Review Draft
United States
Environment?.: Protection
Agency
Off»c« of Air Quality
Planning and Standards*
fl«««arch Tnang.« i>arlc NC 2771
Air
Hazardous Waste
Treatment, Storage,
and Disposal
Facilities (TSDF) —
Air Emission
Models
No/ember 1989
-------
Review Draft
EPA-450/3-87-026
Hazardous Waste Treatment, Storage and Disposal
Facilities (TSDF) —Air Emission Models
Emission Standards Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park NC 27711
-------
CONTENTS
Chapter Page
1.0 Introduction 1-1
1.1 Background 1-1
1.2 Scope 1-2
1.3 Report Organization 1-2
2.0 Description of Pathways 2-1
2.1 General 2-1
2.2 Volatilization 2-2
2.3 Adsorption 2-5
2.4 Migration 2-6
2.5 Runoff 2-7
2.6 Biological Decomposition 2-7
2.7 Photochemical Decomposition 2-8
2.8 Hydrolysis 2-10
2.9 Oxidation/Reduction 2-11
2.10 Hydroxyl Radical Reactions 2-12
2.11 References 2-12
3.0 Importance of Pathways 3-1
3.1 Introduction .* 3-1
3.2 Theoretical Basis 3-1
3.2.1 Surface Impoundments 3-3
3.2.2 Aerated and Nonaerated Wastewater
Treatment 3-3
3.2.3 Land Treatment 3-3
3.2.4 Landfills .' 3-6
3.3 Emission Models 3-6
3.4 References 3-6
4.0 Surface Impoundments and Open Tanks 4-1
4.1 Narrative Description of Emissions and
Model Units . 4-1
4.2 Quiescent Surfaces with Flow 4-3
4.2.1 Emission Model Equations 4-3
4.2.2 Model Plant Parameters for Quiescent
Impoundments 4-11
4.2.3 Example Calculation for Storage
Impoundments 4-12
ill
-------
CONTENTS (continued)
Chapter fjfli
4.3 Biodegradation 4-16
4.3.1 Description of Biological
Active Systems 4-16
4.3.2 Rate of Biodegradation 4-20
4.3.3 Example Calculation for
Quiescent Impoundments 4-27
4.4 Mechanically Aerated Impoundments and
Activated Sludge" Units 4-29
4.4.1 Emission Model Equations 4-29
4.4.2 Model Plant Parameters for
Mechanically Aerated Impoundments 4-30
4.4.3 Example Calculation for Mechanically
Aerated Treatment Impoundments 4-34
4.4.4 Example Calculation for Activated
Sludge Unit 4-40
4.5 Disposal Impoundments with Quiescent Surfaces ... 4-41
4.5.1 Emission Model Equations 4-41
4.5.2 Model Plant Parameters for Disposal
Impoundments 4-45
4.5.3 Example Calculations for Disposal
Impoundments 4-45
4.6 Diffused Air Systems 4-49
4.6.1 Emission Model Equations 4-49
4.6.2 Model Unit Parameters for Activated
Sludge Unit with Diffused Air .-... 4-51
4.6.3 Example Calculation for Diffused
Air Activated Sludge Unit 4-51
4.7 Oil Film Surfaces..." 4-53
4.8 Discussion of Assumptions and Sensitivity
Analysis 4-55
4.8.1 Removal Mechanisms 4-55
4.8.2 Major Assumptions 4-57
4.8.3 Sensitivity Analysis 4-58
4.9 References .' 4-64
5.0 Land Treatment 5-1
5.1 Narrative Description of Land Treatment
and Air Emissions . 5-1
5.2 Land Treatment 5-3
5.2.1 Land Treatment Emission Model
Descriptions 5-3
5.2.1.1 Analytical Correlations 5-3
5.2.1.2 Biodegradation 5-7
5.2.1.3 Estimation of Equilibrium
Coefficient, Keq 5-8
5.2.1.4 Estimation of Effective
Diffusivity 5-9
iv
-------
CONTENTS (continued)
Chapter
5.2.1.5 Waste Partitioning 5-10
5.2.1.6 Emissions at Short Times 5-13
5.2.1.7 Estimating the Fraction
Emitted at Short Times 5-18
5.2.1.8 Estimating the Fraction
Emitted for Longer Times 5-19
5.2.1.9 Tilling 5-21
5.2.1.10-Model Selection 5-25
5.2.2 Waste Application Model 5-26
5.2.3 Oi 1 Fi 1m Model 5-27
5.2.4 Model Inputs 5-27
5.2.5 Estimation of Total VO Emissions 5-36
5.2.6 Example Calculations 5-36
5.2.6.1 Emissions from Land
Treatment Soil 5-36
5.2.6.2 Emissions from Waste
Application .' 5-40
5.2.6.3 Emissions from an Oil Layer
on Soil Prior to Tilling 5-42
5.2.6.4 Examples of the Use of the
Land Treatment Model for
Specific Cases 5-44
5.2.7 Assumptions and Sensitivity Analyses 5-49
5.3 References 5-51
6.0 Landfills and Wastepiles 6-1
6.1 Introduction 6-1
6.2 Closed Landfills 6-1
6.2.1 Emission Model Equations 6-1
6.2.2 Model Plant Parameters for Closed
Landfills 6-14
6.2.3 Example Calculation for Closed
Landf i 11 6-17
6.3 Fixation Pits % 6-21
6.3.1 Emission Model Equations 6-21
6.3.2 Model Plant Parameters for Fixation
Pits 6-30
6.3.3 Example Calculation for Fixation Pit 6-31
6.4 Open Landfills and Wastepiles 6-33
6.4.1 Emission Model Equations 6-33
6.4.2 Model Plant Parameters for Open
Landfills and Wastepiles 6-40
6.4.2.1 Parameters for Open Landfills . 6-40
6.4.2.2 Parameters for Wastepiles 6-41
6.4.3 Example Calculation for Open
Landfill 6-43
6.5 References 6-47
-------
CONTENTS (continued)
Chapter
7.0 Transfer, Storage, and Handling Operations 7-1
7.1 Narrative Description of Model Plants and
Emissions 7-1
7.2 Container Loading 7-1
7.2.1 Emission Model for Container Loading 7-1
7.2.2 Model Parameters 7-2
7.2.3 Sample Calculation for Tank Loading 7-2
7.3 Container Storage 7-5
7.3.1 Emission Model for 55-Gal Drums,
Tank Trucks, and Railroad Tank Cars 7-5
7.3.2 Model Parameters for Drum Storage 7-6
7.3.3 Sample Calculations for Drum Storage 7-6
7.3.4 Emission Model for Open Dumpsters 7-7
7.3.5 Model Parameters for Open Dumpster
Storage 7-7
7.3.6 Sample Calculation for Open Dumpster
Storage 7-8
7.4 Container Cleaning 7-9
7.4.1 Emission Model for Container Cleaning 7-9
7.4.2 Model Parameters 7-10
7.4.3 Sample Calculation for Tank Truck
Cleaning 7-11
7.5 Stationary Tank Loading 7-11
7.5.1 Emission Model for Stationary Tank
Model 7-11
7.5.2 Model Parameters 7-12
7.5.3 Sample Calculation for Tank Loading
Emission Model 7-13
7.6 Stationary Tank Storage 7-15
7.6.1 Model Description 7-15
7.6.2 Model Parameters 7-17
7.6.3 Sample Calculation for Tank Storage
Emission Model 7-17
7.7 Spills .*. 7-18
7.7.1 Model Description 7-18
7.7.2 Model Parameters 7-18
7.7.3 Sample Calculation for Drum Storage
Model 7-18
7.8 Fugitive Emissions 7-19
7.8.1 Emission Model for Fugitives 7-19
7.8.2 Model Parameters 7-20
7.8.3 Sample Calculation for Fugitive
Emission Model 7-20
7.9 Vacuum Truck Loading 7-20
7.9.1 Emission Model for Vacuum Truck
Loading 7-20
vi
-------
CONTENTS (continued)
Page
7.9.2 Model Parameters 7-21
7.9.3 Sample Calculation 7-21
7.10 References 7-23
8.0 Comparison of Model Results with Field Test Data 8-1
8.1 Introduction 8-1
8.2 Surface Impoundments and Open Tanks 8-1
8.2.1 Summary-:- 8-1
8.2.2 Details of Comparisons 8-2
8.2.3 Recommendations for Additional Data 8-17
8.3 Land Treatment 8-19
8.3.1 Midwest Refinery—1985 (Case 1) 8-29
8.3.2 West Coast Refinery (Case 2) 8-34
8.3.3 Commercial Waste Disposal Test (Case 3)... 8-34
8.3.4 Midwest Refinery--1979 (Case 4) 8-34
8.4 Landfills and Wastepiles 8-34
8.5 Transfer, Storage, and Handling Operations 8-39
8.5.1 Models Documented in AP-42 8-39
8.5.2 Fugitive Emissions 8-44
8.5.3 Spillage 8-44
8.5.4 Open Dumpster Storage Emissions 8-44
8.6 References 8-44
Appendix A CHEMDAT7 User's Guide A-l
Appendix B A Guide Through the Literature 6-1
Appendix C Comprehensive Source List C-l
Appendix D Properties for Compounds of Interest 0-1
vn
-------
FIGURES
Number
4-1 Theoretical relationship between concentration
and biodegradation rates normalized by the
amount of biomass as predicted using the Monod
model for phenol, benzene, and chloroform 4-24
4-2 Correlation of limiting first-order rate
constant with the octanol water coefficient 4-27
6-1 Pick's law correction factor Fy as a function
of y* 6"24
8-1 Estimated vs. measured benzene emission flux
rates—Case 1 8'26
8-2 Estimated vs. measured toluene emission flux
rates—Case 1 8'27
8-3 Estimated vs. measured toluene emission flux
rates—Case 2 (data for 4 days only) 8-28
8-4 Estimated vs. measured total VO emission flux
rates—Case 2 8'29
8-5 Estimated vs. measured VO emission flux
rates—Case 3 8~3*
8-6 Estimated vs. measured emission flux
rates—Case 4 8-32
viii
-------
TABLES
Number
2-1 Values of Constants for Use in Equation (2-4) ........ 2-4
3-1 Pathways for Hazardous Waste Area Emission Sources ... 3-2
3-2 Statistics for Surface Water Pathways ................ 3-4
3-3 Pathways for TSDF Sites .............................. 3-5
4-1 Equations for Calculating Individual Mass Transfer
Coefficients for Volatilization of Organic Solutes'
from Quiescent Surface Impoundments .................. 4-6
4-2 Input Parameters —Storage Impoundment ................ 4-13
4-3 Design Parameters for Activated Sludge Processes ..... 4-17
4-4 Impoundments Designed for Biodegradati on ............. 4-19
4-5 Typical or Default Values for Biomass
Concentration ........................................ 4-21
4-6 Equations for Calculating Individual Mass
Transfer Coefficients for Volatilization of
Organic Solutes from Turbulent Surface
Impoundments ......................................... 4-34
4-7 Input Parameters— Treatment Impoundments
(Mechanically Aerated) ............................... 4-38
4-8 Input Parameters— Mechanically Aerated Activated
Sludge Unit ............... ........................... 4-46
4-9 Intermediate and Final Calculation Results for
Activated Sludge Model Unit .......................... 4-47
4-10 Input Parameters— Disposal Impoundments .............. 4-49
4-11 Input Parameters— Diffused Air Activated
Sludge Unit .......................................... 4-55
4-12 Results of Sensitivity Analysis for Quiescent
Storage Impoundment .................................. 4-65
4-13 Results of Sensitivity Analysis for Mechanically
Aerated Impoundments ................................. 4-66
4-14 Results of Sensitivity Analysis for Disposal
Impoundments ......................................... 4-67
5-1 Comparison of the Estimated Fraction Emitted
Using Three Different Equations (Integrated Flux
from Soil) ............................................ 5-6
5-2 Emission Estimates Using Two Different
Equations for the Vapor-Soil Partition Coefficient ... 5-12
5-3 RTI Model for Land Treatment Emissions ................ 5-14
IX
-------
TABLES (continued)
Number Page
5-4 Estimated Air Emission Fraction at Long Times 5-22
5-5 Rigorous vs. Approximate Estimates of Emission
Fractions 5-23
5-6 Waste Application Emission Model 5-28
5-7 Oil Film Surface Emission Model 5-30
5-8 Measured and Estimated Biorates and Decay
Constants for Selected Organic Constituents 5-34
5-9 Estimated Emission Rates and Fractions
Emitted Versus Time for Example Land
Treatment Calculation 5-41
6-1 RTI Closed Landfill Model 6-10
6-2 Input Parameters—Closed Landfill ....._ 6-16
6-3 Pick's Law Correction Factor as a Function of y* 6-23
6-4 Open Dump Model 6-28
6-5 Input Parameters—Fixation Pit 6-32
6-6 RTI Land Treatment Model Applied to Open
Landfills and Wastepiles (No Biodegradation) 6-37
6-7 Input Parameters—Open Landfill 6-42
6-8 Input Parameters—Wastepiles 6-44
7-1 S Factors for Calculating Petroleum Loading
Losses 7-3
7-2 Pertinent Fixed-Roof Tank Specifications 7-14
7-3 Paint Factors for Fixed-Roof Tanks 7-16
7-4 SOCMI Emission Factors for Fugitive Losses 7-19
8-1 Comparison of Results for Reducing Lagoon 1
at Site 5 8-3
8-2 Comparison of Results for Holding Pond 6
at Site 5 8-4
8-3 Comparison of Results for Oxidizing Lagoon 2
at Site 5 8-5
8-4 Comparison of Results for Surface Impoundment
at Site 4 8-5
8-5 Comparison of Results for Wastewater Holding
Lagoon at Site 3 8-7
8-6 Comparison of Results for Primary Clarifiers
at SiteS 8-7
8-7 Comparison of Results for Equalization Basin
at Site 8 8-8
8-8 Comparison of Results for Aerated Stabilization
Basins at Site 8 8-8
8-9 Comparison of Results for Covered Aerated Lagoon
at Site 7 8-10
-------
TABLES (continued)
Page
Description of Petrasek's Activated Sludge
System 8-12
Comparison of Petrasek's Measurements and
Model Predictions 8-13
8-12 Description of Two Chicago Activated Sludge Units .... 8-14
8-13 Comparison of Measured and Predicted Effluent
Concentrations for Chicage Wastewater Treatment
Plants I': 8-15
8-14 Comparison of Measured and Predicted
Biodegradation Rates 8-18
8-15 Summary of Land Treatment Testing and Test Results ... 8-21
8-16 Input Parameters for RTI Land Treatment Model 8-33
8-17 Measured and Estimated Emissions—Case 1 8-34
8-18 Input Parameters for RTI Land Treatment Model 8-36
8-19 Estimated vs. Measured Emissions—Case 2 8-36
8-20 Input Parameters for RTI Land Treatment Model 8-37
8-21 Estimated vs. Measured Total VO Emissions-
Case 3 8-37
8-22 Input Parameters for RTI Land Treatment Model 8-38
8-23 Estimated vs. Measured Emissions—Case 4 8-38
8-24 Model Input Parameters Used in Application of
the RTI Land Treatment Model to an Active
Landfill at Site 5 8-40
8-25 Comparison of Measured and Predicted Emission
Rates for Site 5 Active Landfill 8-42
8-26 Model Input Parameters Used in Application
of the RTI Land Treatment Model to an
Active Landfill at Site 8 8-43
8-27 Comparison of Measured and Predicted Emission
Rates for the Site 8 Active Landfill 8-44
XI
-------
1.0 INTRODUCTION
1.1 BACKGROUND
This report was prepared for the Environmental Protection Agency's
(EPA's) Office of Air Quality Planning and Standards (OAQPS) as part of the
"•»*
effort to develop air emission models for hazardous waste treatment,
storage, and disposal facilities (TSDF). Basic to this effort is the
determination of the means by which volatile organics (VO) escape to the
environment from TSDF.
VO in surface impoundments, land treatment facilities, landfills,
wastepiles, or wastewater treatment (WWT) plant effluents can depart
through a variety of pathways, including volatilization, biological decom-
position, adsorption, photochemical reaction, and hydrolysis. To allow
reasonable estimates of VO disappearance, one must know which pathways
predominate for a given chemical, type of waste site, and set of meteoro-
logical conditions.
Analytical models have been developed to estimate emissions of VO via
various pathways from area emission sources at hazardous waste sites. Some
of these models hav-e been assembled into a spreadsheet that is included in
this report as a floppy diskette for use on an IBM PC, or compatible,
microcomputer. A user's guide for these models is included in the report
as Appendix A. Area emission sources for which models are included on the
diskette are as follows:
• Nonaerated impoundments, which include quiescent surface
impoundments and open top WWT tanks
• Aerated impoundments, which include aerated surface
impoundments and aerated WWT tanks
• Disposal impoundments, which include nonaerated disposal
Impoundments
• Land treatment
Landfills.
1-1
-------
These models can be used to estimate the magnitude of site emissions for
regulatory purposes. Sample calculations using each model also are
included in this report.
1.2 SCOPE
This report briefly describes the chemical and physical pathways for
VO and discusses their importance for different types of sites and sets of
conditions. Models developed for estimating the relative magnitude of
environmental release in the presence of competing pathways are presented,
and physical characteristics of the parameters that serve as inputs to the
models are identified.
