating release is whether water or an organic solvent is
used in cleaning. It should be assumed that water is used unless there is strong
-------
TABLE 5-1. SUMMARY OF RESIDUE QUANTITIES FROM PILOT-SCALE EXPERIMENTAL STUDY,
WT. PERCENT (Source: PEI 1986a)
Surfactant
Unloading method Vessel type Value solution8
Pumping Steel drum Range 3.06
Mean 3.06
Pumping Plastic drum Range Not available
Mean
Pouring Bung-top steel drum Range 0.485
Mean 0.485
Pouring Open-top steel drum Range 0089
Mean. O.OB9
Gravity drain Stope-bottom steel tank Range 0.048
Mean 0.048
Gravity drain Dish bottom stee! tank Range 0.058
Mean 0.058
Gravity drain Dish-bottom glass-feted tank Range 0.040
Mean 0.040
* Surfactant solution viscosity- 3 cenUpofsfl, surface tension* 31.4 dynes/cm2.
b For water, viscosity » 4 centlpofee, surface tension - 77.3 dynes/cm3.
° For kerosene, viscosity « 5 centlpoise. surface tension - 29.3 dynes/cm2.
d For motor ol. viscosity - 87 centipolse, surface tension » 34.5 dynes/cm2.
Water"
1.84-2.61
2.29
2.54 - 4.67
3.28
0266^.458
0;403
0.026 - 0.039
0.034
0.016- 0.024
0.019 .
0.033 - 0.034
0.034
0.020 - 0.040
0.033
Material
Kerosene0
1.93- 3.08
2.48
1.69- 4.08
2.61
0.244 - 0.472
0.404
0.032- 0.080
0,054
0.020^.039
0.033
0.031 - 0.042
0.038
0.024 - 0.049
0.040
e Residue quantities tor high viscosity material were not defined by the study; thus, the quantities presented are esthiwtesof a
based on engineering Judgment.
Source: PEI 1986a.
Material6 with
MotproC1 viscosityi 200 cp
1.97- 2.23 3
2.06
1.70 - 3.48 4
2.30
0.677 - 0.787 1
0.737
0.328 - 0.368 0.5
0.350
0.100 - 0.121 0.1
0.111^
0.133 - 0.191 0.2
6.161
0.112-0.134 0.2
0.127
reasonable worst case scenario
,/l
^O 33 rn
tO CD {/I
CO < C
0> W>" CD
.9 CD Q.
Tl
CD
cn OT
"** cu
ro
00
CD
to
-------
Issued: February 28, 1991
•. , •' • Revised:
Page No.: 5-5
reason to believe otherwise. Once .the use of water for cleaning has been established,
the release of chemical to water is simply the amount of chemical available in the
equipment. Some reactions must be carried out with the complete absence of
moisture, however. For example, it is unlikely that a company would clean using water
if it were then necessary to dry the reactor thoroughly before the next batch. Other
reasons for not using water could include permits that prevent the release of any
process water, extreme reactivity of the product in water, or the existence of a solvent
recovery system for isolating product from the solvent in which it is made (in which
case it can be assumed that the same system could be used to recover solvent from
cleaning wastes). If an organic solvent is used, it should be assumed that all the
chemical available for loss is removed in the solvent and that this material is landfilled
or incinerated with or without separation from the solvent. If information on the solvent
used is not available from the submitter, engineering judgment should be used.
Companies may use water to flush a chemical from equipment, intending to
separate it from the water by settling, filtration, or use of an oil/water separator prior to
further wastewater treatment. The engineer must account for this separation to
present valid estimates of release. The controls section (Subsection VILA) discusses
these wastewater treatment methods. Methods for adjusting release estimates to
account for treatment are also discussed.
B. Tf nk Truck g"H T?nk Car Cleaning
To estimate the quantity of material that is released during tank truck or tank
car cleaning operations, the engineer must consider the cleaning process, the
cleaning frequency, the solubility/miscibility of the material in water, the solvent used
for cleaning (water or organic), the quantity of material residue in the tank truck or car,
and any treatment of the wastewafer.
Cleaning of tank trucks and tank cars generally has four basic steps:
. Any material remaining in the tank is removed through draining or other
means. Sometimes, residue on the sides of the tank may have to be
-------
Issued: February 28, 1991
Revised:
Page No.: 5-6
manually scraped to remove valuable material or to prevent a discharge of
material to wastewater during rinsing with water,
The tank is rinsed with either water or an organic solvent, depending on the
solubility of the chemical in the various solvents. If water is used, the waste
will typically be sent to wastewater treatment. If a solvent is used, it will
typically be recycled, resulting in spent solvent (containing the chemical of
interest) that is likely to be incinerated.
The main wash step is performed with either a caustic solution, a detergent
solution, or simply water. Often, these solutions are filtered and reused in
the washing process. Spent or waste solution resulting from this step is
typically sent to wastewater treatment.
The tank is rinsed with water and the resulting waste is typically discharged
to wastewater treatment.
The engineer must consider at the points at which water will be used in the cleaning
process for the particular chemical in question, and the wastes that will be discharged
to water.
The cleaning schedule of the tank trucks/cars must be known. For tank
trucks/cars dedicated to one chemical, cleaning is often limited either to when the
tank becomes contaminated or prior to tank inspections, repairs, and testing. Some
shippers, however, insist that the tank truck/car be cleaned after every load, whether
the container is dedicated to one chemical or not. Tank trucks/cars used for general
purpose hauling (i.e., nondedicated service) are cleaned after each load. Cleaning
frequency must be clearly stated, and the release per cleaning reported.
The amount of chemical remaining in the tank truck/car and available for
loss should be estimated as a percent of the tank load. This amount depends on the
method used to unload the tank truck/car and the viscosity of the chemical. Studies
have been performed for specific materials and residue data can be found in the
literature. For example, one study determined that an average of 0.15 percent of milk
transported in tank trucks remains on the sides of the truck as residue (PEI 1986a).
-------
Issued: February 28, 1991
Revised:
Page No.: 5-7
A typical tank truck hauls 3.5 loads per week and carries approximately
5,000 gallons per load [CEB n.d.]. A typical tank car carries up to 34,000 gallons per
load (Monsanto 1978). A representative number of loads hauled per week per tank
car cannot be determined since many tank cars are dedicated and, therefore, rarely
cleaned. When data are not available, 0.1 percent (on a mass basis) can be used to
give a rough approximation of the quantity of chemical available for loss.
Tank truck cleaning typically generates 500 gallons of wastewater per tank
truck cleaned, although up to 5,500 gallons of wastewater may be generated from a
full flushing operation (Monsanto 1978). Tank car cleaning can generate as little as
60 gallons of wastewater, although 34,000 gallons may be generated from a full
flushing operation (Monsanto 1978).
Greater than 98 percent of the organic residue is generally removed during
the rinse and main wash steps (Monsanto 1978). This value may be assumed except
for cases involving highly viscous chemicals such as crude petroleum products. The
remaining 2 percent of the residue is expected to be removed during the final
water/caustic wash. ^ _
^^_^^^^
percent of the tank load is assumed to be lost to solvent wastes that will likely be
^^
water rinse. • .,
If water or caustic solutions are used throughout the cleanup, It should be
assumed that 0.1 percent of the tank load is discharged to wastewater.
The solids and immiscibles may be separated from the wastewater prior to
further treatment or discharge. The quantity remaining in the water after separation
should be determined by multiplying the solubility of the chemical by the quantity of
water used. , This approach will typically lead to lower than actual release quantities
because it does not account for suspended solids entrained in the wastewater effluent.
-------
Issued: February 28, 1991
Revised:
Page No.: 5-8
,C. Phase Separation ' ,•
The amount of a PMH chemical or other chemical of concern lost to water
from phase separation will depend on thjs j/aJumejta
the chemical in water. If an organic solvent is also used in the separation, the volume
"ofth'at solventlanTfhe solubifcy of the chemical in the solvent will affect the amount of
chemical lost to water. Since PMN chemicals normally have undergone only bench
scale or pilot plant production when they are reviewed by EPA, much of the infor-
mation that is needed to estinraate loss must be based on professional judgment or
standard chemistry/engineering estimation techniques.
A typical phase separation problem occurs when the "product" PMN
chemical is contained in the organic phase and a small amount remains in the
aqueous phase. This situation may arise when:
Solvent is added to a reaction product to extract the chemical from water
already present (doe to reaction or addition as a carrier for the reactants).
Water is added to a chemical/solvent mixture to extract impurities into the
water.
The chemical constitutes the organic phase and water is the only other
major constituent.
The discussion presented here does not address a fourth case in which the "product"
PMN chemical is removed wfth the water phase.
The following assumptions are implicit in this scenario:
The PMN chemical is present mainly in the organic phase from which useful
product will be recovered. Due to tow solubility of the chemical in water, a
small amount is also present in the aqueous phase.
• The process is batch. Sufficient residence time is allowed to achieve the
interphase in the phase separator. Separation is assumed to be perfect (i.e.,
the water can be drawn off with no entrapment of undissolved organics).
Losses from auxiliary operations such as cleaning and distillation, are not
considered. ' . '
-------
Issued: February 28, 1991
Revised:
Page No.: 5-9
To estimate release, the engineer must determine the amount and type of solvent
present, the amount of water present, and either the partition coefficient for the
chemical in solvent/water or, more simply, the water solubility of the chemical.
If the volume of the aqueous phase cannot be obtained from the submitter,
the engineer has several approaches to estimating the amount:
. A mass balance on the reaction can be used to determine the amount of
water either added with reactants or produced in the reaction. If there is
insufficient information to perform a mass balance (e.g., processing steps
are speculative), it can be assumed that water equals 10 to 50 percent of
total batch volume.
. If water is added to wash impurities from the organic phase, it can be
assumed water added equals 10 to 100 percent of batch volume.
If the size of decantin^aqui^ment is known, the amount of^aqueousjphase
'
Va * Va(DL)(IL) Equation 5-1
where: Va = Volume of aqueous phase
Vd = Volume of decanting vessel
DL = Decanter fill level, % of decanter height
IL = Interphase level, % of decanter height
If only the decanter volume is known, assume:
V. - 0.30 Va ' Equation 5-2
based on an assumed fill height of 75 percent and an aqueous phase height of 40
percent of decanter height.
A similar approach can be used for estimating the volume of solvent
present, if it is not specifically provided by the submitter. If only decanter volume is
known, it can be assumed that:
V, m 0.45V, Equations*
based on a fill height of 75 percent, and a solvent height of 60 percent.
This approach does not apply if the PMN 'chemical is the only major organic.
-------
Issued: February 28, 1991
. Revised:
Page No.: 5-10
Two approaches can be used depending on the constituents:
The PMN is the only organic present. If the PMN chemical is the only major
constituent other than water, the amount dissolved in water is simpfy the
volume of water times the water solubility of the PMN chemical, which can
be obtained from ICB if needed.
PMN chemical, solvent, and water are present. The preferred method for a
chemical/solvent/water system is to use the partition coefficient for the
solvent/water system using Equation 5-4:
WL
*
WL
*
,
Equation 5-4
where: Wta = Weight of substance in aqueous phase, kg
Wtb = Weight of substance in batch to be separated kg
^-~- — —~ ^ • ^-~~ - •— ^ . • " - ,,- " x-/ ' '""^"^-•^ -
= JJartitiorTcdefficlent of substance inolvent/water, dimensionless
vT =• Volume of solvent phase
Va = Volume of aqueous phase.
This equation is derived from the relationship:
K _ Concentration of PMN in Qrg^nic phase
5W Concentration of PMN in aqueous phase
Equation 5-5
where:
concentration in organic
moles/liter
Wt
*
concentration in aqueous = *^v > moles/liter
M = molecular weight of PMN
V0 = Volume of the organic phase
Partition coefficients may be obtained from ICB. These may be based on measured
values or on estimation techniques. Often, it will not be possible to estimate the
-------
Issued: February 28, 1991
Revised:
Page No.: 5-11
coefficient for the specific solvent. The engineer may be abie to use the octanol/water
partition coefficient for solvents judged similar to octano! by ICB. if the octanol/water
partition coefficient is not appropriate and the solvent/water coefficient cannot be
determined, the engineer should default to the water solubility of the compound to
calculate the amount dissolved in water. The water solubility alone also can be used
when the PMN chemical is highly soluble in the solvent and only slightly soluble in
water.
D. Condensers and Scrubbers
Condensers and scrubbers may be used as integral parts of chemical
processing or manufacturing operations for product recovery or as air pollution control
devices. Estimation of control efficiency of the equipment is presented in Section
VII.B. Chemical release to water is likely to occur from use of condensers and
scrubbers (assuming water is used in the processes). The following sections discuss
estimation of chemical release to water from these operations.
in estimating water release, the engineer should consider volume and
source of the water, water solubility of the chemical, the relative volatility of the
chemical, the presence of organics and their miscibility in water, process conditions
such as temperature and pressure, the type of equipment, the process configuration,
and process variations (i.e., whether the purpose is to remove all the water or to
evaporate some solvent).
The relative volatility, «, is the ratio of the distribution coefficients of
components A and B in the vapor and liquid. It is a measure of the degree of
separability of the substances. Assuming constant relative volatility (over the range of
rnole fractions being considered) and ideal solutions, or that Raoult's law is valid (the
equilibrium partial pressure of a constituent is the product of the vapor pressure at the
temperature and the mole fraction in the liquid), the relative volatility is given by
Equation 5-6:
-------
issued: February 28, 1991
Revised:
Page No.: 5-12
P
« . _* Equation 5-6
where: PA = Vapor pressure of more volatile component (A)
PB = Vapor pressure of less volatile component (B)
Thus, « can be estimated independently from physical properties.
1. Condensers
Three condenser scenarios are presented below. The first scenario
involves water and an immiscible chemical. The second involves water and a miscible
chemical. The third scenario involves a chemical, water, and an organic solvent. In
the absence of information, it shall be assumed that there is total condensation of the-
overhead and no column ahead of the condenser. If steam is used as a stripping
agent, the amount of steam used can be added to the amount of water in the process.
a. Case No. 1
Water and an immiscible chemical are present in a batch, as is
true of polymerizations yielding water as a byproduct. If the substances are
completely insoluble, each component exerts its own vapor pressure expressed by
Equation 5-7:
PT = Pft + pe Equation 5-7
The mixture produces a vapor of constant composition until all the more volatile
component is volatilized. The equilibrium vapor composition of the most volatile
compound, A, may be estimated using Raoult's law, Equation 5-8:
-------
Issued: February 28, 1991
Revised:
Page No.: 5-13
y . A Equation 5-8
where: PA = Vapor pressure of compound A
PT = Total pressure
Similarly, the equilibrium vapor pressure of the less volatile component, B, may be
estimated using the relationship,
y: = fl = 1 -yA* Equation 5-9
B D
MT
where: PB = Vapor pressure of compound B
The total weight of the less volatile material driven off can be calculated using Equation
5-10.
• wt - m* PB MB '•• Equation 5-10
B" M. PA
where: WtA = Weight of the more volatile component
WtB « Weight of the less volatile component
MA = Molecular weight of the more volatile component
Ms = Molecular weight of the less volatile component
An example of the applicability of this method is in the
manufacture of unsaturated polyester. The amount of water to be removed can be
estimated based on stoichiometry.
b. Qa$e No.2
When a reaction is carried out in water and no immiscible
phase is present, the composition of the condenser overhead and thus the quantity of
PMN chemical can be calculated based on a one-plate distillation. The relative
-------
Issued: February 28, 1991
Revised:
Page No.: 5-14
volatility and the amount of water to be evaporated must be determined to complete
the calculation. . . •
Assuming constant relative volatility and a binary mixture, one form
of the equation is given in Equation 5-11 0"reybal 1980):
,n FX? Soc/n£(1 "XF) Equation 5-11
W~X~ W.(1 -Xw)
where: F = Number of moles of the charge of composition X,
W = Number of moles of residual liquid of composition Xw
Xp = Mole fraction in feed
Xw = Mole fraction in residue
« = Relative volatility . -
If mole fraction of chemical is to be. used as the basis of the calculation, then the
relative volatility must represent the ratio of chemical to water.
Typically, the engineer will know the moles of chemical produced
in a batch and, batch composition. The terms in the denominator are therefore known.
If the amount of water is not known, 0.1 mole percent can be assumed. The moles of
charge and charge composrtion can be determined by a trial and error solution of the
above equation if the amount of water in the charge has been determined. If mass
balance data are insufficient to determine this value, it should be assumed that water
constitutes 50 percent by weight of the charge. Once the amount of charge has been
determined, the amount of chemical lost overhead is the difference between the
amount in the charge and the amount in the residual (product).
c.
If a PMN chemical (or other chemical of interest), water (other
than from reaction), and organic solvent are present in the batch, it should be
assumed that phase separation occurs before batch distillation. The phase separation
procedures in Section V.C should be used to calculate the amount of PMN going to
the waste stream, whether organic or aqueous, from the separation. If the chemical is
then to be recovered from the aqueous phase by driving water off overhead, the batch
-------
Issued: February 28, 1991
. Revised:
Page No.: 5-15
distillation method for binary mixtures should be used to calculate the amount of
chemical lost overhead.
Azeotropic mixtures of the solvent and water complicate
estimation because significant amounts of both solvent and water may remain to be
driven off. In the absence of information to the contrary, it should be assumed that
the water will be separated from the solvent after condensation for subsequent
treatment/disposal and that the solvent is returned to the reactor. The amount of
chemical lost in the wastewater may be calculated by the phase separation procedure
in Section V.C.
When the water solubility of the chemical is low, (or its
partition into the solvent versus water is high), an adequate assumption is that
sufficient chemical will be volatilized to support saturation in the water. That is, the
amount of water driven off and the solubility of the chemical in water will provide an
upper bound estimate of the amount of chemical lost. The engineer can very roughly
calculate the amount of chemical driven overhead with solvent to determine the
adequacy of this assumption.
2. Scrubbers
Assuming 100 percent efficiency of the scrubber, the amount
of chemical released in the effluent can be calculated from distillation equations if there
is no condenser before the scrubber. The calculation of the amount of material driven
overhead is thus the same; the only difference is that the material is scrubbed instead
of condensed out of the exit gases.
If the condenser precedes the scrubber, it Should be
assumed all the chemical is condensed (and therefore discharged from the
condenser) and none reaches the scrubber. This is based on the assumption that
total condensation is achieved.
E. po-lyelectroivtes
Polyelectrolytes are used in wastewater treatment operations to remove
suspended solids. They also may be used as stabilization and strengthening agents
-------
Issued: February 28, 1991
Revised;
Page No.: 5-16
in pulp and paper manufacture. A polyelectrolyte is either a negatively (anionic) or
positively (cationic) charged polymer chain (CEB 1989b). The ionic nature of these
polymers makes them useful in removing suspended solids from wastewater. They
are added to wastewater to bind with charged solid particles, resulting in neutralized
solids that are subsequently removed from the wastewater by settling, clarification,
filtration, or other means. Release of the polyelectrolytes to water is likely to occur
from their use in wastewater treatment. The following, section discusses estimation of
the release of polyeiectrolytes to water from wastewater treatment operations.
The use of settling, clarification or filtration does not result in 100 percent
removal of suspended solids from wastewater. Some solids, and thus, some
polyelectrolyte will pass through these treatment systems. Several studies claim that
99 to 100 percent of the polyelectrolytes added to wastewater is absorbed onto the .
solids and that essentially none is bibavailable through the liquid phase (CEB 1989b),
Simple analytical methods cannot detect the presence of polyelectrolytes in effluent
streams. Thus, there are no definitive data on the amount of-free or bound
polyelectrolytes in wastewater (CEB 1989b). In the absence of data specific to the
case being evaluated, the following assumptions may be-made:
* Polyelectrolyte dose concentration typically between"! and 5 ppm.
. - Essentially complete absorption of the polyeleetrolyte to solids.
. •; 90 percent removal of solids from wastewater by clarification (carrying 90
percent of the dose of polyelectrolyte along);
. The remaining 10 percent of the polyelectrolyte (primarily absorbed to
solids) is carried to a stream, POTW, or to further on-srte treatment.
The assumption of essentially complete absorption of the .polyelectrolyte-to solids does
not imply that zero "free" polyelectrolyte leaves the elarifier, nor that the material is
permanently bound to the solids. Thus, from the above assumptions, 10 percent of
the polyelectrolyte dose is released to water or sent to further on-srte treatment. To
estimate the polyelectrolyte dose, the volume of wastewater generated by the process
of interest must be estimated or obtained from the submitter. This volume of
wastewater generated (daily basis) is then multiplied by the typical dose concentration
-------
Issued: February 28, 1991
. Revised:
Page No.: 5-17
(1 to 5 ppm) to calculate the quantity of polyelectrolyte added to the system per day.
This polyelectrolyte daily quantity may then be multiplied by 10 percent to estimate the
quantity of polyelectrolyte (bound or free) in the daily clarifier effluent. This quantity
may be adjusted to account for further wastewater treatment if applicable. Use of the
preceding estimation method will provide a reasonable worst case estimate in the
absence of actual data. Volume II contains the CEB document "Pofyelectrolytes" that
presents a more detailed description of polyeiectrolyte use and release to water.
F. Mst^l Wgrking Operations
Cutting fluids are used in metal cutting and finishing operations to lubricate,
cool, and flush the zone of contact between the tool and the workpiece. They also
coat and protect finished surfaces from corrosion. Cutting fluids include oil-based
systems, miscible aqueous fluids, and water-based emulsion systems. The choice of -
cutting fluid is a function of the type of cutting or metal finishing operation, the
intended function (i.e., cooling, flushing, or lubricating), cost, required surface finish,
quantity required, and reliability of the fluid. Types of cutting fluids and chemical
additives are described in "Industrial Process Profiles to Support PMN Review: Metal
Treatment Chemicals" (Walk n.d.).
The inhouse report "pisposal of Metalworking Fluids" (CEB 1984c) describes
two scenarios developed by CEB to describe potential water releases from small
metalworking shops and from metalworking large shops. These scenarios are
frequently used for PMN cases when the submitter is unable to provide specific details
about use operations. The following is a summary of the information provided in the
report.
Metalworking fluids are generally intended to be recycled within a single
machine or a central cooling system serving several machines. During ust, cutting
and grinding fluids become contaminated with metal chips and dirt that must be
removed continuously or periodically by using strainers, filters, centrifuges, or settling
tanks. Fluid is lost as. carry-off on the workpieces or by evaporation. Thus, frequent
makeup is necessary to maintain the proper fluid concentration for good performance.
-------
Issued: February 28,1991
Revised:
Page No.: 5-18
Eventually, the spent fluid must be pumped out and the system cleaned out to be
refilled with fresh fluid.
About 80 percent of all metalworking fluids is used in small shops while the
balance is used in large shops. CEB defines a small shop as a facility using a single
machine system. Assuming a capacity of 25 gallons of fluid, a useful life of the cutting
fluid as 1 to 3 months, and discharge of entire volume after the fluid's useful life,
releases of PMN chemical can be estimated based on its expected use concentration.
CEB describes a large shop as likely to use a large coolant system serving groups of
, machines. Assuming a capacity as large as 24,000 to 80,000 gallons, a useful life of
about 6 months, and discharge of entire volume after the fluid's useful life, releases of
PMN chemical can be estimated based on its expected use concentration.
Disposal methods vary according to the type of industry the shop is
associated with and the size of the shop. For water-based cutting fluids, small shops
may discharge to municipal sewer. Wastewater from large shops is likely to be treated
onsite before discharge. Treatment may include oil skimming, settling, neutralization.
Some shops drum wastes out and dispose of them off-site. Depending on the volume
of fluid used, spent oil-based fluids may be sold for reclamation or incinerated.
G. Filtering Solids From Water
The major factors governing release to water from separation of a solid
chemical substance (usually the product) from water through filtration are its solubility
in water and the volume of water used. Solubility is affected by temperature, pH,
presence of dissolved salts or minerals, and presence of dissolved organic matter.
Thus, it may be controlled by the manufacturer by adding materials such as salts or
pH adjusters to the water.
The chemical release quantity can be determined from solubility data at
processing temperatures of the filtrate and from the volumes of water that will be
directed to filtration operations. The amount of chemical lost to wastewater may be
estimated from the volume of water used multiplied by the chemical's solubility in.
water. This calculation assumes that 100 percent of suspended solids are filtered from
-------
Issued: February 28, 1991
Revised:
Page No.: 5-19
wastewater. If the necessary data cannot be obtained from the submitter, the
engineer can use the following steps to arrive at an estimate:
Obtain estimates of water solubility from ICB. These estimates may be
based on analogues or standard estimation techniques.
Estimate volumes of water at:
- Filtrate = 2 to 4 times chemical volume
Rinse water * 2 times chemical volume
if, based on professional judgment, it is expected that the chemical will be
salted out (that is, a salt will be added that depresses the solubility of the
chemical), reduce solubility in filtrate by 90 percent
Estimate quantity of chemical spilled as 0.05 percent of chemical volume.
. Calculate chemical dissolved in filtrate-and rinse water (assuming both are ;
appropriate for the scenario).
The sum of quantities released in rinse water, filtrate, and from spills and
leakage is the total released to wastewater. Total losses should not be assumed to
exceed 10 percent of production. This can be used as the upper bound for losses
based on the assumption that in the event that losses exceed 10 percent, the
company will take steps such as salting out or cooling to reduce solubility to avoid
larger losses.
Additionatlosses can occur when disposing of the filter. Assume, unless
otherwise indicated, that this, may occur once every 3 to 6 months, and that up to 0.5
percent of a batch can be lost to landfill or incineration in this way.
H. Spray Coating
One use scenario which CEB frequently assesses is industrial spray
application of coatings. Since submitters of PMN's frequently do not have much
information about the operations at potential use sites, CEB developed a report
entitled "Generic Engineering Assessment Spray Coating" (CEB 1987a) that describes
several generic spray coating scenarios. The report, which is included in Volume II, is
organized according to the industry of application: automotive finishing and refmish-
-------
Issued: February 28, 1991
Revised:
Page No.: 5-20
ing, wood and metal furniture finishing, large appliance finishing, railroad car finishing,
light aircraft finishing, and heavy machinery finishing. For each industry, a.matrix
provides information which can be used to estimate typical usage rate of coating,
coating composition, number of use sites, and transfer efficiencies.
CFB generally expects releases from overspray. The percentage of coating
that is oversprayed is dependent on the spraying atomization technique, the type and
composition of the coating, and the type of substrate. For spray coating operations
performed in spray booths, the overspray may be removed from air through capture
by a continuous curtain of moving water or by dry filter.
Water curtains are more likely to be used to control overspray generated in
large scale spray painting operations such as the finishing of automobiles, appli-
cances, and heavy machinery, Spent water from water wash systems can be a
source of water release if the water is disposed to a POTW or stream. To estimate the
release of PMN chemical to wastewater, the quantity of PMN chemical in the spent
water must be estimated based on the expected amount of overspray and the capture
efficiency of the curtain. If we do not have this information, we will make worst case
estimates of water releases based on water solubility. If the PMN chemical is
insoluble, we assume that little remains in the water wash and that the PMN chemical
is disposed of as part of the sludge. If the PMN is soluble, we estimate the amount
that will remain in the water using the following information (Duff 1985):
• ' Spray booths with water curtains typically contain 1,000 to 2,000
gallons of water.
The water is dumped out once a week or less frequently (up to once
yearly) depending on the efficiency of the paint sludge removal.
Paint sludge is drummed and sent to landfill.
The engineer should compare this estimated release to the estimated
amount of overspray as a check to see if the release makes sense (namely, the
release cannot be greater than the amount of PMN expected in the overspray).
-------
Issued: February 28, 1991
• Revised:
Page No.: 5-21
For dry booths, we generally assume that the overspray Is disposed of as
part of the dry filter wastes which are likely to be landfilled.
I. Leather Dyeing
Approximately 80 to 90 percent of all leather produced in the U.S. is dyed
(Rutland 1986). Leather dyeing is performed after leather hides have been tanned with
chrome or vegetable tanning agents, cleaned, defatted, trimmed, shaved, graded,
weighed, split, and tested for quality. The hides are then placed on wooden drum
dyeing wheels.
Dyeing or coloring of the hides is performed by rotating the dyeing wheels
while dyeing compounds are mechanically,pumped from a bath into the wheel through
an inlet port. The dye usually consists of a mixture of three different dyes. The
tanneries use a large number of dyes per day; however, only a few dyes are used in
large quantities (CEB I987c). Primarily, acid-based, metal complex dyes are used;
however, direct and solvent-based metal complex dyes are also used in the leather-
dyeing process. These dyes are generally manufactured in the powder form. Acid-
based dyes contain approximately 50 percent active colorant.
Following dyeing, the spent dye bath solution is typically discharged to the
plant sewer. This is followed by rinsing of the hides and subsequent discharge of the
rins'ewater to the sewer. Fat liquoring is the next step, It consists of addition of
chemicals to the hides to replace the natural oils lost during processing. The spent
fat-liquoring solution is also typically discharged to the plant sewer. The process is
concluded with removal of the hides from the dye wheels followed by drying and
finishing. Finishing consists of buffing, polishing, or other surface treatment
The main source of release of PMN chemicals to water is discharge of spent
dye baths to water. The bath is likely to contain some unexhausted PMN active
colorant. To estimate release of PMN colorant to wastewater from dye baths, the type
of dye, composition of the dye, and approximate quantity of dye used per batch must
be determined. Also, the degree of uptake or exhaustion of the dye in the bath must
be estimated to determine what percentage of the dye remains in the bath. If this
-------
Issued: February 28, 1991
. Revised:
Page No.: 5-22
information is.not available from the submitter, CEB generally assumes that 10 to 30
kg of PMN active colorant is used per site day for dyeing of grain and suede, respec-
tively, and that the degree of exhaustion of dye onto the leather ranges from 60 to 95
percent (CEB 1987c).
The document provides expected ranges of values for quantity of dye used
per year at an average tannery, type of dye used, and percent of active colorant in
dyes. From this information, a water release estimate can be made. The release
quantity should be adjusted to account for further wastewater treatment if applicable.
The removal efficiency of the chemical by the designated treatment method can be
estimated using methods outlined in the Water Controls section, Section VILA.
J. prilling Operations
Drilling fluids are essential to oil well drilling operations. In commercial oil
and gas exploration and production well drilling, rotary drilling equipment is generally
used A rotary bit is attached to and rotated by a hollow drill through which drilling
fluid is pumped. The drilling fluid is metered through the bit and subsequently fills the
volume of the hole or drill casing up to the surface. Upon reaching the surface, the
used drilling fluid is removed, cleaned, treated, and recycled back to the operation.
Drilling fluids perform the following functions: 1) lubricate and cool the drill
bit, 2) remove cuttings from beneath the bit to the surface for removal, 3) prevent
formation fluids from flowing out of the formation, 4) coat the well to prevent migration
of fluid into permeable formations, 5) suspend cuttings when circulation is interrupted,
and 6) help support the weight of the drill apparatus. They are generally composed of
liquids and suspended solids, and are either water- or oil-based. Water-based fluids
use water as the solids-suspending medium. They are comprised of approximately 70
to 90 percent water by volume, while the remainder consists of additives. Oil-based
fluids are comprised of oil, water, blown asphalt, and additives.
Drilling fluid additives include barite, clays, lignosulfonates, lignites, lime,
caustic soda, soda ash, silicates, salt, and phosphates. These additives are used to
control a wide range of fluid properties, including alkalinity, bacteria growth, calcium
-------
Issued: February 28, 1991
' .. Revised: November 8, 1991
, . . ' Page No.: 5-23
buildup, corrosion potential, foaming, friction, viscosity, and density. More information
on specific drilling fluid additives is provided in "Drilling Fluids • Environmental Release
Analysis" (CEB 1987d), which is contained in Volume II of this manual. .
The use of chemical additives in drilling fluids can result in releases to water.
The major sources of release to water are disposal of spent drilling fluids and produced
water. Spent drilling fluids or mud is generated during treatment of the drilling fluid.
Water-based drilling fluids are continuously removed from the well, passed through solids
control equipment, and circulated back to the well. The solids control equipment varies,
but it generally consists of a combination of screens, settling pits, hydrocyclones, and
centrifuges. These control systems remove sand and silt; however, they generally do not
have the capacity to remove fine particles.
As drilling fluid recirculates, the concentration of fine particles increases and
eventually the fluid becomes too viscous for further use. At this point, a portion of the
fluid (mud) is removed. It is replaced with water and additives in order to adjust the
properties of the fluid for continued use. In the case of oil-based fluids, the spent fluid is
usually sent back to the supplier for reconditioning. For water-based fluids used in on-
shore drilling, the spent fluid or mud is disposed in mud pits which are covered with dirt
or it is reinjected into the ground below the water table (CEB 1987d). In off-shore
drilling, the drilling mud is usually discharged into the ocean, provided an oil film is not
left on the water surface (CEB 1987d). Therefore, it is likely that an attempt is made to
remove only oil prior to discharge of the mud to the ocean (CEB 1987d).
A second potential source of discharge to water is disposal of produced water.
Produced water is a combination of formation water and drilling fluids, and is the
highest-volume waste source in the oil and gas industry (U.S. EPA 1985c). The quantity
of produced water is related to fluid production and varies from site to site. Prior to
disposal, produced water is sent through treatment to remove oil. Disposal is the same
as for drilling mud. For on-shore drilling, produced water is either reinjected into the
ground or placed in pits. For off-shore drilling, it is disposed to the ocean, provided that
no oil sheen can be detected on the water surface.
-------
Issued: February 28, 1991
Revised: November 8,. 1991
' Page No.: 5-24
Releases of specific chemicals to the ocean from disposal of produced water
and drilling mud can be estimated using the concentration of the chemical in the mud or
produced water multiplied by the mud or produced-water discharge rate. Where PMN
chemical concentration and mud or produced-water discharge rates are not available,
typical data can be used. The document "Drilling Fluids - Environmental Release
Analysis" presents the EPA Generic Drilling Fluids List, which provides concentration
data for eight generic types of mud (as defined by EPA). This document also provides
drilling fluid discharge rates by geographical location and off-shore produced-water
discharge rates.
K. R ecirculating Water-Cooling Towers (Kunz 1977, Perry 1984)
PMN chemicals are frequently used as water treatment additives in recir-
culating cooling systems to prevent corrosion, scaling, and growth of microorganisms. -
Typical concentrations of additives in the recirculating water are listed in Table 5-2.
TKEfE 5-2. OOCUM3 fOXER AEESTEVEi
Typical concentration
: Type of Additive _ (gpn) _ ___ • *
Scale control 1 to 5
Corrosion control. 50 to 1000
Microorganism control* ^...^?r^
3 This is a flfM use «nd not generally covered by TSCA. .
Source: Walk (n.d)b.
CEB typically assumes a recirculation rate of 2,000 gallons per minute (gpm)
for a moderatley-sized tower. A large cooling system may have a recirculation rate- of up
to 100,000 gpm.
Makeup fresh water must be added to the system to replace losses from
evaporation, entrainment (drift or windage), and blowdown. Releases from a
recirculating cooling tower are shown in Figure 5-1.
-------
Cooling «««P Circulatiart
Blowdown
tO **K
Issued: February 28,1991
Revised:
Page No.: 5-25
Air
TrMtmsnt Ctvwnical*
wattr
Figure 5-1. Schematic Of Cooling Tower System.
-------
Issued: February 28,1991
Revised:
Page No.: 5-26
Description of Release
Evaporative losses typically range from 0.5 to 3% of the recirculation
rate. If the PMN chemical is-non-volatile, the losses of PMN chemical due to
evaporation are expected to be negligible. , - '
Evaporation of water causes the nonvolatile materials to accumulate.
To keep the salt concentration at a predetermined level, a small amount of water is
deliberately discarded (biowdown). CEB generally assumes that blowdown is about
0.5 to 0.6% of the recirculation rate. Typically, blowdown is sent to either an on-site
treatment plant or POTW.
Windage losses are a function of the mist eliminator design and
generally range from less than 0,1 to up to 0.2% of the recirculation rate. Some
cooling tower manufacturers warrant as low as 0.008% for windage losses. In the
absence of other information, CEB assumes windage losses to be 0.1% of the
recirculation rate.
Calculations
Assuming that blowdown is equal to 0.6% of the recirculation rate, and
converting to the appropriate units, the water releases are estimated as follows:
B - 0.6% x Xr x Rx (5760 x 10~6 min-kg/hr-gat)
where B = blowdown (kg/site/day)
Xf = concentration of PMN chemical (ppm)
R = recirculation rate (gpm)
Assuming that the recirculation rate is 2000 gpm and that releases will
occur over 360 days/yr, releases to onsite wwt or POTW are estimated as:
-------
Issued: February 28, 1991
Revised: November 8, 1991
Page No.: 5-27
B = 0.07 xXr
where B = blowdown (kg/site/day)
X = concentration of PMN chemical (ppm)
R = recirculation rate (gpm)
Assuming that windage is equal to 0.1% of the recirculation rate and
converting to the appropriate units, air releases are estimated as:
W = 0.1% x Xf x R x (5760 x 10"6 min-kg/hr-gal) ,
where W = windage (kg/site/day)
Xr ~ concentration of PMN chemical (ppm) .
R = recirculation rate (gpm)
Assuming that the recircuiation rate is 2000 gpm and that releases will
, occur over 360 days/yr, releases to atmosphere are estimated as;
W = 0.012 xXf
where W = windage (kg/site/day)
X = concentration of PMN chemical (ppm)
R = recirculation rate (gpm)
Makeup or throughput of the PMN chemical is estimated as the total of
the windage and blowdown losses.
-------
-------
Issued: February 28, 1991
Revised:
Page No: 6-1
VI. MODELING RELEASE TO AIR
Air releases can occur from many sources. Typical sources include process
vents, tank working and breathing losses, fugitive releases, and secondary sources. A
vapor generation rate will often have been calculated for some processes and been
used to estimate worker exposure using the procedures described in Section IV. If a
vapor generation rate was calculated, the estimated release from the process should
be calculated using the same assumptions. Some equations presented in Section IV
are also used in this section with the modification that vapor generation rates are
typically calculated in grams/second and retease is typically calculated in kg/day.
When a chemical such as a PMN substance is used in low volumes and is of low
volatility, it may not be necessary to quantify release to air. In these cases a
qualitative estimate should be substituted such as "transfer retease is expected to be
negligible."
A. Process Vents
Process vents include the main air exhaust from the manufacturing process
including pressure relief valves. The methods that can be used to estimate releases to
air from a process vent include measurement, mass balance, emission factors,
engineering calculations, or a combination of these methods,
1, Measurement
Measurement is the most straightforward means of estimating releases.
EPA emission test procedures for regulated compounds are described in 40 CFR 60,
Appendix A, July 1986, The pollutant concentration and flow rate from a process vent
during typical operating conditions, if available, can be used to calculate releases.
Total annual releases are based on the plant operating schedule for the year. The
actual emission rate is derived by making a volume correction to account for the
difference between standard and actual vent gas absolute temperatures. The concen-
tration is then multiplied by the vent gas flow rate.
-------
Issued: February 28, 1991
Revised:
Page No: 6-2
This calculation assumes that the measured emissions always
represent the actual emissions. This may not always be the case. Ideally, using a
continuous emission monitor to measure and record releases would provide the most
representative data and therefore a basis for calculating an average concentration.
Gaseous concentrations also are frequently expressed in parts per
million (ppm) by volume (i.e.) a volume of the constituent in a million volumes of vent
gas). In this case, the vent gas volume must be multiplied by the concentration. The
resulting value is divided by the molar volume (adjusted to the vent gas temperature}
and multiplied by the compound's molecular weigh! to obtain the mass emission rate.
Some vent streams contain large amounts of water vapor (10 to 20
percent by volume), and the actual vent gas rate includes this volume of vapor.
Concentrations of chemicafs in the gas, however, are frequently expressed on a dry
basis. For an accurate release rate, the vent gas rate should be corrected for its
moisture content by multiplying by {1 minus the fraction of water vapor). The resulting
dry volume can then be multiplied by the chemical's concentration,
2. Mass Balance
Mass balance provides a means of accounting for all the inputs and
outputs of chemical in a process. A mass balance is useful for estimating releases
when measured release data are not available or when other inlet and output streams
are quantified. The amounts entering or leaving a process are either measured or
estimated. A mass balance can be performed on the process as a whole or on a
subprocess. Individual operations within the process usually must be evaluated.
3. Emission Factors
A third technique for estimating air releases from process vents involves
the use of emission factors. One type of emission factor relates a quantity of a
pollutant to some pro cess-related parameter or measurement The amount of
pollutant per quantity of product is frequently used.
Many air emission factors are expressed in terms of total volatile
organic compounds (VOC) or particulates rather than a single chemical compound.
-------
Issued: February 28, 1991
Revised:
Page No: 6-3
Emission factors for VOCs are available in "VOC Emission Factors for the NAPAP
Emission Inventory" (USE PA 1986c). These data can be used with actual process vent
measurements of volatile organscs or particuSates to estimate emissions of a specific
compound. The "Volatile Organic Compound (VOC) Species Data Manual" (USEPA
1980} also provides information on many air emission sources. This allows the user to
estimate releases of specific toxic compounds based on the total amount of VOC's
emitted from a particular source. Similarly, the "Receptor Model Source Composition
Library" (Carl 1984) provides information relating rnefaSs emissions to total paniculate
emissions for different release sources. Another good source of information is 'Toxic
Air Pollutant Emission Factors - A Compilation for Selected Air Toxic Compounds and
Sources" (USEPA 1988a),
4. Engineering CBlculations
When parameters related to emissions cannot be directly measured,
emissions can be estimated or inferred through engineering calculations or measure-
ment of other secondary parameters (i.e., physical/chemical properties of the mate-
rials involved, design information on the unit operation for which the estimate is being
made, or emission information from similar processes). Engineering calculations are
generally used to "fill in" information needed for other emission estimation methods.
Information derived from equipment design, such as fan curves, vessel
capacities, operating temperatures* and operating pressures, can be used to estimate
gaseous flow rates. Physical/chemical information derived from the ideal gas law,
vapor pressure, and equilibrium relationships can frequently be applied when
estimating gaseous concentrations of a particular compound.
B. Tank Working and Breathing Losses
Releases of chemicals from material handling, storage, and loading may
result from both breathing and working losses. Breathing losses are due to vapor
expansion and contraction, which force vapor from a tank or vessel. Expansion and
contraction are caused by temperature and atmospheric pressure fluctuations.
Working losses occur when the tank or vessel is filled or emptied.
-------
Issued; February 28, 1991
Revised:
Page No: 6-4
These types of releases are generally estimated by using emission factors
and engineering calculations. The EPA publication AP-42 provides equations for
estimating air emissions from organic liquid storage and handling operations {USEPA
19B5b). These equations contain factors that depend on tank parameters and service
conditions, The CEB engineer should consult Section 4.3 of AP-42 for specific
equations and typical factors to use to calculate working and breathing losses. More
specific information on storage tank emissions including example calculations for
horizontal tanks and chemical mixtures can be found in "Estimating Air Toxics Emis-
sions From Organic Liquid Storage Tanks" {USEPA 1988c}.
C. Fugitive Releases
Fugitive emissions are those emissions that are not released through a
stack, chimney, vent, or other confined vent stream. These releases include process
leaks, evaporation from open processes and spills, and raw material and product
loading and unloading losses. Whenever possible, fugitive emissions should be
calculated by using data available from direct measurement,
Fugitive emissions, however, often have to be estimated by using emission
factors or engineering calculations because they are too diffuse or dilute to be
measured directfy, or they are too small relative to the amounts of material processed
to permit the use of mass balance, This is particularly true of hazardous or toxic air
pollutants.
One basis for estimating process fugitive releases is the use of plant air
measurement data or worker exposure estimates provided by the PMN submitter.
Health and safety regulations may require measurements or regulated air pollutant
concentrations on either an absolute or a not-to-exceed basis. These data could
provide a basts for determining fugitive emissions. Occupational standards them-
selves, however, should not be used to calculate emissions. Only actual measure-
ments taken to ensure compliance with the standards should be used.
The accepted method of estimating releases from leaks in vessels, pipes,
and valves is to use emission factors. Various factors are available to estimate
-------
Issued: February 28, 1991
Revised:
Page No: 6-5
releases due to leaks in process streams carrying hydrocarbon vapors, light liquids
(more volatile than kerosene, i.e., a vapor pressure greater than 0.1 psia at 100'F), or
heavy liquids (equal to or less volatile than kerosene). These factors also can be used
to estimate fugitive emissions in other industries that process hydrocarbon streams.
Table 6-1 presents a summary of average fugitive emission factors in the synthetic
organic chemicals manufacturing industry (SOCMI). These data are based on
information from "Emission Factors for Equipment Leaks of VOC and HAP" (USEPA
1986a). This report addresses fugitive emissions and reductions due to scheduled
operation and maintenance procedures.
TABLE 6-1. AVERAGE FUGITIVE EMISSION FACTORS
FOR THE SYNTHETIC ORGANIC CHEMICALS
MANUFACTURING INDUSTRY (SQCMl)a'b
Emission factor
Fugitive emission source
Pump seals
Light liquids
Heavy liquids
Valves (in-line)
Gas
Light liquid
Heavy liquid
Gas safety-relief valves
Open-ended lines
Flanges
Sampling connections
Compressor seals
g/s
0.014
0.0059
0.0015
0.002
0.000064
0.029
0.0004?
0.00023
0.0028
0.063
These factors take into account a leak frequency determined
from field studies in the synthetic organic chemicals
manufacturing industry. Light liquids have a vapor pressure
greater than 0,1 psia at 10QT.
Factors have been converted from ib/h In the original
source.
Source; USEPA, 1986a
-------
issued: February 28, 1991
Revised:
Page No: 6-6
D. Secondary Sources
Secondary emissions of volatile compounds to the air can occur from the
on-site treatment of aqueous or solid waste. The bulk of secondary emissions are
estimated to result from the handling, pretreatment, and final treatment (primarily
biological treatment) of aqueous wastes,
Other sources include surface impoundments, landfilling, and incineration of
liquid and solid waste. Estimating releases of volatile compounds from disposal is
complex and requires detailed knowledge of the compound's parameters and the
disposal procedure.
Analytical models have been developed by EPA's Office of Air Quality Plan-
ning and Standards (QAQPS) to estimate emissions of volatile organic compounds via
various pathways from emission sources at hazardous waste disposal sites. The
report entitled "Hazardous Waste Treatment, Storage, and Disposal Facilities (TSDF) -
Air Emission Models" (USEPA 1988b) discusses these models. To make reasonable
estimates of volatile releases, one must know which pathways predominate for a given
chemical, type of waste site, and set of meteorological conditions. Models have been
developed for the following emission sources:
Nonaerated impoundments (which include quiescent surface impoundments
and open-top tanks);
Aerated impoundments (which include aerated surface impoundments and
open-top tanks);
Disposal impoundments (which include nonaerated disposal impoundments);
Land treatment; and
Landfills.
Computerized methods for applying these emission rnodeis are being devel-
oped by EPA. Models for aerated and nonaerated impoundments, lagoons, landfills,
wastepiies, and land treatment facilities have been installed in an integrated spread-
sheet program, CHEMDAT4, This program allows a user to calculate the partitioning
of volatile compounds among various pathways depending on the particular parame-
ters of the facility of interest.
-------
Issued: February 28, 1991
Revised:
Page No.: 7-1
VII. EVALUATING RELEASE CONTROLS
A. Water Controls
1. Gravity Separation
Gravity separation is widely used as a waste treatment process for the
removal of settle able solids, oils, grease and other material from water, Ciarifiers, API
separators, and inclined plate settlers are commonly used for gravity separation.
For this situation it is necessary to estimate the amount of water used
in cleaning to calculate the amount of chemical dissolved in the water and not readily
separated. The amount of water should be estimated as equivalent to 10 to 100
percent of batch volume ( all material in the batch, not just the chemical under study),
using the higher range for high viscosity materials in multi-component systems (i.e.,
several pieces of equipment). The amount of chemical lost in water is then the volume
of water times the water solubility of the material. Material not dissolved (i.e., amount
available minus amount dissolved) should be assumed to be incinerated or landfilled
after separation. This estimation method ignores variation of chemical solubility in
water, which may vary with temperature, pH» and the presence of dissolved salts and
minerals. In addition, the method is based on the assumption that 100 percent of the
suspended solids are removed during filtration or settling, resulting in a release of
dissolved solids only. For liquid/liquid systems, the method does not consider
formation of emulsions, Although this method does not describe actual systems, it
provides an order of magnitude estimate in the absence of specific data.
This approach should be used whenever immiscible materials are
washed from equipment and subsequently removed from rinsewater prior to waste-
water treatment or discharge (other than in a large holding pond in which dilution may
be assumed sufficient to dissolve most of the chemical).
2. Carbon adsorption
Carbon adsorption is a physical separation process in which organic
and inorganic materials are removed from wastewater by the attraction and
-------
issued: February 28, 1991
Revised:
Page No.: 7-2
accumulation of impurities on the surface of activated carbon. This method is typically
used when a material that is present in concentrations of less than 10 percent is to be
removed from a water stream that is otherwise relatively clean and free of suspended
matter, Activated carbon is either added directly to the wastewater stream and subse-
quently filtered out or it is placed in a column through which the wastewater passes.