The models provide an estimate of the relative magnitude of VO
pathways on a compound-specific basis. Models for aerated and nonaerated
impoundments, lagoons, landfills, wastepiles, and land treatment facilities
have been installed in an integrated spreadsheet program, CHEMOAT7, which
allows a user to calculate the partitioning of VO among various pathways
depending on the particular parameters of the facility of interest. The
program is structured to allow new data (e.g., compounds and model facility
parameters) to be added (see Appendix A for user's guide). The results of
the calculated partitioning may be used to identify those characteristics
that are important in determining relative VO loss rates.
Source variability will significantly influence the relative impor-
tance of the pathways. For highly variable sources, it may be possible to
exclude insignificantly small pathways from consideration. The relative
magnitude of these pathways then can be compared by applying the methodol-
ogy to a model facility to determine relative differences among various
compounds.
1.3 REPORT ORGANIZATION
Chapter 2.0 describes each of the potential pathway mechanisms that
determine the fate of various chemical species. Chapter 3.0 discusses the
importance of the pathways for surface impoundments and aerated and non-
aerated WWT facilities, land treatment sites, and landfills/wastepiles.
Chapters 4.0, 5.0, and 6.0 describe the emission models applicable to these
sites. Models for estimating emissions from transfer, storage, and han-
dling operations are described in Chapter 7.0. Chapter 8.0 compares
emission model predictions with the field data that are available.
1-2
-------
This report compares relative rates of VO destruction and volatiliza-
tion to determine the most significant pathways. The rate of VO volatili-
zation destruction for any one pathway is calculated so that it can be
expressed as a fraction of the loss/destruction from all pathways.
Appendix B contains supplementary material, and Appendix C presents a
comprehensive source list that includes pertinent literature in addition to
that cited in the sections and appendixes of this report.
Properties of compounds of interest to TSDF pathways and emission
estimation are presented in Appendix D. A subset of these compounds is a
part of CHEMDAT7. The user's guide, Appendix A, describes the procedures
that are used in estimating emissions using CHEMDAT7 and other procedures
presented in the body of the report. The user's guide also contains
instructions for modifying CHEMDAT7 to include additional compounds using
the compound characteristics presented in Appendix D.
1-3
-------
•2.0 DESCRIPTION OF PATHWAYS
2.1 GENERAL
A pathway is considered here to be any process that removes volatile
organics from a site. The removal may be physical (as in volatilization of
a solvent from a surface impoundment) or chemical (as in oxidation of an
alcohol in a wastewater treatment plant).
Pathways may be considered as rate processes, with rate often strongly
dependent on concentration of the disappearing species and temperature of
the system. Rates vary in order from zero to mixed, with perhaps first
order predominating, that is:
rate = - g| = kyc (2-1)
where
c = concentration of disappearing substance, g/L
t - time, s
kv - volatilization constant, s'1
Half-life, the time required for one-half of the substance to disap-
pear, is a useful concept. It provides an easily visualized measure of the
time required for disappearance. For a first-order rate process:
t1/2 » (In2)k/l • 0.693 k/1 (2-2)
where
tj/2 * half-life, s.
The half-life of a second-order equation 1s as follows:
t1/2 • k^1 C/1 (2-3)
2-1
-------
where
k' » second-order volatilization constant, L/(g«s)
C - initial concentration, g/L.
o
Note that first-order half-lives are independent of initial concentration
while higher order half-lives are not.
Much of the following material is taken from ICF.1 The pathways
described are physical (volatilization, adsorption, migration, and runoff)
and chemical (biological decomposition, photochemical decomposition,
hydrolysis, oxidation/reduction, and hydroxyl radical reaction).
2.2 VOLATILIZATION
Volatilization occurs when molecules of a dissolved substance escape
to an adjacent gas phase. The driving force for this process in nonturbu-
lent liquids is molecular diffusion. Equation (2-1) shows the rate of
volatilization of an organic chemical from water. For this case, the rate
constant can be estimated:2
-i _1
(2-4)
where
L « mixing depth of water, cm
k? « mass transfer coefficient of oxygen in water, cm/s
diffusion coefficient of the chemical (c) or oxygen (o) in water,
cm2/s
liquid turbulence exponent, 0.5 to 1, dimensionless, from Table
2-1
R « ideal gas constant, atm
T * temperature, K
H « Henry's law constant, atm m^/rnol
kw « mass transfer coefficient for water vapor in air, cm/s
9
D * diffusion coefficient of the chemical (c) or water (w) in air,
9 cm2/s
2-2
-------
n - gas turbulence exponent, 0.5 to 1.0, dimensionless, from
Table 2-1.
Equation (2-4) requires values of diffusion coefficients and Henry's
law constants. If tabulated values are not available, the following esti-
mations can be used. For the diffusion of a chemical in air:3
D - 0.0067T1'5 (0.034 + M'1) ' M'0'17 [(M/2.5d)0'33 * 1.81]"2 (2-5)
9
where
M - molecular weight of chemical, g/g mol
d * density of liquid chemical, g/cm3.
For diffusion coefficients in water:
DT - 1.518 (ID'4) V'°-6 (2-6)
where
Van * molar volume of chemical, cm3/g mol.
This equation assumes the system temperature to be 300 K. For other
temperatures, a more rigorous form of the equation should be used, as in
Perry.4 Molar volume is estimated as the ratio of molecular weight to
liquid density at room temperature.
If ideal gases and solutions are assumed, Henry's law constant can be
estimated from:
H = P/(14.7s) (2-7)
where
P * pure component vapor pressure, psia
s * solubility of chemical in water, g mol/m3.
Values for other terms in Equation (2-4) have been tabulated by ICF
and are given in Table 2-1.
In general, equations are available to estimate volatilization from
wastewater treatment systems and surface impoundments.6,7 in the case of
land treatment and landfills, the models for volatilization are much less
well developed and the supporting data are more limited than those of the
2-3
-------
FABLE 2-1. VALUES OF CONSTANTS FOR USE IN EQUATION (2-4)5
Value
Constant Rivers Lakes
L (cm) 200 200
k° (ows-1) 0.0022 0.0005
m 0-7 1-0
T (K) 293 293
RT (m3.atra-mol-l) 2.40 x 10'2 2.40 x 10'2
K« (cm-s-1) 0.58 0.58
n 0.7 0.7
2-4
-------
aqueous systems. The rate of volatilization at a soil-air interface is a
function of the concentration and properties of the escaping chemical, soil
properties (moisture, temperature, clay, and organic content), and proper-
ties of the air at soil level (temperature, relative humidity, and wind
speed).8
2.3 ADSORPTION
Adsorption takes place when molecules of a dissolved chemical (in a
liquid-solid system) become physically attached to elements of the solid
phase. Chemical bonding may also "occur (chemisorption). An example of
adsorption is molecules of solvent being sorbed by particles of silt in a
surface impoundment. If the adsorptive capacity of the solid material is
reached, no further net sorption will occur. With reductions in concentra-
tion in the bulk liquid of the chemical being sorbed (adsorbate), desorp-
tion may take place. The amount of material adsorbed depends on (1) the
concentration of adsorbate, (2) the amount of solid phase (adsorbent), and
(3) the temperature. For systems with constant adsorbent properties, pri-
marily surface area per unit mass, the amount of material adsorbed at a
particular concentration and temperature is proportional to the mass of
adsorbent. For example, the Freundlich adsorption isotherm equation allows
prediction of amount adsorbed as follows:
i • V (?'8)
where
X - mass of chemical adsorbed, g
m * mass of adsorbent, g
Kf - Freundlich adsorption coefficient, (g sorbate/g sorbent)/
(g sorbate/g solution)
C1 » concentration of chemical in solution at equilibrium, g
sorbate/g solution
n = empirical constant, ranging from 0.7 to 1.1,. typically 1.0 for
soils, dimensionless.
A Langmuir adsorption isotherm can be derived from a kinetic rate
theory describing the adsorption and desorption rates. The rate of adsorp-
tion is proportional to the rate of collisions between adsorbate molecules
2-5
-------
and free adsorbent surface. The rate decreases with lowering adsorbate
concentration and with decreasing surface sites available for adsorbing
molecules. The following rate equation applies:
Rate of adsorption = kj C* (1-f) (2-9)
where
kj » rate constant for adsorption, g/s
f * fraction of adsorption sites occupied, dimensionless.
For desorptlon:
Rate of desorptlon * k-f (2-10)
where
k2 » rate constant for desorptlon, g/s.
At equilibrium the two rates are equal, and
k.c'
f - } . (2-11)
k C + k
K « W ^ **^
Adsorption rates are usually rapid compared to the other processes
discussed here. However, mass transfer limitations may reduce effective
rates, especially for poorly mixed systems. Lack of sorbent and its satur-
ation may also reduce the effectiveness of adsorption.
For estimating adsorption partitioning, a linear relationship is
assumed (n = 1 or kjC « kg). The equilibrium relationship for biomass is
estimated from an equation of Matt er-Mu Her,9 based on the logarithm of the
octanol-water partition coefficient, LOW. For land treatment and land-
fills, the only partitioning of Importance to fate predictions is gas-
liquid partitioning.
2.4 MIGRATION
Migration occurs when chemicals applied to soils are transported
through the soils to groundwater. Leaching and percolation are the mecha-
nisms that physically remove chemical molecules from a point of deposit and
2-6
-------
carry them toward a water table. Capillary flow is a resisting mechanism
that moves the molecules upward through the soil. The Teachability of a
chemical is a function of soil texture and cation exchange capacity, amount
of soil organic content, amount and intensity of rainfall, and mechanical
placement and adsorptive properties of the chemical.10
2.5 RUNOFF
Chemicals at or near the soil may be washed away by rain. The rate
depends on soil and chemical characteristics and on rainfall rates and
frequency. Clark, Viessman, and Hammer11 state that runoff in any drainage
area is a function of climate and the physical characteristics of the area.
Significant factors include precipitation type; rainfall intensity, dura-
tion, and distribution; storm direction; antecedent precipitation; initial
soil moisture conditions; soil type; evaporation; transpiration; and, for a
given drainage area, its size, shape, slope, elevation, directional orien-
tation, and land use characteristics. If rainfall is heavy shortly after
application of a chemical, runoff and erosion can physically remove it.
The chemical may be dissolved in runoff water, carried along by it, or
adsorbed on eroding soil particles that move with runoff. For pesticide
applications, about 3 to 10 percent of the applied material appears in
runoff water. Below a certain intensity, rainfall will promote leaching of
nonadsorbed chemical into the ground rather than result in runoff.
2.6 BIOLOGICAL DECOHPOSITION
Biological decomposition takes place when microbes break down organic
compounds for metabolic processes. The rate of decomposition depends on
the structure of the compound and on the needs of the microbes. If the
compound is present in excess, the rate of population increase is as
follows:
dx/dt » Rx (2-12)
where
x = concentration of biomass, g/L
R * specific growth rate coefficient, s'1.
If the compound is present in limited amount, the rate becomes a hyperbolic
saturation function of the compound (substrate) concentration:12
2-7
-------
where •
R = maximum specific growth rate coefficient (where substrate
max is in excess), s~*
S s concentration of substrate, g/L
K » substrate concentration at which the rate of substrate
s utilization is one-half of the maximum rate, g/L.
•»•
Because the microbial population increases at the expense of the compound,
the growth rate is proportional to the compound's rate of disappearance.
The rate process may be of zero, first, or mixed order depending on concen-
tration of the substrate. In the presence of multiple substrates, kinetics
become complex.
For the case of S much greater than Ks, the equation approaches zero
order, and Equation (2-13) becomes:
For S much less than Ks, the equation approaches first order:
with Rmax/Ks being the first-order rate constant.
For intermediate values of S, the equation is mixed order, with the
order dependent on values of the constants Rroax and Ks-
2.7 PHOTOCHEMICAL DECOMPOSITION
»
Photochemical decomposition may occur in two ways. A chemical may
absorb light and react (direct photolysis), or the chemical may react
because of light absorption by surrounding elements (indirect photolysis).
For direct photolysis, the rate of reaction of a dilute solution of
chemical in pure water is as follows:
2-8
-------
Kp = b * I ex!x[C], (2-16)
where
K = rate of direct photolysis, g/(L s)
b = unit conversion constant, 3.8 x 10~21 g mol cm4/(L photon)
* = reaction quantum yield, dimensionless
cx = light absorption coefficient at wavelength interval X,
L/(g mol'cra)
Ix = light flux at wavelength interval X, photons/ (cm3«s)
C = concentration of the chemical in water, g/L.
Lymanl3 refers to Zepp and Cline;14 Zepp;15 and Mabey, Mill, and Hendry16
for details of rate calculations in aquatic systems. In these systems, the
rate constant Kp varies with the distribution of sunlight and its inten-
sity. Time of day, season, cloud cover, and latitude all affect Kp so that
a reference condition must be stated; e.g., a light flux of photons per
second corresponding to a cloudless yearly average at a latitude of 40'N.
Reactions may be photocatalyzed. For example, a Ti02 catalyst can be
photoexcited by light at wavelengths less than 360 mm. Oil is17 examined
the degradation of halogenated hydrocarbons with this catalyst and found a
rate equation of. the form:
dl?7dt * £
where
k » photolysis rate constant, g chemical/(L«s«g catalyst)
K^ = apparent binding constant of* a reaction intermediate adsorbed
on the illuminated catalyst surface, L/g chemical.
For 11 halocarbons, values of k ranged from 5.8 x 10'8 to 2.3 x 10'6
g/L»s-g of catalyst, with most about 2.8 x 10~7 to 1.7 x 10'6. A twelfth
halocarbon had a k value of 2.3 x 10'4. Values of K& for the 12 compounds
ranged from 2 to 20 L/g.
2-9
-------
2.8 HYDROLYSIS
Hydrolysis occurs when a chemical reacts with water. For organic
compounds, the reaction usually replaces a functional group (X) with a
RX + H20 = ROH + HX . (2-18)
Reaction rate constants may be pH-dependent; for a specific pH:
kH a ka [H+]" + kn * kb [OH~J (2
where
k., * first-order hydrolysis rate constant, s~*
k * second-order rate constant for acid-promoted hydrolysis,
a L/(g mol«s)
[H+] » hydrogen Ion concentration, g mol/L
n
first-order rate constant for pH- independent neutral
hydrolysis, s'l
kh » second-order rate constant for base-promoted hydrolysis,
. D L/(g mol«s)
[OH~] = hydroxyl ion concentration, g mol/L.
If
kw • [H+] [OH"] (2-20)
where
k * ionlzatlon constant for water s 10-1* g mol2/L2.
Equation (2-19) can be transformed to:
k ^ Tu 1 x u ^ if \f I Tu I i *)
•j — l\a \jn I " K " KL K / in I • \^"
n a now
The rate constant k^ depends on system pH and on the relative values of
kg, and kp.
2-10
-------
2.9 OXIDATION/REDUCTION
Organic compounds in aquatic systems may be oxidized by oxygen (par-
ticularly as singlet oxygen, ^3) or other oxidants such as hydroxyl radi-
cals (OH) and peroxy radicals (R02). The OH radicals tend to be very
reactive, but present only in low concentrations. The R02 radicals are
less reactive than the OH radicals, but are present in greater concentra-
tions. Singlet oxygen is highly reactive, but also selective. It has an
affinity for electron-rich structures such as dienes and substituted
olefins.
The oxidation rate can be calculated as:19
37 - C (kon [RO-] + k«.n [1OJ + k [X]) (2
Qt KU** £ JV £ A
where
kon » rate constant for peroxy radicals, L/(g mol»s)
n v M
[R02] * concentration of peroxy radicals, g mol/L
k-0 » rate constant for singlet oxygen, L/(g mol»s)
[102] « concentration of singlet oxygen, g mol/L
k = rate constant for "other" oxidants, L/(g mol-s)
A
[X] » concentrations of "other" oxidants, g mol/L.
In anaerobic environments, reduction reactions may take place.
Organochlorines are particularly affected. The reduction rate can be
calculated as:20
where
k. » rate constant for reductant 1, L/g mol«s
[R,] » concentration of reductant 1, g mol/L.
2-11
-------
2.10 HYDROXYL RADICAL REACTIONS
Hydroxyl radical reactions may occur through addition of a hydroxyl
radical, abstraction of a hydrogen atom, or both. In the addition, reac-
tion molecules with high electron density portions attract electrophilic
hydroxyl radicals. Hydrogen abstraction takes place when a carbon-hydrogen
bond in an organic molecule is easily broken; it is controlled by elec-
tronic configuration and number of hydrogen reactions in the molecule. The
rate constant for the reaction is often in the range of 6 to 60 x 10^
L/(g mol»s).
A hydroxyl radical reaction rate can be calculated as:21
I - "OH t°H~l c <2'24)
where
kftu » rate constant for hydrogen abstraction or hydroxyl addition,
OM L/(gmol-s).
2.11 REFERENCES
1. ICF, Inc. The RCRA Risk-Cost Analysis Model Phase III Report, Appen-
dix E. Chemical and Physical Processes Affecting Decay Rates of
Chemicals in Aquatic Environments. Draft. Economic Analysis Branch,
U.S. Environmental Protection Agency Office of Solid Waste.
Washington, DC. 1984.