Carbon adsorption can be used as a wastewater treatment method or for recovering
valuable organics and inorganics from water streams. In either case, a portion of the
chemicals is likely to remain in wastewater effluent that may be released to a stream or
POTW.
Organic chemicals differ widely in their treatability by carbon adsorption.
A survey performed by Troxfer, Parmlee, and Barton (CEB 1984a) resulted in carbon
adsorption removal efficiencies that varied from 42 to 99 percent for a variety of
organic chemicals. This report, which is included in Volume II, presents results of the
survey for chemicals that have been treated on an experimental level with carbon
adsorption and for chemicals removed in toll-scale treatment operations. The efficien-
cies presented in the tabies may be used to estimate water releases of the chemicals
that are included in the tabies; however, for chemicals not included in the survey (e.g.,
new chemicals), a different release estimation method must be used.
To estimate water releases of chemicals not included in the survey, the
chemical's treatability must first be estimated. Some properties used to assess a
chemical's treatability are molecular weight, structure, polarity, solubility, and water
conditions {pH and temperature). If adequate information is available, an assessment
of the effect of these properties and conditions on a chemical's treatability must be
made by the CEB engineer. CEB I984a discusses the effect these parameters have
on treatability. This document may be used to make reasonable worst case release
estimates for new chemicals. If a judgment about a new chemical's treatability cannot
be made from available information, data found in CEB 1984a for a similar existing
chemical should be used.
-------
Issued: February 28, 1991
Revised:
Page No.: 7-3
8. Air Controls
Air pollutants entering an air control device can undergo one or more of the
following: 1) they can be transferred from the air stream to another medium, 2) they
can be modified to a less toxic state, 3} they can be destroyed through combustion or
dissociation, or 4) they can pass through untreated. The physical characteristics of
the pollutant to be removed generally determine which type of control device is used.
Estimates of releases to air must consider the control equipment efficiency.
This efficiency should be based on the amount of pollutant removed from the air inlet
stream of the control device by destruction, modification, or transfer to another
medium.
V — X
Percent efficiency = inle! ^1 x 100 Equation 7-1
^inlet
where: XinlB1 = Total mass of pollutant X flowing to the air inlet of the control
device in a given year
^ouiiet = Totaf mass °* pollutant X flowing from the air outlet of the
control device in a given year
The amount of pollutant transferred to and subsequently released in another medium
(solid or water) must be included in the releases of that particular pollutant in that
medium.
The best basis for an efficiency estimate is a measurement or test, a mass
balance calculation, or a combination of measurement and mass balance calculations.
if such data are not available, comparison of "controlled" and "uncontrolled" emission
factors for the pollutant (chemical) of concern, engineering calculations, data on the
operation parameters of the control device, or vendor data or guarantees that reflect
actual operating conditions may be used. It is important to use data that reflect
efficiency obtained during typical operations, not the theoretical optimum efficiency.
fn the absence of typical operating data, treatment efficiency data cited in
the open literature for a similar process may be used as an approximate guide.
Without actual source test data for a specific emission stream and control system, the
-------
Issued: February 28, 1991
Revised;
Page No.: 7-4
removal efficiency can be assumed to equal total VOC removal efficiency if the
chemical is volatile organic compound (not a participate, metal, PCB, etc.),
Typical types of asr release controls include incineration, adsorption,
absorption, condensation, cyclones/mechanical collectors, fabric filters, electrostatic
predpitators, and wet scrubbers.
1. Jrjcjn§ration
Incineration is used to control air release of chemicals that can be
oxidized. Incineration is also used to reduce liquid and solid waste generated in a
process. A discussion of incineration as a control for both air streams and waste
streams is presented in Section VILC.1.
2. Adsorption
In an adsorption process, a pollutant is adsorbed on the surface of the
adsorbent until its capacity is reached. Common adsorbent materials used are activat-
ed carbon, resins, and molecular sieve materials. The adsorbent can then be regener-
ated. The pollutant is released in a more concentrated from, which is recovered or
treated by further processing. The particular adsorption/regeneration process and the
pollutant and its associated process parameters determine further processing steps.
These can include incineration or condensation and decantation so the chemical can
be recovered for recycling or disposal. Although adsorption is effective in the removal
of various toxic chemicals from air, the regeneration and further processing steps may
transfer some toxic substance to water or to solid waste streams, which must be
considered releases to these media. Typically, the adsorption capacity increases with
the molecular weight of the VOC being adsorbed. In addition, unsaturated com-
pounds are generally more completely adsorbed than saturated compounds, and
cyclical compounds are more easily adsorbed than linearly structured materials. Also,
the adsorption capacity is enhanced by lower operating temperatures and higher
concentrations. The VOCs characterized by low vapor pressures are more easily
adsorbed than those with high vapor pressures.
-------
Issued: February 28, 1991
Revised:
Page No.: 7-5
In the absence of process specific data, carbon adsorption can be
assumed to have maximum DREs in excess of 50 percent for VOC inlet concentrations
over 200 ppmv, over 95 percent for VOC inlet concentrations over 1,000 ppm, and
over 99 percent for VOC inlet concentrations over 5,000 ppnw (USEPA 1984).
Adsorption can be used for VOC recovery for relatively small industrial VOC sources.
Commercial adsorption systems are available for small flow rates (several hundred to
tens of thousands cfm) and low VOC concentrations (usuaily several hundred up to
several thousand ppm} (Spivey 1986).
3, Absorption
Absorption as a method of treating an emission is a physical or chemi-
cal process that transfers a components from a gas stream to a liquid. Although often
used to recover products or raw materials, absorption also can serve as an emission
control device. In this capacity, absorption has been used to control alcohols, acids,
chlorinated and fluorinated compounds, aromatics, esters, and aldehydes (USEPA
1984). Absorption devices can be used separately or with other air pollution control
equipment (e.g., to provide additional pollutant removal after incineration or condensa-
tion). Liquids are used as the absorbent; therefore, a media transfer of toxic pollutants
can occur. Liquid-to-gas ratios, liquid temperature, and column height are also
important parameters affecting efficiency.
In the absence of process specific data, absorption can be assumed to
have maximum DREs in excess of 90 percent for VOC inlet concentrations over 250
ppmv, over 95 percent for VOC inlet concentrations over 500 ppmv, and over 98 per-
cent for VOC inlet concentrations over 5,000 pprnv (USEPA 1984).
4. Condensation
Condensation is used as a control technique for some organic
compounds. It cools the gas stream and transforms the gaseous compound to a
liquid. Like absorption, condensation is one of the primary techniques used for
product recovery; however, it is also used as an air-pollution control. Control of
storage and process emissions is a common application. Condensers are frequently
-------
Issued; February 28, 1991
Revised:
Page No.: 7-6
used in series with other control equipment, including absorbers, incinerators, and
adsorbers.
In the absence of process specific data, condensation can be assumed
to have maximum. DREs in excess of 50 percent for VOC inlet concentrations over
500 ppnnv, over 80 percent for VOC inlet concentrations over 2,500 ppmv, and over
95 percent for VOC inlet concentrations over 5,000 ppmv (USEPA 1984),
5. Cjctones/'Mechanical Collectors
Cyclones are seldom used as the sole or primary means of particulate
collection, but they often serve as "first stage" air-cleaning devices that are followed by
other method of particle collection. Cyclone collection efficiency is probably more
susceptible to changes in particulate characteristics (i.e., process variation) than are
other types of devices. Therefore, care should be taken in the use of design efficiency
to estimate actual operating conditions. Although very little compound-specific collec-
tion data are available, cyclone operation depends on physical parameters (particle
size, density, velocity) as opposed to the chemical nature or properties of the material
being collected. Thus, within reason, it may be possible to obtain and transfer
efficiency data from known applications to unknown applications on processes with
physically similar particulate and gas flows.
Cyclones are good as precieaners removing large particles because of
their inability to capture particles smaller than 5 microns. Removal efficiencies range
between 80 and 99 percent in a conventional cyclone for particles greater than
15 microns. A high-efficiency cyclone will capture particles greater than 5 microns
(Stern 1977),
6, Fabric Filters
When properly designed and operated, fabric filters or baghouses are
efficient collection devices, even for small particles. Vendor information is often a
good source of collection efficiency information, as most units are designed for
specific applications. As is true of cyclones, fabric filter performance is affected by
process variations that affect the gas stream and by other variables such as
-------
Issued: February 28, 1991
Revised:
Page No.: 7-7
temperature and gas dew point. The particle collection mechanisms of these filters
(like those of cyclones) usually depend solely on physical as opposed to chemical
properties. Thus, data from known applications may be transferable.
Baghouses can be used with heavy grain loadings, but should not be
used for oily, hydrascopic, or explosive dusts. Fabric filters can collect particle sizes
ranging from submicron to several hundred microns in diameter. Removal efficiencies
are generally greater than 99 percent for particles greater than 1 micron (Perry and
Green 1984). Design parameters which affect the removal efficiency include the air-to-
cloth ratio, the inlet particulate concentration, the temperature of the air, and the
physical characteristics of the solids being removed,
7. Electrostatic Precipitators
Electrostatic precipitators (ESPs) remove from gas streams particles
that have been electrically charged. They are not used to collect organic solids
because of combustibility potential. Efficiency data are limited except for ESPs applied
to combustion processes. The collection efficiency of an ESP depends on the
physical characteristics of the particulate and the gas stream, and on the electrical
resistivity of the pollutant to be collected. Electrical resistivity, in turn, can be affected
by temperature, which may vary in some processes.
Theoretically, there is no minimum limit to the size of the particles that
can be collected by an ESP. High-efficiency ESPs have efficiencies greater than 90
percent for particles in the 0.1- to 10.0-micron range (Stern 1977), However, the
collection plates must be periodically cleaned in order to maintain a high overall
efficiency, since the resistivity of the dust cake affects the voltage and corona current,
and thereby the performance of the precipltator.
8, Wet Scrubbers
Wet scrubbers are used to collect organic and inorganic particulate
matter and reactive gases. Scrubbers which often use water as the scrubbing medi-
um, have the inherent potential of creating releases in the liquid medium. Like some
other particulate collection equipment devices, scrubber designs are based on
-------
issued: February 28, 1991
Revised:
Page No.: 7-8
physical parameters, so available efficiency data may be transferable. The key factors
in scrubber performance are partide size and scrubber pressure drop. As shown in
Figure 7-1 for a venturi-type scrubber, a high particle removal efficiency can be
achieved for larger particles and at higher pressure drops across the device (USEPA
1986d). The removal efficiency of a scrubber in removing gases or vapors depends
on both the solubility of the gas or vapor in the liquid and the degree of saturation.
Removal efficiency also depends on the contact time between the contaminate in the
gas phase and the surfaces of the liquid phase.
C. Liguid and Solid Waste Controls
Liquid (nonaqueous) and solid waste (including sludges and slurries) may
be reduced by incineration, chemical treatment, physical treatment, recovery/reuse,
and solidification/stabilization.
1. incineration
incineration is one of the most widely used technologies for hazardous
waste disposal. Industrial wastes consisting of a PMN chemical fraction are often
incinerated. These wastes include solvent cleaning/degreasing wastes, distillation and
reactor bottoms, separator/detergent sludges, skimmer refuse, waste oils, polymer
wastes, paint/ink sludges, thinners, pesticides/herbicides/insecticides, filter cakes,
and contaminated fiber drums/boxes.
The quantity of chemical or other chemical of interest released to air
from Incineration of waste (containing the chemica!) can be estimated using the
Destruction and Removal Efficiency (ORE) of the incineration device. The DRE
combines the efficiencies of both the destruction of the organic chemical of interest in
the combustion chamber and, if applicable, the efficiency of the subsequent removal of
any residual material from the stack effluent using control devices.
There are four types of incineration devices: incinerators, flares,
industrial boilers, and process furnaces. For each of these four types of incineration
there are many variations; however, ail incineration can be grouped into one of the
four types.
-------
Issued: February 28, 1991
Revised:
Page No,: 7-9
*
e
I
x;x" ^ v^ _v^_>^_^L^^^^ >*
0.2
0,3 0,4 0-6 0.6 0,7 0.8 0.9 1,0
Sin o< Pinicies iAefodynimic Mttn Oi«rn,J,
20 3.0 4.0 5.0
Figure 7-1. Venturi scrubber collection efficiencies (USEPA 1986d).
-------
Issued: February 28, 1991
Revised:
Page No,: 7-10
Incinerators are designed for the destruction of materials at high
temperatures. Incinerators may be thermal or catalytic. Thermal incinerators rely on
high temperatures, sufficient residence time, and adequate turbulence to insure high
destruction efficiencies. Catalytic incinerators operate at somewhat lower tempera-
tures as a catalyst promotes the oxidation. Although most VOCs are rapidly oxidized
at temperatures over 1400'F, some compounds (e.g., haiogenated hydrocarbons)
require higher temperatures. While destroying one air pollutant, incineration may
create other pollutants that require further treatment for removal from flue gases. For
example, an incinerator that effectively destroys trichloroethylene may create hydrogen
chloride that must then be removed by flue gas scrubbing. Flares are used to destroy
purged gaseous organic compounds when it is not economical to recover the heat
value of the gases. Industrial boilers are designed to generate steam through
combustion of fossi! fuel and may use hazardous wastes as a supplementary fuel.
Process furnaces are integral components of a manufacturing process and are used
for the recovery of material or energy.
Typically, the concentration of organics in the incinerator stack gas is
low, resulting in high DREs. The best basis for estimating efficiency is actual measure-
ment, which in PMN scenarios is unlikely, or the use of data that reflect typical
efficiency during similar operations.
EPA compiled efficiency data for three of the four types of incineration
devices from trial burns and other performance tests (CEB 1989a). From these data,
well-operated full scale incinerators averaged between 99.994 and 99.99997 percent
ORE, Boiler data averaged between 99.98 to 99.999 percent ORE. Data compiled for
process kilns ranged from an average of 99.2 to 99.998 percent ORE, The ORE
averages are difficult to interpret for PMN chemicals since trial burns occur under
steady-state conditions and the actual ORE is dependent on the concentration and
chemical class. These averages may be used for well-operated incineration devices.
A ORE of 99 percent is suggested for release estimation to represent a worst case
estimate.
-------
Issued; February 28, 1991
Revised:
Page No.: 7-11
Although flaring is widely used, information on the ORE for flares is
limited . A 98 percent ORE can be achieved for flares provided they operate under the
conditions listed in Table 7-1.
TABLE 7-1. OPTIMAL OPERATING CONDITIONS FOR FLARES
Exit velocity, V Heating value of HT of
Type of flare _ (ft/sec) _ gas stream" (Btu/scf)
Steam-assisted V < 60 HT > 300
60 1000
Non-assisted V < 60 HT > 200
60 < V b 200 < HT < 1000
V < 400 HT > 1000
Air-assisted V < V max ^ HT > 300
Heating value of total gas stream (not just listed chemical).
n n 1 n!"! I'M 11
T UlU1 ° {n^n or log Vma,fi; = 1424 + 0.0018 H-
° VhliK,:?, = 28.54 + 0..0087 HT
Source: CEB 1989a, 40 CFR 50.18 (July 1966)
HT should be calculated at conditions of 25 *C (77 *F) and 1
atmosphere (14.7 psia). For information on measurement and calculation of operating
exit velocity and heating value of the gas stream, consult 40 CFR 60.18 (July 1986).
Flares with values of less than 300 Btu/scf (steam- or air-assisted flares) or 200
Btu/scf (nonassisted flares) may or may not achieve 98 percent destruction. For
example, a steam-assisted flare burning a volatile organic compound could be
considered to have a 98 percent efficiency for that compound rf its exit velocity and
Btu value of the gas stream were within one of the three operating conditions listed in
-------
Issued: February 28, 1991
Revised:
Page No.: 7-12
Table 7-1 for this type of flare. This would allow an estimate of the control efficiency in
absence of other data for the compound.
To estimate release from an incineration device, the volume of waste
containing the PMN chemical or chemical of interest must be determined. Some
typical wastes include spent cleaning solvents, filter cakes, still bottoms, sludges, and
drum residues. Estimates of waste volume of the material of interest must be obtained
from the PMJM submitter or from manufacturers of existing chemicals.
To estimate the amount of chemical emitted to air from incineration,
volume of chemical fed to the incineration device should be adjusted with the expected
efficiency. This method provides an order of magnitude estimate only.
-------
Issued: February 28, 1991
Revised:
Page No.: 8-1
VIII. REFERENCES
ACGIH. 1988. American Conference of Governmental Industrial Hygienists, Industrial
Ventilation. A Manual of Recommended Practice, 21st Edition.
A, D. Little. 1985. Arthur D. Little, Inc. Guidelines for the Selection of Chemical Pro-
tective Clothing, Los Alamos National Lab. U.S. Environmental Protection Agency.
A. D. Little. 1989. Personal communication with Rosemary Goydan of Arthur D. Little,
Cambridge, MA, and Dawn Weiner of PE! Associates, Inc., Cincinnati, OH. December
1989.
Clement Associates, Inc. 1982. Methods for Estimating Workplace Exposure to PMN
Substances. Washington, D.C.: Office of Toxic Substances, U.S. Environmental
Protection Agency. Prepared under subcontract to Walk, Haydel, and Associates,
Contract 68-01-6065.
Carl, J. E., et a!., 1984. Receptor Mode! Source Composition Library. EPA-45Q/4-85--
002.
CEB (n.d.) Chemical Engineering Branch. CEB Research Project Effluent Guidelines
Information: Part A - Tank Truck Cleaning Operations. Washington, D.C,: Office of
Toxic Substances, U.S. Environmental Protection Agency.
CEB. 1984a. Chemical Engineering Branch, Carbon Adsorption. Washington, D.C.:
Office of Toxic Substances, U.S. Environmental Protection Agency,
CEB. 1984b. Chemical Engineering Branch. Exposure to N-nitroso diethanolamine in
Machine Shops. Washington, D.C,: Office of Toxic Substances, U.S. Environmental
Protection Agency,
CEB. 1984C. Chemical Engineering Branch, Exposure to N-nitroso diethanoiamine in
Selected Mela I working Operations. Washington, D.CX: Office of Toxic Substances,
U.S. Environmental Protection Agency.
CEB. 1985. Chemical Engineering Branch. Disposal of Metal-Working Fluids. Wash-
ington, D.C.: Office of Toxic Substances, U.S. Environmental Protection Agency,
CEB, 1987a, Chemical Engineering Branch. Generic Engineering Assessment Spray
Coating Occupational Exposure and Environmental Release, Washington, D.C.: Of-
fice of Toxic Substances, U.S. Environmental Protection Agency.
CEB, 1987b, Chemical Engineering Branch. CEB Research Project: Engineering
Standards. Washington, D.C.: Office of Toxic Substances, U.S. Environmental Protec-
tion Agency.
-------
issued: February 28, 1991
Revised:
Page No.: 8-2
CEB, 1987c. Chemical Engineering Branch, Generic Engineering Assessment *
Leather Dyeing: Occupational Exposure and Environmental Release, Washington,
D.C.: Office of Toxic Substances, U.S. Environmental Protection Agency.
CEB. 1987d. Chemical Engineering Branch, CEB Research Project: Drilling Fluids,
Environmental Release Analysis. Washington, D.C.: Office of Toxic Substances, U.S.
Environmental Protection Agency.
CEB. 1989a. Chemical Engineering Branch. CEB Research Project: Industrial Haz-
ardous Waste Incineration, Washington, D.C.: Office of Toxic Substances, U.S. Envi-
ronmental Protection Agency.
CEB. 1989b. Chemical Engineering Branch. PoSyefectrolytes, Washington, D.C.:
Office of Toxic Substances, U.S. Environmental Protection Agency.
Duff. 1985. Personal communication between W. W. Duff (Mitre) and R. Kachkuda
(EPA) concerning findings of discussions with industry representatives (unpublished).
January 15, 1985.
Geomet. 1989. Geomet Technologies, Inc., Multi-Chamber Consumer Exposure
Model (MCCEM) Version 2.1. Office of Research and Development, U.S. Environ-
mental Protection Agency. Report No, 1E-2130.
Gikis, B. 1983. SRI International. Industrial Process Profiles to Support PMN Review.
Final Report: Printing Inks, Washington, D.C.: Office of Toxic Substances, U.S. Envi-
ronmental Protection Agency. Contract No. 68-01-6010,
Girman, J. R., and Hodgson, A. T. 1985. Source Characterization and Personal Ex-
posure to Methyiene Chloride From Consumer Products. Washington, D.C.: Con-
sumer Products Safety Commission. Contract CPSC-IAO-84-1171.
Heath. 1984. The Dyeing and Printing of Textile Fibers Relative to Worker Exposure
and Environmental Release. Washington, D.C.: Office of Toxic Substances, U.S. Envi-
ronmental Protection Agency.
Heath, G. 1988, Memo from George Heath, OTS-ETD, to CEB Staff entitled Textile
Drug Room Monitoring Study Assessment of Workplace Dust Inhalation Exposures.
Kunz, R. G., et al. 1977. "Cooling-Water Calculations." ghemieaLErigineering Vol. 84,
No. 16, August 1, 1977, p 61-71.
Mitre. 1984, The Mitre Corporation, Information on the Loading and Unloading of
Chemicals Under Nitrogen Blanket, Washington, D,C.: Office of Toxic Substances,
U.S. Environmental Protection Agency. Contract 68-01-6610.
-------
Issued: February 28, 1991
Revised:
Page No,: 8-3
Mitre, 1985. The Mitre Corporation, Particulates in the Workplace. Washington, D.C.:
Office of Toxic Substances, U.S, Environmental Protection Agency. Contract
68-01-6610.
Monsanto, 1978. Monsanto Research Corporation. Source Assessment: Rail Tank
Car, Tank Truck, and Drum Cleaning, State-of-the-Art. Dayton, Ohio,
MRl. 1986. Occupational Exposure from Bagging and Drumming Operations. MRI
Project 85Q1-A(10). Washington, D.C.: Office of Toxic Substances, U.S. Environmen-
tal Protection Agency. Contract 68-02-3938.
NIOSH. 1973. National Institute for Occupational Safety and Health. Industrial Envi-
ronment, Its Evaluation and Control. Cincinnati, OH; NIOSH, U.S. Department of
Health and Human Services.
NIOSH. 1976. Nationa! Institute for Occupational Safety and Health. A Guide to
Industrial Respiratory Protection. Cincinnati, OH: NIOSH, U.S. Department of Health
and Human Services. HEW Pub. 76-189.
NIOSH, 1984, W, N. McKinnery and W. A. Heiibrink, National Institute for Occupa-
tional Safety and Health. Control of Air Contaminants in Tire Manufacturing.
Cincinnati, OH: NIOSH, U.S. Department of Health and Human Services. DHHS Pub,
84-111.
NIOSH. 1987, National institute for Occupational Safety and Health, Respirator Deci-
sion Logic, Cincinnati, OH: NIOSH, U.S. Department of Health and Human Services.
DHHS Pub. 87-108,
O'Brien, D. M. and D. E, Hurley, 1981, National Institute for Occupational Safety and
Health, An Evaluation of Engineering Control Technology for Spray Painting.
Cincinnati, OH. NIOSH, U.S. Department of Health and Human Services. DHHS Pub.
81-121.
Pace Laboratories. 1989. Evaporation Rates of Volatile Liquids, Second Edition.
Washington, D.C.: Office of Toxic Substances, U.S. Environmental Protection Agen-
cy, Contract 68-D8-Q112,
PEI Associates, Inc. 1986a, Releases During Cleaning of Equipment. Washington,
D.C.: Office of Toxic Substances, U.S, Environmental Protection Agency, Contract
68-02-4248.
PEI Associates, Inc. 1986b. Use of Oii Separators in Drum Reconditioning and Trans-
portation Vessel-Cleaning Facilities. Washington, D.C.: Office of Toxic Substances,
U.S. Environmental Protection Agency. Contract 68-02-4248,
-------
Issued: February 28, 1991
Revised:
Page No.: 8-4
PEl Associates, Inc. 1987. Effectiveness of local Exhaust Ventilation for Drum-Filling
Operations. Washington, D.C.: Office of Toxic Substances, U.S. Environmental Pro-
tection Agency. Contract 68-02-2947,
PEI Associates, Inc. 1989. Guidelines for Statistical Analysis of Occupational Expo-
sure Data, Washington, D.C.: Office of Toxic Substances, U.S. Environmental Protec-
tion Agency. Contract 68-D8-0112,
Perry, R. H., and D. Green, 1984, Perry's Chemical Engineers' Handbook, Sixth Edi-
tion. McGraw-Hill Book Company, New York, NY.
Popendorf, W. J., and Leffingwell, J, T. 1982. Regulating OP Pesticide Residues for
Farmworker Protection. Residue Review, Vol. 82, pp. 156-157, New York, NY.
Rutland. 1986. Leather Industries of America. Letter to Jerry Borbach, Re: Com-
ments on July 1985 Draft of Generic Engineering Assessment of Leather Dyeing, as
cited in CEB 1987c.
Schrov, J. M., Wu, J. M. 1979. Emissions from Spills. Proceedings: Air Pollution
Control Association/Water Pollution Control Association Joint Specialty Conference:
Control of Specific (Toxic) Pollutants. Gainesville, FL
Schrov, J, M. 1981. Prediction of Workplace Contaminant Levels. NIOSH Symposi-
um Proceedings: Control Technology in the Plastics and Resins Industry. February
27-28, 1931. National Institute for Occupational Safety and Health, Division of Physical
Sciences and Engineering, Cincinnati, OH. p. 190. DHHS Pub. 81-107,
Spivey, J. J. 1986. Recovery of Volatile Organic Compounds From Small Industrial
Sources. Research Triangle Institute.
Stern, A. C. 1977. Air Pollution, Third Edition,, Volume IV: Engineering Control of Air
Pollution, Academic Press, New York, NY.
Traynor, G. W., Girnnan, J, R., Apte, M. G., et al. 1985. Indoor Air Pollution Due to
Emissions From Unreacted Gasfireo* Space Heaters. Air Pollution Control Assoc J 35:
231-237,
Treybal, R. E. 1980. Mass-Transfer Operations, 3rd Edition, New York: McGraw-Hill,
Inc. pp. 369-371,
Turk, A. 1963, Measurements of Odorous Vapors in Test Chambers. ASHRAE J
5:55-88.
-------
Issued: February 28, 1991
Revised;
Page No.: 8-5
USEPA. 1980. U.S. Environmental Protection Agency. Volatile Organic Compound
(VOC) Species Data Manual. Second Edition, EPA-450/4-8D-015. Research Triangle
Park, 'NC. 465pp.
USEPA. 1984. U-S. Environmental Protection Agency. Hazardous/Toxic Air Pollutant
Control Technology, A Literature Review, Research Triangle Park, NC,
EPA-600/2-84-194.
USEPA. 1985a. U.S. Environmental Protection Agency. Survey of Perchloroethylene
Emission Sources. Research Triangle Park, NC. EPA-45Q/3-85-017.
USEPA. 1985b. U.S. Environmental Protection Agency. Compilation of Air Pollutant
Emission Factors. Volume 1: Stationary Point and Area Sources. Research Triangle
Park, NC. (AP-42).
USEPA. 1986a, U.S. Environmental Protection Agency. Emission Factors for Equip-
ment Leaks of VOC and HAP, Washington, D.C. EPA 450/3-86-002.
USEPA, 1986b, U.S. Environmental Protection Agency,, New Chemical Review
Process Manual. Washington, DC. EPA 560/3-86-002.
USEPA. 1986c. U.S. Environmental Protection Agency. VOC Emission Factors for
NAPAP Emission inventory. Research Triangle Park, NC. EPA 600/7-86-052,
USEPA. 1986d. U.S. Environmental Protection Agency. Control Technologies for
Hazardous Air Pollutants. Cincinnati, OH, EPA/625/6-86/014.
USEPA. 1987. U.S. Environmental Protection Agency. Estimating Releases and
Waste Treatment Efficiencies for the Toxic Chemical Release Inventory Form,
Washington, D.C. EPA 560/4-88-002.
USEPA. 1988a. U.S. Environmental Protection Agency. Toxic Air Pollutant Emission
Factors - A Compilation of Selected Air Toxic Compounds and Sources, Research
Triangle Park, NC. EPA 45D/2-88-QQ6a,
USEPA. 1988b. U.S. Environmental Protection Agency, Hazardous Waste Treatment,
Storage, and Disposal Facilities (TSDF) - Air Emission Models. Research Triangle
Park, NC. EPA 450/3-87-026.
USEPA, 1988c. U.S. Environmental Protection Agency, Estimating Air Toxics
Emissions From Organic Liquid Storage Tanks. Research Triangle Park, NC. EPA
450/4-88-004.
-------
Issued: February 28, 1991
Revised:
Page No.: 8-6
Van En et al. 1980, Worker Exposures to Chemical Agents in the Manufacture of
Rubber Tires: Solvent Vapor Studies. American Industrial Hygiene Association
Journal (41), March 1980, pp. 212-219,
Versar. 1984, Versar, Inc. Exposure Assessment for Retention of Chemical Liquids
on Hands. Washington, D.C.: Exposure Evaluation Division, U.S. Environmental Pro-
tection Agency. Contract 68-01-6271,
Wadden R. A., Franke, J, F, 1985. Eddy Diffusivities Measured Inside a Ugh! Indus-
trial Building. Poster No. 10? presented at the American Industrial Hygiene Confer-
ence, Las Vegas, NV. May 23, 1985.
Wadden, R. A., and Berrafato, L P. 1988. Predicted vs. Measured Air Emissions of
Volatile Organic From a Simulated Hazardous Liquid Waste Lagoon. Paper to be pre-
sented at the 18th Annual Mid-Atlantic Industrial Waste Conference.
Walk. (n.d,)a Walk, Hayde!, & Associates, inc. Industrial Process Profiles to Support
PMN Review: Metal Treatment Chemicals. Washington, D.C.: Office of Toxic Sub-
stances, U.S. Environmental Protection Agency. Contract 68-01-6065.
Walk. {n,d)b Walk, Haydel, & Associates, Inc. Industrial Process Profiles to Support
PMN Review: Waste Treatment Chemicals. Washington, D.C,: Office of Toxic Sub-
stances, U.S. Environmental Protection Agency. Contract 68-01-6065,
Williams, T. M. 1980. Worker Exposures to Chemical Agents in the Manufacture of
Rubber Tires: Particulates, American Industrial Hygiene Association Journal (40),
March 1980, pp. 204-211,
Wong, K, F. 1988, Memo from Kin F. Wong, OTS-ETD to CEB Staff titled Interpreta-
tion of Nonisolated Intermediates.
-------
APPENDIX A
SAMPLE INITIAL REVIEW ENGINEERING REPORT
A-l
-------
ENGINEERING REVIEW FORM CBI Y/N
c-1« , -, ,<-., CE3' Focus Representative:
PMN; I ME.'I . _________
PV fkQ'Vf'' SUBMITTER' — —
MSDS ~ LABE_ H, USE:
aloves/goggies/giasses/iocai exhaust ventilation/general mechanical ventilatiorVother.
respirator: air punfymg/arganic vapor/dust/paint mist/supplied air/other
Health Effects: corrosive/flammable/other
irritant to skin/eyes/lungs/mucous membranes
TLV/PEL (PMN c-f raw materials) ... _ ——
CRSS:
Chemical Name/Category
VP: torr@25degC S-H2O: Phys State: .
MW: ^ < 500 % < 1000
SAT (concerns): Health: _ — ——
Eco:
Assumptions:
Poiution Prevention Considerations:
EXPOSURE BASED REVIEW
1) # workers exposed:
2) > 100 workers with > 10 mg/day inhalation exposure LJ
3) (a) > 100 workers w/1-10mg/day for >100days/yr Q
(b) Routine Dermal Contact: > 250 workers &. > 100 days/yr
FOCUS
Date:
Decision/Comments:
A-2
-------
ENGINEERING REV;EW FORM
CBI Y/N
PMN/TME/I
MFG/P ROC/USE
Site/Location
P'ocess Description
Days/yr
Occupational Exposure
Total No. of Workers
days/yr
Inhalation; negligible.'vapor/'mist/particulate
# workers witfi inhalation exposure
Dermai: contact with liquids or solids
Dermal: contact negligible r^~;
Basis
mg/day over
Basis
_days/yr
% concentration
Environmental Releases;
. kg/site-day WATER over days/yr from -
kg/si:e-day AIR over days/yr from.
Total.
Total.
kg/yr INCINERATION from
kg/yr LANDFILL/DEEP WELL from
P ROC/USE
Site/Location
Process Description
Days/yr
to
Occupational Exposure
Total No. of Workers
days/yr
Inhalation: negligible/vapor/mist/particulate
# workers with inhalation exposure
Dermai: contact with liquids or solids
Dermai: contact negligible I I Basis
mg/day over
Basis
Environmental Releases:
kg/site-day WATER over,
— kg/site-day AIR over
jdays/yr
% concentration
jjays/yr from .
. days/yr from.
Total.
Total,
kg/yr INCINERATION from
kg/yr LANDFILL/DEEP WELL from
Consumer Use? Y/N
.to
A-3
-------
APPENDIX B
GUIDELINES FOR COORDINATED ETD PMN
STANDARD REVIEW
B-l
-------
Guidelines for Coordinated ETD
PKN Standard Review
March 1989
fi-Z
-------
E . I n t rodLIC t, i on
The following guidelines have been, designed in response to
the increased resource constraints on ETD PMN standard Review
efforts. Previously, a member of each Branch prepared a separate
Branch report for each PMH that was placed in Standard Review.
Consequently, there was duplication of effort among the Branches.
The new review process eliminates duplication, draws upon the
most appropriate expertise from each Branch and expedites the
transfer of relevant data to the target audience. These
guidelines outline not only each Branch's responsibilities
during Standard Review but also discusses Pre-Focus and Focus
activities in which the data being collected for PMNS can be used
later for those cases that are designated for Standard Review.
II. Pre-Focus Activities
A. ICB
ICB's pre-Focus"activities include providing information for
the Chemistry Review and Search Strategy meeting (CRSS), at which
all PKN submissions are evaluated on a chemical basis for
completeness, consistency and accuracy. The Chemistry Report
generated for CRSS contains the following information when
available or obtainable:
i- Chemical identity, including chemical name, trade
name, categorical name, molecular formula, and CAS
Registry Number;
ii. Physicai/chemical properties, molecular weight,
production volume and physical state;
iii. Associated substances;
B-3
-------
~"~7 • Use as listed in ?MT-r subr.iss i on ; nn - u ^ t >•• e r --«=•• = - •
necessary for CRSS ' "
v. Chemical struct,;., re information :or ZMD's database
r' F.., I. A ;
''' i . Analogs ;
vii. Pertinent chemical reactions;
v i i i .. Add i 11 or.a 1 i n format i on, i nc lud i ng other uses and
ix. References,
For those* cases that go to focus the "Additional
Information" section will contain information on possible other
uses of the ?l-!N chemical, based on chemical or technical
feasibility. The other uses information win be obtained from
readily available sources, such as CAS Online and the chemist's
direct knowledge. If no information Is available through such
sources, other uses listed win be based on expert judgement,
This other use information for the CRSS report is to be
considered preliminary information and should only consist of a
list of potential other uses that can be gathered relatively
easily and should constitute little additional time spent on any
case,
3. CEB
CEB will prepare the Initial Engineering Report, addressing
potential releases and exposures associated with the
manufacturing, processing and use of the compound.
The sections of the engineering report detail manufacturing,
process, and use. Manufacture is the method by which the PMH
chemical is produced. Process specifies any operations the PMN
B-4
-------
chemical win undergo in becoming part of the end-product. This
section will include to whom or to what types of companies -ihe
PKK substance will be sold, concentration of the PKN chemical in
the product and application techniques. Use win describe the
various uses of the PMN chemical or the FMN chemical based
product including a percentage of total production volume
allocated to each use if there will be more than one use. Any
industrial use will also be described in this section.
C, RIB
For all cases identified at the Structure Activity Team {SAT)
meeting with health and ecological concerns of greater than
"l/l", RIB win assess the reported production volume to
determine whether there is potential for increased growth in the
use(s) listed and whether the volume submitted appears reasonable
or valid. Those cases that meet the criteria established for
exposure based review will be evaluated in accordance with that
review. Determination of valid volume will be through the use of
the RIB historical PMN database, "Herman", and the RIB Volume-Use
Matrix.1
This RIB^FMN database, "Herman", contains historical
information, on a per case basis, on variables such as
the use, substrate, industry and volume of the PMN
substance. This data is stored on hard disk (RIB,
1989). The RIB Volume-Use Matrix (RIB, 1988a) contains
a percentage of total population for any given use and
volume in each cell of the matrix. Use of the RIB PMN
Information Program (PIP) (RIB, I988b), a menu driven
information retrieval system designed for use with
Herman, in conjunction w;th the Volume Use Matrix will
provide an accurate assessment of expected volume of a
particular market.
B-5
-------
The volume information win be collected and — P^.,,,^ ?n ,
£-- _ ..» O V_ * i -,„, ,w. ,_j, j. j | ,;3
standardized format on the back of the FOCUS sheet that RIB
currently draws up for each PHN the Agency receives. In case of
multiple uses, the search win be performed on each use code, at
the appropriately prorated volume, and separately presented. In
cases of too little historical data, professional judgement will
determine volume reasonableness at the Focus Meetino.
III. FOCUS Meeting
h, ICB
The ICB representative win present the chemical identity,
physical/chemical properties and associated substances
information. The representative will also present the potential
other uses listed In the Chemistry Report, as discussed in
Section II A, Pre-Focus Activities, ICB,
B. CEB
The engineering representative win report the initial
engineering assessment of the PMN.
C. RIB
The RIB representative win discuss reasonableness of
reported volume and potential for increased volume of the
reported use, based on the information that has been collected as
discussed in Section XI, B, Pre-Focus Activities, RIB.
D. Initial Review Report
The chemistry initial review and the engineering initial
review reports will comprise the ETD Initial Review Report, The
B-6
-------
Initial Review will be collated at the FOCUS meetina for cases
placed into standard review. The integrated ETD Initial Review
vi11 be red-dotted and submitted to the second floor CBic room
for immediate tracking arid filing. if the case is dropped or
placed into a different review category (ie. Exposure-based
review) the initial assessments will be finalized in accordance
with current procedures for those reviews.
IV. Standard Review - Branch Reports
A. ICB
The ICB Standard review report will be 1-2 pages and contain
the most pertinent information to the case. Standard Review
information will include the PMM #, chemist, submitter, chemical
name, categorical name, chemical structure, and other chemistry
information that the chemist deems pertinent to the case.
Additional information on intended use and other use will
also be collected at this time. "Intended Use" will address
chemical functionality (how the PMH substance works from a
chemical perspective) especially for those cases in which the
functionality of the molecule is a unique, unusual or complex
"Other Uses" will be expanded if other sources are located that
can enhance the original listing developed for CRSS.
The standard review section of the chemistry report will be
completed no later than a week before mid-course for inclusion in
the ETD Standard Review report.
B-?
-------
3. CE3
The CEB standard review will be tailored to the specific
engineering or exposure issues that arise as a result of FOCUS.
The CEB standard review will contain more extensive information
on the particular points addressed in the Initial Review,
including intended use. Information on intended use win include
function and application of the chemical substance (e.g., fiber-
react ive dye on cotton) and tier distribution.
The "Intended Use" information win not reiterate the ICB
input but, rather, is intended to address aspects of use
relevant to the engineering assessment, which shall include the
production volume of the PMN chemical and formulation
information. Thus, while ICB win address the chemical
functionality of the substance in its intended uses, the
engineering report win deal with the more physical use oriented
aspects of the substance.
If preliminary information on chemical substance, volume and
related subjects is necessary for managerial review of the draft
CEB report, the ICB initial report can be given simultaneously
for F'fi purposes,
The engineering report should be completed and given to the
ETD lead branch for coordination no later than a week before mid-
course. Although, the report win still be completed in the same
time frame as before, it will not be available in report form
until the entire ETD report is submitted to the fourth floor HERD
CBIC room. However, exposure and release estimates will be
8-8
-------
available to other divisions, as it is assumed that the current
policy of relaying this information through personal contact will
continue.
C, RIB
The economics section of each Division standard review win
consist of benefits, substitutes, and production volume
verification. Other uses will be dealt with almost exclusively
by ICB. Only in rare cases where additional other use
information is requested by HERD or CCD, will RIB contribute
additional input on other uses, specifically the market
feasibility of other uses. Production volume verification will
be available, in the majority of cases from the RIB Focus sheets.
In those cases where no definitive Information was collected,
additional RIB input Is necessary.
It is the RIB analyst's responsibility to retrieve the Focus
Sheet (mentioned in II.c.) from the RIB files and pass this
information along with the RIB report to orient RIB management
for review purposes,
V. Collation of Information for Division Report
One of the three branches on any given case will be elected
as the lead branch, Lead branch for any case would be determined
by the last number of the PMH case in review. If the case ends
with a zero, the last non-zero digit in the case number, will
then be the determinant. See the following chart.
B-9
-------
1,2 or 3 JCB
-' ,5 or 6
L.HJ3
7'8 °r 9 T-n
RIB
For example, for cases P-89-14G7, P-89-100 arid P-89-6 RIB, ICB
and CEB would be the lead branches, respectively. in the event
of a consolidated PW , or several cases that have been combined
because of similar concerns, the number of the first case In the
series win determine the lead branch,
The analyst from the lead branch win be the coordinator for
that particular case who win collect and collate reports from
the branches. The coordinator's job is solely one of: (1)
assuring consistency of assumptions and. data in the reports
(e.g., are the intended uses listed by ICB, for the initial
report that are consequently being used for the final report, the
same as those that CEB is assessing from an exposure
perspective), most often this will mean reading the consolidated
report for consistency only, and (2) physically assembling (i.e.,
stapling!) all three branch reports together for one Division
package. The coordinator will not edit or rearrange information,
Each branch is responsible, as is currently the policy, for the
technical quality of each report.
it is each analyst's responsibility to insure that all
information presented in any report is in agreement with the
information presented in the other reports. Consequently, each
analyst must coordinate with her/his branch counterparts on any
B-10
-------
particular case to insure that the information presented in each
section is'consistent with the other two sections. One copy of
each branch report will be due to the coordinator no later than
one week before mid-course. It is expected that the current
scheduling format produced by CCD win be revised to reflect the
reporting changes proposed here. The coordinator win submit .the
collated report to the fourth floor CBI room in HERD.
VI. ETD D i s po s i t i on
ETD Disposition (Dispo) will be held according to the
current scheduling routine. At Dispo, each analyst will be
responsible for reporting the salient points of their portion of
the standard review. ICB can use charts prepared for CRSS while
CEB can either create overheads of the process-flow diagrams
present in their report or draw up separate charts.
If additional factors arise as a result of ETD Dispo that
entail additional analysis from one or more branches, the new
information will be incorporated as an addendum by the branch
analyst responsible for providing more data.
-------
REFERENCES
Regulator Impacts Branch, 1988a. Memorandum with Attachment from D.
Qzolins, Economics and Technology Division to Anna Coutlakis Chemical Control
Division, April,
Regulatory Impacts Branch, 1988b. "PKN Information Programs (PIP) User's
Fanual". Prepared for Office of Pesticide and Toxic Substances, U.S.
Environmental Protection Agency. Prepared by Mathtech, Inc. Contract No.
68-02-4240. June,
Regulatory Impacts Branch, 1989. Computerized Use Information on all PHNs
Submitted to the Office of Toxic Substances up to FY89.
B-12
-------
APPENDIX C
SAMPLE PRODUCTION EXPOSURE PROFILE (PEP)
NOTE: Sections I through VIII of the PEP are completed by RIB and ICB
c-i
-------
ACRYLOHITRILE
IX. PROCESS METHODS
A- Manufacture
The domestic production volume of acrylonitrile (CAS No. 107-13-1) in
I98E was 1,170 mi 11 ion kilograms (2,580 million pounds){CMR, 89). Imports of
acrylonitrile are very small {US. EPA, 89), Nine companies manufacture
acrylonitrile and four import acrylonitrile, as reported in the TRI data
base.
The entire production of acrylonitrile in the United States is obtained
by the ammoxidation of propylene (Kirk-Qtteer, 78a; US, EPA 77a). Figure 1
presents a schematic diagram of this process. The process is based on the
vapor-phase catalytic air oxidation of propylene and ammonia. In this pro-
cess, refinery propylene (90+ percent), fertilizer grade ammonia (99.5+
percent), and air are combined in a fluidized bed-reactor at a temperature of
450°C and a pressure of 2 atm (US EPA, 77a). The reaction is catalyzed by a
Sohio developed catalyst (50 to 60 percent bismuth phosphonohydrate on A1203)
which increases the yield of acrylonitrile and decreases the production of
acetonitrile and HCN.
Owing to the high conversion of propylene to acrylonitrile, a once
through process with a residence time of a few seconds is employed
(Kirfc-Othmer, 78a). The heat of reaction is recovered in the form of steam.
Commercially recoverable quantities of acetonitrile and hydrogen cyanide are
produced as by products.
The reactor effluent is cooled and scrubbed with water in a countercurrent
absorber. The absorber off-gas consisting chiefly of nitrogen is vented
while water, acrylonitrile, and byproducts are removed from the bottom of the
absorber and sent to the acrylonitrile recovery column. Crude acrylonitrile
C-2
-------
Steam
Water-
Catalyst f : =>
_J
Ammonls
N-—-^
^ Re;
xyr 1
^Scrubbed vent gas
to atmosphere
Water
Reactor
Absorber
T
Water
Recovery
column
Crude
acQtonitrito
A
Acetonitrile
fractionatar
__ Byproduct HCN to
Storage or disposal
Acrylonftrfte
*s
Lights
columr
i
>,
'
T
Product
column
Heavy
impurities
Figure 1. Simplified Schematic Diagram of the Production of Acrytonilrito by the
Ammoxidalion of Propylana Using
-------
and HCM are recovered overhead while water and acetonitrile are removed at
the column bottom. The crude acrylonitrlle is then sent to the lights column
for removal of the HCH. It is then further purified in the product column to
obtain fiber-grade acrylonitrile by fractionation at atmospheric pressure,
Aqueous wastes can be disposed of by a variety of means including
deep well disposal and incineration.
Plants are assumed to operate with a 7-day work week, 50 weeks per year.
B. Manufacture of Acrylic and Modacrylic Fibers
An acrylic fiber is a manufactured fiber-forming long-chain synthetic
polymer composed of at least 85 wt percent acrylonitrile units (Kirk-Othmer,
78b). Modacrylic fibers contain from 35 to 85 mass percent acrylonitrile.
Commercially, modacrylics contain from 25 to 60 percent of monomers such as
vinyl chloride or vinylidene chloride and thus possess a high degree of flame
resistance.
Acrylic copolymers are generally made by either heterogeneous or solution
polymerization (Kirk-Othmer, 78b). Modacrylics are made by these two methods
plus emulsion polymerization. Both batch and continuous processes are used.
Acrylonitrile is moderately water soluble, but the polymer is insoluble in
both acrylonitrile and water.
In a typical continuous process monomer, comonoroer, water, and initiator
are fed to a continuously-stirred, overflow reactor at atmosphere pressure
and a temperature from 30 to 70°C. Figure 2 is a schematic diagram of this
process. For modacrylic fibers, a halogen-containing monomer is usually also
added to increase the flame resistance of the polymer. The slurry of polymer,
water, and unreacted monomer is filtered. The polymer is washed and dried
while the monomer is recovered from the filtrate and returned to the reactor.
C-4
-------
Recovered monomer
Acrytonitrite
Gomanomer
Water
Catalyst
Chilled water—^
Polymerization
reactor
Filter/
Monomer
recovery
Washer/
dryer
Polymer
storage
(water and catalyst)
Figure 2, Simplified Schematic Diagram of the Production of Acrytte/Modacrylic Fibers
by the Continuous Process (Kirk-Othmer, 78b).
-------
Use of monomers containing halogen (modacrylic) introduces certain polymeri-
zation complexities not generally encountered in preparing acrylic polymers.
Their low boiling point (-12CC for vinyl chloride and 38CC for vinylidene
chloride) Riay require pressurized polymerization vessels, and their lower
reactivity requires that an excess be added necessitating that a higher
percentage be recovered for reuse.
Solution polymerization is used to prepare acrylic polymers directly in
a form suitable for wet or dry spinning. Solvents include dimethyl sulfoxide,
dimethylformanide, and aqueous solutions of zinc chloride or various thio-
cyanates.
Assuming the plants operate a 7-day work week, 50 weeks per year, acrylic
fibers are manufactured 350 days per year.
C. Manufacture of ABS and SAN Resins
The domestic production volume of acrylonitrile-butadiene-styrene (ABS)
copolyrners (CAS No. 9003-56-9} in 1987 was 572 million kilograms (1,261
million pounds) (USITC-SOC, 88). The domestic production volume of styrene-
acrylonitrile (SAN) copolymers (CAS No. 9003-54-7) in 1987 was 95 million
kilograms (209 million pounds) (US1TC-SOC, 88).
Commercially, SAN copolymers are manufactured by three processes:
emulsion, suspension, and continuous mass, ABS resins are also produced by
three processes: emulsion, suspension, and bulk.
C.I. Emulsion Process
The production of ABS resin by the emulsion process will be used at the
example since it involves one more step than SAN resin. Figure 3 is a
schematic diagram of the process, which consists of three distinct poly-
merizations. A polybutadiene substrate latex is prepared, styrene and
-------
Butadfam,
Emulsifiers.