2. Reference 1, p. E-18, Equation (14).
3. Splvey, J. J., C. C. Allen, D. A. Green, J. P. Wood, and R. L.
Stall ings. Preliminary Assessment of Hazardous Waste Pretreatment as
an Air Pollution Control Technique. Research Triangle Institute.
Research Triangle Park, NC. EPA Contract No. 68-03-3149, Task 12-5.
1984.
4. Perry, R. H., and C. H. Chllton. Chemical Engineers' Handbook, Fifth
Edition. New York, McGraw-Hill. 1973.
5. Reference 1, p. E-18 - E-19.
6. Allen, C. C., D. A. Green, and J. B. White. Preliminary Assessment of
Aerated Waste Treatment Systems at TSDFs—Phase I. Draft. Research
Triangle Institute. Research Triangle Park, NC. EPA Contract No. 68-
03-3149, Task 54-01F. 1985.
2-12
-------
7. Farino, W., P. Spawn, M. Jasinski* and B. Murphy. Evaluation and
Selection of Models for Estimating Air Emissions from Hazardous Waste
Treatment, Storage, and Disposal Facilities. GCA/Technology. EPA
450/3-84-020. 1984.
8. Hornick, S. B. In: Land Treatment of Hazardous Waste, Parr, J. F.
(ed). Noyes Data Company. Park Ridge, NJ. 1982.
9. Matter-Muller, C., W. Gujer, W. Geiger, and W. Stumm. The Prog. Wat.
Tech. (Toronto) 12:299-314. lAWPR/Pergamon Press, L.td., Great
Britain. 1980.
10. Reference 8.
11. Clark, J. W., W. Viessman, Jr., and M. J. Hammer. Water Supply and
Pollution Control. Scran'on, PA, International Textbook Company.
1971.
-12. Reference 1, p. E-16, Equation (11).
13. Lyman, W. J., et al. Research and Development of Methods for Esti-
mating Physicochemical Properties of Organic Compounds of
Environmental Concern. Phase II, Part I. NTIS AD 11875A. 1981.
14. Zepp, R. G., and D. M. Cline. Rate of Direct Photolysis in Aquatic
Environment. Environ. Sci. Technol. U.(4):359-366. 1977.
15. Zepp, R. G. Quantum Yields for Reaction of Pollutants in Dilute
Aqueous Solution. Environ. Sci. Technol. 12(3):327-329. 1979.
16. Mabey, W. R., T. Mill, and D. G. Hendry. Photolysis in Water. In:
Laboratory Protocols for Evaluating the Fate of Organic Chemicals in
Air and Water. Draft. U.S. Environmental Protection Agency. EPA
Contract 68-03-2227. 1980.
17. Ollis, D. F. Contaminant Degradation in Water. ES&T. 19(6):480-484,
1985.
18. Reference 13.
19. Reference 1, p. E-12, Equation (2).
20. Reference 1, p. E-12, Equation (3).
21. Reference 1.
2-13
-------
3.0 IMPORTANCE OF PATHWAYS
3.1 INTRODUCTION
The importance of the nine pathways described in Chapter 2.0 for
surface impoundment, open tanks, land treatment facilities, landfills, and
wastepiles is described in this section. The discussion centers on the
pathways used in the emission models described in subsequent sections. The
pathways described in Chapter 2.0 are repeated below for convenience:
• Volatilization
• Adsorption
• Migration
• Runoff
• Biological decomposition
• Photochemical decomposition
• Hydrolysis
• Oxidation/reduction
• Hydroxyl radical reaction.
Section 3.2 presents the relative importance of these pathways based
on the theoretical discussions appearing in Chapter 2.0, the data appearing
in the literature, and engineering judgment. Section 3.3 summarizes in
tabular form the results of the emission model analyses in Chapters 4.0
through 6.0 and the pathways forming the basis for the emission models.
3.2 THEORETICAL BASIS
The relative importance of the nine pathways for TSDF is discussed in
the following text and summarized in Table 3-1. These data were used as
3-1
-------
TABLE 3-1. PATHWAYS FOR HAZARDOUS WASTE AREA EMISSION SOURCES*
Wastewater
treatment plants
Surface
Pathway impoundments
Volatilization
Biodegradation
Photodecompos i t i on
Hydrolysis
Oxidation/
reduction
Adsorption
Hydroxyl
radical
reaction
Migration^
Runoffb
I
I
S
S
N
N
N
N
N
Aerated Nonaerated
I
I
•Jl
S
N
S
N
N
N
I
I
N
S
N
S
N
N
N
Land
treatment
I
I
N
N
N
N
N
N
N
Landfill
I
S
N
N
N
N
N
N
N
I » Important.
S * Secondary.
N * Negligible or not applicable.
aindividual chemicals in a given site type may have dominant pathways dif-
ferent from the ones shown here.
bwater migration and runoff are considered to have negligible effects on
ground and surface water in a properly sited, operated, and maintained RCRA-
pemritted hazardous waste treatment, storage, and disposal facility.
3-2
-------
the basis for the emission models contained in CHEMDAT7. Results of exer-
cising these models to identify pathways of importance are discussed in
Chapters 4.0 through 7.0 and are summarized in Section 3.3. A short dis-
cussion of the theoretical basis for pathways selection follows. Appendix
B presents a more detailed discussion.
3.2.1 Surface Impoundments
Data reported by ICF show predominant removal mechanisms and half-
lives for 71 chemicals. Table 3-2 lists the mechanisms and statistics for
six surface water pathways. Average half-lives range from about 1/2 to 8
days, with predominant mechanisms being volatilization and biodegradation.
The rate of photodecomposition depends on the depth of the surface impound-
ment.. The rate is negligibly low for depths as great as 3 meters and is
indicated in Table 3-1 as S for a secondary effect.
3.2.2 Aerated and Nonaerated Wastewater Treatment
As in the case of the surface impoundments, volatilization and bio-
degradation are potentially significant mechanisms. The relative rates of
these mechanisms depend on the particular component and treatment system.
Photodecomposition is not expected to be a significant pathway due to the
opacity of the system, the depth of the liquid, and the residence time of
the processes. Adsorption is not expected to be significant except for
large loadings of suspended solids and oils in the wastewater. The concen-
trations for many VO are expected to be roughly the same in the biomass as
in the aqueous phase.
3.2.3 Land Treatment
Based on available emission data and literature sources, volatiliza-
tion and biodegradation are expected to be important in land treatment.2-6
For highly volatile constituents, volatilization is expected to be the
predominant pathway; for low volatile constituents, biodegradation is
expected to be the predominant pathway. Adsorption of organic compounds
onto organic carbon in the soil also occurs at land treatment sites. How-
ever, calculations of land treatment air emissions both with and without
consideration of adsorption show a difference of only 10 percent. There-
fore, adsorption is not considered a major pathway for organics removal.
The method of waste application and incorporation into the soil
influence the importance of photochemical reactions in the degradation of
3-3
-------
TABLE 3-2. STATISTICS FOR SURFACE WATER PATHWAYS
Pathway
Oxlda-
Vola- Photo- tlon/
tiliza- Biodegrada- decompo- reduc-
tion tion sitiona Hydrolysis tion Adsorption
Range of
half-lives,
days
Average
half-life
Standard
deviation
Number of
chemicals
0.9-15
2.24
2.85
38
0.04-96
8.05
19.4
26
0.04-900
"••
76.3
1.37
259.0
1.82
12
0.0003-35
5.39
10.8
11
0.1-5
2.05
2.40
<»
4
0.04-1.5
0.55
0.83
3
4Stat1st1cs are given for chemicals with and without an outlier.
3-4
-------
TABLE 3-3. PATHWAYS FOR TSDF SITES
Type of faci1i ty
Pathways included in model
Quiescent storage and treatment impoundments
Mechanically aerated impoundments
Quiescent disposal impoundments
Land treatment facilities
Closed landfills
Active landfills
Wastepiles
Volatilization
Volatilization
Biodegradation
Volatilization
Volatilization
Biodegradation
Volatilization
(diffusion
through cap)
Barometric pumping
Volatilization
(diffusion through
waste)
Volatilization
3-5
-------
organic wastes in land treatment facilities.7 Photodecomposition can occur
in land treatment between application and tilling (usually 24 hours),
although exposure to sunlight is limited to daylight hours. While exact
rates of photodegradation are not known, they are expected to be low. The
oil in which the hazardous materials are suspended is semiopaque to sun-
light, which would tend to keep photodecomposition low. After tilling,
photodegradation is nonexistent because sunlight does not penetrate the
soil surface.8 Consequently, photodecomposition is not expected to be
significant.
3.2.4 Landfills
Volatilization is expected to be a primary VO pathway for landfills.
Biodegradation is expected to be negligible for hazardous waste landfills.
The toxic properties of the water are expected to inhibit biological proc-
esses and therefore biodegradation.9
Rates of diffusion in the gas phase may be important. Components can
diffuse through unsaturated soils (air pockets present). Control of liquid
infiltration into the landfill is expected to keep migration into the soil
at a negligible level.
3.3 EMISSION MODELS
Based on the exercise of CHEMOAT7 in predicting and comparing pathways
for TSDF processes, the pathways shown in Table 3-3 are used as the basis
of the models. Insignificant emissions or inadequate data upon which to
develop the model relationships are the principal reasons for limiting the
models to the pathways shown in Table 3-3.
It should be noted that CHEMOAT7 includes provisions to activate the
unused pathways should further investigations and field tests indicate the
desirability of incorporating additional pathways in the emission models.
3.4 REFERENCES
1. ICF, Inc. The RCRA Risk-Cost Assessment Model Phase III Report,
Appendix E. Chemical and Physical Processes Affecting Accurate Rates
of Chemicals in Aquatic Environments. Draft. Economic Analysis
Branch, U.S. Environmental Protection Agency Office of Solid Waste.
Washington, DC. 1984.
2. American Petroleum Institute. Land Treatment—Safe and Efficient
Disposal of Petroleum Waste. Undated.
3-6
-------
3. Bossert, I., et al. Fate of Hydrocarbons During Oily Sludge Disposal
in Soil. Applied and Environmental Microbiology. 47(4):763-767.
1984.
4. Pelter, P. Determination of Biological Degradability of Organic Sub-
stances. Water Research. 10:231-235. 1976.
5. Dupont, R. Ryon, and J. A. Reinemon (Utah Water Research Laboratory).
Evaluation of Volatilization of Hazardous Constituents at Hazardous
Waste Land Treatment Sites. Prepared for U.S. Environmental Protec-
tion Agency. Ada, OK. August 1986. 157 p.
6. Eklund, B. M., T. P. Nelson, and R. G. Wetherold (Radian Corporation).
Field Assessment of Air Emissions and Their Control at a Refinery Land
Treatment Facility. Prepared for U.S. Environmental Protection
Agency. Cincinnati, OH. DCN 86-222-078-15-07. September 12, 1986.
330 p.
7. Kaufman, D. D. Fate of Toxic Organic Compounds in Land-Applied
Wastes. In: Land Treatment of Hazardous Wastes, Parr, J. F., et al.
(eds). Park Ridge, NJ, Noyes Data Corporation. 1983. p. 77-151.
8. Reference 7.
9. Shen, T. T. Estimation of Hazardous Air Emissions from Disposal
Sites. Pollution Engineering, pp. 31-34. August 1981.
3-7
-------
4.0 SURFACE IMPOUNDMENTS AND OPEN TANKS
This chapter discusses the approach used to estimate air emissions
from surface impoundments and open'top tanks. The emission models are
described, model facilities are defined, and example calculations are
presented.
4.1 NARRATIVE DESCRIPTION OF EMISSIONS AND MODEL UNITS
Emissions from surface impoundments and open tanks originate from the
uncovered liquid surface that is exposed to the air. The model used to
estimate emissions from the liquid surface is based on an overall mass
transfer coefficient that incorporates two resistances to mass transfer in
series—the liquid-phase resistance and the gas-phase resistance. Numerous
correlations are available to estimate the individual mass transfer coeffi-
cients (or resistances), and they depend upon the compound's properties and
the system's parameters. The recommended correlations and their applica-
bility are described in subsequent sections. The emission estimating
procedure also incorporates a flow model that describes the method of oper-
ation. For flowthrough systems, the impoundment's or tank's contents may
be completely mixed, plug flow, or somewhere in between with varying
degrees of backmixing or axial dispersion. Biologically active impound-
ments and aeration tanks can be designed for either completely mixed or
plug flow, and both types of flow models are discussed for these types of
systems. For disposal impoundments, the contents are assumed to be well
mixed, and the bulk concentration is expressed as a function of time. An
expression for biodegradation is incorporated for those units specifically
defined for biodegradation, such as treatment impoundments or wastewater
treatment tanks. For these units, the relative rates of air emissions and
biodegradation are determined to assess the predicted extent of each
mechanism.
4-1
-------
The general approach tha is used to estimate emissions compares the
relative rates of air emissions, biodegradation, and removal with the
effluent. Several different types of model units are presented and include
mass transfer to the air from quiescent, mechanically aerated, diffused-
air, and oil-film liquid surfaces. The other major difference among the
types of model units is the type of flow model that is used. For flow-
through systems, the degree of mixing can range from complete mixing to
plug flow (no mixing), and both cases are presented. For disposal units
with no flow out, emissions are a function of time, and average emissions
are estimat for some specified time since disposal. The major difference
in the emiss jn equations is the liquid-phase concpntration that is used
for the driving .force for mass transfer to the air. The simplest case is
represented by well-mixed systems in which the driving force is represented
by C|_, the liquid-phase concentration in the bulk liquid, which is also
equal to the effluent concentration. Relative removal rates can be com-
pared for this well-mixed case from a simple material balance.
For plug flow, integration is required because the driving force for
mass transfer changes as the liquid flows through the system. This concen-
tration is a function of location or time (which are equivalent in plug
flow) and is expressed as Ct (denoting a dependence on time). The effluent
from a plug flow system is denoted as Ce. For disposal impoundments, the
driving-force concentration changes with time and is also denoted as Ct;
however, there is no effluent from a disposal impoundment. The integration
required for plug flow is from t » 0, when the material first enters the
unit, to t » residence time, when the material leaves the unit. For
disposal units, the integration is from t * 0, when the material is first
placed in the unit, to t » time since disposal, which must be specified to
estimate average emissions. The integrated forms of these emission equa-
tions are very similar.
The well-mixed flow model is recommended and is the model used in the
computer program accompanying this report. This flow model is more gener-
ally applicable than plug f'sow -he calculations > more straightforward,
and the two types give simila- suits. The on' ception is a flow-
through impoundment with an oil 1m surface, wh . uses the plug flow
4-2
-------
model because the oil film inhibits mixing. Both models yield an estimate
of air emissions, biodegradation, and the quantity leaving with the efflu-
ent. It is important to recognize that the quantity leaving with the
effluent may also eventually contribute to air emissions, especially for
treatment units in series or for discharges to streams or publicly owned
treatment works.
Equations are presented for estimating the various removal rates, and
example calculations for different types of impoundments are also provided.
Example calculations are not presented separately for open tanks because
the procedure is analogous to that used for impoundments. In general, open
tanks will have different input parameters that will account for
differences in emission rates compared to impoundments. For example, the
liquid surface area for open tanks will be less, and the fetch-to-depth
(F/D) ratio will be much lower for tanks. If the open tank has a wind
barrier to reduce the wind velocity, the reduced wind velocity can be used
in the mass transfer correlations. In addition, the modeling approach
accounts for the shorter retention times in tanks (on the order of hours)
compared to impoundments (on the order of days). For open tanks, the mass
transfer correlation of Springer is recommended for windspeeds less than
3.25 m/s, and the correlation of MacKay and Yeun is recommended for wind-
speeds greater than 3.25 m/s. Both are discussed in the following section.
4.2 QUIESCENT SURFACES WITH FLOW
4.2.1 Emission Model Equations
The primary focus on emissions from impoundments and wastewater treat-
ment tanks is on aqueous solutions contaminated with organics because
aqueous waste is the most common waste type handled in these facilities.
For aqueous systems, the basic relationship describing mass transfer of a
volatile constituent from the open liquid surface to the air is:
E = KACL (4-1)
where
E » air emissions from the liquid surface, g/s
K = overall mass transfer coefficient, m/s
4-3
-------
A * liquid surface area, m2
CL a concentration of constituent in the liquid phase, g/ra3.
The overall mass transfer coefficient (K) is estimated from a two-
phase resistance model that is based on the liquid-phase mass transfer
coefficient (k|_ in m/s) , the gas-phase mass transfer coefficient (kg in
m/s), and Henry's law constant in the form of a partition coefficient
(Keq). The two resistances act in series and the overall resistance is
expressed as:
where
K * overall mass transfer coefficient, m/s
k|_ » liquid-phase mass transfer coefficient, m/s
kQ » gas-phase mass transfer coefficient, m/s
Keq » equilibrium constant or partition coefficient, concentration in
gas phase/concentration in liquid phase where both concentrations
are in the same units.
Henry's law constant (H in atm»m3/g mol) is estimated for the consti-
tuents of interest by dividing the constituent's vapor pressure (in atmos-
pheres) by its solubility in water (in g mol/m3). The equilibrium constant
is estimated by:
Keq - H/RT (4-3)
where
H » Henry's law constant, atm*m3/g mol
R » universal gas constant, 8.21 x 10-S atm«m3/g mol»K
T • temperature, K.