Initiators,
Water
n
i
Polybufidten*
lalnx reactor
Emulsfftors,
Initiators
Steam
ABS
latex
n
o
>
K
r-JC
O
r**^
\
I
Go
Goagulator
Water
wasft
Hof air
Dry main
Figure 3. Schematic Diagram ol (he Production of ABS Resin by Ihe Emultion Process (Kirk-Olhrntr, 78c),
-------
acrylonitrile are grafted onto the polybutadiene substrate, and the styrene-
acrylonltrile copclymer is formed (Kirk-Othmer, 78c). The latter two
reactions may take place simultaneously in the safne vessel, followed by
blending of the latex.
The emulsion process usually takes place in a batch reactor at between
5° and 70°C depending on the desired structure of the polymer. The initiator,
activator, and emulsifier solutions are prepared in separate vessels ar.d
added to the reactor which has been purged of oxygen. The dimineralized
water and butadiene are then added, the temperature is increased, and the
reaction cycle begins. Heat of polymerization is removed by use of a water
circulating jacket and the reaction vessel is designed to withstand pressures
up to 145 psi. Reactors range in capacity from 13 to 30 m3 (3,400-7,900 gal)
and reaction times range from 12 to 24 hours (Kirk-Qthmer, 78c).
Styrene and acrylonitrile are then grafted on the polybutadiene sub-
strate. This can be done by the addition of SAN copolymers or with the
formation of the SAN copolymer in situ. These reactions are run from 55° to
75°C at atmospheric pressure in vessels of up to 20 m3 (5,300 gal) (Kirk-
Qthmer, 78c), The reactor is heated and reaction times range from 1 to 6
hours. The resins are then coagulated at elevated temperatures (80° to
100°C) to promote agglomeration of the resin particles. The slurry is then
dewatered and dried in a hot air dryer such as a rotary fluid-bed or flash
dryer. The dry resin may then be pneumatically conveyed to silos.
The emulsion copolymerization of SAN is similar to that described for
ABS except that it can be either a batch or continuous process. The copoly-
merization is carried out between 70° and 10Q°C. The copolymer later may be
C-8
-------
used to make ABS or it may be coagulated, washed, and dried to recover the
SAN copolyrners, Cycle time is about 1 to 3 hours (Kirk-Othmer, 78c).
C.2. Suspension Process
Figure 4 presents a schematic diagram of the suspension process. In
contrast to the emulsion process, the suspension process begins with a poly-
butadiene rubber which is so lightly trosslinked that it is soluble in the
monomers. It is then heated to 8Q~12Q°C for a period of 2-3 hours with
shearing agitation sufficient to prevent cross!inking and maintain the desired
polymer particle size. This prepolymer syrup is then transferred to a suspen-
sion reactor where it is dispersed in water with agitation. The reactor is
heated to 100° to 170°C depending on the initiator used until polymerization
is essentially complete (6 to 8 hours) in a reactor of up to 40 m3 (10,600)
gal and pressures up to 3.5 atm.
When the batch has reached the desired conversion, it is cooled, dewatered,
and dried. This part of the process may be continuous. Dry beads are stored
in silos prior to compounding.
The suspension process to produce SAN is similar with copolymerization
carried out at temperatures between 60° and 15CTC. The polymer spheres
formed are much larger than with the emulsion process which makes dewatering and
drying easier.
Plants are assumed to operate with t 7-day work week, 52 weeks per year.
X. OCCUPATIONAL EXPOSURE AND ENVIRONMENTAL RELEASE
A. Occupatl..o.naj Exposure
Acrylonitrile is a colorless liquid in the pure state with a characteristic
odor of peach seeds. Acrylonitrile has a molecular weight of 53, a vapor
pressure of 109 ran Hg at 25°C» and a water solubility of 7.4 percent at 25°C.
C-9
-------
o
I
o
s
r
Rub
dtssi
tymm
vbbei
C
tor
liver
»(
r
Z>
Pr«j
trw
>
(^ —
C2
1
O
*^^_ .-?
soly^
'iztr
>
r^"
c
1
O
k^__ ^>
\
^Neu
traliza
20^30% Suspensbn
conversion reactor
Hc?f a/r
1 Exhaust
t Cenirffuge X i
x^"^ ^"V _,, 1
/ \ ^ inifinfi __^. i
^ J UUuUU ^^ J
~~~~ ^^S,VJJ''
T Dryar -^P
Efftimt
Dry rwsto
Figure 4. Sdwmatfc Diagram of the Production of ABS Resin by the Suspensbn Process (Kirk-Othmor, 78c).
-------
The OSHA Permissible Exposure Limit (PEL)'s 2 ppm (10 ppm ceiling, 15 min),
arsd the N10SH Recommended Exposure Limit (REL) are 1 ppm (8 hr TWA} and 1C
ppm (15 minute ceiling). KIOSH recommends that acrylonitn'le be handled as a
suspected human carcinogen. The exposure level determined to be Immediately
Dangerous to Life or Health (IDLH) is 4,000 ppm. The ACGIH Threshold Limit
Value (TLV) is 2 ppm with a skin notation. According to the NOES Survey
(1981-1983), 61,534 workers were exposed to acrylonitrile. A breakout of the
number of workers by SIC code from the HOES Survey is given in Table 1.
A.I, Manufacture
During acrylonitrile manufacture, workers may be exposed to acrylo-
nitrile during sampling, maintenance activities, cleanup of spills, drumming,
and bulk loading of the final product, transfer of waste off-site for in-
cineration or landfill, and disposal of waste on-site by underground injec-
tion or to land.
P£I estimates that approximately 45 workers per plant may be exposed to
acrylonitrile during acrylonitrile manufacture based on the total employees
and total number of plants reported for SIC 28 in the NOES survey for acrylo-
nitrle. Due to the closed nature of the manufacturing process, inhalation
exposures are expected to be controllable at or below the OSHA PEL of 2 ppm
although the use of respirators may be required for some job activities.
Dermal exposures are expected to be low, since the workers are expected to
wear gloves while handling acrylonitrile containing products or waste. It is
also expected that workers will wear safety glasses or goggles.
A.2. Production of Acrylic/Modacrylic Fibers and ABS/SAM Resins
During the manufacture of acrylic and modacrylic fibers and ABS/SAN
resins, workers may be exposed to acrylonitrile during receipt of the acrylo-
nitrile in bulk, during transfer of the acrylonitrile, during maintenance
C-11
-------
TABLE 1. WORKERS EXPOSED TO ACRYLONITRILE ACCORDING TO THE NOES SURVEY
SIC
15
17
20
22
23
28
30
3?
49
73
75
80
Total
Description
General building contractors
Special trade contractors
Food and kindred products
Textile mill products
Apparel and other textile products
Chemicals and allied products
Rubber and misc. plastics products
Transportation equipment
Electric, gas, and sanitary services
Business services
Auto repair, services, and garages
Health services
Plants
303
282
23
16
47
151
89
7
47
24
485
17
Total
employees
3,233
5,131
1,174
6,265
27,720
6,806
4,991
50
372
2,451
971
2^371
61,534
C-12
-------
activities on. the acryloni trile storage or transfer lines and during clear- up
of spills of acrylonitrile. No information was found estimating residual
acrylonitrile monomer in acrylic or modacrylic fibers or ABS/SAN resins.
PEI estimates that approximately 45 workers per plant may be exposed to
acrylonitrile during the manufacture of acrylic and modacrylic fibers or
ABS/SAN resins based on the total employees and total number of plants
reported for SIC 28 in the NOES survey for acrylontrile. Due to the closed
nature of the manufacturing processes, inhalation exposures are expected to
be controllable at or below the OSHA PEL of 2 ppm, although the use of res-
pirators may be required for some job activities. Dermal exposures are
expected to be low, since workers are expected to wear gloves while handling
acrylontrile. It is also expected that workers will wear safety glasses or
goggles.
A.3 0 S HA Expo sure_Measureren t s
Table 2 presents a summary of OSHA monitoring data for acrylonitrile.
There was only one facility where OSHA took a personal sample for acryloni-
trile. This was at SIC 2283, yarn mills-wool, and the measured values were
0.31 and 0.33 mg/m3. There were no measurements by OSHA at any of the facil-
ities identified as acrylonitrile manufacturers or producers of acrylic and
modacrylic fibers. Two facilities identifed as ABS/SAN resin producers,
American Cyanamid and Sybron Chemical had screen values reported by OSHA,
The American Cyanamid facility had a screen value reported of
0.0 mg/m3 while the screen measurement for Sybron Chemical was 260 mg/m3 (120
ppm}. All other OSHA measurements were screen values and all were 0.0 mg/m1,
B. E rs v i ronmenta 1 Release
Acrylonitrile is a RCRA priority pollutant. The RCRA waste number for
acrylonitrile is U009. Acrylonitrile is also regulated under the Clean Hater
C-13
-------
TABLE 2, OSHA MONITORING DATA FOR ACRYLQNITRILE EXPOSURE VALUE, fig/m3
SIC JOB TITLE PERSONAL TWA SCREEN
Manufacture: None
Production of acrylic and modacrylie fibers; None
Production of ABS/SAN resins:
2821 --- — 0
2821 — --- 260
Other:
1721 — --- 0
2263 Spinner 0,33
Spinner 0,31 - —
3069 --- — 0
3079 -— — 0
0
3674 — — 0
0
0
6531 — — 0
C-14
-------
Act (307{a/» 311 (bjI, CERCLA (100 Ib reportable quantity) anc the Toxic
Substances Control Act (TSCA, 8s, 8e). Acrylonitrile Is not regulated at
present by the following EPA Acts: Clean Air Act, Safe Drinking Water Act,
and FIFRA. Acrylonitrile is reportable under SARA Title III, Section 313.
The environmental releases for 1987 reported under SARA, Title III, Section
313 are presented in Table 3,
B.I. HajTijfac_ture of Acryloim'trile
In Table 3, the companies identified in this PEP as manufacturers in
Section I were separated in the 1987 TRI data base and the reported informa-
tion for these facilities is presented in Table 3 as acrylonitrile
manufacturers. All reported manufacturers were located in the 1987 TR! data
base for the acrylonitrile manufacture category. All acrylonitrile
manufacturers identified in the SRI Directory as manufacturers reported
sizeable release of acrylonitrile to air (both fugitive and stack) and to
underground injection.
6.2. Production of Aerylie/Nodacrylic Fibers
SIC 2824 (Synthetic Organic Fibers, except Cellulosic - acrylic fibers)
was used to define the production of acrylic and roodacrylic fibers. While
there may be some overlap with other uses, this SIC seems to be primarily
concentrated in this use. Comparison with the SRI Directory of Chemical
Producers (SRI, 86) resulted in the transfer of one facility from SIC 2821 to
this category. These facilities are presented in Table 3 as producers of
acrylic and modacrylic fibers. All facilities that produce acrylic and
uiodacrylic fibers reported sizeable release of acrylonitrile to air (both
fugitive and stack). Only one facility reported a sizeable release or
off-site transfer to another media (underground injection).
C-15
-------
Table 3. TR! Data for Acrytenitrile.
Cwnpwjy
Manufacturing sites:
AMERICAN CYANAM
BORG WARNER CHE (a*)
BP CHEMICALS -
BP CHEMICALS AM
DUPQNTBEAUMON
GENERAL ELECTRI fab)
MON5ANTQCOMPAN
STAR-GLO«0USTfb)
STERLING CHEMC
Stale
IA
MS
TX
OH
TX
NY
TX
W
TX
SfCoxte
2819
2821
sees
2873
2822
2821
2869
3069
2865
FWM&es, tofy
_F^ta_
i?,ooo
58,439
24,000
K.ooo
8,100
91,000
499
65,000
PrwfcKtfort of actytic a«7 modsraj^fe Jtoers,1
AMERICAN CYANAM
BASFCORPQRATO
OU PONT MAY PIA
QypQNTWAY1«S8
MONSANTOCQMPAN
a
VA
sc
VA
AL
Prt*dt*ftart of MBS/SAN iws/rtt:
AKRON POLYMER P
AKBQN POLYMER P
AMERICAN CYANAM
AMQGQ PERPQfiMAN
BFQQQ0WCHAVQ
BF GOODRICH. LO
BOflS-WARNiRCHE
BQRG-WARNERCHE
DOCK RESWS OOfl
DOWCHEMOL
OOWWOuHCAL
oa«OlEMK*i
ICt RESINS US.
MONSAffTOCOMPAN
MORTON CHEMICAL
MOfiTON THWKOL
OH
OH
CT
X
OH
KY
WV
1
Ml
OH
CA
CT
MA
OH
SC
IL
2824
2824
2824
2821
2824
2821
2821
MM
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
51,920
25,810
75,000
S.MO
07,000
4,832
25,870
4,509
4,189
1,788
1,700
270,000
13,000
40
10
1,300
10
16,000
499
999
Start
132,000
27,643
11,000
110,000
29.000
160,000
499
39,000
55,971
482,722
249,000
120,000
170,000
68,900
1,387
5,214
49
100,000
870,000
581,000
6
1,700
580
510
22
110,000
499
499
Water
499
499
4»
240
4.138
499
Uwtef.lnj.
1,100,000
730,000
1.900.000
210,000
200,000
41,000
35,712
Land
499
1,498
499
9,983
499
S
TO
11
Cd-site iransta , b^
POTW
45,231
4
499
130.000
48
499
Lareflill
9M
1,497
750
499
3.B8Q
24
49i
86
9W
1
3,200
499
irrarwration
61,206
499
§0,588
499
3?
9,800
5,800
9,083
700
32,000
38,000
499
Water
12,000
OHwf
14
Wasfs
Sreatmwt
L
AW
AW
A
AL
AW
A
W
w
W
WL
AW
WL
AW
AW
AW
A
AL
AL
AL
A
AW
AW
AW
-------
Table 3. TR! Data for Acrylonitrile.
Company
Production of ABS/SAN r
NATIONAL STARCH
NATIONAL STARCH
PARA-CHEW SOUTH
PPG INDUSTRIES,
RECHHOLQ CHEW
ROHM AND HAAS,
AOHM AND HAAS.
mm AND HAAS,
ROHM AND HAAS,
SYBRON CHEMICAL
THESfQQQORlCB
THE DERBY CGMPA
THEDQWCHEMICA
UCAR EMULSION S
LICAR EMULSKJN S
UNION CARBIDE C
UNION CARBIDE C
UNION CARBIDE C
UNIROWLCH£M»C
UNDCAL CHEMICAL
VALCHEM POLYMER
WALSH CHEMICAL
Other:
3M COMPANY
ABCG INDUSTRIES
AIR PRODUCTS &
AIGQ CHEMICALS
ALOO CHEMICAL C
AMERICAN CYANAM
AMERICAN CYANAM
AMERICAN SYNTHE
BASFCORPORATIO
BASFOOHPORATO
BENKBER ELECTS
Stale
sslns (co
SC
It
SC
OH
Oi
KY
CA
TN
PA
NJ
W
MA
Ml
TX
1L
HI
CA
QA
OH
NC
SG
NC
KY
^
KS
IL
TN
WV
NJ
KY
Ml
NC
CA
SIC code
nTj:
2®1
2^1
2821
^
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2821
2121
2641
2899
2865
2869
2199
2819
2669
2822
2989
2989
3629
Releases, toyr
ZBjgWws__,
1,339
149
250
499
28.439
270
220
800
1,104
330
530
499
18,198
2
1
2
22
10
24,400
499
499
499
499
4,100
499
999
3,700
108
4,000
4M
Sack
499
71
100
1,100
12,117
160
6
90
499
2
499
9,786
2
8
9
25,000
999
999
499
499
999
499
999
2,238
4
3,700
499
Warer
14
1
499
2
499
499
499
UfKter. Inf.
499
Land
12
499
GK-siie irareief, b«Yr
POTW
6
320
5
700
499
10,000
499
499
12
Landfl
230
6
499
f
1
Inciwralion
499
30
155
499
Waist
Ottwf
7
114
1
1
499
1.400
Wastfl
troalmwil
AW
W
AW
AL
AW
AW
AW
AW
AW
AW
AW
AS
A
A
A
A
A
A
AW
WS
-------
Table 3, TRI Data for Acrytonitriie.
Qjmyny
Other (oont,}:
BFQQQORICHCQM
BUCKMANLABORAT
CHEM-FOUR FIRST
COPOLYMER RLBBE
DIXIE CHEMICAL
OU FONT FRONT R
DU PONT PARL1N
DUPQNTSASINE
DUPONTVICTOR1
EXXON CHEMICAL
E»ON CHEMICAL
FAIRPRENE WDUS
GOODYEAR TIRE*
GRAIN PROCESSIN
JETCQCHEMtCALS
LAUREL RUBBER D
LUBRBOL PETROL
MERCK* CO., W
MONSANTQCQMPAN
NALCOCRiMICAL
MATKDMAL STARCH
ORGANIC PIGMENT
PETfiOUTECQHPQ
PPQ MOUSTRES.
QUANTUM CHEM1CA
RHQNE-PQULENC!
ROHM TEC 1C.
SHEREX CHEMICAL
SIGMA CHEMICAL
TEXACO CHEMICAL
TEXAS PETROCHEM
THE DOWCKEM1CA
THE SHERWIN-W1
TRWOLWELLCAB
UNION CARBIDE C
W.R. GRACE ICO
W.RGfiACE/EVAN
Slate
OH
TN
AL
LA
TX
VA
W
TX
TX
W!
NJ
CT
TX
1A
TX
NJ
TX
VA
1A
LA
m
NC
TX
OH
PA
WV
MA
IL
MO
TX
TX
TX
KY
KS
WV
KY
NY
SlCoxte
2868
2899
2869
2822
2S13
2851
3861
2989
20E9
?m
2869
3068
2822
SJOffi
?869
3069
2869
2834
2871
y$f)
2869
2815
2899
2»1
2965
28S9
2851
2643
2868
?869
2S6i
2812
2851
3351
2969
mz
2980
Rflteases, Mf
fu^l'iVH
5,000
1.226
499
2.20)
999
499
1.262
6
1.600
2,000
499
7,600
499
499
410
9.BOO
200
756
499
1,900
499
33,230
499
499
499
3,500
499
49
3,664
26,000
1,900
920
Stack
1,600
499
499
400
3,156
1,985
2
310
380
499
15,890
21
499
1,570
MQ
730,000
200
7,067
1,900
1,540
3.059
499
499
499
550
2
3,664
3,800
3.200
530
Watw
450
2
26
6W
18
8
Under, hij.
92,000
21,000
Land
499
2,589
28
Off-silo Irarrstw, to/yr .
POTW
91.000
499
499
5
473
1,304
25
999
9.079
499
495
400,000
LandW
3.600
310
1.6(30
2.589
1
Indnefalnn
1,648
499
1,996
497,836
394
3,428
Walef
Ollw
4 AA
100
6,392
trealmwH
AW
AW
f^. *T
AW
A
lit g>
ws
A W
W
W L
Jt t*J
A W
W
L
L
A
W
(•) Tl»» soytef**! manufaefurefi «re abo in SK 2821 (ABS/SAN mdns),
(b) Thssa fadltes ire not feled as aoytonltrfi mamlatfiiraR In th* SRI Dtoawy d OwmW Produews,
-------
B.3. Production of ABS/SAJvJRes|ns
SIC 2821 (Plastics Materials, Synthetic Resins, and Nonvu I can-liable
Elastomers - ABS resins, S.AN resins) was used to define the production of SAN*
and ABS resins. Two facilities listed in this SIC in the 1987 TRI data base
were also acrylonitrile manufacturers. These facilities were deleted from
the SAN/ABS resin list and so footnoted in Table 3. While there may be some
overlap with other uses, the SIC seems to be primarily concentrated in this
use. Comparison with other information provided by EPA resulted in the
identification of several facilities in SIC 2821 as actually nitrile rubber
producers (US EPA, 88), These facilities were not deleted as PEI did not
have a complete list of nitrite rubber producers.
Some release to acrylonitrile to air (both fugitive and stack) was
reported by all ABS/SAN resin producers but the release estimate is highly
variable between facilities. Reported air release ranged from 1 pound to
over 1 million pounds per facility. Release to other media and off-site
transfers is generally low with a few specific exceptions.
8.4. Other Uses
Release of acrylonitrile from other facilities follow the same pattern
as for ABS/SAN resin production with almost all reporting release to air but
with the estimate highly variable. Release to other media and off-site
transfer was generally low with a few specific exceptions.
At one acrylic polymer facility studied for EPA, OAQPS, sources of
acrylonitrile release included acrylonftrile storage tanks, monomer blend
tanks, monomer feed tanks, monomer recovery strippers, recovered monomer feed
tanks, recovered monomer weight tanks, the equalization basin, filter
emissions and fugitive emissions (US EPA, 87).
C-19
-------
Industrial Process Profiles for Environmental Use (US EPA, 77a) esti-
mated release of 5,0 kg/mg of acrylonitrile product released to air from the
manufacture of acrylonitrile.
XI, RECOMMENDATIONS FOR ADDITIONAL INFORMATION
It is likely that additional investigation into both occupational expo-
sure and environmental release would produce better estimates. SpecificaHy^
further investigation of the one high screen value at Sybron Chemical may
identify what the purpose of the screen measurement was and if this screen
measurement means that there may be exposure of workers to levels of acrylon
itrile above the PEL.
Categorization of acrylonitrile manufactures by size and control types
or other factors could help to explain differences in releases in the TRI
data base. Additional investigation into which facilities in SIC's 2821 and
2624 actually produce the SAN/ABS resins and acrylic fibers would improve
estimates made in these categories. Contact of exceptionally high or low
release estimates could also help to identify the reasons for the variant
estimates.
C-20
-------
REFERENCES
CMR. 1989. Chemical Profile: Acrylonitrile. Chemical Marketing Reporter,
p. 50. March 6, 1989.
Kirk-Othrner» 78a. Encyclopedia of Chemical Technology. Volume 1.
Acrylonitrile. Wi ley-Interseience. 1978.
Kirk-Qthmer. 78b. Encyclopedia of Chemical Technology. Volume i. Acrylic
and Modacrylic Fibers, Hiley-Interscience. 1978.
Kirk-Othmer, 78c. Encyclopedia of Chemical Technology, Volume 1.
Acrylonitrile Polymers (SAN and A8S}. Wiley Interscierice. 1978.
SRI. 1988 Directory of Chemical Producers, United Stated. SRI International
1988.
US EPA, 1977a. Industrial Process Profiles for Environmental Use: Chapter
6. The Industrial Organic Chemicals Industry. Prepared for the U.S.
Environmental Protection Agency. Cincinnati, Ohio. February 1977.
US EPA. 1987, Assessment of Acrylonitrile Emissions From American Cyanamid
Company, Hilton, Florida. Prepared for the U.S. Environmental Protection
Agency, Office of Air Quality Planning and Standards, Research Triange Park,
H.C. Karen 1987.
US EPA. 1988. A Review of the Acrylonitrile Pilot Program. Prepared for
the U.S. Environmental Protection AGency, Office of Air Quality Planning and
Standards, Research Triangle Park, N.C. February 1988.
US EPA, 1989. U.S. Environmental Protection Agency, Toxic Chemical Release
Inventory (TRJ) Data Base. Washington. D.C.
US ITC-SOC, 88. U.S. International Trade Commission, Synthetic Organic
Chemicals, United States Production and Sales, 1987. Washington, D.C.
USITC pub. 2118.
C-21
-------
APPENDIX D
SAMPLE TR1 DATA
D-l
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
TOS DC 2O6C
MEMORANDUM
SUBJECT: Engineering Assessments of ?RI Petiti
TOXIC suss
ons
FROM:
TO:
Larry E. Longanecker, Section cfiief
Chemical Engineering Branch
CEBStaff
EB
is required to prepare a preliminary release assessment
«,fi?nS.t? deUSt SeCtl°n 313 <*emicals within ab"t 3
weeks of the initiation of the review period. Sufficient data
must De provided to EAB to allow them to complete a orel Jnarv
exposure assessment one week later. This memo lays out the tasks
the engineer _ should complete with respect to the database for
1987 suomissions to prepare this report.
,h . dev^loPed a ^nned database search and report format
that should expeditiously give the engineer the most important
information to quickly evaluate before transmittal to EAB
Essentially, a table will be printed for releases from the
facility (fugitive air, stack air, water to receiving streams,
underground infection, and on~site land disposal) and one for
of, -site transfers (to POTK, landfill, incineration, other water
treatment, and other treatment). water
o water to POTW
o other off-site transfers
Each table will present:
O DCN
o facility name
o SIC Code (reports
o quantity .released
listed in ascending SIC Code order)
At the end of each table, data on * of sites, total release, and
average release per site will be presented.