For a standard temperature of 25 °C, the expression for Keq reduces
to:
Keq » 40.9 x H (4-4)
4-4
-------
The units associated with Keq in Equation (4-4) are the ratio of gas-phase
to liquid-phase concentrations and require that both be expressed in the
same units of mass/volume.
Several mathematical models have been developed to estimate the indi-
vidual liquid- and gas-phase mass transfer coefficients. The models are
based on different systems, constituents, and sometimes different
theoretical considerations. Many of these models yield similar results.
The procedures used in this section to estimate the individual mass trans-
fer coefficients rely primarily on-existing mass transfer correlations that
are believed to be generally applicable.
The liquid-phase mass transfer coefficient (ki_) has been .shown to be a
function of the constituent's diffusivity in water, windspeed, and liquid
depth.1|2 Work performed at the University of Arkansas by Springer et al .3
confirmed these effects and resulted in the correlations given in Table
4-1. Springer used simulation studies in a wind tunnel water tank of a
constant fetch (2.4 m) and variable depth (4.7 cm to 1.2 m). Fetch is
defined as the linear distance across the liquid surface in the direction
of the wind flow, and the F/D ratio is defined as the fetch divided by the
depth of the impoundment. Ethyl ether was used as the volatile component
in the desorption experiments, in which the wind velocity and F/D ratio
were varied. Springer's results shown in Table 4-1 yield three different
correlations for k|_ that depend upon the combination of windspeed and F/D
ratio of interest. Springer's model implies that k|_ is constant for wind-
speeds of 0 to 3.25 m/s. Although Springer examined only the mass transfer
of ethyl ether, his results are extrapolated to other compounds by the
ratio of the compound's and ether's diffusivities in water to the 2/3
power. The windspeed in Springer's correlation is defined as the windspeed
10 m above the liquid surface. For practical application of his
correlation, typically reported values of windspeed are used. Springer's
model does not include the case in which the F/D ratio is less than 14 and
the windspeed is greater than 3.25 m/s. For this specific case, k[_ was
estimated from MacKay and Yeun's correlation shown in Table 4-1.?«8 MacKay
and Yeun9 did not address the effect of depth; however, their correlation
is based on data from 11 organic compounds in a well-mixed system, the
4-5
-------
TABLE 4-1. EQUATIONS FOR CALCULATING INDIVIDUAL MASS TRANSFER
COEFFICIENTS FOR VOLATILIZATION OF ORGANIC SOLUTES FROM
QUIESCENT SURFACE IMPOUNDMENTS
Liquid phase
Springer et al.4 (for all cases except F/D<14 and Uio>3.25 m/s):
2/3
(0 < U1Q<3.25) (m/s)
(All F/D ratios)
kL » 2.78 x 10
-6
r o.
w
'ether
"9
"7
kL « [2.605 x 10" (F/D) * 1.277 x 10"] u
r D
'ether
2/3
(U1(J>3.25) (m/s)
(143'25>
etherj
where
windspeed at 10 m above the liquid surface, m/s
ON * diffusivity of constituent in water, cm2/s
Dether * diffusivity of ether in water, cm2/s
F/D * fetch-to-depth ratio (fetch is the linear distance across the
impoundment).
Gas phase
MacKay and Matasugu (in HwangS):
,0.78 c--0.67.4-0.il *
4.82 x 10 J U'
p_—«.W/ J'
5cr d
6 e
(m/s)
where
U * windspeed, m/s
» Schmidt number on gas side
* viscosity of air, g/cm«s
(continued)
4-6
-------
TABLE 4-1 (continued)
PQ * density of air, g/cm3
Da = diffusivity of constituent in air, cm2/s
f I0'5
4A
de * effective diameter of impoundment = — , m
I J
A = area of impoundment, m2.
Liquid phase
MacKay and Yeun6 (for F/D <14 and Uio>3.25 m/s):
kL = 1.0 x 10-6 + 34.1 x io-4 u* scL-0.5 (u*>0.3) (m/s)
kL = 1.0 x ID'6 + 144 x 10'4 U*2-2ScL-0-5 (u*<0.3) (m/s)
where
U* s friction velocity (m/s) = 0.01 UIQ (6.1 + 0.63
s windspeed at 10 m above the liquid surface, m/s
/*.
Sc. = Schmidt number on liquid side = — s-
L />L
/t, - viscosity of water, g/cm»s
p. = density of water, g/cm3
Dw = diffusivity of constituent in water, cm2/s.
4-7
-------
compounds represent a broad range of Henry's law constants, and their
general correlation is applicable for the case described above that is not
covered by Springer's correlation.
The gas-phase coefficient (kg) was estimated from the correlation of
MacKay and Matasugu as shown in Table 4-1.10 This correlation was devel-
oped from experiments on the evaporation of isopropyl benzene, gasoline,
and water into air. These researchers verified that previous work, which
assumed that the wind velocity profile follows a power law, could be used
to quantify the rate of evaporation from a smooth liquid surface. The
result was a correlation that expressed kg as a function of windspeed and
the fetch or effective diameter of the liquid surface.
The individual mass transfer coefficients estimated from the correla-
tions in Table 4-1 are used in Equation (4-2) to estimate the overall mass
transfer coefficient. The equilibrium constant for a constituent dissolved
in water at 25 °C is estimated from Equation (4-4). However, an estimate
of the concentration in the liquid phase (CL) is needed in Equation (4-1)
to estimate emissions.
The concentration CL in Equation (4-1) is the driving force for mass
transfer. For an impoundment that is instantly filled with waste, the
driving force (CL) is the initial concentration in the waste. However,
this concentration will decrease with time as the constituent is lost to
the air, which suggests that emissions may also decrease with time (assum-
ing constant K and A). For flowthrough systems, the concentration may be
cyclical if the loading of the process is cyclical. Continuous flowthrough
systems may attain some equilibrium concentration.
The flow model assumed for quiesdent impoundments and tanks with no
biodegradation is that the contents of the system are well mixed and that
the bulk concentration (driving force) in the system is equal to the
effluent concentration. A material balance around this system yields:
QC0 - KACL + QCL (4-5)
or
CL - QC0/(KA+Q) (4-6)
4-8
-------
where
Q * volumetric flow rate,
C0 s initial concentration in the waste,
C|_ = equilibrium or bulk concentration in the impoundment, g/m3
K - overall mass transfer coefficient, m/s
A = liquid surface area, m2.
The well-mixed assumption is.made for the sake of simplicity and
assumes that bulk convection and wind-induced eddies combine to mix the
basin contents. Axial dispersion in the flow direction is also possible,
and some systems may be designed specifically for plug flow (e.g., some
biological treatment tanks). An assumption of plug flow instead of well-
mixed flow would yield slightly higher estimates of emissions; however, the
difference is small. Calculations presented by Thibodeaux for an aerated
basin that was well-mixed or had plug flow showed that the plug-flow
assumption yielded estimates that were higher by 11 percent for acetalde-
hyde, 5 percent for acetone, and 0 percent for phenol.H
The approach described to estimate emissions from quiescent impound-
ments with no biodegradation includes the following steps:
1. Estimate the individual mass transfer coefficients from
Table 4-1.
2. Estimate the equilibrium constant from Equation (4-3).
3. Estimate the overall mass transfer coefficient from Equation
(4-2).
4. Estimate the liquid-phase concentration from Equation (4-6).
5. Estimate emissions from Equation (4-1).
The major assumptions associated with this procedure are:
• The two-resistance model and the correlations for the
individual mass transfer coefficients are applicable to the
system of interest.
• The impoundment's contents are well mixed.
• There is no significant removal by biodegradation, seepage,
adsorption, or other forms of degradation.
4-9
-------
• The waste material of interest is aqueous waste with no
separate organic phase.
• The estimate of Henry's law constant (equilibrium partition-
ing between the vapor and liquid) is reasonably accurate.
The recommended procedure for quiescent impoundments is to assume that
the liquid is well mixed. This assumption is used in the computer model
accompanying this report and is illustrated in the example calculations.
However, impoundments and tanks with quiescent surfaces can also be
designed for plug flow with the use of baffles or other design techniques
to reduce the extent of backmixing. In a plug-flow system, the rate of air
emissions at any point in the system changes as the material flows through
the system. There is no uniform liquid concentration within the plug-flow
unit as there was in the well -mixed system, and the lowest concentration
occurs in the effluent (i.e., there is no backmixing of the effluent with
the influent). For plug flow, the rate of disappearance of a compound by
air emissions is given by:
.KAC (4-7)
dt
where
Ct s concentration after the plug has traveled t seconds
t * time, s
V * volume, m3
and with the other symbols as previously defined.
Rearranging Equation (4-7) yields :
H P
- £ * (-KA/V)dt (4-8)
Ct
Integrating Equation (4-8) from Ct « C0 at t * 0 to Ct - Ce at t = V/Q (one
residence time) gives:
Ce/CQ - exp (-KA/Q) (4-9)
4-10
-------
where
Ce * effluent concentration,
and with the other symbols as previously defined.
The residence time, r in seconds, equals V/Q and V = AD (area times
depth); consequently, A/Q = r/D. Substituting into Equation (4-9) yields
an equivalent expression:
Ce/CQ « exp (-Kr/D) . (4-10)
The ratio Ce/Co represents the fraction removed with the effluent; there-
fore, 1 - Ce/C0 represents the fraction that is emitted (fair) from the
plug-flow system:
fair « 1 - Ce/CQ « 1 - exp (-Kr/D) . (4-11)
The average emission rate is calculated from:
E - fa,r Q C0 (4,12)
where
E * emissions, g/s
f . = fraction emitted from Equation (4-11)
air
Q = flow rate, m^/s
C = influent concentration, g/m3.
4.2.2 Model Plant Parameters for Quiescent Impoundments
A model facility was developed for quiescent impoundments to illus-
trate the emission estimating procedure. A 1981 survey compiled by
Westat*2 showed that the median surface area for storage impoundments was
approximately 1,500 m2 and that the median depth was 1.8 m. Detention
times ranged from 1 to 550 days, with over half of the values at 46 days or
less. For this example, a detention time of 20 days was chosen. The area
and depth yield a total volume of 2,700 m3, and the detention time of 20
days yields a flow rate of 1.6 L/s (0.0016 m3/s).
Meteorological conditions are also needed as input parameters for the
emission models. For this emission estimate, a standard temperature of
25 °C and a windspeed of 4.47 m/s (10 mi/h) were used. Benzene was chosen
4-11
-------
as an example constituent at a concentration of 10 g/m3 (10 ppm) to
estimate emissions from the model facility. The properties of benzene that
are used include Henry's law constant (5.5 x 10'3 atm«m3/g mol), diffusiv-
ity in air (0.088 cm2/s), and diffusivity in water (9.8 x 10'6 cm2/s).
Table 4-2 lists the input parameters for the estimate of emissions given in
Section 4.2.3.
4.2.3 Example Calculation for Storage Impoundments
This section presents a step-by-step example calculation for emissions
from storage impoundments. The equations described in Section 4.2.1 are
used with the model unit parameters given in Section 4.2.2 to estimate
emissions from an aqueous waste containing 10 g/m3 of benzene.
a. Calculate liquid-phase mass transfer coefficient, k|_. Use Springer's
model (see Table 4-1):
Effective diameter
•M
0.5
,0.5
x 2
F/D * Effective diameter/depth
43.7
1.500
I ' J
* 24.3
x 2 * 43.7 m
Windspeed = 4.47 ra/s (UIQ I 3.25 m/s)
F/D = 24.3
"9
"7
kL - [2.605 x 10" (F/D) + 1.277 x 10"] U1Q
where
UHJ • windspeed » 4.47 m/s
0* « 9.8 x 10'6 cm2/s (benzene)
Dether • 8-5 x 10'6 - cm2/s (ether)
F/D - 24.3.
w
'ether
0.67
m/s
Then
k. - [2.605 x 10'9 (24.3) + 1.277 x 10"7] (4.47)
9.8 x 10
-6
8.5 x 10
-6
0.67
4-12
-------
TABLE 4-2. INPUT PARAMETERS—STORAGE IMPOUNDMENT
Area 1,500 m2
Depth 1.8 m
Volume 2,700 m3
Retention time 20 days
Flow 0.00156 m3/s
Temperature 25 °C
Windspeed 4.47 m/s
Constituent Benzene in water
Concentration 100 g/m3
Henry's law constant 5.5 x 10-3 atm«m3/g mol
Diffusivity in air (benzene) 0.088 cn»2/s
Oiffusivity in water (benzene) 9.8 x 10-6 cm2/s
Diffusivity in water (ether) 8.5 x 10-6 cn»2/s
Viscosity of air 1.81 x 10~4 g/cm-s
Density of air 1.2 x 10'3 g/cm3
4-13
-------
kL » [2.605 x 10-9 (24.3) + 1.277 x 10'7] (4.47)2 (1.1)
kL a 4.2 x 10-6 m/s
b. Calculate gas-phase mass transfer coefficient, kg. Use MacKay and
Matasugu (see Table 4-1):
kG = 4.82 x 10-3 uO-78 $c-0.67 ^-0.11^)
where
U « windspeed, 4.47 m/s
^ Schmidt No. viscosity of gas _
b G * for gas (gas density) (diffusivity of i in gas)
Gas = air
Viscosity (air) = 1.81 x 10"4 g/cm»s
Density (air) = 1.2 x 10*3 g/cm3
Diffusivity (benzene in air) = 0.088 cm2/*
Sc 1.81 x 10 "4g/cm»s s
>
G (1.2 x 10"3 g/cm3) (0.088 cm2/s)
de * effective diameter * 43.7 m
Then
kG = (4.82 x 10-3) (4.47)0-78 (l.71)-0.67 (43.7)-O.H
= 7.1 x 10-3 m/s
c. Calculate overall mass transfer coefficient (K) from Equation (4-2)
11. 1
K k[ Keq
where
Keq -. -H s _Li.' 10'3. n»3*atm/mol
(8.21 , -O'5) ~- (298 K)
4-14
-------
Then
5
» - - - - - ,-
K 4.2 x 10'6 (0.225) (7.1 x 10'J)
2.39 x 10
K = 4.2 x 10-6 ra/s .
d. Estimate emissions for a well-mixed system:
QC - KC.A + QCL (from material balance of Equation (4-5))
c
CL
Detention time * 480 h
Volume = 2,700 m3
where
Q = flow rate - - 0-00156 »,3/s
C0 « 10 g/m3
K - 4.2 x 10-6 m/s
c = _ (0.00156 m3/s)(10 q/m )3 _ = ig 8 g/n,3
L (4.2 x 10"6 m/s) (1, 500 m2) + (0.00156 m3/s)
A * 1,500 m2 .
Air emissions - KCtA (Equation 4-2)
> (4.2 x 10-6% m/s) (19.8 g/m3)(l,500 m2) = 0.012 g/s
= 3.9 Mg/yr .
e. Estimate emissions for a plug-flow system:
f.i.. s 1 - exP (-K^/D) (Equation 4-11)
Cll P
K = 4.2 x 10"6 m/s (Step c)
r * 480 h = 1.73 x 106 s
D - 1.8 m
4-15
-------
fair a l " exp ("4'2 x 10"6 m/s-1*73 x 106s/1.8 m) * 0.98
E s f*i* Q Crt (Equation 4-12)
a 11 u
fair ' °'98
Q =» 0.00156 m3/s
CQ » 100 g/m3
E - (0.98)(0.60156 m3/s)(10 g/m3)
E » 0.15 g/s - 4.7 Mg/yr .
4.3 BIODEGRAOATION
This section identifies some of the major design features of biologi-
cal treatment processes, such as activated sludge units and impoundments
designed for biodegradation. Mathematical models for biodegradation are
also presented and incorporated into predictive fate models.
4.3.1 Description of Biological Active Systems
The activated sludge process is an aerobic biological treatment in
which the pollutants are degraded by microorganisms suspended uniformly in
the reaction tank. Oxygen is introduced by mechanical means, and the
microorganisms are maintained by recycling the activated sludge that is
formed. In most units, the sludge is removed by settling in a separate
unit, a portion of the sludge is recycled, and a small portion is wasted
(removed from the system) on a continuous basis. Oxidation or stabiliza-
tion impoundments and aerated impoundments are used to treat entire plant
wastes as well as to polish the effluent from other treatment processes.
Solids usually settle out in the impoundment or are removed in a separate
vessel. Generally, the solids are not recycled; however, if the solids are
returned, the process is the same as a modified activated sludge process.13
Typical design parameters for an activated sludge process are given in
Table 4-3. Two of the most commonly used parameters are the food-to-
microorganism (F/M) ratio and residence time. The F/M ratio describes the
organic loading on the biological system and is calculated as the weight of
BOOs (biochemical oxygen demand from a 5-day test) that enters the system
in a 24-hour period divided by the total weight of biological solids in the
system. The biological solids may be roughly estimated from the mixed
4-16
-------
TABLE 4-3. DESIGN PARAMETERS FOR ACTIVATED SLUDGE PROCESSES14
Process
Convent ionalc
CSTRd
Contact
stabilization
Extended aeration
02 systems
F/M,a
kg BOD/ kg
biomass»day
0.2-0.4
0.2-0.6
0.2-0.6
0.05-0.15
0.25-1.0
Loading,
kg BOD/nH»day
0.3-0.6
0.8-2.0
1.0-1.2
0.1-0.4
1.6-3.3
MLSS,b
9/L
1.5-3.0
3.0-6.0
1.0-3.Q6
4.0-1QT
3.0-6.0
6.0-8.0
Retention
time, h
4-8
3-5
0.5-16
3-6f
18-36
1-3
aF/M * Food to microorganism ratio.
&MLSS * Mixed liquor suspended solids.
cPlug flow design.
dCSTR - Continuous stirred-tank reactor.
eContact unit.
^Solids stabilization unit.
4-17
-------
liquor suspended solids (MLSS) if substantial quantities of inorganics
(such as silt) are not present. If inorganic solids are present, the
biological solids may be better approximated by the mixed liquor volatile
suspended solids (MLVSS).15 For municipal wastewater systems, the volatile
solids comprise about 60 to 80 percent of the total suspended solids in the
sludge; consequently, in the absence of a direct measurement of MLVSS, the
biological solids in municipal wastewater can be estimated as 60 to 80
percent of the total suspended solj.ds.16 Conventional plants, which use an
activated sludge process that has long and narrow basins designed to
approach plug flow, operate with an F/M ratio of 0.2 to 0.4, but values as
low as 0.05 are not unusual. High F/M values indicate a high loading, as
from a sudden influx of organics or the loss of biological solids, and will
lead to a deterioration in effluent quality.17
Aeration tanks are usually constructed of reinforced concrete, are
open to the atmosphere, and are usually rectangular in shape. Treatment
plants may consist of several tanks, operated in series or parallel. Some
of the largest treatment plants may contain 30 to 40 tanks arranged in
several groups or batteries.^
Typical parameters associated with biologically active impoundments
are given in Table 4-4. The loading parameter is expressed in terms of
area or volume, and typical retention times in aerated impoundments range
from 7 to 20 days. The level of suspended solids in these impoundments is
over an order of magnitude less than the level in activated sludge proc-
esses. Although the parameters in Table 4-4 are listed as "typical." large
variations exist among real facilities, and at a single facility the values
may change with time. For example, a'study conducted over 12 months at an
aerobic impoundment used to treat municipal wastewater reported suspended
solids levels of 0.02 to 0.1 g/L and volatile suspended solids of 0.01 to
0.06 g/L.21 Another study of eight quiescent impoundments at four differ-
ent sites with confirmed biological activity estimated active biomass
concentrations from the rate of oxygen consumption that ranged from 0.0014
to 0.22 g/L with an average of 0.057 g/L.22
The bioniass concentration is an important parameter in estimating
biodegradation rates. The best value to use for a specific site is a
4-18
-------
Typ*
Aerated
Aerobic
TABLE 4-4. IMPOUNDMENTS DESIGNED FOR BIODEGRADATION"<2*
Appl1cation
Facultative
Raw municipal wastewater
Effluent from primary
treatment, trickling
filters, aerated ponds,
or anaerobic ponds
Industrisl wastes
Overloaded facultative
ponds
Situations where limited
land area is available
Generally used to trest
effluent from other
processes, produces
effluent low in soluble
BODg and high in algae
solids
Typical daily
loading,
kg BOOB/m3»d.y
0.0011 - 0.0034*
0.008 - 0.32
0.021 - 0.043b
Retention
time, d
Typical depth, m
26-180
1.2-2.6
Suspended
so11 da, g/L
0.11-0.40
7-20
2-0
0.26-0.30
10-40
0.3-0.46
0.14-0.34
Anaerob i c
Industrial wastes
•Based on a typical depth of 2 m.
^Based on a typical depth of 0.4 m.
0.16 - 0.80
20-60
2.6-6
0.08-0.16
-------
direct measurement such as volatile suspended solids for the system of
interest. In the absence of site-specific data, a number may be chosen
from the ranges for suspended solids given in Tables 4-3 and 4-4. Alter-
natively, typical or default values for biomass concentration given in
Table 4-5 may be used.
The major mechanisms of organic removal in biologically active systems
include biodegradation, volatilization, removal with the effluent, and
removal by adsorption on the waste sludge. A study by Petrasek et al. of
purgeable volatile organics in a p'l lot-scale wastewater treatment system
showed that less than 0.4 percent (generally less than 0.1 percent) of the
volatiles were found in the waste-activated sludge.23 Bishop, in a study
of municipal wastewater treatment, concluded that only a modest amount of
purgeable toxics were transferred to the sludge,24" Hannah et al.25 found
that the concentrations of volatile organics in sludges from pilot-scale
systems were generally comparable to or less than the corresponding concen-
trations in the process effluent. This indicated that volatile organics do
not have a high affinity for wastewater solids and do not concentrate in
the sludges. Kincannon and Stover found that 0 to 1 percent of three
compounds (1,2-dichloroethane, phenol, and 1,2-dichlorobenzene) was
adsorbed on the sludge.26 Melcer, in a review of biological removal
studies, concluded that polycyclic aromatic hydrocarbons, pyrene, anthra-
cene, fluoranthene, and chrysene were the most commonly occurring priority
pollutants found in sludges.27 These studies suggest that the compounds
most likely to be emitted to the air (volatiles) do not concentrate on
sludges; however, some of the relatively nonvolatile organics may be
adsorbed. Consequently, the model ing,approach presented in this section
assumes that the removal of volatile organics with the waste sludge is not
significant. The major removal mechanisms that are considered include
volatilization, biodegradation, and removal with the effluent.
4.3.2 Rate of Biodeqradation
Numerous models have been proposed for the removal of organic com-
pounds by biodegradation and include design equations for activated sludge
systems and stabilization or oxidation impoundments.28,29 There is general
agreement in the literature that, for high organic loadings relative to
4-20
-------
TABLE 4-5. TYPICAL OR DEFAULT VALUES FOR
BIOMASS CONCENTRATION
Unit Biomass concentration, g/L
Quiescent impoundments 0.05^
Aerated impoundments 0.25C
Activated sludge units 4.0d
^These values are recommended for use in the emission equa-
tions when site-specific data are not available.
&8ased on the range (0.0014 to 0.22) and average (0.057)
from actual impoundments as discussed in the text.
cprom the data in Table 4-4 for aerated impoundments.
Assumes biomass is approximated by the suspended solids
level. Range is typically 0.05 to 0.30.
dMidrange value from Table 4-3 for CSTR based on mixed
liquor suspended solids.
4-21
-------
biomass, the biodegradation rate is zero-order with respect to
concentration (i.e., the rate is independent of organic concentration).
For lower residual levels, the rate becomes first order with respect to
concentration or follows Monod-type kinetics.30-31.32 The Monod-type
biodegradation rate equation can be written as follows:
where
rg = biodegradation rate, g/s
V » volume, m3
bj » biomass concentration, g/m3
Kfflax * maximum rate constant, g/s-g biomass
C * component concentration, g/m3
Ks » half saturation constant, g/m3.
The Monod model was originally developed to describe microbial growth
rates for a single microbial population based upon a single, rate-limiting
substrate. A yield coefficient was subsequently employed to determine the
utilization rate of that substrate. For convenience of use, the biodeg-
radation rate model given in Equation (4-13) has been written directly for
component disappearance in terms of overall biomass concentration. It is
assumed that Equation (4-13) applies to each organic constituent in the
waste (although the rate constants wi\l be different for each constituent),
and that the biodegradation of any one constituent is independent of the
concentrations of other constituents. Subsequent references to the Monod
or the Monod-type model in this report refer specifically to Equation
(4-13). The significant features of this model are:
1. At high concentrations (specifically, C » Ks), C dominates
the denominator and can, therefore, be eliminated from Equa-
tion (4-13). The biodegradation rate is then independent of
(i.e., zero order with respect to) the component concentra-
tion.
4-22
-------
At low concentrations (C « Ks), Ks dominates the denomi-
nator, and the biodegradation rate becomes directly propor-
tional (i.e., first order with respect) to the component
concentration. The apparent first-order rate constant is:
Theoretical Monod curves for several different compounds are presented
in Figure 4-1 to illustrate these features.
A literature review was conducted to determine appropriate rate
constants for the Monod model. References that served as primary sources
of biodegradation rate data included: Pitter,33 Kincannon et al.,34
Petrasek et al.,35 and Hannah et al.36 Data obtained from each reference
included rate constants as reported, influent concentrations, effluent
concentrations, biomass concentration, retention time (RT), and fraction of
the amount of component removed by biodegradation (Fg). Using this infor-
mation and field data collected during specially designed biodegradation
rate studies, Coburn et al . developed a base of component-specific biologi-
cal removal rates that contains nearly 500 entries and removal data for 90
different organic constituents. 37
For most compounds, there were inadequate biodegradation rate data to
determine the Monod rate constants using traditional methods (e.g.,
Lineweaver-Burke plot). However, when reported, values for Ks were
generally between 1 and 10 mg/L for a variety of different compounds.
Thus, the Monod constant, Kmax, was calculated from organic removal data
when high concentrations (C > 10 mg/L) were employed by assuming strict
zero-order kinetics as follows:
Kmax • Wo - CL)/[(RT)bi] (4-14)
»
where
FB * the fraction of component removal attributed to biodegradation
C0 = inlet concentration, g/m3
CL = bulk liquid and effluent concentration, g/m3
(RT) = residence time, s.
Note that, with zero-order kinetics, Equation (4-14) applies to both
continuous, well-mixed systems and to plug-flow and batch systems.
4-23
-------
ro
CD
a*
01
Chloroform
-3
Natural Log of Concentration (in mg/L)
Figure 4-1. Theoretical relationship between concentration and btodegradatlon rates normalized by
the amount of biomass as predicted using the Monod model for phenol, benzene, and chloroform.
-------
The half-saturation constant Ks was estimated (knowing Kmax) from the
apparent first-order rate constants when low concentrations (C < 1 mg/L)
were present (specifically, Ks - Kmax/Ki). The equation used to calculate
the apparent first-order rate, Kj, depends on the type of experimental
system that was employed. For continuous, well-mixed systems, KI was
calculated as follows:
K! - FB(Co - CL)/[(RT)biCL] (4-15)
For batch systems and for continuous, plug-flow systems, the equation used
to calculate KI was:
KI - FB ln(C0/CL)/[(RT)bi] . (4-16)
Using this approach, rate constants for specific compounds in the
biodegradation rate data base were determined. These rate constants are
provided in Appendix D, Table 0-1. Upon evaluating the biodegradation rate
data from several different laboratory and field studies, it is recognized
that biodegradation rates can vary widely from site to site. Therefore,
the following priority schedule is provided as guidance in determining the
appropriate biodegradation rate constants to be employed in the emission
models:
• Use site-specific biodegradation rate data in experiments
controlled for air emissions where available.
• Use the rate constants suggested in Appendix D, Table 0-1,
as available. ,
• Estimate the biodegradation rate constants using the follow-
ing methodology:
Approximate Kmax from available data for Kmax for com-
pounds of similar structure and/or functional groups;
and
Approximate KJ either by using the correlation:
K! = 3.75 x 10-8 K0wO-38 (4-17)
4-25
-------
where
Kow = octanol -water partitioning coefficient,
or by using the default (average) value for Kj, which
is: iq = 1 L/h/g (2.78 x 10'7 m3/s/g) , and then
calculate Ks as: Ks = Kmax/Kj.
The correlation provided in Equation (4-17) was developed based upon
the assumption that biodegradation was primarily an intracellular phenom-
enon. As such, the first-order bi'oxlegradation rate can be limited either
by the rate of the internal reaction or by the rate of diffusion of the
chemical through the cell membrane and into the cell. If the internal
component concentrations are assumed to be proportional to the concentra-
tion of components absorbed onto the cell membranes, then, regardless of
what limits the first-order biodegradation rate, the limiting first-order
biodegradation rate will be directly proportional to the concentration of
constituent absorbed onto the membrane. Because the octanol-water parti-
tioning coefficient has been used to correlate the absorption partitioning
of organic chemicals onto biomass.38,39 it follows that the octanol-water
partitioning coefficient may also be used to correlate the limiting first-
order biorate constant since the observed b iorate is based on bulk liquid
concentrations. To that end, the limiting first-order rate constants for a
variety of compounds were plotted versus their corresponding octanol-water
partitioning coefficient. The results, presented in Figure 4-2. indicate a
fair correlation between the octanol-water partitioning coefficients and
the limiting first-order rate constants for most compounds. The primary
discrepancies are for ionizable or poUr compounds.
Assuming continuous, steady-state operation for a system that is well-
mixed, a mass balance on the system can be written as follows:
QC0 * QCL + VKnjaxbiCt/fKs+CL) + Kotner V CL (4-18)
where
Q « flow rate, m^/s
K0ther = sum °f apparent first-order rate constants for competing
mechanisms, 1/s.
and the other symbols are as previously defined.
4-26
-------
ro
22
20 -
18 -
16 -
| '^
£10-
£ 8-
o
o
e-
4 -
•S 2-
1 •
e -2-
1 -4-
-6
BIS(2-ETHYLHEXYl.)PHTHALATE
DIN(TROPHENOL(S
Correlation:
.4)
0.135 K
0.38
ow
CHU)ROBEN2ENE
* CRESO
PHENOL1
ETONE
ACRYLONITRILE
-3-11 3
Natural Log of First Order Rate Constant (L/h/g biomass)
Note: Dashed lines represent a factor of 5 error.
Figure 4-2. Correlation of limiting first-order rate constant with the octanol water coefficient.
-------
Note that Equation (4-18) was written in a general fashion so that, if
desired, the rate of removal via adsorption onto biomass solids can be
included. For most volatile organics, however, the adsorption pathway is
negligible so that K0tner is dominated by the volatilization rate. Conse-
quently,
Kother * KA/V (4-19)
where
K » overall mass transfer coefficient, m/s
A * area, a?.
To determine the fraction of VO emitted or biodegraded using the Monod
model, one first has to solve for the effluent concentration. The effluent
concentration can be determined by rearranging Equation (4-18) as follows:
K'CL2 * [KSK' + (V/Q)Kraaxbi - C0]CL - KSC0 * 0 (4-20)
where
K' * (Kother)(V/Q) + 1, dimensionless.
Equation (4-20) is easily solved using the quadratic formula as
follows:
CL * [-b + (b2 - 4ac)0-5]/2a (4-21)
where
« • K' - (Kother)(V/Q) + 1
b - KSK' * (V/Q)Kmaxbi - C0
c - -KSC0.
The plus sign is selected in Equation (4-21) to ensure positive
effluent concentrations. Note that, because all of the rate constants and
concentrations mist have positive values, the constant, c, must be negative
so that the quadratic equation always has real, positive roots.
4-28
-------
Once the effluent concentration is calculated, the fraction of the
component feed emitted to the air (fair) is:
fair = K A CL/Q C0 . (4-22)
Emissions (E, g/s) are calculated from:
E * fairQCo • (4~23)
Similarly, the fraction of the component feed biodegraded (fbio) is:
io = VKmaxbi/[(Ks^L)QC0] (4-24)
If the biological system is operated with plug flow, the treated
wastewater does not mix with the influent. The biodegradation rate and air
emission rate change as the .treatment progresses toward completion. For
plug flow, the rate of disappearance of a compound by biodegradation and
air emissions is given by:
Wl^ K P
, max i t + KA c (4.25)
dt (Ks + Ct)
where
Ct - concentration at time - t
t = time, s
and with the other symbols as previously defined. Due to the nonlinear
nature of the biodegradation rate term, Equation (4-25) cannot be directly
integrated. Therefore, it is further assumed that first-order kinetics
dominates the system's biodegradation. Equation (4-25) can then be
rearranged as follows:
£ = (-Kjb. - KA/V) dt (4-26)
Ct
where
K! * Kmax/Ks, m3/g biomass.
4-29
-------
Integrating Equation (4-26) from Ct = C0 at t = 0 to Ct = Ce (effluent
concentration) at t - V/Q (one residence time) gives:
Ce/CQ = exp (-Kjb.V/Q - KA/Q) (4-27)
The ratio Ce/C0 represents the fraction leaving with the effluent; conse-
quently, 1 - Ce/C0 represents the sum of the fractions that are biodegraded
and emitted to the air. The fractions of component feed emitted to the air
and biodegraded are calculated from their relative rates:
fair = (1 * VCo)(KA)/(KA * Klbiv) <4-28>
fbio = (1 * Ce/CQ)(K1b1.V)/(KA + Kjb.V) (4-29)
The average emissions rate (E, g/s) is:
E • fair QC0 - (4-30)
4.3.3 Example Calculation for Quiescent Impoundments
The application of the biodegradation model to quiescent impoundments
is presented in the form of an example calculation. The calculation is
based on the quiescent impoundment's operating parameters from Table 4-2.
For other types of impoundments, the application of the biodegradation
model is illustrated in subsequent sections.