^ The_engineer will review these tables to determine, based on
engineering judgment, whether release numbers are reasonable, A
D-2
-------
column will be provided on the table for the enaineer t
questionable data such as: '
o low air releases of a volatile;
o lack of water releases for some facilities in
where water release is expected;
o 46,0i0,000 pounds of A12Q3 to POTW.
l f nf1based on f««her review of the entire form,
calls _ to the facility, engineering judgement/calculations, or
data in the petition, enter a reasonable estimate if necessary
lhandwriting it on the table). A column will also be included
for the engineer to estimate days of release, if necessary for
tne assessment. J
_ The engineer will highlight the largest (or otherwise most
nrr^i^f3 rel"s«s in each media for EAB attention. since the
DCN wiK be provided, it should not be necessary for CEB to
provide additional data from the database to EAB (e g^ name of
POTK, lat/long). HoweverT^hoqjIdTKr^hHose ?o use one or
several of the reported values in a site-specific assessment of
exposure, the engineer may be required to do further work to
confirm the release amount or establish the release scenario.
The printouts will also indicate whether, for example a
uty reP°rted on-site wastewater treatment. The engineer
should look at these forms in evaluating whether releases are
reasonably reported and, in the case of water treatment, whether
s-hould evaluate reported efficiencies.
It should be possible to search the database within a couple
of days of assignment. Work with the EAB assessor to determine
early which numbers warrant further evaluation on your part.
Finally, the written report (not due until a month after the
preliminary assessment} should rely heavily on these tables
Process descriptions, conclusions about the data, and alternative
estimates made by CEB should be the focus of text.
cc; William Burch Susan Hazen
Liz En* an Bob israei
Lynn H^lpire CEB File; Petitions 300080
D-3
-------
DOCUHCNT
CONTROL I
us re i
OOOU33TX
0796570TX
04595 09PA
061073 WJ
0096503IN
OS7J879CA
03«4tSW
0777HMWtl
OOJA16JME
05^591 WI
06 SfcOO 36*
FACILITY HAH1
ATLAS HAHUFACTU
XL CORPORATION
COLE DISTBIBUTI
CAPITOL CASTING
THE ELECTRON CO
ITEM INDUSTRIES
CERTAIN tECO CO
ECC AMERICA AMS
APEX CHEMICAL C
BOBOEN, INC, -
MILK SPECIALTIf
AMERICAN CRYSTA
HOLLY SUGAP COR
AftEBICAN CRYSTA
AMERICAN CSY3TA
HINH-DAK FARriER
AHERICAN CRT5TA
FMC CORPORAT10H
L. HATRICK IMC.
HEST POIHT PiPP
KEALTH-TEX IHC.
HEALTH-TIX IMC,
TME mm CWPAM
SPRINGS 11CU5TR
SPRINGS IHDU^IR
StAHITIVILLE CO
J. CHARLES SAW
REPORTED 11S7 OFF-SITE TRANSFERS OF FOSMALDEHTDE I OOOD50000 1
SIC TOTM OTHfR OFF-Site
CODE LANDFILL INCINERATE HATER TB. OTHER DAYS/YH NOfES
(LB/YR) ILB/YR) tLB/YRI (LB/YPJ I16/YRI
14S5
29
202&
2063
2065
2063
2063
206 J
Z061
2099
2JU
2257
2257
JZftl
12*9
376
821
(,99
5065
2900
21000
ZOO
s
997
5000
-------
DOCUMENT
CONTROL t
113701
D09J333TX
06947J46A
0706570TX
Ol6ft7!»aAZ
02776670K
0 44066 30H
04595S9PA
86£9Qa3GA
Q6I0739NJ
0096503IN
03199S9HJ
0520M6M)
Q573B79CA
0394498HN
Q034938rW
00061CDHD
0777164W
Q036162HE
0041 *ttSC
«7fi6SAL
OS«90?SC
OSWIWI
S6SMW&A
C346249SC
034fl47aSC
«24794SC
OTM&2ZNC
FACItirV MAM£ SIC FUGITIVE
COOE 1LB/TR)
ATLAS MAHUFACTU
XL CORPORATION 40
COLE OISTfJIBUTl 499
CAPITOL CASTING 8510ft
THE ELECTRON CO 51
ITEM INDUSTRIES 999
CERTAIN TEED CO tt<)Q
ECC AMERICA AHS 1455
APEX CHEH1CAL C I® IS
80ROEN, IMC. - £026
MILK SPECIALTIE 2040 499
AMERICAN CRtSTA 2063 499
HOLLY SUGAfi COS 2043 499
AMERICAN CR¥STA 2SJ65 499
AHfRICAN CTtSTA 20&3 499
rtlHH-DAK FARMER 2063
AMERICAM CBYSTA 20*3 499
FfIC CORPOfiATlON 2099
1, MAMRICK IMC. 2211
WEST POIMT PEPP 2211
BCALTH-TEX IMC. 2257 499
HCALTH-TEX INC. 2257 499
THE BIBB CQMPAM 2261
3TOIMG3 INDU5TI? 2262 499
SPHINSS IWU3TR £262 499
etAMlffVILLi CO 2269
J. O-AfiLtS SAUH £262
STACK
1 IH/YH 1
5J
193^7
30798
1100
4^9
499
499
499
499
200G
3600
2809
499
499
12578
37000
2 OS 76
76«S
REPORTED 19fl/ RELEASES OF rORttALDEHTDE
WATER U IHJ
ILB/TB) (IB/TR)
( 000050000 1
D4YS/TB H/fVy TREATMENT MfllfS
IIB/TR i
7*00
499
H
A
A
999
-------
APPENDIX E
INDUSTRIAL PROCESS PROFILES AND OTHER COMPLETED STUDIES
E-1
-------
Issued: February 28, 1991
Revised:
Office of Research and Development (ORD)
Battelte, 1988. Final Report on Biosafety in Large-scale rDNA Processing Facilities.
Volumes 1,2,3,4, Cincinnati, OH: Office of Research and Development, U.S. Environ-
mental Protection Agency, Contract 68-03-3248.
Moskowitz PD, Kalb PD, Lee JC, and Fthenakis VM. 1987. Brookhaven National
Laboratory, An Environmental Source Book on the Photovollaics Industry. Final
Report. Washington, DC: Office of Research and Development, U.S. Environmental
Protection Agency. Interagency agreement 89931812-01-0.
Brown A, et al. 1985. Southern Research Institute. Predicting the Effectivenessof
Chemical Protective Clothing: Model and Test Method Development. Draft Report.
Washington, DC: Office of Research and Development, U.S. Environmental Protection
Agency, Contract 68-03-3113.
JACA Corp/ MITRE Corp, 1985. Preliminary Assessment of Predictive Techniques for
Unit Operations.
1, Filtration, drying, size reduction, mixing, sampling
2. Maintenance, cleanup, centrifugation, unloading
3. Extraction, flaking, agglomeration
4. Distillation, absorption, flotation, solids transfer
5. Decantaiion, adsorption
Draft report, Washington, DC: Office of Research and Development. U.S. Environ-
mental Protection Agency. Contracts 68-03-3186 and 68-01-6610.
JACA Corp/ MITRE Corp. 1985. Preliminary Assessment of Predictive Techniques for
Unit Operations.
1. Alkylation, halogenation, hydrohalogenation, polymerization
diazotization, and cleaning
2. Arnination, phosgenation, and nitration.
Draft report. Washington, DC: Office of Research and Development. U.S. Environ-
mental Protection Agency. Contracts 68-03-3186 and 68-01-6610.
Soklow R. 1984. S-Cubed. Paper Productionand Processing-Occupational Exposure
and_Enyjrpnmental Reje_a_se_Stud_y. Final report. Cincinnati, OH: Office of Research and
Development, U.S. Environmental Protection Agency. Contract 68-03-3015.
Blackwel! CD, Blackard Al, Stackhouse CW, and Alexander MW. 1983. TRW Energy
Development Group. JJMc^cieJly.^^jl J^^ Studv. Final
report. Washington, DC: Office of Research and Development, U.S. Environmental
Protection Agency. Contract 88-02-3174.
E-2
-------
Issued: February 28, 1991
Revised:
The following series of Industrial Process Profiles for Environmental JUjse has been
issued by ORD:
Chapter Title
1 Introduction
2 Oil and Gas Production
3 Petroleum Refining Industry
4 Carbon Black Industry
5 Basic Petrochemicals Industry
6 The Industrial Organic Chemicals Industry
7 Organic Dyes and Pigments Industry
8 Pesticides Industry
9 The Synthetic Rubber Industry
10 The Plastics and Resins Production Industry
10a The Plastics and Resins Processing Industry
I0b Plastic Additives
11 The Synthetic Fiber Industry
12 The Explosives Industry
13 Plasticizers Industry
14 (Not published)
15 Brine and Evaporate Chemicals Industry
16 The Fluorocarbon-Hydrogen Fluoride Industry
17 The Gypsum and Wallboard Industry
18 The Lime Industry
19 The Ctay Industry
20 The Mica Industry
21 The Cement Industry
22 The Phosphate Rock and Basic Fertilizer
Materials Industry
23 Sulfur, Sulfur Oxides and Sulfuric Acids
24 The Iron and Steel Industry
25 Primary Aluminum industry
26 Titanium Industry
27 The Primary Lead Industry
28 The Primary Zinc Industry
29 The Primary Copper Industry
30 The Electronic Component Manufacturing
Industry
Report Number
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
EPA/600/2-
77/023a
77/023b
77/023C
77/023CJ
77/0236
•77/023f
77/023g
77/023h
77/023I
85/085
85/086
85/087
77/Q23K
77/023!
77/023m
EPA/600/2-77/0230
EPA/600/2-77/023p
EPA/6QQ/2-77/Q23q
EPA/600/2-77/023r
EPA/600/2-77/023S
EPA/600/2-77/023t
EPA/600/2-77/023U
EPA/600/2-77/023V
EPA/600/2-77/023W
EPA/600/2-77/023X
EPA/600/2-77/023y
EPA/600/2-77/023Z
EPA/600/2-80-168
EPA/600/2-80-169
EPA/600/2-80-170
EPA/600/2-83-033
E-3
-------
Issued: February 28, 1991
Revised:
Office of Air Quality Planning and Standards
1. Locating and Estimat^Q:Air::,.ErnjssjQQ§ fr.Q.mjSpurces
Acrytonitrile
Carbon Tetrachlorsde
Chloroform
Ethyiene Dichforide
Formaldehyde
Nickel
Chromium
Manganese
Phosgene
Epichlorohydrin
Vinylidene chloride
Ethylene oxide
Chlorobenzenes
Polychbrinated
Biphenyl's (RGB's)
Polycyctic Organic
Matter (ROM's)
Benzene
Perchloroethytene and
Tricloroethylene
Chromium (Supplement)
1,3 Butadiene
EPA 450/4-
EPA 450/4-
EPA 450/4-
EPA 450/4'
EPA 450/4.
EPA 450/4-
EPA 450/4'
EPA 450/4
EPA 450/4
EPA 450/4
EPA 450/4
EPA 450/4
EPA 450/4
EPA 450/4
•84-007a
•84-GQ7b
•84-007C
•84-007d
•84-0076
•84-007f
•84-007g
•84-OG7h
-84-007i
•84-QG7J
-84-007K
•84-007!
-84-007m
-84-QQ7n
EPA 450/4-84-007p
EPA 450/4-84-007q
EPA 450/2-89-013
EPA 450/2-89-002
EPA 450/2-89-021
Mar 1984
Mar 1984
Mar 1984
Mar 1984
Mar 1984
Mar 1984
July 1984
Sept 1985
Sept 1985
Sept 1985
Sept 1985
Sept 1986
Sept 1986
May 1987
Sept 1987
Mar 1988
Mar 1988
Aug 1989
Dec 1989
These reports may be useful for estimating releases for existing chemicals.
They contain information concerning chemical/physical properties, overview of
production and uses, amount consumed per end use, major industrial source
categories, process descriptions and flow diagrams, potential emission points,
emission factors, number of sites and facility names, and references for source
sampling and analysis procedures.
2.
QraanicChernical Manufacturing Vol. 6 through 10: Selected Processes:
.Organic. Chemical Manufacturing Vol. 6: Selected Processes
PB 81-220550 EPA 450/3-80-0283 Dec 1980
Cyclohexane
Ctorabenzenes
Styrene
E-4
-------
Issued: February 28, 1991
Revised:
Cyclohexanol
Cyciohexanone
Maleic anhydride
Ethytbenzene
Capralacton
Adipic acid
.Organic Chemical Manufacturing VskJLJiiteMgQ' Processes
PB 81-220568 EPA 450/3-80-028b Dec 1980
Nitrobenzene
Toluene dissocyanate
Dimethyl terephthalate
Phenol
Aniline
Cumene
Crude terephthalic acid
Purified terephthalic acid
Acetone
Linear alkylbenzenes
Organic Chemical Manufacturing Vol. 8: Selected Processes
PB 81-220576 EPA 45Q/3-8Q-Q28C Dec 1980
Ethylene dichloride
Perchloroethylene by hydrocarbon chlorinolysis process
Fluorocarbons
Trichloroethylene
Chloromethanes by methane chlorination process
ChIoromethanes by methane! hydrochlorination and methyl chloride
chlorination process
Carbon tetrachloride
1,1,1-Trichloroethane
Perchloroethylene
Vinylidene chloride
Organic Chemical Manufacturing Vol. 9: Selected Processes
PB 81-220584 EPA 450/3-80-028d Dec 1980
Formaldehyde
Ethylene
Acetaldehyde
Methanol
Ethanol amines
E-5
-------
Issued: February 28, 1991
Revised:
Ethylene oxide
Vinyl acetate
Ethylene glycol
Organic Chemical Manufacturing Vol.10: Selected Processes
PB 81-220592 EPA 450/3-80-028e Dec 1980
Propyiene oxide
Glycerin and intermediates (ally! chloride, epichiorohydrin, acrolein, allyl alcohol)
Chloroprene
Formic acid
Waste sulfuric acid treatment for acid recovery
Acrylonitrile
Acetic anhydride
Acetic acid
Ethyl acetate
Methyl ethyl ketone
These reports contain industry and process descriptions, process flow
diagrams, emissions data, applicable control systems and impact analysis.
3. AjMg^Compilation of Air Pollutant Emission Factors
industries; External Conbustion Sources, Solid Waste Disposal, Internal Combustion
Engine Sources, Evaporative Loss Sources, Chemical Process Industry,
Food and Agricultural Industry, Metallurgical Industry, Mineral Products
industry, Petroleum Industry, Wood Products Industry, Miscellaneous
Sources.
The futl table of contents appears in AP-42. The chapters most useful to CEB are:
Chapter 4 Evaporation Loss Sources
4.1 Dry Cleaning
4.2 Surface Coating
Nonindustria! Surface Coating
General Industrial Surface Coating
Can Coating
Magnet Wire Coating
Other Metal Coating
Flat Wood Interior Panel Coating
Paper Coating
Fabric Coating
E-6
-------
Issued; February 28, 1991
Revised;
Automotive and Light-Duty Truck
Surface Coating
Pressure Sensitive Tape and Label Industry
Meta! Coil Surface Coating
Large Appliance Surface Coating
Metal Furniture Surface Coating
4.3 Storage of Organic Liquids
4.4 Transportation and marketing of Petroleum Liquids
4.5 Cutback Asphalt, Emulsified Asphalt, and Asphalt Cement
4.6 Solvent Degreasing
4.7 Waste Solvent Recovery
4.8 Tank and Drum Cleaning
4,9 Graphic Arts
4.10 Commercial/Consumer Solvent Use
4.11 Textile Fabric Printing
Chapter 5 Chemical Process Industry
5,1 Adipic Acid
5.2 Synthetic Ammonia
5,3 Carbon Black
5.4 Charcoal
5.5 Chior-Alkaii
5.6 Explosives
5.7 Hydrochloric Acid
5.8 Hydrofluoric Acid
5.9 Nitric Acid
5,10 Paint and Varnish
5.11 Phosphoric Acid
5.12 Phthatic Anhydride
5.13 Plastics
5.14 Printing Ink
5.15 Soap and Detergents
5.16 Sodium Carbonate
5.17 SuIfuricAcid
5.18 Sulfur Recovery
5.19 Synthetic Fibers
5,20 Synthetic Rubber
5.21 Terephthalic Acid
5.22 Lead Atkyl
5.23 Pharmaceuticals Production
5.24 Maleic Anhydride
E-7
-------
Issued: February 28, 1991
Revised:
4, New Source Performance Standard (40 CFR 601
Subpart Standards of Performance for -
D Fossil-Fuel Fired Steam Generators for Which Construction is
Commenced After August 17, 1971
Da Electric Utility Steam Generating Units for Which Construction is
Commenced After September 18, 1978
Db ' Industrial-Commerical-lnstitutional Steam Generating Units
DC Small Industrial-Commercial-institutional Steam Generating Units
E incinerators
F Portland Cement Plants
G Nitric Acid Plants
H Sulfuric Acid Plants
I Asphalt Concrete Plants
J Petroleum Refineries
K Storage Vessels for Petroleum Liquids for Which Construction, Recon-
struction, or Modification Commenced after June 11, 1973 and prior to
May 19, 1978
Ka Storage Vessels for Petroleum Liquids for Which Construction, Recon-
struction, or Modification
Kb Volatile Organic Liquid Storage Vessels (including Petroleum Liquid
Storage Vessels) for which Construction, Reconstruction, or Modification
Commenced after July 23, 1984
L Secondary Lead Smelters
M Secondary Brass and Bronze Ingot Production Plants
N Primary Emissions from Basic Oyxgen Process Furnaces for Which
Construction is Commenced After June 11, 1973
Na Secondary Emissions From Basic Oxygen Process Steetmaking Facilities
for Which Construction Commenced After January 20, 1983
O Sewage Treatment Plants
P Primary Copper Smelters
Q Primary Zinc Smelters
R Primary Lead Smelters
S Primary Aluminium Reduction Plants
J Phosphate Fertilizer Industry: Wet-Process Phosphoric Acid Plants
U Phosphate Fertilizer Industry: Superphosphoric Acid Plants
V Phosphate Fertilizer Industry: Diammonium Phosphate Plants
W Phosphate Fertilizer Industry: Triple Superphosphate Plants
X Phosphate Fertilizer Industry: Granular Triple Superphosphate Storage
Facilities
Y Coal Preparation Plants
Z Ferroalloy Production Facilities
AA Steel Plants: Electric Arc Furnaces
E-8
-------
Issued: February 28, 1991
Revised;
AAa Steel Plants: Electric Arc Furnaces and Argon-Oxygen Decarburization
Vessels Constructed After August 17, 1983
BB Kraft Pufp Mills
CC Glass Manufacturing Plants
DD Grain Elevators
EE Surface Coating of Metal Furniture
FF (Reserved)
GG Stationary Gas Turbines
HH Lime Manufacturing Plants
KK Lead-Acid Battery Manufacturing Plants
LL Metallic Mineral Processing Plants
MM Automobile and Light-Duty Truck Surface Coating Operations
NN Phosphate Rock Plants
PP Ammonium Sulfale Manufacture
OO (Reserved)
QQ Graphic Arts Industry: Publication Rotogravure Printing
RR Pressure Sensitive Tape and Label Surface Coating Operations
SS Industrial Surface Coating: Large Applicances
TT Metal Coil Surface Coating
UU Asphalt Processing and Asphalt Roofing Manufacture
W Equipment Leaks of VOC in Synthetic Organic Chemicals Manufacturing
Industry
WW Beverage Can Surface Coating Industry
XX Bulk Gasoline Terminals
AAA New Residential Wood Heaters
B8B Rubber Tire Manufacturing Industry
CCC (Reserved)
ODD (Reserved)
EEE (Reserved)
FFF Flexible Vinyl and Urethane Coating and Printing
GGG Equipment Leaks of VOC in Petroleum Refineries
HHH Synthetic Fiber Production Facilities
Sit Volatile Organic Compound Emissions from the Synthetic Organic
Chemical Manufacturing Industry (SOCMi) Air Oxidation Unit Processes
JJJ Petroleum Dry Cleaners
KKK Equipment Leaks of VOC From Onshore Natural Gas Processing Plants
LLL Onshore Natural Gas Processing: S02 Emissions
MMM (Revised)
NNN Volatile Organic Compound Emissions from Synthetic Organic Chemical
Manufacturing Industry Distillation Operations
OOO Nonmetallic Mineral Processing Plants
PPP Wool Fiberglass Insulation Manufacturing Plants
QQQ VOC Emissions from Petroleum Refinery Wastewater Systems
RRR (Reserved)
E-9
-------
Issued: February 28, 1991
Revised:
SSS Magnetic Tape Coating Facilities
TTT Industrial Surface Coating: Surface Coating of Ptastic Parts for Business
Machines
UUU (Reserved)
VW Polymeric Coating of Supporting Substrates Facilities
E-10
-------
Issued: February 28, 1991
Revised:
Office of Toxic Substances (Contractor Reports)
Pace Laboratories. 1989, Evaporation Rates, of Voiatile Liquids. Final Report. Second
Edition. Washington, DC: Office of Toxic Substanes, U.S.Environmental Protecton
Agency. Contract 6S-D8-0112.
PEl Associates. 1939. .Respirator and Engineering Control Costs. Washington, D.C.:
Office of Toxic Substances, US EPA. Contract 68-02-4248.
PEl Associates. 1988. Effectiveness of Local Exhaust Ventilation for Drum-filling
Operations. Washington, D.C.: Office of Toxic Substances, U.S. Environmental
Protection Agency. Contract 68-02-4248.
PEl Associates. 1988. Releases During.CJejning of Equipment. Washington, D.C.:
Office of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-02-
4248.
PEl Associates, 1987. Exposure Assessment of Acrylates/Methacrylates in Radiation-
cured Applications. Washington, D.C.: Office of Toxic Substances, U.S. Environ-
mental Protection Agency. Contract 68-02-4248.
MRI. 1986. Occupational Exposures from Bagging and Drumming. Finaf Report. (2
volumes). Washington, DC: Office of Toxic Substances, U.S. Environmental Protection
Agency. Contract 68-02-3938.
Myers WR. 1986. NIOSH. Strategy for Recommending Respirators for_C_o_ntroj_oJ
Exposures to Substances Undergoing Premanufacturjng Notice Review. Washington,
DC: Office of Toxic Substances, U.S. Environmental Protection Agency. Contract EPA
DW 75932235.
PEl Associates. 1986. Occupational Exposure andJ-nyjronmental Release Assessment
joLAcryjates / M eth aery I ate s. Washington, DC: Office of Toxic Substances, U.S.
Environmental Protection Agency. Contract 68-02-4248.
PEl Associates. 1986. Occupational Exposure and Environmental Reiease Assessment
of Diisocvanates. Washington, DC: Office of Toxic Substances, U.S. Environmental
Protection Agency. Contract 68-02-4248.
PEl Associates. 1986. _CQsLoi Selected Engineering Controls. Washington, DC: Office
of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-02-4248.
E-11
-------
Issued: February 28, 1991
Revised:
PEI Associates. 1986. Use of Oil Separators in Drum Reconditioning and
TransQOrtation Vessel Cleaning Facilities. Washington, DC: Office of Toxic Substances,
U.S. Environmental Protection Agency. Contract 68-02-4248,
PEI Associates, 1990. Process Flow Diagram Users Manual. Washington B.C.:
Office of Toxic Substances, U.S. Environmental Protection Agency, Contract No. 69-
D8-0112.
Development Planning & Research Associates,Inc. 1985, Generic Assessment of the
Electronics. Industry. Washington, DC: Office of Toxic Substances, U.S. Environmental
Protection Agency. Contract 68-02-3952.
Versar, 1985. An_Oyeryjew of Carbon Adsorption. Washington, DC: Office of Toxic
Substances, U.S. Environmental Protection Agency. Contract 68-02-3968.
Bomberger DC, Brauman SK, and Podoil RT. 1984. Southern Research institute.
Studies to Support PMN Review: Effectiveness of Protective Gloves. Washington,
DC:Office of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-01-
6016.
MITRE, 1984. information on the Loading and Unloading of Chemicals Under Nitrogen
Blanketing. Washington, DC: Office of Toxic Substances, U.S. Environmental Protec-
tion Agency. Contract 68-01-6610.
Versar. 1984. _Bcp_osjjre_Assessment for Retention of CheroMaLyg^uids^n^Handg.
Washington, DC: Office of Toxic Substances, U.S. Environmental Protection Agency,
Contract 68-01-6271,
Dryden FE and Keifer LC. 1983. Walk, Haydel & Associates, Inc. Industriai Process
Profiles to Support PMN Review: Oil Fields Chemicals. Final report. Washington, DC:
Office of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-01-
6065.