The waste stream for the example calculation is defined as containing
benzene at 10 ppm with a total organic content of 250 ppm (0.25 g/L). The
resultant organic loading on the impoundment on a daily basis is
12.8 kg/1,000 m3. The active biomass is assumed to be 0.05 g/L from a
reported range from eight quiescent impoundments of 0.0014 to 0.22 g/L.
a. Calculate the effluent concentration of benzene for a well-mixed
system from Equation (4-21):
CL « [-b «• (b2 - 4ac)0-5] /2a
where
a = K1 = (KA/V) (V/Q) + 1 = KA/Q + 1
b = K5K' «• (V/Q) Kmax bi - Co
4-30
-------
c * -KSC0
K = 4.2 x 10-6 m/s (Section 4.2.3, Step c)
A = 1,500 m2
Q = 0.00156 m3/s
Kmax s 19 mg/g/L » 5.28 x 10-6 g/g/s (from Appendix D, Table
D-l)
Ks - 13.6 mg/L = 13.6 g/n»3 (from Appendix D, Table D-l)
bi = 0.05 g/L = 50 g/m3
V = 2,700 m3
Co - 100 ppm - 100 g/m3
KA » (4.2 x 10-6 m/s)(1,500 m2) * 6.3 x 10-3 ra3/s
a » K1 » (6.3 x 10-3 ra3/s)/(0.00156 m3/s) + 1 » 5.0
b = (13.6 g/m3)(5.0) + (2,700 ra3/0.00156 m3/s) (5.28 x
10-6 g/g/s) (50 g/m3) - (100 g/m3)
= 425 g/m3
c = -(13.6 g/ra3)(100 g/ra3) = -1,360 g2/m6
CL s {-[425 g/m3] + [(425 g/ra3)2 . 4(5.0)(-1,360 g2/m6)]0.5}/
[2(5.0)]
= (-425 g/m3 * 517.6 g/m3)/10
- 3.08 g/m3
b. Calculate the fraction emitted for a well-mixed system from
Equation (4-22):
fair - KACL/(QC0)
where
»
fair - (6.3 x 10-3 m3/s)(3.08 g/m3)/[(0.00156 m3/s) (100 g/n»3)]
fair s 0.124 .
c. Calculate benzene emissions for well-mixed system:
E(g/s) = fair Q C0
» (0.124)(0.00156 m3/s)(100 g/m3)
= 1.93 x 10-2 g/s = 0.61 Mg/yr
d. For a plug-flow system, calculate fraction removed with the
effluent from Equation (4-27):
4-31
-------
Ce/Co - exp (- KI bi V/Q - KA/Q)
where
KI » 1.4 L/g-h » 3.89 x 10-7 m3/g-s (from Appendix D,
Table D-l)
bi » 0.05 g/L * 50 g/m3
V - 2,700 m3
Q - 0.00156 m3/s ...
Co * 10 ppm « 10 g/m3
K » 4.2 x 10-6 m/s
A • 1,500 m2
KibiV - (3.89 x 10-7 m3/s/g biomass)(50 g/m3)(2,700
- 5.25 x 10-2 m3/s
KA - (4.2 x 10-6 m/s)(1,500 m2) « 6.3 x 10-3 m3/s
C /C - exp f -5.25 x 1Q-2 m3/s m 6.3 x 1Q-3 m3/s
e ° I 1.56 x 10-3 ni3/s 1.56 x 10'3 m3/s
Ce/CQ - exp (-37.7) » 0.00 .
e. Calculate fraction emitted from Equation (4-28):
fair « (1 - Ce/Co)(KA)/(KA + KI bi V)
fair - (1 - 0)(6.3 x 10-3 ni3/s) / (6.3 x 10-3 m3/s +
5.25 x 10-2 m3/s)
fair - 0.107
f. Calculate benzene emissions .for plug flow:
E(g/s) « fair Q C0
- (0.107)(0.00156 m3/s)(10 g/m3)
- 1.67 x 10-2 g/s - 0.53 Mg/yr
4.4 MECHANICALLY AERATED IMPOUNDMENTS AND ACTIVATED SLUDGE UNITS
Some impoundments and tanks are mechanically agitated to improve.
mixing or to transfer air to the liquid (e.g., treatment tanks designed for
biodegradation). The agitation creates a turbulent liquid surface that
4-32
-------
enhances mass transfer to the air. A significant difference from the
approach for quiescent surfaces discussed in Section 4.2 is the appropriate
correlations for the individual mass transfer coefficients.
4.4.1 Emission Model Equations
The calculation of the overall mass transfer coefficient for mechani-
cally aerated systems considers that the liquid surface is composed of two
zones, quiescent and turbulent. The individual mass transfer coefficients
for the turbulent zone are based on the correlations of Thibodeaux^O and
Reinhardt.41 Thibodeaux's model was developed from accepted interphase
mass transfer concepts, published rate coefficient correlations, and exist-
ing operating data on 13 aerated basins at 11 pulp and paper mills. The
basins represented a wide range of design and operating parameters, in
spite of being from only one industry type. The simulation employed 11
organic chemical species common to industrial wastewater.
Reinhardt absorbed ammonia in aqueous sulfuric acid to measure the
gas-phase mass transfer coefficient associated with flat-blade surface
agitators in developing his correlation to calculate the gas-phase mass
transfer coefficient.^
Table 4-6 summarizes the correlations developed by Thibodeaux and
Reinhardt. These correlations are used to estimate the individual mass
transfer coefficients for the turbulent portion of the liquid surface. The
individual coefficients are then used in Equation (4-2) to calculate an
overall mass transfer coefficient for the turbulent zone. An overall mass
transfer coefficient for the quiescent zone is calculated as described in
Section 4.2. The two overall coefficients are combined to obtain a single
coefficient for the system based on the relative areas of the turbulent and
quiescent zones. For example, if 25 percent of the surface of the impound-
ment is turbulent, the overall coefficient would be the sum of 25 percent
of the value for the turbulent area coefficient plus 75 percent of the
value for the quiescent zone.
The model for mechanically aerated systems also incorporates biodegra-
dation as a competing mechanism. The extent of biodegradation is difficult
to predict in a generally applicable form because it is very dependent upon
the constituent of interest, the waste matrix, the design and operation of
4-33
-------
TABLE 4-6. EQUATIONS FOR CALCULATING INDIVIDUAL MASS TRANSFER
COEFFICIENTS FOR VOLATILIZATION OF ORGANIC SOLUTES FROM
TURBULENT SURFACE IMPOUNDMENTS
Liquid phase
Thibodeaux:43,44
°'5
kL - [8.22 x 10-9 j (POWR)(1.024)t'20 ot 10$ MWL/(VaypL)] (D^DQ |W)' (m/s)
|W
where
J » oxygen transfer rating of surface aerator, Ib 02/h«hp
POWR » total power to aerators, hp
T « water temperature, *C
Ot " oxygen transfer correction factor
MHj_ " molecular weight of liquid
V « volume affected by aeration, ft3
ay « surface-to- volume ratio of surface impoundment, ff1
p. * density of liquid, g/cnP
Dyi * diffusivity of constituent in water, cra2/s
-5 2
Dn M - diffusivity of oxygen in water * 2.4 x 10 , cm /s.
^2'
Gas phase
Reinnardt:45,46
ks » 1.35 x 10'7 R^'42 p°'4 Scg'5 .F;°'21 DaMWa/d (m/s)
where
Re * d2w^a//«a * Reynold's number
d » Impeller diameter, cm
w » rotational speed of impeller, rad/s
(continued)
4-34
-------
TABLE 4-6 (continued)
/»a » density of air, g/cra3
/ia » viscosity of air, g/cm«s
= 4.568 x 10-7 T(°C) + 1.7209 x 10'4
p 3 Pj gc/(/>Ld*5w3) = power number
P! = power to impeller, ft«lbf/s
= 0.85 (POWR) (550 ft*lbf/s«hp)/number of aerators,
where 0.85 * efficiency of aerator motor
gc « gravitation constant, 32.17 Ibra»ft/s2/lbf
p. » density of liquid, Ib/ft3
d* » impeller diameter, ft
SCQ - Schmidt number on gas side » Ma//>a °a
Fr » d*w2/gc = Froude number
Da « diffusivity of constituent in air, cm2/s
MWa * molecular weight of air.
4-35
-------
the biodegradation unit, and the concentrations and properties of the
microorganisms.
4.4.2 Model Plant Parameters for Mechanically Aerated Impoundments
The dimensions of the treatment impoundment used as an example to
estimate emissions were derived from the Westat data as described in
Section 4.2.2 for storage impoundment. A median area of 1,500 m2 and a
depth of 1.8 m were chosen, which yields a total volume of 2,700 m3. The
retention time in treatment impoundments is expected to be less than the
retention time in storage impoundments. Two design manuals listed typical
retention times for aerated (biologically active) ponds as 7 to 20 days47
and 3 to 10 days.48 For the example case, a retention time of 10 days was
chosen from the design range of 3 to 20 days. The resulting flow rate is
3.1 L/s (0.0031 m3/s).
The correlations of Thibodeaux and Reinhardt given in Table 4-3
require values for the parameters that describe the mechanical aeration-
system. Metcalf and Eddy, Inc.,49 suggest a range of 0.5 to 1.0
hp/1,000 ft3 for mixing in an impoundment. However, more power may be
needed to supply additional oxygen or to mix certain treatment solutions.
A review of trip reports showed power usage as high as 3.5 hp/1,000 ft3 at
a specific TSDF impoundment.50 For this analysis, a midrange value of
0.75 hp/1,000 ft3 from Metcalf and Eddy was used to generate an estimate of
75 hp required for mixing in the model unit.
Data from Reference 51 indicated that five aerators with 15-hp motors
and 61-cm diameter propellers turning at 126 rad/s would agitate a volume
of 441 m3 (15,590 ft3). Assuming a uniform depth in the impoundment of
1.8 m, the agitated surface area was estimated as 245 n£ (441/1.8). The
agitated surface is assumed to be turbulent and comprises 16 percent
(245/1,500 x 100) of the total area. The balance of the surface area of
the impoundment (84 percent) is assumed to be quiescent. As a comparison,
Thibodeaux reported a turbulent area of 5.22 m2/hp and investigated a range
of 0.11 to 20.2 m2/hp. The value of 5.22 m2/hp and a total of 75 hp yields
an estimated turbulent area of 392 m2 (26 percent), which is neater than
the 16-percent turbulent area calculated by the above procedure.52 (Very
4-36
-------
few data are available on the distribution of turbulent areas for aerated
impoundments. The extent of turbulence depends in part on the number
size, and placement of aerators. The example is based on typical aerator
requirements to mix the contents of the impoundment.)
Typical values were chosen for the oxygen transfer rating of the
aerator and the oxygen transfer correction factor. A value of 3.0 Ib
02/hp/h was chosen for oxygen transfer rating from a range of 2.9 to 3.0.53
A value of 0.83 was used for the correction factor from a typical range of
0.80 to 0.85.54 The transfer of power to the impeller was assumed to be
85 percent efficient, yielding an estimate of 64 hp for the impeller power.
The model for biodegradation requires the system's biomass concentra-
tion as an input parameter. The concentration of biomass in real systems
can be highly variable depending upon the system's design and method of
operation. For this analysis, the specified biomass is assumed to be
actively degrading the constituent of interest. A value of 250 g/m3
(0.25 g/L) of biomass was chosen from the values presented in Table 4-5.
The example constituent (benzene) and the meteorological conditions
chosen for the example calculation are the same as those chosen for storage
impoundments. Input parameters for the mechanically aerated model unit are
summarized in Table 4-7.
4.4.3 Example Calculation for Mechanically Aerated Treatment Impoundments
The example calculation for emissions from a mechanically aerated
impoundment includes an estimate of the overall mass transfer coefficient
for the turbulent zone. The overall mass transfer coefficient for the
quiescent zone for storage impoundments is calculated as illustrated in
Section 4.2.3 and will not be repeated here. Biodegradation is included as
a competing removal mechanism.
a. Calculate turbulent liquid-phase mass transfer coefficient, kj_. Use
Thibodeaux (Table 4-6):
k|_(m/s) = [8.22 x 10"9 J(POWR)(1.024)T"20 Ot 106
V.
4-37
-------
TABLE 4-7. INPUT PARAMETERS—TREATMENT IMPOUNDMENTS
(MECHANICALLY AERATED)
Area: 1,500 is2
Depth: 1.8 m
Volume: 2,700 m3
Retention tine: 10 days
Flow: 0.0031 m3/s
Turbulent area: 240 n£ (16%)
Quiescent area: 1,260 m2
Number of impellers: 5
Total power: 75 hp
Power to Impeller: 13 hp
Impeller speed: 126 rad/s
Impeller diameter: 61 cm
02 transfer: 3 Ib/h/hp
02 correction factor: 0.83
Temperature: 25 *C
VHndspeed: 4.47 m/s
Viscosity of air: 1.8 x 10'4 g/cm«s
Density of air: 1.2 x 10-3 g/cm3
Diffusivity of 02 in water: 2.4 x 10-5 cm2/s
Density of liquid: 1 g/c»3
Molecular weight of liquid: 18 g/g*mol
Molecular weight of air: 29 g/g*mol
Constituent: benzene with other biodegradable organics in water
Concentration (benzene): 100 g/m3 (100 ppm)
Concentration (total organics): 250 g/m3 (250 ppm)
Henry's law constant (benzene): 5.5 x 10-3 atra»m3/g mol
(continued)
4-38
-------
TABLE 4-7 (continued)
Diffusivity in air (benzene): 0.088 cm2/s
Diffusivity in water (benzene): 9.8 x 10"6 cra2/s
Maximum biorate (benzene and other organics): 19 mg/h/g of bioraass
5.28 x 10'6 g/g biomass«s
Limiting first-order biorate constant: 1.4 L/h/g * 3.89 x 10'?
biomass
Biomass concentration: 0.3 g/L s 300 g/m3
4-39
-------
where
J » 02 transfer rating, use 3.0 Ib 02/h»hp
POWR * 75 hp
T * water temperature * 25 °C
Ot » 02 transfer correction factor, use 0.83
molecular wt of liquid (water) * 18 g/g mol
(Vav) - agitated area in ft2 - 240.0 m2 [0;0g29 ^J -^,583 ft2
p. « water density « 1 g/cm3
D - 9.8 x 10-6 cm2/s
ri
DO w- 2.4 x 10-5 ci»2/s
k » (8 22 x 10'9)(3)(75)(1.024)5 [^-83)(106)(18)1 J9.8 x 10^1
L * ' I (2,583) (1) J 12.4 x 10"5J
» 7.7 x 10"3 m/s .
b. Calculate turbulent gas-phase mass transfer coefficient, k&. Use
Reinhardt (see Table 4-6):
kG(m/s) - 1.35 x 10'7 Re1'42 p0'4 ScG0'5 Fr'0*21 Oa MWa/d
where
I2-''-
Re * Reynold's number a
/•a
d * impeller diameter - 61 cm
w * impeller speed * 126 rad/s
/»a - 1.2 x 10-3 g/cm3
pa • 1.81 x 10-4 g/cn»s
4-40
-------
Re . (126) (1.2 x IP"3) „ 3<1 x 1Q6
1.81 x 10
PI gc
p * power number s
550 ft lbf
12.8 hp - -!- - 7,040
,,, ,, lb«ft
g = 32.17 -*
s' lbf
pL » 62.37 lb/ft3
d* « impeller diameter in feet =2.0
w » 126 rad/s
. (7r040? (32.17) . 5>6 x 1Q-5
(62.37)(1 )(126f
SCQ s 1-71 (from Section 4.2.3, part b)
Fr = Froude-number = &£ - ^r^ s 9'9 x l^
Da » 0.088 cm2/s (benzene)
MWa - 29 g/g mol
d « impeller diameter in cm a 61 cm
kG - (l.SSxlO-^O.lxlO6)1'42 (5.6xlO-5)°-4(1.71)°-5(9.9xl02)-0-21
(0.088)(29)761
k » 5.7 x 10"2 m/s .
4-41
-------
c. Calculate overall mass transfer coefficient for turbulent area, K:
1 , * + . 1 * - 1 - , + - 1 - «- = 2.4 x 102
IT *T KeqkG 7 ; x 1Q-3 (0.225) (5. 7 x 10"Z)
K = 7.71 x 10-* ra/s
d. Calculate overall mass transfer coefficient for combined quiescent and
turbulent areas, K:
From Section 4.2.3, K for quiescent area » 4.2 x 10-6 ra/s
From Part C, K for turbulent area - 7.7 x 10~4 «/s
Turbulent area • 240 ra2
Quiescent area * 1,260 m?
K (»/s) • . , . 1Q-) (240) , ? ? „ ,0-4
Righted by area) «l'»° * 366'
e. Calculate the effluent concentration for benzene for a well-mixed
system from Equation (4-21):
CL * [-b + (b2 - 4ac)0.5] /2a
where
a - K1 - (KA/V) (V/Q) + I » KA/Q + 1
b - KSK' + (V/Q) Knax bi - C0
c » -KsCo
K - 7.7 x 10-4 m/s
A - 1,500 *2
Q - 0.0031 m3/s
5.28 x 10-6 g/s/g biomass
i » 0.3 g/L • 300 g/m3
V « 2,700 m3
4-42
-------
C0 - 100 ppm - 100 g/m3
KS a Kmax/Kl - (5.28 x 10-6 g/s/g)/(3.89 x 10-7 m3/s/g)
= 13.6 g/ra3
KA » (7.7 x 10-^ m/s)(1,500 ra2) = 1.15 m3/s
a = K1 = (1.15 m3/s)/(0.0031 m3/s) + 1 * 373
b * (13.6 g/m3)(373) + (2,700 m3/0.0031 «3/s) (5.28 x
10-6 g/s/g) (300 g/n»3) - (100 g/n»3)
» 7,965.6 g/ra3
c - -(13.6 g/m3)(l6b g/m3) = -1,360 g2/m6
CL - {-[7,965.6 g/«3] + [(7,965.6 g/m3)2
- 4(373) (-136 g2/m6)]0.5}/ [2(373)]
- (-7,965.6 g/m3 + 7,982.2 g/m3)/970
» 0.22 g/m3 .