Zak FJ, Melton R, Byeriy J, and Keifer LC. 1983. Walk, Haydel & Associates, Inc.
industrial Process Profiles to Support PMN Review: Lube and Fuel Additives. Final
report. Washington, DC: Office of Toxic Substances, U.S. Environmental Protection
Agency. Contract 68-01-6065.
Gikis B, Fowler, Strauss E, and Boughton R. 1983. SRI Int. Industrial Process Profiles
to Support PMN Review: Printing Inks. Final report. Washington, DC: Office of Toxic
Substances, U.S. Environmental Protection Agency. Contract 68-01-6010.
E-12
-------
issued: February 28, 1991
Revised:
Westbrook EJ, Schneider P, and Keifer LC, 1983. Walk, Hayde! & Associates, Inc.
.Industrial Process Profiles to Support PMN Review^MjM Treatment ChemicgSs. Final
report. Washington, DC: Office of Toxic Substances, U.S. Environmental Protection
Agency. Contract 68-01-6065,
Gikis B, Fowler, Connolly E, and Boughton R. 1983. SRI Int. industrial Process Profiles
to Support PMNReyJewi Paints. Varnishes, and Coatings. Final report. Washington,
DC: Office of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-
01-6016.
Bomberger D, Ferguson A, Fowler D, et al. 1983. SRI Int. Profile of Release and
Exposure for Chemicals Used in Processing Qrgs_ajrj^_MiQgrgi£. Final report.
Washington, DC: Office of Toxic Substances, U.S. Environmental Protection Agency.
Contract 68-01-6016.
Bomberger D, Boughton R, Endiich R, et a!. 1983, SRI Int. !ndustrial Process PfofilejQ
Support PMN Review: Filling of Drums and Bags. Draft Report. Washington, DC: Office
of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-01-6016.
Rand J and Dennis R. 1984. GCA Corporation. Textiie and Leather pyejng_gnoM3ye
Manufactyring, and Processing: Occupational Exposures and Environmental Releases
.of Dyestuffs. Finai report (in three volumes). Washington, DC: Office of Toxic
Substances, U.S. Environmental Protection Agency. Contract 68-02-3165.
Keifer L, Dryden FE and Seifert M. 1983. Walk, Haydel & Associates, Inc. Industrial
£ro_£ess_P_rofiles to Support PMN Review: Water IregimenjJ^ejTTigaJs. Final report.
Washington, DC: Office of Toxic Substances, U.S. Environmental Protection Agency.
Contract 68-01-6065.
Berman DW. 1982. Walk, Clement Associates. Methods for Estimating Workplace
Exposure to PMN Substances. Washington, DC: Office of Toxic Substances, U.S.
Environmental Protection Agency. Contract 68-01-6065.
Waik, Haydel & Associates. 1981. .generLc Polymer Study. Final report. Washington,
DC: Office of Toxic Substances, U.S. Environmental Protection Agency. Contract 68-
01-6065.
Clement Associates, 1981. Exposure^najy5isj3fJbe_Fla^ Industry. Rnal
report. Washington, DC: Office of Toxic Substances, U.S. Environmental Protection
Agency. Contract 68-01-6065.
Clement Associates. 1981. Mathematical Models fQiJEstimating Workplace
Concentration LeveJsLA Literature Review. Washington, DC: Office of Toxic
Substances, U.S. Environmental Protection Agency. Contract 68-01-6065.
E-13
-------
Issued; February 28, 1991
Revised:
Office of Toxic Substances (In-house Reports)
Vorbach J. Undated. CEB Rejejrcli Project^ Effluent Guideline I.QfQrmatj,QnL__P^rtjt\-
Tank Truck Cleafrir^J3perations. Washington, DC: Office of Toxic Substances, U.S.
Environmental Protection Agency.
Vorbach J. Undated. CEB Research PjQJegLJEffluent .Guideline Information: PartJk
Content of Development Documents Produced by US EPA Office of Water. Industrial
Technology Division. Washington, DC: Office of Toxic Substances, U.S. Environmental
Protection Agency.
Vorbach J. Undated, .CJEB_Research. Project: Effluent Guideline Information: Part C-
Abstract of Final Development Document for Effluent Urnitations, Guidelines, and
Standards for LeatherJLaoning & Finishing Point Source Category. Washington, DC:
Office of Toxic Substances, U.S. Environmental Protection Agency.
Netson, 1990. Prgcess Flow Diagrams.
Jackson E. 1989. Polyetectrolytes- Their Application and Estimation of Releases.
Washington, DC: Office of Toxic Substances, U.S. Environmental Protection Agency.
Reilfy B. 1989. Industrial .H^3.nd^us_^Va^teJn^inM3tiQji. Washington, DC: Office of
Toxic Substances, U.S. Environmental Protection Agency.
Franklin K. 1988. Memorandum to CEB Staff from Kathy Franklin concerning Office of
..... Reference List. Dated April 1, 1988.
Heath G. 1988. Memorandum to CEB Staff from George Heath concerning Textile
Drug Room Monitoring Study (TDRMS). Assessment of Workplace Dust Inhalation
Exposures. Dated February 17, 1988.
Wong K. 1988, Memorandum to CEB Staff from Kin Wong concerning interpretation of
jsSon-lsplated Intermediates. Dated October 4, 1988.
Reilly B. 1988. Catalogue of Databases. Washington, DC: Office of Toxic Substances,
U.S. Environmental Protection Agency.
Kumar V. 1987. Drilling Fluids: Environmental Release Analysis. Washington, DC:
Office of Toxic Substances, U.S. Environmental Protection Agency.
Macek G. 1987. CEB Research Project: Engineering Standards. Washington, DC:
Office of Toxic Substances, U.S. Environmental Protection Agency.
E-14
-------
Issued: February 28, 1991
Revised:
Rodriguez V, 1987. Generic Engineering Assessment: Spray Coating - Occupational
Exposure and Environmentai Release (Revised October , 1987), Washington, DC:
Office of Toxic Substances, U.S. Environmental Protection Agency.
Vorbach J, 1987, Generic Engineering Assessment: Leather Dyeing-
Exposure and Environmental Release. Washington, DC: Office of Toxic Substances,
U.S. Environmental Protection Agency.
Heath G. 1986. Memorandum to Craig Matthiessen and Larry Longanecker from
George Heath concerning Generic Exposure Assessment- The Dying and Printing of
Textile Fibers. Dated November 17, 1986.
Franklin K. 1986. Memorandum to CEB Staff from Kathy Franklin concerning Air
ErQQram Information - CEB Research Project. Dated March 28, 1986.
Wong K. 1985. Disposal of Metalworking Fluids. Washington, DC: Office of Toxic
Substances, U.S. Environmental Protection Agency.
Chafmon M. 1984. Carbon Adsorption Report. Washington, DC: Office of Toxic
Substances, U.S. Environmental Protection Agency.
Heath G, 1984. The Dyeing and Printing of Textile Fibers Relative to Worker Exposure
and Environmental Release. Washington, DC: Office of Toxic Substances, U.S.
Environmental Protection Agency.
Wong K. 1984. Exposure to N-nitrosodiethanolamin_ajQjMa.chine Shops. Washington,
DC: Office of Toxic Substances, U.S. Environmental Protection Agency.
Wong K. 1984. Exposure to N-nitrosodiethanolamine in
. Washington, DC: Office of Toxic Substances, U.S. Environmental
Protection Agency.
Wong K, 1983. Zjnc_DiaJkyldrthiophosphates- Industrial Exposure ..... andRele.ase
Washington, DC: Office of Toxic Substances, U.S. Environmental
Protection Agency.
E-15
-------
Issued: February 28, 1991
Revised;
Title III Section 313 Release Reporting Guidance Estimating Releases from:
Monofiiament Fiber
Manufacture
Printing Operations
Electrodeposition of
Organic Coatings
Spray Application of
Organic Coatings
Semiconductor Manufacture
Formulating Aqueous Solutions
Electroplating Operations
Textile Dyeing
Presswood and Laminated
Wood Products
Roller, Knife, and Gravure
Coating Operations
Paper and Paperboard
Production
Leather Tanning and
Finishing Processes
Wood Preserving
Rubber Production and
Compounding
Food Processers
EPA 560/4-88-0043 Jan 1988
EPA 560/4-88-004b
EPA 56Q/4-88-004C
EPA 560/4-88-0046
EPA 560/4-88-004!
EPA 56Q/4-88-Q04g
EPA 560/4-88-004h
EPA 560/4-88-004i
EPA 560/4-88-004J
EPA 560/4-88-004k
EPA 56Q/4-88-0Q4I
EPA 560/4-88-004p
EPA 56Q/4-88-OQ4q
Jan 1988
Jan 1988
EPA 560/4-88-004d Jan 1988
Jan 1988
Mar 1988
Jan 1988
Feb 1988
Mar 1988
Feb 1988
Feb 1988
Feb 1988
Feb 1988
Mar 1988
EPA 560/4-90-014 June 1990
These reports contain brief descriptions of the industry, identify potential release
points, and model calcualtions for estimating releases.
E-16
-------
APPENDIX F
SUMMARY OF GUIDELINES FOR STATISTICAL ANALYSIS OF
OCCUPATIONAL EXPOSURE DATA
-------
The following guidelines are a summary of the report, "Guidelines for
Statistical Analysis of Occupational Exposure Data (CEB 1989)". This
reference should be consulted for the detailed procedure before any
occupational exposure monitoring data is analyzed.
The procedure is designed to be used by CEB engineers with the assis-
tance of industrial hygienists and statisticians. The procedures provide a
systematic methodology for performing an occupational exposure assessment
based on the types of data which are most commonly available for such analysis,
Figures 1, 2 and 3 provide a flow diagram of the procedure.
The methodology is based en dividing the data into three broad types of
occupational exposure data:
0 Type 1 data consist, of measurements For which all important
variables are known. The data consist of studies that contain
individual measurements and include all backup and ancillary
information.
0 Type 2 data consist of measurements where important variables are
not known but for which assumptions can be made for their estimation.
The data consist of individual monitoring measurements, but backup
and ancillary information is inconsistent,
0 Type 3 data consist of measurement summaries, anecdotal data, or
other data for which the important variables are not known and
cannot be estimated. Individual monitoring measurements are
typically not available.
Once the data has been classified into one of these three types, the
data types are only combined as specifically described in the procedure.
Within each data type, the variables that are potentially important to worker
exposure are identified and the data categorized by three variables. The
traditional categorization of data by the industrial hygienist or engineer is
supplemented by statistical analysis of the categorizations. The goal of the
procedure to the combination of similar categories producing larger data sets
for analysis. Because the size of the data set being analyzed has a large
effect or the confidence that can be placed in the analysis, this procedure
F-l
-------
t
ro
NJOSH
Other fodtrml *g#nd#&fofKc#
ST»f» •flienetoj
Trade nmoflaftan.i
ffaws Jn Jruteffy
©
t ®
Oflftiadala
rveeds
fe*s
— ^
ktenlly
exposure
variaWas
®
Wartly
uncBrtBWies
and biases
Rank variables
In wiporrjiiri€$
©
S»pof»*9 Typ*
3 data
^
Treal
d.
r ®
TypeS
iia
— >•
C?)
Croala
preliminary
B>tp(55ure malri*
-^
PreKminwy
exposure
malnx
Non-stafe(lc8l
rcp
-------
Definition of
data needs
from Step 3
Preliminary
exposure
matrix
Check for
oonstelency and
reasonabten ess
Collect
addilionat
missing
information
Estimate
additional
rrtisstmg
Informaliort
Assess ability to
m««rt oset
needs
Yes
Can data
m«e! user
needs?
con»l»l«Tf untta
Treat non-
detected valuo*
Tfeat
unoBrtairrtles,
assurr^Mions and
biases
Sepwala Into
Typa 1 ctafa and
Type ? data
Figure 2. Ftaw diagram for creation of a compteted exposure matrix.
-------
Calculate
descriptive
statistics lor each
primary category
•*^
• ^"
Combtna primary
categories
*te
Present results
Figure 3, Ftow diagram for the statistical analysis of Type 1 and Type 2 data.
-------
allov.s e higher confidence to he placed in the descriptive statistics produced
by the analysis,
When analyzing occupational exposure data, the CEB engineer should be
aware of the variability in most monitoring data. Studies of occupational
exposure are rarely found which are developed based on a statistical approach
to providing representative information for an individual facility; it is
even less likely to find a study which represents a particular industry
subsecto*- or group of facilities. While random sampling is preferred, "worst-
case sampling" during a 1- to 3- day sampling campaign is common industrial
hygiene practice for compliance with regulatory standards.
Even in statistically-selected well-done studies, there may be high
variability in the characterization of worker exposure. Measurements at a
plant made over a period of no more than a few days may be all that are
available to characterize exposures over an entire year or a period of years.
Seasonal variability, interday and intraday variability, and changes in the
process or worker activities can cause the exposure to vary from that measured
on a single day. Temperature changes can affect evaporation rates, and
seasonal changes in natural ventilation affect exposure. Sampling methods
and time periods can also vary. Seldom can all these variables be measured
and accounted for. However, if important variables are identified and quant-
ified, it is hoped the influence of less important variables on the
overall measure of central tendency will be minimized. Variables that may
not be obvious may also arise between plants in the same industry category.
Variables such as the age of the plant, the age of the control equipment,
whether the plant, is in e volatile organic compound (VOC) non-attainment
area, and operation and maintenance (O&M) practices at the plant should be
investigated.
When analyzing sample data, it is important to understand the sources of
variation in exposure sample results that combine to create the observed
variability (Patty 1981). The size of the variations is a function of both
the exposure levels and the measurement method. Both random and systematic
errors should be considered.
Random variations in workplace exposure levels can result in intraday
variations, interday variations, or variations in exposures of different
workers within a job group or occupational category (Patty 1981), Variability
F-5
-------
in the measurement procedure can be caused by random changes In pump flow
rate, collection efficiency, or desorption efficiency. It is important to
realize that random variation in real workplace exposure levels will usually
exceed measurement procedure variation by a substantial amount, often by
factors of 10 or 20 (Patty 1981),
Systematic variations in the determinant variables affecting workplace
exposure levels will lead to systematic shifts in the exposure results.
Variability in worker exposure levels include changes in worker job opera-
tions during a work shift or over several days, production process changes,
or control system changes. Systematic errors in the measurement procedure
can result form mistakes in pump calibration, use of sampling devices at
temperature or altitude substantially different from calibration
conditions, physical or chemical interferences, samples degradation during
storage, internal laboratory errors, and interlaboratory errors (Patty 1981).
These errors may be identified and their effects minimized with use of
quality assurance programs.
It is also important to ascertain the objectives of the monitoring study
to identify potential biases in the data. For example, if the objective was
to sample only we!1-controlled facilities, then the results would probably
not represent the exposure in the industry as a whole. If the monitoring
resulted from worker complaints, then exposures may not represent typical
exposures. If the monitoring was conducted tc evaluate engineering controls
or as a preliminary screening of exposure, the results may not represent
actual employee exposure. It is important that all potential variables be
identified and evaluated.
Once the data have been analyzed, the results must be presented clearly
to allow the user to properly interpret the results. All assumptions and
uncertainties associated with the data must be clearly identified. The
descriptive statistics should be accompanied by graphic presentation of the
data such as probability plots or box-and-whisker plots, where possible.
Finally, the original data used in the analysis should be presented in tabu-
lar form to allow the user to calculate additional statistics when necessary.
F-6
-------
REFERENCES
CER 1989. Guidelines for Statistical Analysis of Occupational Exposure Data,
Draft Final. Prepared for the U.S. Environmental Protection Agency, Chemical
Engineering Branch, August 19, 1989,
Patty 1981. Patty, F. A, Patty's Industrial Hygiene and Toxicology, 3rd
Edition. Volumes 1 through 3, General Principles, Statistical Design and
Data Analysis Requirements. John Wiley and Sons, New York, NY, 1981.
F-?
-------
APPENDIX G
DERIVATION OF FORMULAS FOR CALCULATION OF
WORKPLACE AIRBORNE CONCENTRATION
S-l
-------
APPENDIX 6
FORKiiLAS DrRnfED BY, CLEMENT ASSOCIATES FOP CALCULATION OF
1. Derivation of Genejra__tipn Rate:
The simplest form for describing evaporation of a liquid into stagnant
air is (Thibodeaux 1979) ;
M K A (Pc - P)
G =
R
where:
G = vapor generation rate, g/sec
M = molecular weight, g/g role
K = mass transfer coefficient, cm/sec
A = surface area of the liquid, cm2
Pc= equilibrium vapor pressure, atm
P = actual partial pressure in the gas phase, atrr
R = universal gas constant, 82.05 ere3 atm/mol K
T( = liquid temperature, °K
For most cases Pc (equilibrium vapor pressure) is much greater than P (actual
partial pressure in the gas phase^ and the equation reduces to:
G = l Equation 6-1
R TL
This form is used for determining the vapor generation rate for sampling,
cleaning and maintenance.
When liquids are transferred, vapors are also generated from the
displacement of saturated vapors in the vessel while the liquid is filling
it. The volumetric rate at which saturated air is displaced may be expressed
as V r. If this is put into the same form as above to describe the rate at
which material enters the gas phase during filling, then:
H V P° r M K A P°
3600 R TL R TL Equation G-?
where:
V • fill volume, cm3
r = duration of filling operation
G-2
-------
The equation accounts for both evaporation and displacement generation.
Usu?.Tly evaporation is much lower than displacement end can be ignored during
filling.
The amount of vapors generated by saturation can be affected by the
method In which the material is loaded. Some vessels are loaded by allowing
the liquid to drop or splash into the vessel and some are loaded by pumping
the liquid in under the surface of the liquid present in the tank. To adjust
the generation rate depending on the method of filling, a factor, f is added
to the equation:
f v v r P
G = .LJ_L_r_I Equation G-3
3600 R TL
where f = 1.0 for splash and 0.5 for subsurface,
2. Derivation of Airborne Concentration Formula:
The most common model for workplace contaminant calculation describes an
overall mass balance of contaminant as it is generated end removed from an
enclosed space:
V ~ = G - kCQ Equation 6-4
where:
C = airborne contamination,
dC/dt = the change in concentration over time,
G = generation rate of the chemical,
V = room volume,
K = mixing factor,
Q = ventilation rate.
If the generation rate (G) and ventilation rate (Q) are assumed to be
constant, this equation may be integrated and reduced by assuming that the
concentration remains the same over long periods of time (steady state):
C = G/kQ Equation G-5
To obtain C in ppm, from Q in cubic feet per minute, and Gin grains per
second, multiply through by R, M, and T:
fi-3
-------
- G fg/s} x 82 .OS f a tin-cm3/g-mole Kelvin) x T (Kelvin) x 106
k x q (ft3/min) x (min/60 s) x 2.832 x ID4 fee/ft3) x M (g/g-mole)
C - U x 105 T G Equation 3
iRi Section IV A)
R-4
-------
APPENDIX H
CHART OF BODY AREAS AND ESTIMATION OF SKfN AREA
H-l
-------
TABLE H-1. ESTIMATION OF THE DISTRIBUTION OF SKIN AREA ON AN
EXPOSED BODY {ASSUMED 1,9 m2 TOTAL AREA}
Body Part
Anatomic Model
Area
Head
Neck
Upper arms
Forearms
Hands
Shoulder
Chest
Back
Hips
Thighs
Calves
Feet
Fingers
5.7
1,2
9,7
6,7
6,9
6,8
8.0
8.0
9.1
18.0
13.5
6,4
3.3
0.11
0.023
0.18
0.13
0.13(includes fingers)
0,13
0.15
0.15
0.17
0,34
0.26
0,12
0.083
Source: Anatomic Model - 1976,
r^
l F°f«- • '-'SP*? !
,
*'*
8J% ' 97%
X p*TG*niU« enufi
fct - (73 ew
S* - T.S3 ml
N«* ',2%
~N I
4 I i
N r
• t N»J
,' *
/, - A
|'i •(')
\( li
i
ShauJsw 4.IX
JL
FIGURE H-1, HUMAN DERMAL SURFACE AREA MODELS
3 Derived from mensuration formula and anatomic dimensions. Each percentage
corresponds to the proportion of the total surface area(SA) for each location.
Source: Popendorff and Letting well, 1982.
H-2
-------
APPENDIX I
OTHER FACTORS TO BE CONSIDERED IN RESPIRATOR SELECTION
1-1
-------
APPENDIX
OTHER FACTORS TO BE CONSIDERED IK RESPIRATOR
In the assessment of workplace exposure for ail new chemicals
ana often fot existing chemicals, no OSHA Permissible Exposure
Limits (PEL's) or ACGIH Threshold Limit Values (TLV's) have been
established. Since information about the workplace is limited,
the presence of other contaminants, such as dusts, solvents,
etc., is taiely known and questions of oxygen deficiency and
confined spaces cannot be addressed. Similarily, sufficient
information on then workers' activities and workplace
environment is not available to fully consider the worker's
ability to wear a device, its comfort, wear time, and other site-
specific factors. These are all important factors in selecting a
specific respirator{s) for the situation of concern. Thus, the
review of respiratory protection alternatives by CEB should not
focus on individual respirators but rather on the degree of
protection assigned to various classes ot respirators, judgements
about likely conditions of use of the respirator, an estimated
potential workplace concentration and, the physical and chemical
properties of the contaminant.
Conditions_j:>f Respirator Use_
Conditions of use may be categorized as routine ot non-
routine. Non-routine conditions include potential escape or
emergency situations, confined space entry, oxygen deficient
atmospheres, and immediately dangerous to life and health (IDLH}
T-Z
-------
atsmosphtit es . Routine conditions include no trial ever, ycay tasks
white there is a potential foi inhalation exposure, such as
sampling, mate i3.il tt a listers and first b teaks into closed
equipment for maintenance. Alt no ugh CEB engineers generally
focus on toutine exposures, potential contact in non-toutine
situations should be kept in mind.
Allboi ne Concentration
For all new chemicals and often tot existing chemicals, no
actual measurements of airborne concentrations of the specific
contaminant in the workplace ate available. Thus, estimated
workplace concentrations are usually based on data for simiiat
compounas in similar uses 01 are calculated based on methods
described elsewhere in this Manual.
When a PEL has been established Cor chemical ot concetn, a
surrogate chemical, ot a chemical used in the same workplace
setting, it is often usea as the reasonable worst case estimate
unless information is available demonstrating exposure levels to
De typically higher 01 lower. When considering the application
of a respirator, credit _c a_n nojt be given for reducing exposure
levels below the PEL unless it is demonstrated that the PEL can
be met without the use of a respirator on a continuous basis. If
exposure levels are predicted below the PEL (with or without a
respirator), the assessment must indicate that without additional
regulatory controls, the enforcsble level of exposure is the PEL.
Respirator Per fotrm
Research has shown that the degree of protection afforded by
a particular respiratoiy protective device depends primarily upon
the type of device and its fit. Many types of respiratory
1-3
-------
protective devices a:e available ranging from inexpensive filter
masks to Lather costly seif-cor.ta i ned breathing apparatus
-------
Ft opei SeJectiQ_n_ ana Use ofJ
Respirators must, be properly maintained and properly used in
QLciei to provide the protection associated with the respirator.
CE3 has no control over whether a respirator is properly used,
environmental conditions, etc. The industrial hygienist (or
other person) in charge of the respirator progiam at the site
must consider these factors and others in selecting the
appropriate respirator for an individual. OSHA requites that a
comprehensive respiratory protection program be implemented when
iespiia tots ate worn. The respirator that is least disruptive to
the task but provides the best protection should be selected to
enhance the probability that 1) the respirator will be worn and
that 2) it will be worn properly. in selecting a respirator, in
addition to limitations imposed by the tespirator itself, one
must consider the operation or process, environmental conditions
and work area characteristics, the materials used and the
worker's duties and actions. Distress associated with the work
environment is accentuated by wearing a respirator: vision is
restricted, breathing is mote difficult, equipment may be
cumbersome and restrict movement, and wearing the respirator may
add to the adverse effects of temperature extremes.
Er\ v i r o nm e n t a 1 _Coj>d_i^t_ic>_r)s_
High temperature environments are stressful and wearing a
respirator applies additional stress on the worker. In selecting
a respirator to be used in a high temperature environment,, it is
important to reduce the volume of expired air near the worker's
breathing zone. In addition, the respitator should be light-
weight and have & low resistance to breathing. A supplied-air
1-5
-------
lespiratoi may be equipped with a vo: tex tube to cool the air-
supplied to the facepiece. A powered air-purifying respirator or
half mask negative-pressure respirator can also be used in hot
env i torments .
Cool tempeiatutes may cause fogging on full facepiece
respirators/ valve sticking and rubber stiffness that prevents a
good faceseal . A nose-cup installed in a full facepiece
respirator will eliminate facepiece fogging. Coating the inside
surface of the lens may prevent fogging as well. A vortex tube
may be used with some respirators to warm the air-supplied to the
facepiece.
Human Facto t s C o n s _i oe r _a_tIP n s
Powered air-pur ifying respirators (PAPRs) have good
application in many industries because the worker has total
mobility and is provided with a stream of air to the breathing
zone. PAPRs are lighter in weight (less than 10 Ibs) than
supplled-air respirators or self contained breathing apparatus
(SCBA). The battery pack roust be fully charged to provide the
protection afforded by the respirators associated with the
respirator. With higher work rates, the protection may be
reduced, depending on the type of PAPR worn. These respirators
are advertised to be positive pressure devices. Recent studies
have found that two helmet PAPRs could not maintain positive
pressure inside the faceraask all or even part of the time.
However, the half-mask device tested came close to being a
positive pressure device, and was able to maintain a positive
pressure for 100% of the time at 7cfm as opposed to 50% of the
time at 3cfm. It has good applicability to abrasive blasting,
1-6
-------
tound:. it?s, grinding, pesticide spraying, etc. OS HA nas spc-citieci
that PAPRs rnuBL be given to asbestos wot nets if requested (29 CFR
1910.1001) although the EPA ana M1OS1I do not recommend PAPRs in
abatement woik due to the protection factors associated with
them. PAPRs may be equipped with a facepiece, hood, or helmet.
Negative pressure ait-putifying respirators generally weigh
less than 2 IDs, and also ofte: enhanced mobility. However,
because these operate under negative pressure, the wearer must
overcome the negative pressure in the device while bteathing in,
which may cause some d iscornf ot t, These respirators ate simple
devices and Can be readily used by workers that have been
pic3pei.lv trained and tit-tested.
Suppiied-air respirators enable longer work periods than
SCBAs and are less .bulky. Supplied air respirators weigh less
than 5 Ibs. However, tne airline impairs worker mobility and
tequites that a worker retrace his steps when leaving an area,
The airline is vulnerable to being punctured. Airlines should be
kept as short as possible when in use. The longest length of
airline approved is 300 feet. With increasing length of the
airline, the approved airflow to the facepiece may be decreased.
SCBAs are probably the most cumbersome respirator to wear.
The SCBA with tank can weigh up to 35 Ibs and is limited to a
maximum of 30 minutes of breathing air per tank. Of course,
under heavy work rates, a full tank of ait will be used at a much
faster rate. SCBAs are approved for escape only and for entry
into and escape from a hazaidous atmosphere. All breathing gas
cylinders must meet DOT requirements and supply Grade D breathing-
ait or better. SCBAs are the most complex respirator in use
1-7
-------
today, Ttaining in respirator use is essential
1-8
-------
APPENDIX J
STANDARD LANGUAGE FOR 5{e) ORDERS AND SNURS
-------
MAR -3 !983
''• ( e ) Orders and SNUBS
Fro~: Cathy Fehrenbacher, Industrial Hygier.ist
Chemical Engineering Branch (TS-779)
To; Paul Matthai, Section Chief
Fremanufacture Notification Branch (TS-794)
Attached is a revised guideline for using the respiratory
protection standard language when writing 5{e} Orders and SN'JRs.
This revision should replace the earlier version as it
incorporates the numbering system used in the revised 5{e)
language and the SWUR proposal. In addition, I have included a
statement that a CEB Industrial Hygienist (nyself} should review
all 5(e) Orders and SNURs, as a means of verifying that the
proper respiratory protection language is used. 1 have given
this guideline to CEB Engineers for their use as well, I do not
anticipate another revision unless the standard language is
revised.
As always, I am available to answer any questions regarding
respiratory protection or other industrial hygiene concerns,
cc: Paul Quillen
Roy Seidenstein
Crate Spears
0-2
-------
th-
the
STANDARD LANGUAGE FOR 5(e)BORDERS ..AJ?D SKURS
^ EFA generally has insufficient information to focus
inaividusl respirators for a particular scenario, but can f.^cus
the degree of protection assigned to various clac
respirators, judgments about likely conditions of us/*
respirator, an estimated potential workplace concentration
physical and chemical properties of the contaminant. ' A "ore
detailed ciscussion can be found in the CEB Engineering Manual.
The standard language presented below reflects the use of the
NIOSH Assigned Protection Factor (APF) values and describes th«
general class(es) of respiratory protection that should be
recommended m 5(e) or other Orders and in SNURs. Mcre pr^tec-< "e
respirators are included in each of the lesser protect*
categories. For example, for particulate exposure where a
protection factor of 50 or greater is needed, three {3 \ types ~*
respirators have APF values of 50 or greater, and will fu'l *i 11 "~h^s
requirement. The Order or SNTR would list these three Yve*s" o*
respirators and the Company would have the option of selecting thp
most appropriate individual respirator from these three tyces,"""
The standard language is also based on consideration c*
information on respiratory protection developed at the Kc^'sh-p
(held February 12 and 13, 1936} , by experts in this fie1^" and
professional judgement. Special circumstances nay recui-e
modification of the standard language. A CEB Industrial HygieVst
should review all 5 (e) Orders and SNURs to determine that "he-
appropriate respirator classes have been selected fcr the
individual substance. Any questions regarding respiratory
protection should be referred to a CEB industrial hygienist,
(Note: Roman numerals in parenthesis refer to the list of
respirator types in the regulatory language of the Significant New
Use Rule, Section 721.63(a)(5) and the 5(e) Order language).
I. STANDARD LANGUAGE FOR A 2000-FOLD REDUCTION IN EXPOSURE:
Select respirator type (i) only.
(i) Category 19C Type C supplied-air respirator operated - ~.
pressure demand or other positive pressure node and
a full facepiece.
II. STANDARD LANGUAGE FOR A 50-FOLD REDUCTION IN EXPOSURE (APF of
50 or greater):
J-3
-------
A. Fart ic-j late Exposure: Select; respirator types (i
and (v) only.
ype C supplier-air respirator crserEteci iri
pressure der.and cr continue..,^ flow node and equipped with a tigr.r-
fitting facepiece.
(iv) Category 2 1C air-purifying respirator equipped with a
full facepiece and high efficiency particulate filters,
(v) Category 21C powered air-purifying respirator equipped
with a tiqht-f itting facepiece and high efficiency particulate
filters,
B. Pa_lnt_S pray ...... M i s t ...... Expos ure: Select respirator types (ii}r
(viii), and (ix) only.
(ii) Category 19C Type C supplied-air respirator operated in
pressure demand or continuous flow mode and equipped with a tight-
fitting facepiece,
(viii) Category 23C air-purifying respirator equipped with
a full facepiece and combination cartridges approved for paints,
lacquers and enamels. (Approval label may preclude use for some
paints, lacquers or enamels).
(ix) Category 2JC powered air-purifying respirator equipped
with a tight-fitting facepiece and combination cartridges approved
for paints, lacquers and enamels. (Approval label may preclude use
for some paints, lacquers or enamels).
C . CjT^a;vi.c_G_as_/ya_gor_ Exposure
If .no data on cartridge ......... p_e_rf_o_rTTia.nce is available: Select
respirator type (ii) only.
(ii) Category 19C Type C supplied-air respirator operated in
pressure demand or continuous flow mode and equipped with a tight-
fitting facepiece.
The preamble to the order should state that the company can
petition EPA to modify the order to allow the use of air-purifying
respirators which will give the same reduction in exposure, The
company should consider the warning properties of the substance and
must demonstrate the effectiveness of the respirator cartridge;
this information should be submitted for EPA evaluation. The
company should consult EPA for guidance on test methodology and
protocol. In addition, if the PMN chemical has poor or unknown
warning properties, the company must develop a change-out schedule
for cartridges based on service life data.
-------
If data c- cartridge per for-ance IE accer.*^r_^& • •--. - • .- - -
resriratcr types -e
filters. ' * " "~ *""
B. Paint 5_B.ray Mist_Exposurg : Select respirator tvces ' i ; -' '•
(viiij, (ix), and (x) only,
(iii) Category ISC Type C supplied-air respirator operated
in pressure demand or continuous flow mode and equipped with a per--5
or helmet, or tight-fitting facepiece.
(viii) ^Category 23C air-purifying respirator equipped with
a full facepiece and combination cartridges approved for caz~tsr
lacquers and enamels. (Approval label may preclude use for'so-s
paints, lacquers or enamels).
J-5
-------
(ix) Category 23C powered air-purifying respirator equipped
with a tight-fitting facepiece and combination cartridges approved
fcr paints, lacquers and enamels. (Approval label nay preclude use
for some paints, lacquers cr enamels).
(x) Category 23C powered air-purifying respirator equipped
with a loO'Se fitting hood or helmet and combination cartridges
approved for paints, lacquers and enamels, (Approval label may
preclude use for some paints, lacquers or enamels).
C. Organic Gas/Va_p_gr Exposure
.If ..no .datia _on_ c_artrLldae__Ee_r_f.Q_£Tnance is a\/a_ij^a_b!le_ : Select
respirator type (iii) only.
(Hi) Category 19C Type C supplied-air respirator operated
in pressure demand or continuous flow mode and equipped with a hood
or helmet, cr tight-fitting facepiece.
The preamble to the order should state that the company can
petition EPA to ir.odify the order to allow the use of air-purifying
respirators which will give the same reduction in exposure. The
company should consider the warning properties of the substance and
must demonstrate the effectiveness of the respirator cartridge;
this information should be submitted for EPA evaluation. The
company should consult EPA for guidance on test methodology and
protocol. In addition, if the PMN chemical has poor or unknown
warning properties, the company must develop a change-out schedule
for cartridges based on service life data.
If dataoncartridge performance is acceptable: Select
respirator types (iii), (xii), (xiii), and (xiv) only.
(iii) Category 19C Type C supplied-air respirator operated
in pressure demand or continuous flow mode and equipped with a hood
or helmet, or tight-fitting facepiece.
(xii) Category 23C air-purifying respirator equipped with a
full facepiece and organic gas/vapor cartridges,
(xiii) Category 23C powered air-purifying respirator equipped
with a tight-fitting facepiece and organic gas/vapor cartridges.
(xiv) Category 23C powered air-purifying respirator equipped
with a loose fitting hood or helmet and organic gas/vapor
cartridges.
IV. STANDARD LANGUAGE FOR A 10-FOLD REDUCTION IN EXPOSURE (APF of
10 or greater):
5-6
-------
(vi)
v i t h a 1 o o s e
filters.
A. Farticulate Exposure: Select respirator tv-^
<-±i), (iv) , (v), (vi), and (vii) only,
'ill) Category 19C Type C supplied-air resr.ir2trr ±r-^-----
in £r-= = £-ure aera^a cr continuous flow -r.ode and ecru itcad w-^- ^ •--,-.;
cr he.-e", cr tight-fitting facepiece.
(iv; Category 21C air-purifying respirator equipped with a
full facepiec-e and high efficiency particulate filters.
(v) Category 21C powered air-purifying respirator equ^peed
v:th a tight-fitting facepiece and high efficiencv ^,ar-i'c,.iA*c
f 1 1 t P r c; . '
Category 21C powered air-purifying respirator equipped
2 fitting hood or helmet and high efficiency particulate
(vii) Category 2IC air-purifying respirator equipped -,.••-'- 3
high efficiency particulate filter"] including disposables!
B* EaJ-^t Spray Mist Exposure; Select respirator i-vc<=s M-M
(vni), fix), (x), and (xi} only. " '"* ''
(iii) Category ISC Type C supplied-air respirator ccerated
in pressure demand or continuous flow mode and equipped with a heed
or hel~et, cr tight-fitting facepiece.
Jviii) category 23C air-purifying respirator equipped with
a full facepiece and combination cartridges approved 'for' pair.te
lacquers and enamels. (Approval label may preclude use "for so-te
paints, lacquers or enamels).
(ix) Category 23C powered air-purifying respirator ecuipped
with a_tight-fitting facepiece and combination cartridges approved
for paints, lacquers and enanels. (Approval label nav creclude use
for sore paints, lacquers or enamels).
•;x) Category 23C powered air-purifying respirator equipped
with a loose fitting hood or helmet and combination cartridaes
approved for paints, lacquers and enamels. (Approval label ray-
preclude use for sorae paints, lacquers or enaroels),
(xi) Category 23C air-purifying respirator equipped with
conbination_cartridges approved for paints, lacquers and "enanels,
including disposables. (Approval label may preclude use for sore
paints, lacquers or enamels),
C. Organic Gas/Vapor Expos_u_re_
J-7
-------
Lf_no data_ on Cartridge perf crnajicg^jis ava_i 1 s 11 r: ;
respirator type (iii) only.
fill) Category 19C Type C sur::!ied-air respirator
ir. pressure demand cr continuous flow node and equipped vi:
cr helr.et, cr tight-fitting facepiece.
The preamble to the order should state that the cor.pany car,
petition EPA to modify the order to allow the use of air-purifying
respirators which will give the same reduction in exposure. The
company should consider the warning properties of the substance and
nust demonstrate the effectiveness of the respirator cartridae;
this information should be submitted for EPA evaluation. The
company should consult EPA for guidance on test methodology and
protocol. In addition, if the PMN chemical has poor or unknown
warning properties, the company must develop a change-out schedule
for cartridges based on service life data.
I f_ d a t a o r c_a r t r idge p e r f o rr> a n c e_ig_ accept a b_Ie : Select
respirator types (111} , (xnl, (xiii) , (xivj , and (xv) only.
fiii) Category I9C Type C supplied-air respirator operated
ir. pressure demand cr continuous flow mode and equipped with a hood
or helmet, cr tight-fitting facepiece.
(xii) Category 23C air-purifying respirator equipped with a
full facepiece and organic gas/vapor cartridges.
{xiii) Category 23C powered air-purifying respirator equipped
with a tight-fitting facepiece and organic gas/vapor cartridges.
(xiv) Category 23C powered air-purifying respirator equipped
with a loose fitting hood or helmet and organic gas/vapor
cartridges.
(xv) Category 23C air-purifying respirator equipped with
organic gas/vapor cartridges, including disposables.
V, OTHER COMBINATIONS OF CONTAHINAHTS:
Consult a CEB Industrial Hygienist.
J-8
-------
APPENDIX K
DERIVATION OF EQUATION FOR
EVAPORATION FROM OPEN SURFACES
(REVISED)
K-1
-------
Evaporation of a Liquid in a Flowing Air Stream
Evaporation can be defined as the change of state of a liquid into a gas at the
cost of a specific amount of energy. Evaporation of a liquid in a flowing air stream can
be idealized as in Figure I, in which the air flow direction is defined to be along the z
axis, and the flow is always parallel to the liquid surface,
air flow
Figure 1
Once the evaporation system reaches equilibrium, namely, the surface
temperature of a sufficiently large pool is constant and the heat of evaporation is
provided by the surroundings, the evaporation rate will also be a constant. In this case,
we can establish a mass balance on a differential element above the liquid pool (normal
to the liquid surface, or in the x direction) and along the air flow (z direction). A similar
system has been considered in a standard text of Transport Phenomena by Bird Stewart
and Lightfoot (Section 17.5).
Along a volume element, the following equation applies:
6W,
= 0
(1)
where NA^ and N^ are the molar fluxes of chemical A in the z or x direction
(moles/unit time/unit area).
K-2
-------
For NA7, the flux in the z direction, we assume no changes in concentration
CA(X,Z) as a function of time (equilibrium), and neglect the contribution from diffusion of
A in the z direction (small compared to the air velocity). In addition, diffusion at the
edges of the pool in the y direction is neglected. Geometry of the surroundings may
influence these assumptions (such as physical barriers above and to the side of the pool),
but may be neglected if the concentration of the vapor is relatively low at that boundary'.
In order to simplify the analysis for the overall evaporation rale, the air velocity is also
assumed constant as a function of x and flow only in the 2 direction. Thus,
NK - 0AV2 (2)
For NAili, the flux above the pool in the x direction, we neglect convective
transport (no net flow in that direction). This essentially means that CA is small, and that
there is essentially no mixing in the area above the pool where the greatest portion of
the concentration gradient exists. Thus,
where DAB is the diffusion coefficient of chemical A in B (air). To be rigorous, the
diffusion coefficient should be defined at the local temperature of the flowing air,
However, this temperature will vary between the liquid surface temperature and the
source air temperature. In this analysis, the liquid temperature is used,
Thus, equation (1) becomes:
&r* z.2r.
(4)
The boundary conditions are .as follows:
at z = 0 CA = 0
x = 0 CA = CAO
x = » CA = 0
CAO is assumed to be the concentration represented by the vapor pressure of the liquid at
the liquid surface temperature. In reality, a temperature gradient will usually exist
between the liquid surface and the air temperature some distance away. Using the liquid.
vapor pressure neglects the effect of this gradient on the evaporation.
K-3
-------
The solution to this partial differential equation is:
" 1 "T3 / e^c
CAO V« o (5)
x
I ""''
The complementary error function (erfc) is well known as shorthand for the integral in
equation (5) above.
This equation gives the vapor concentration as a function of z and x. The
evaporation rate at any point on the liquid surface is given by;
The total evaporation rale is therefore the sum of all these points over the entire
liquid surface, given by:
Evap.Rate = i)(Breaofpooty f JN^zdy
Solving the partial derivative in equation (7) by differentiating equation (5) yields:
(7 2D^^~dzd¥
zJ4«D^Kz
(8)
This equation is in units of moles per unit time and area. In units of weight, we
multiply by the molecular weight of the evaporating substance:
K-4
-------
Eva.p.Rate(mass/time*area) - 2 MW CAO
(9)
n Az
The concentration of the evaporating solute at the interface is given by c^, which
can be approximated by:
R T
where "vapor pressure" refers to the saturated vapor pressure above a liquid at
temperature 'T' in Kelvin, and "R" is the gas constant (39.381 in. Hg * ft* / { Ib moles *
Kelvin }.
Equation (9) can therefore be written in the following dimensions:
Evap.Rate(lbfkr j^2) =
13.3792 M.W. V.P.
\
(11)
Az
where:
M.W. = molecular weight of evaporating liquid
V.P. = vapor pressure (in. Hg)
T = temperature (Kelvin)
DAB = diffusion coefficient of liquid vapor in air, ft2/sec
vz - air velocity, ft/mi n
Az = poo! length along flow direction, ft.
Equation (11) is completely general for a dilute evaporating substance in a
flowing air stream.
K-5
-------
The diffusion coefficient may noi be available for many cases. According to
Chapman-Enskog kinetic theory (Bird, Stewart and Light fool section 16,4), the diffusion
coefficient is ejven bv:
O0018583.j'Fo/W41"+ 1/A/j)
P ® AB AS
(12)
where:
T = temperature, Kelvin
MA = chemical A molecular weight
MB = chemical B (air) molecular weight (= 29)
p = pressure (atmospheres)
aAB = Lennard-Jones distance, angstroms
n.AB = function of Leonard-Jones potential
The Lennard-Jones parameters are tabulated (appendix B, Bkd.Stewart and
Lightfooi) and can be roughly fit against temperature and molecular weight. This is
shown in Figures 2,3 and 4.
QUO
70C -
OJ
V
E
o
tn
ci
us
403
30C
SOD
1DQ
D
30 100 150 EDO 25O
molecular w&lght
©ps r lon/k dets __eps! lon/k predTcT.ec!
3OD
Figure 2
Figures 2 and 3 show the fit of the function Omega against molecular weight
(Figure 2) and temperature (Figure 3). In Figure 2, the Lennard-Jones potential
epsiSon/k for a variety of compounds is shown versus their molecular weight. A simple
exponential regression was fit against this graph with the results shown, Figure 3 shows
K-6
-------
O)
o
1 . S
Q 3
V"
y
Omega A3 dat.o
eps I I on
_ Omega AB predicted
Figure 3
the plot of the Omega function against the Lennard-Jones potential divided by
temperature, and a simple exponential fit of this function, Combining these two fits
gives an approximation of the fl function in Equation 12,
5. 5
1
tfj
O>
c
CD
ra
E 3.5
LO
1DO ISO
moJe-cular w
200
Z3D
300
:.: Sigma AB dana
Sfgma AB predicted
Figure 4
-------
Similarly, Figure 4 shows the relationship and fit of data for the a function in
Equation 12, The exponential fits of these functions versus molecular weight and
temperature are:
(M.W.)
™1
_
I
Substituting into equation 12, this gives;
5 Fw 71/29 + \}M.W.t
^
where:
p =• pressure (atmospheres)
T = temperature (Kelvin)
M.W.A = molecular weight of evaporating liquid
Converting DAB to ft2/sec, and then inserting into equation (11), an approximate
expression for the evaporation rate of a liquid in a flowing air stream is given by:
Evap.Rate(lb}hr
2.79jdCT3 (M.W.fm (F.P.) IJM.W.A + 1/29
jO.05
v
*_ (IS)
where:
M.W. = molecular weight of evaporating liquid A
p = overall pressure, atmospheres
Az = length of pool along air flow, feet
T = surface temperature of pool, Kelvin
V.P. = vapor pressure of substance A (in. Hg)
vz = velocity of air, ft/min
Equation 15 is a completely general equation that can be used to predict the
evaporation rate of a liquid in a flowing air stream. In addition, it lakes into account the
effect of changes in temperature, pressure, pool size and air velocity. Since it is rare to
K-8
-------
find completely stagnant air conditions in most spill situations, an approximation of the
air speed, spill size can be made, and this equation will work well for esiimating the
evaporation rate. Although many assumptions were made to arrive at Equation 15, they
are not at all unreasonable for a variety of situations. In general, the equation assumes
low concentrations of the evaporating solvent, as compared to the air concentration in
the room. As a result, this equation may begin to show errors for strongly evaporating
solvents, long pool lengths or slow air wind velocities.
This equation can be tested by comparing it to experimental data provided by
Pace Laboratories. In an extensive study using a specially-built apparatus, Pace
measured the evaporation rate of 15 different compounds at several different
temperatures and air velocities, and fit the data against "power law" regression against
molecular weight, vapor pressure and air velocity, with generally good results. They
performed an overall regression analysts for all chemicals except the "low vapor pressure"
alcohols (1-hexanol, 1-hepianoi and 2-ocunol) and obtained the following equation:
= 0.000237XMW)(KP)(F,0-625) U6)
where:
MW = molecular weight of evaporating substance
VP = vapor pressure at liquid temperature, In. Hg
Vz = air velocity, ft/min
The EPA. equation, based on work by Mackay is:
*3 ,.~
(17)
In some cases, the wind velocity' term has been assumed to be 10 mph, which
corresponds to 880 feet/min. For the comparison below, however, the velocity term was
included.
Figures 5, 6 and 7 compare Equations 15, 16 and 17 as applied to the Pace data,
including the low boiling alcohols. The only differences between the figures are the
scales, showing the experimentally measured evaporation rates from 0 to 0.06, 0 to 0.7
and 0 to 6 Ibs/hr fr.
K-9
-------
r
,' \
CM
t-1
(0
o
o/
t-'
(J-
0.)
Q_
D 01 D 02 D C3 Q 0* Q 05
Experimental Evsp Rate CId/fir/fta}
• Pace Equation 0 Theory/Frtied Dab
x EPA Equatlen
0.06
Figure 5
D,1 0.2 0.3 0.4 O.S 0.6
Experimental Evop Rate Clb/hr/ft23
m Pace Equat ion
x EPA Eauatlon
0,7
o Theory/ Fitied Dab
Figure 6
A perfect fit of experiment and equation would be on the straight line along the
diagonal. As can be seen in the three figures, both the Pace equation and the
theoretical Equation 15 does a good job over a large range of evaporation rates. Since
the Pace equation was a regression fit, it should naturally fit the closest. Remarkably,
the theoretical equation is almost as good, and seems to be more accurate at the lower
evaporation rates, which would correspond to normally encountered situations in EPA
PMN analysis. The MacKay equation, however, is extremely poor, universally
underestimating the actual evaporation rate by an order of magnitude for many measurements.
K-10
-------
1 2 3
Experimental Evep Rste Clb
m Pace Equation o Theory/Fitted Dab
x Ep-A Equal i on
Figure 7
Closer inspection of ihe data shows that the theoretical equation 15 does the best
job of the three equations at the low velocity gas rates of 100 feet/minute, but like the
other two equations, it underestimates the experimentally measured evaporation rate for
all of the low flow rate data. It is not -clear whether this is from a deficiency in the
equations or in the experiment. However, since these low rates would be more
frequently encountered when judging evaporation rates inside ventilated buildings, the
theoretical equation may be the best choice of the three. In addition, the assumptions
are clear in the theoretical development, and cases where they may be violated can be
easily identified. The agreement between the Pace results and the theory shows that
these assumptions also apply to the Pace experiment.
Dr. Albert A, Hummel
Himont Research & Development
800 Greeobank Rd.
Wilmington, DE 19808
August 18, 1990
(Formerly of the Chemical
Engineering Branch)
K-n
------- |