Calculate the fraction emitted for a well-mixed system from Equation
(4-22):
fair « KAC|_/(QC0)
where
fair * (1.15 m3/s)(0.22 g/m3)/[(0.0031 «3/s)(100 g/m3)]
fair = 0.816 .
Calculate benzene emissions for well-mixed system:
E(g/s) - fair Q C0
* (0.816)(0.0031 m3/s)(100 g/m3)
- 2.5 x 10-2 g/s « 8.0 Mg/yr .
»
For a plug-flow system, calculate the fraction removed with the
effluent from Equation (4-27):
Ce/Co • exp (-KI bi V/Q - KA/Q)
KI = 3.89 x 10-7 m3/s/g biomass
bi - 0.3 g/L * 300 g/m3
V » 2,700 m3
Q = 0.0031 m3/s
4-43
-------
Co - 100 ppm - 100 g/m3
K « 1.0 x 10-3 m/s
A - 1,500 m2
iV = (3.89 x 10-7 m3/s/g biomass) (300 g/m3)(2,700 m3)
- 0.315 m3/s
KA - (7.7 x 10-4 »/s) (1,500 «2) - 1.15
c/c ,exp -0.315 q/s ^1.1
60 0.0031 «3/s 0.0031 m3/s
1. Calculate fraction emitted from Equation (4-28):
fair « (1 - Ce/C0)(KA)/(KA + KibiV)
fair - (1 - 0)(1.15 m3/s)/(1.15 «3/s + 0.315 m3/s)
fair * 0.78 .
j. Calculate benzene emissions for plug flow:
E(g/s) « fair Q C0
- (0.78) (0.0031 m3/s)(100 g/m3)
- 0.24 g/s « 7.7 Mg/yr .
4.4.4 Example Calculation for Activated Sludge Unit
As discussed in Section 4.2, an activated sludge unit usually consists
of a concrete tank that is aerated and contains a relatively high concen-
tration of active biomass. A model unit is defined in this section for
this process, and the results of intermediate and final calculations are
given. Detailed example calculations ^are not presented because the
approach is exactly the same as that used for the mechanically aerated
impoundment. The only significant difference in the method of operation is
the recycle of solids back to the activated sludge unit, which results in a
higher biomass concentration. For this model unit, a biomass concentration
of 4 g/L (4,000 g/m3) was chosen from the range of 1.5 to 6 g/L in
Table 4-3 and the recommended values in Table 4-5. Other differences
between the aerated impoundment and activated sludge tank include, for the
tank, a smaller surface area, a shorter retention time, a greater turbulent
area, and a smaller F/D ratio. The aerated surface area was estimated as
4-44
-------
described in Section 4.4.2. An aerator with a 7.5-hp motor will agitate a
volume of 56.9 m3 (2,010 ft3). For a uniform depth of 4 m, the agitated
volume yields an agitated surface area of 14.2 m2 (56.9 m3/4 m). The input
parameters are defined for this model unit in Table 4-8, and the results of
the calculations are presented in Table 4-9.
4.5 DISPOSAL IMPOUNDMENTS WITH QUIESCENT SURFACES
4.5.1 Emission Model Equations
A disposal impoundment is defined as a unit that receives a waste for
ultimate disposal rather than for storage or treatment. This type of
impoundment differs from the storage and treatment impoundments in that
there is no liquid flow out of the impoundment (seepage into the ground is
neglected). For this case, the well-mixed system with a bulk concentration
that is at equilibrium (i.e., the bulk concentration does not change with
time) is not applicable. The quantity of a constituent in a disposal
impoundment will decrease with time after the waste is placed in the
impoundment because of the loss of volatiles to the air.
The calculation of the overall mass transfer coefficient is the same
as that presented for impoundments with quiescent surfaces. If the
disposal impoundment is aerated, K is calculated as described for aerated
impoundments in Section 4.4. The emission estimating procedure differs in
the calculation of the liquid-phase concentration that is the driving force
for mass transfer to the air. For a disposal impoundment that is filled
with a batch of waste, the rate of disappearance of a compound by biodegra-
dation and air emissions is described by Equation (4-26).
Integrating Equation (4-26) from Ct s C0 at t = 0 to Ct » Ct at t = t
gives: *
Ct/CQ - exp (-K^.t - KAt/V) . (4-31)
For an impoundment with a uniform depth, V/A = D. Substituting V/A = D
into Equation (4-31) yields:
Ct/C0 = exp (-iqbi t - Kt/D) . (4-32)
4-45
-------
TABLE 4-8. INPUT PARAMETERS—MECHANICALLY AERATED
ACTIVATED SLUDGE UNIT
Area: 27 m2
Depth: 4 m
Volume: 108 m3
Retention time: 4 h
Flow: 0.0075 m3/s
Turbulent area: 19 m2 (70%)
Quiescent area: 8.0 m?
Total power: 7.5 hp
Power to impeller: 6.4 hp
Impeller speed: 126 rad/s
Impeller diameter: 61 cm
62 transfer: 3 Ib/h/hp
62 correction factor: 0.83
Temperature: 25 *C
Windspeed: 4.47 m/s
Viscosity of air: 1.8 x 10'4 g/cm«s
Viscosity of water: 9 x 10-3 g/cra-s
Density of air: 1.2 x 10-3 g/cm3
Diffusivity of 0? in water: 2.4 x 10-5 cuK/s
Density of liquid: 1 g/cn3
Molecular weight of liquid: 18 g/g»mol
Molecular weight of air: 29 g/g»mol
Constituent: benzene with other biodegradable organics in water
Concentration (benzene): 10 g/m3 (10 ppm)
Concentration (total organics): 250 g/nP (250 ppra)
Henry's law constant (benzene): 5.5 x 10-3 atra-nrf/g-mol
Diffusivity in air (benzene): 0.088 cm2/s
Diffusivity in water (benzene): 9.8 x 10-6 cm2/s
Maximum biorate (benzene and other organics): 5.28 x 10-° g/s/g biomass
Limiting first-order biorate constant » 3.89 x 10-7 m3/s/g biowass
Biomass concentration: 4.0 g/L » 4,000 g/m3
4-46
-------
TABLE 4-9. INTERMEDIATE AND FINAL CALCULATION RESULTS
FOR ACTIVATED SLUDGE MODEL UNIT
Quiescent zone:
k(_ - 6.5 x 10-6 m/s
kg * 8.9 x 10-3 m/s
K = 6.5 x 10-6 m/s
Turbulent zone:
RL = 9.7 x 10-2 m/s
kg = 4.3 x 10-2 m/s
K = 4.88 x 10-3 m/s
Overall mass transfer coefficient = 3.4 x 10-3 m/s
For well-mixed system:
CL - 3.17
fai-r * 0.391
Emissions =* 0.30 g/s * 9.3 Mg/yr
For plug-flow system:
fair a 0.391
Emissions = 0.30 g/s = 9.3 Mg/yr
4-47
-------
When Equation (4-32) is evaluated after some fixed time t, the ratio Ct/Co
represents the fraction of the compound remaining in the impoundment;
consequently, 1 - Ct/C0 represents the fraction that has been removed by
biodegradation and air emissions. The fractions emitted to the air and
biodegraded after some time (t) are calculated from their relative rates:
fair * (1 - Ct/C0)(KA)/(KA + K^iV) (4-33)
ffaio * 0 - Ct/C0)(Kibi\0/(KA + K!biV) (4-34)
The quantity emitted after some time (t) is given by:
Emitted quantity (g) » fair v C0 (4-35)
The average emission rate over the period of time * t is:
E (9/5) * fair V C0/t (4-36)
Alternatively, a simplifying assumption may be made that, because the
impoundment is designed for disposal, all significantly volatile compounds
are eventually emitted to the air. Emissions under this assumption would
simply be QC0 where Q equals the disposal rate in cubic meters/second.
This assumption is probably valid for volatile compounds; however, com-
pounds that are relatively nonvolatile may be removed slowly and the
assumption may result in an overestimate of emissions.
4.5.2 Model Plant Parameters for Disposal Impoundments
The Westat data summary for impoundments indicated that disposal -
impoundments generally have higher surface areas and shallower depths than
storage and treatment impoundments. The median surface area for disposal
impoundments was approximately 9,000 m2 (compared to 1,500 m2 for storage
"impoundments), and the median depth was approximately 1.8 m. The disposal
impoundment is assumed to be filled with waste every 6 mo (two turnovers
per year).
The meteorological conditions and type of waste (water containing
benzene and other organics for the example calculation are the same as
those used for quiescent and aerated impoundments with biodegradation. The
inputs for the example calculation of emissions from disposal impoundments
are summarized in Table 4-10.
4-48
-------
TABLE 4-10. INPUT PARAMETERS—DISPOSAL IMPOUNDMENTS
Area:9,000 m2
Depth: 1.8 m
Volume: 16,200 m3
Turnovers per year: 2
Temperature: 25 8C
Windspeed: 4.47 ro/s
Diffusivity in water (ether): 8.5 x 10-6 cm2/s
Viscosity of air: 1.81 x 10-4 g/cm«s
Density of air: 1.2 x 10-3 g/cm3
Constituent: benzene with other biodegradable organics in water
Concentration (benzene): 100 g/m3 (100 ppm)
Concentration (total organics): 250 g/m3 (250 ppm)
Henry's law constant (benzene): 5.5 x 10-3 atm-m3/g mol
Diffusivity in air (benzene): 0.088 cn»2/s
Oiffusivity in water (benzene): 9.8 x 10-6 Cm2/s
Limiting first-order biorate constant*: 3.89 x 10-6 m3/s/g biomass
Biomass concentration: 0.05 g/L s 50 g/m3
4-49
-------
4.5.3 Example Calculations for Disposal Impoundments
Example calculations are presented below for the model unit defined to
represent disposal impoundments.
a. Calculate liquid-phase mass transfer coefficient, kj_. Use Springer's
model (see Table 4-1):
Effective diameter
Area
0.5
x 2 =
,0.5
9.000
r
x 2 = 107 m
F/D » Effective diameter/ depth
Windspeed » 4.47 m/s (UIQ > 3.25 m/s)
F/D - 59.5
59.5
2.611 x 10
"7
w
'ether
0.67
m/s
where
UIQ s windspeed - 4.47 m/s
Dw - 9.8 x lO'6 cm2/s (benzene)
Dgther s 8-5 x 10~6 = Cm2/s (ether) .
Then
k = 2.611 x 10"7 (4.47)1
kL « 5.7 x 10-6 m/s .
9.8 x 10-°"
8.5 x 10
-6
0.67
b. Calculate gas-phase mass transfer coefficient, kg,
Matasugu (see Table 4-1):
kG = 4.82 x 10-3 u°'78 See0'67 de-0-U(m/s)
where
U a windspeed = 4.47 m/s
Use MacKay and
4-50
-------
s s Schmidt No. _ viscosity of gas
G for gas " (gas density)(diffusivity of i in gas)
Gas = air
Viscosity (air) = 1.81 x 10'4 g/cm«s
Density (air) = 1.2 x 10"3 g/cm3
Diffusivity (benzene in air) = 0.088 cm2/s
Sc = 1.81 x 10"4 q/cm»s = K/J
G (1.2 x 10"3 g/on3) (0.088 cm2/s)
de = effective diameter = 107 m .
Then
kG - (4.82 x 10-3) (4.47)0-78 (i.71)-0.67 (107)-0.11
» 6.5 x 10-3 m/s
c. Calculate overall mass transfer coefficient, K:
1 .
K kL Keq
where
_H , 5.5 x ID"3 m3.atm/m0l -
(8.21 x 10'5) n_ (298 K)
Then
I = - ^ - ^ + - i - ^ « 1.76 x 105
K 5.7 x 10'b (0.225) (6. 5,x 10"^)
K » 5.7 x 10'6 m/s .
d. Calculate the fraction remaining from Equation (4-32). The impound-
ment is filled with waste initially, and 6 mo later it will be filled
again. Calculate the fraction remaining after the initial 6-mo
period:
Ct/C0 3 exp (-Kib, t - Kt/D)
KI » 3.89 x 10-7 m3/s/g biomass
b-j * 50 g/m3
4-51
-------
t « 6 mo » 1.58 x 107 s
C0 - 100 g/m3
K » 5.7 x 10-6 m/s
D * 1.8 m
t - (3.89 x 10-7 m3/s/g bioroass)(50 g/m3) (1.58 x 107 s)
» 307
Kt/D - (5.7 x 10-6 m/s)(1.58 x Ifl7 s) / 1.8 m - 50.0
Ct/Co - exp (-307 - 50) - 0 .
f. Calculate the fraction emitted from Equation (4-33):
fair • (I - Ct/Co)(KA) / (KA + Kib,V)
Ct/C0 » 0
KA - (5.7 x 10-6 m/s)(9,000 m2) - 0.051/m3/s.
KibfV - (3.89 x 10-7 m3/s/g biomass)(50 g/m3)(16,200 m3)
» 0.315 m3/s
fair " (1 - 0)(0.051 m3/s) / (0.051 m3/s + 0.315 «3/s)
fair a 0.14 .
g. Calculate the average emission rate over the 6-mo period from Equation
(4-36):
E (g/s) - fair V C0/t
- (0.14)(16,200 m3)(100 g7m3)/1.58 x 1Q7 s
« 1.4 x 10-2 g/s.
4.6 DIFFUSED AIR SYSTEMS
4.6.1 Emission Model Equations
Some Impoundments and open tanks (e.g., activated sludge units) are
sparged with air to promote biodegradatlon or air stripping. To estimate
emissions from diffused air systems, the model assumes that the air bub-
bling through the liquid phase reaches equilibrium with the liquid-phase
concentration of the constituent. The emissions leaving with the diffused
air are estimated by:
4-52
-------
E » QaKeqCL (4-37)
where
E - emissions, g/s
Qa ~ air flow rate, n>3/s
Keq = equilibrium constant
CL = concentration in the liquid phase, g/m^.
"* •**
Emissions can also occur from wind blowing across the surface. If the
air sparging creates a very turbulent surface similar to the surface of
mechanically aerated systems, then the emission rate should be based on
values of K typical for mechanically aerated systems. If the air sparging
rate does not result in a turbulent surface, then K can be estimated from
the correlations given for quiescent surfaces in Section 4.2.
The approach to estimate total emissions for flowthrough tanks and
impoundments sparged with diffused air is similar to that described for
quiescent and aerated systems. Because the unit is sparged with air, the
liquid phase is assumed to be well mixed and the plug-flow model is not
used. A material balance around this well-mixed system is identical to
Equation (4-18) in Section 4.3.2, but now K0ther 1S:
Kother « (*A + QaKeq)/V (4-38)
where all of the symbols have been previously defined. The steady-state
liquid phase concentration (C|_) is then calculated using Equation (4-21).
Air emissions are estimated as the sum from wind blowing across the surface
and from the diffused air: .
E » KACL + Qa Keq CL . (4-39)
The fraction of the component feed emitted to the air (fair) is:
fair = (KCLA + QaKeqCL)/QC0 . (4-40)
For disposal impoundments with diffused air systems, the steady-state
assumptions of the flowthrough models do not apply. Emissions are greatest
4-53
-------
when the waste is first placed in the impoundment and gradually decrease
with time. To incorporate the contribution to mass transfer from diffused
air, an equivalent mass transfer coefficient is defined:
KD = KeqQa/A (4-41)
where
Kg * equivalent mass transfer coefficient for diffused air, m/s
and all of the other symbols are a,s previously defined.
The mass transfer coefficient for wind blowing across the surface (K)
is calculated as described previously for flowthrough systems. A combined
overall mass transfer coefficient (Kc) is defined as:
Kc = K0 + K . (4-42)
The overall mass transfer coefficient (Kc) is used in the equations for
disposal impoundments (Section 4.5.1) to estimate the fraction emitted
(Equation 4-33) and the average emission rate (Equation 4-36). The
combined overall mass transfer coefficient defined above includes the mass
transfer effects from both removal mechanisms (wind and diffused air).
4.6.2 Model Unit Parameters for Activated Sludge Unit with Diffused Air
A model unit for the activated sludge process was defined in Section
4.4.4 and Table 4-8. The same dimensions are used here to define an acti-
vated sludge unit that uses diffused air instead of mechanical aeration.
The only additional parameter that must be specified is the diffused air
rate, which typically ranges from 0.3 to 0.5 ra^/s per 1,000 m3 of volume
(20 to 30 ft^/min per 1,000 ft^ of volume).55 For the model unit with a
volume of 108 n»3, an estimate of 0.04,m3/s is recommended based on the mid-
point of the design range. The model unit input parameters are summarized
in Table 4-11.
4.6.3 Example Calculation for Diffused Air Activated Sludge Unit
An example calculation is presented below for the model unit defined
in Table 4-11.
a. Calculate the liquid-phase, gas-phase, and overall mass transfer
coefficients. This procedure was illustrated for quiescent sur-
faces and the results for this model unit are given in Table 4-9:
4-54
-------
TABLE 4-11. INPUT PARAMETERS-DIFFUSED AIR ACTIVATED SLUDGE UNIT
Area: 27 m2
Depth: 4m
Volume: 108 m3
Retention time: 4 h
Flow: 0.0075 m3/s
Quiescent area: 8.0 m2
Diffused air rate: 0.04 n>3/s
Temperature: 25 °C
Windspeed: 4.47 m/s
Viscosity of air: 1.81 x 10~4 g/cm»s
Density of air: 1.2 x 10-3 g/ciiP
Diffusivity of 02 in water: 2.4 x 10-5 cm2/s
Density of liquid: 1 g/ctn3
Molecular weight of liquid: 18 g/g«mol
Molecular weight of air: 29 g/g»mol
Constituent: benzene with other biodegradable organics in water
Concentration (benzene): 100 g/m3 (100 pom)
Concentration (total organics): 250 g/m3 (250 ppm)
Henry's law constant (benzene): 5.5 x 10-3 atm»m3/g«mol
Diffusivity in air (benzene): 0.088 cm2/s
Diffusivity in water (benzene): 9.8 x 10-° cm2/s
Maximum biorate (benzene and other organics): 5.28 x 10-° g/s/g biomass
Limiting first-order biorate constant: 3.89 x 10"' m3/s/g biomass
Biomass concentration: 4.0 g/L = 4,000 g/m3
4-55
-------
k|_ • 6.5 x 10'6 m/s
kg » 8.7 x ID'3 m/s
K » 3.42 x 1C-3 m/s (weighted by area).
Calculate the equilibrium constant, Keq. The compound is benzene
in water, and Keq has been presented as 0.225 in the previous
sample calculations (from Equation 4-4).
Calculate the equilibrium liquid concentration in the unit (C[J
from Equation (4-21):
Q = 0.0075 m3/s
C0 » 100 g/m3
K = 3.42 x lO-3 m/s
A' « 27 m2
Qa 3 0.04 m3/s
Keq * 0.225
5.28 x lO'6 g/s/g biomass
a Kmax/Kl s <5-28 x 10"6 9/s/g)/(3.89 x 10'7 mVs/g)
- 13.6 g/m3
bi « 4,000 g/m3
V » 108 m3
QC0 - (0.0075 «3/s)(100 g/m3) » 0.75 g/s
KA » (3.42 x 10-3 m/s)(27 m2) » 9.23 x 10'2 m3/s
QaKeq • (0.04 m3/s)(0.225) - 9.0 x 10'3 m3/s
Kother = (KA + QaKeq)/V (from Equation 4-38)
* [(9.23 x 10-2 m3/s) * (9.0 x 10'3 m3/s)]/(108 m3)
* 9.39 x 10'4 1/s
V/Q - (108 m3)/(0.0075 rn^/s) - 14,400 s
a « K1 « (9.29 x 10'4 1/s)(14,400 s) + 1 - 14.5
b - (13.6 g/m3)(14.5) * (14,400 s)(5.28 x 10'6 g/s/g)
(4,000 g/m3) - 100 g/m3
» 40.1 g/m3
c * -(13.6 g/m3)(100 g/m3) - -1,360 g2/m6
CL = {[-401 g/m3] + [(401 g/m3)2 - 4(14.5)
(-1.360 g2/m6)]0.5}/[2(14.5)]
» [(-401 g/m3) + (490 g/m5)]/29
« 3.06 g/m3 .
4-56
-------
d. Calculate air emissions from Equation (4-39).
E - (9.23 x 10-2 m3/s)(3.06 a/m3) + (9.0 x 10-3 m3/s)(3.06 g/«3)
* 0.31 g/s = 9.7 Mg/yr
4.7 OIL FILM SURFACES
Some impoundments may have a floating film of oil on the surface. A
rigorous approach to estimating emissions from this type of source would
consider three resistances acting in series:
• From the aqueous phase to the oil
• Through the oil
• From the oil to the air.
Such an approach would require estimates of these three resistances and
estimates of the equilibrium partitioning between both the aqueous and oil
phases and the oil and air phases. Because these estimates are not gener-
ally available, a simplifying assumption is that the oil film is relatively
thin and that mass transfer is controlled by the gas-phase resistance. For
this case, Equation (4-2) reduces to:
K = kQ Keq (4-43)
where kQ is calculated from the correlation of MacKay and Matasugu (Table
4-1) and Keq is calculated from Raoult's law by:
Keq = P*/»a
where
Keq » dimensionless equilibrium constant
P* * vapor pressure of the volatile compound of interest, atm
P0 « total pressure, 1 atm
pa - density of air, g/cm3
pi * density of oil, g/cm3
a molecular weight of oil, g/g mol
MWa - molecular weight of air, 28.8 g/g mol.
4-57
-------
The value of K calculated above is substituted into the equations for flow-
through systems to estimate emissions. For the well-mixed flow models, C0
and CL in Equations (4-1) and (4-6) represent the VO concentration in the
oil phase (entering and leaving the impoundment, respectively), and the
flowrate Q is the volumetric flow rate of oil. Biodegradation is neglected
because the oil film inhibits the transfer of oxygen.
The procedure described above assumes that the oil layer in the
impoundment is well mixed. For example, changes in wind direction in units
with retention times on the order of days may tend to move the oil layer in
different directions and result in mixing. However, some systems may be
designed for or characterized by plug flow. This flow model assumes that
the oil film moves across the impoundment's surface without backmixing.
For plug flow of the oil film in flowthrough impoundments and tanks, the
fraction of VO in the oil layer emitted to the air is given by Equation
(4-11), and air emissions are estimated from Equation (4-12). In these
equations, Ce is the VO concentration in the oily effluent, Co is the
initial concentration in the oil layer entering the impoundment, r is the
residence time, D is the oil-film thickness, and Q is the volumetric flow-
rate of oil.
For an oil film on a disposal impoundment, emissions are calculated as
described in Section 4.5. However, biodegradation is neglected and Equa-
tion (4-32) reduces to:
Ct/Co - exp (- Kt/D) (4-45)
and the fraction emitted to the air is:
fair • 1 -*exp (-Kt/D) (4-46)
where
Ct " concentration in the oil film at time a t
Co * initial concentration in the oil film
0 » oil-film thickness
and with the other symbols as previously defined. The average emission
rate over the period of time equal to t is:
E (g/s) - fair V C0/t
4-58
-------
where
V = volume of oil in the impoundment, m^
and with the other symbols as previously defined. An example calculation
of this approach is given in Section 5.0 for applying an oil film to soil,
which is analogous to an oil film on a disposal impoundment because there
is no flow out in either case and emissions are a function of the time
since disposal.
4.8 DISCUSSION OF ASSUMPTIONS AND"SENSITIVITY ANALYSIS
4.8.1 Removal Mechanisms
The organic constituents present in wastes that are treated, stored,
or disposed of in surface impoundments and open tanks may leave the unit by
any of several mechanisms. Because of the large open surface area and
relatively high volatility of many organic constituents, emissions to the
air may be a primary removal mechanism for certain constituents. Other
constituents may be destroyed in impoundments and tanks specifically
designed for biodegradation. Aeration is often used to supply oxygen to
biologically active systems. Unfortunately, aeration also greatly enhances
the mass transfer of organic constituents to the air. Other removal
mechanisms include adsorption on solids, seepage through the ground, or
degradation (e.g., by photolysis or hydrolysis). For f.lowthrough systems,
the organic constituents may leave the unit with the effluent that will
subsequently be treated, stored, or disposed of.
Initial studies suggest that emission to air is a primary removal
mechanism, especially for volatile constituents. Biodegradation in
specific systems, particularly for semivolatiles, may also be significant.
For flowthrough systems, the removal of semivolatiles with the effluent may
also be a primary removal mechanism. Other forms of degradation, adsorp-
tion, and seepage are neglected in this analysis for several reasons.
These mechanisms are not believed to be significant for most systems and
most constituents; however, they may be removal routes in a specific system
or for a specific constituent. For example, an open tank may be designed
specifically for liquid-phase carbon adsorption. These mechanisms are also
difficult to model in a manner that is generally applicable considering the
relatively sparse data on such removal mechanisms, especially in hazardous
4-59
-------
waste impoundments and tanks. Consequently, the modeling effort focuses on
mass transfer to the air and some consideration of biodegradation.
Numerous studies have been conducted to assess mass transfer to the
air; these include theoretical assessments, correlations based on labora-
tory and bench-scale measurements, and field measurements at actual
sources. Additional data on specific wastes have been collected in air-
stripping studies as more air-stripping columns have been used to remove VO
constituents from water. The resuTt is that the state of knowledge of mass
transfer from the liquid to the gas phase (e.g., ambient air) is probably
advanced compared to the state of knowledge of other removal mechanisms.
The level of confidence in the air emission models is probably highest for
the volatile constituents because of very high mass transfer rates. The
level of confidence is somewhat lower for the relatively nonvolatile
constituents because of potentially significant rates of removal by other
mechanisms.
Much of the data on the performance of systems designed for biodegra-
dation are reported as total removal from measurement of the influent and
effluent concentrations. This total would include removal to the air and
biodegradation. Some studies have been conducted in closed systems in
which the biodegradation rate may be measured directly (loss to the air is
deliberately prevented). These data are useful for comparing the relative
rates of removal by biodegradation among constituents and make possible a
ranking of these constituents with respect to biodegradability. In addi-
tion, the estimated rate of biodegradation may be compared to the estimated
rate of air emissions to assess the relative extent of each.
The biodegradation model has not been validated and is used in this
report as an approximate measure of the extent of biodegradation. For any
specific treatment system, measurements of actual biodegradation rates
should be used if available. Any user of the biodegradation model should
be aware that the predicted rate is very sensitive to the choice of values
for the biorate, biomass concentration, and the concentration of organic
constituents in the waste. An environmentally conservative approach with
respect to air emissions would be to neglect biodegradation (assume the
rate is zero). This approach is probably valid for volatile constituents
-------
in aerated systems; however, the approach may tend to overestimate emis-
sions of relatively nonvolatile constituents that are destroyed in treat-
ment systems specifically designed for biodegradation.
4.8.2 Major Assumptions
An inherent assumption in the emission estimating procedure is that
the mass transfer correlations chosen earlier are generally applicable. A
paper that compares several different models concludes that, in most cases,
many different models yield comparable results for volatile constituents.56
The choice of models may affect the estimated mass transfer coefficients
for semivolatiles more than those for volatiles. The calculations indicate
that emissions of volatiles are controlled by the liquid-phase resistance.
Consequently, the value for the overall mass transfer coefficient (K) is
primarily determined by the correlation used for the liquid-phase mass
transfer coefficient (k|_). For constituents with decreasing volatility,
both the liquid-phase and gas-phase resistance begin to contribute to the
overall resistance to mass transfer. For these constituents, the choices
of correlations for both kg and kj_ become important, and the choice of
correlations may significantly affect the emission estimates.
The flow model chosen for storage and treatment impoundments assumes
that the impoundment's contents are well mixed and that the system is oper-
ated at steady-state conditions. The flow for specific facilities may be
better represented by plug flow or a model that accounts for axial disper-
sion. The choice of flow model does not make a significant difference in
the estimated emissions. However, if the loading of the impoundment is
cyclical or intermittent instead of continuous, the emissions from the
impoundment are likely to be cyclical.or intermittent. Estimates of short-
term emission rates are very dependent upon the method of operation of the
system. For disposal impoundments, peak emissions occur when the waste is
first placed in the impoundment and then decrease with time. The approach
used in this report estimates the average emission rate over a given period
of time and does not provide an estimate of the initial peak emissions.
The calculation of Henry's law constant also contains inherent assump-
tions. The approach is valid for dilute solutions and has been applied
successfully in the design of air-stripping columns. However, specific
4-61
-------
•Ixtures may deviate from Henry's law because of component interactions or
because of concentrations outside the range of applicability. Errors in
applying Henry's law are generally environmentally conservative; i.e., the
actual gas-phase concentration is not likely to be underestimated.
For concentrated mixtures of organics in a separate oil layer, the use
of Raoult's law is recommended. This approach is.valid for mixtures of
constituents with similar properties, especially when the concentration of
the component of interest is very-high. A preferred approach would be to
avoid the use of Henry's law or Raoult's law and actually measure the equi-
librium partitioning between the liquid and gas phase of a waste. However,
very few data are available for equilibrium partitioning that can be
applied generally to hazardous waste mixtures.
4.8.3 Sensitivity Analysis
The emission correlations were evaluated for sensitivity to each of
the input parameters.57 in the analysis, each input parameter was varied
individually over the entire range of reasonable values. The effect on
emissions was noted, and the most sensitive parameters were identified.
Detention time is an important parameter that affects emissions from
the impoundment. The emission estimates for volatile constituents are
sensitive to short detention times, and the estimates for semivolatiles are
sensitive to long detention times. Essentially all of the volatile consti-
tuents are emitted for longer detention times (several days), and very
little of the semivolatiles are emitted for short detention times (a few
days). However, significant emissions of the semivolatiles may occur for
long detention times in storage impoundments or in disposal impoundments.
The value of Henry's law constant was not important for volatile
constituents. The correlations indicated that these constituents are con-
trolled by the liquid-phase resistance, which is not affected by Henry's
law constant. The value of Henry's law constant has a direct effect on the
emissions of semivolatiles (such as phenol), and the greatest effect is on
those relatively nonvolatile compounds for which mass transfer is con-
trolled by the gas-phase resistance.
Windspeed has a direct effect on the emission estimates for quiescent
surfaces and has little effect on those from aerated systems. The results
4-62
-------
showed that a standard windspeed of 4.5 m/s was reasonable compared with
the results for windspeed distributions at actual sites.
Temperature did not affect the emission estimates for the volatile
constituents. However, temperature did affect the emission estimates for
nonvolatile constituents with mass transfer controlled by the gas phase.
The temperature dependence of Henry's law constant accounts for this
effect.
The diffusivity in air and water for a wide variety of constituents
spans a relatively narrow range of" values. The analysis showed that the
emission estimates were not sensitive to the choice of values for diffusiv-
ity.
For mechanically aerated systems, the choice of values for impeller
diameter, impeller speed, oxygen transfer rate, and oxygen correction
factor did not affect the emission estimates significantly. The total
horsepower and turbulent area had a direct effect on emissions of semi-
volatiles (e.g., phenol). However, there was no significant effect on
emissions of volatile constituents because the models predicted that they
would be stripped almost completely from the water over the full range of
aeration values.
The biodegradation model was very sensitive to all parameters investi-
gated. The sensitive parameters include organic concentration, biomass
concentration, and biorate.
Two meteorological parameters required in the models are temperature
and windspeed. The emission estimates are based on a standard temperature
of 25 °C and a windspeed of 4.47 m/s (10 mi/h). These standard values were
evaluated by estimating emissions for.windspeed/temperature combinations at
actual sites based on their frequency of occurrence. Over a 1-yr period,
the results from site-specific data on windspeed and temperature were not
significantly different from the results using the standard values. Conse-
quently, the standard values were judged adequate to estimate annual emis-
sions. For short-term emissions, the actual temperature and windspeed over
the short-term interval should be used to avoid underestimating emissions
during high-windspeed/high-temperature conditions.
4-63
-------
A sensitivity analysis was performed for three impoundment model units
(storage, mechanically aerated, and disposal) presented in the example
calculations in this section. Three compounds were chosen to represent
relatively nonvolatile compounds (p-cresol), moderately volatile compounds
(acetone), and relatively volatile compounds (benzene). Each of these
compounds can be biodegraded. The results are given in Tables 4-12, 4-13,
and 4-14. The key input parameters identified in the tables were increased
by 50 percent from the base case ta determine the effect on the percent of
the compound in the waste that is emitted to the air.
For each of the different types of impoundments, the volatility
appears to be important only for the low volatility category. As discussed
previously, the windspeed (air turbulence) has a direct effect for each of
the compounds in a storage impoundment and does not affect the mechanically
aerated unit's results. The low volatility compounds are the most sensi-
tive to changes in depth and biomass concentration for all three types of
impoundments. An assumption of no biodegradation also has the most drama-
tic effect on the low volatility compound with smaller effects observed for
the higher volatility compounds. The effects of retention time are small
except for the results shown for the disposal impoundment after 5 days.
The disposal impoundment results show that for short times, the time since
disposal is an important parameter affecting emissions.
4-64
-------
TABLE 4-12. RESULTS OF SENSITIVITY ANALYSIS FOR QUIESCENT
STORAGE IMPOUNDMENT
Percent emitted for given Henry's
law constant, atm«m3/mole
Key emission model inputs
Base casea
50-percent increase from base case*3
Volatility
Air turbulence
Retention time
Depth
Biomass concentration
No b i odeqradat i ond
10-7
2.9
4.2 (45)c
4.0 (38)
3.2 (10)
2.1 (-28)
2.1 (-28)
10 (245)
10-5
58
61 (5)
72 (24)
62 (7)
50 (-14)
52 (-10)
74 (28)
10-3
59
59 (0)
76 (29)
62 (5)
49 (-17)
52 (-12)
80 (36)
aThis corresponds to the model unit for storage impoundments used in the
example calculation.
bEach parameter is increased individually by 50 percent from its base case
value.
cValues in parentheses are percent change from the base case.
dBase case with no biodegradation.
4-65
-------
TABLE 4-13. RESULTS OF SENSITIVITY ANALYSIS FOR MECHANICALLY
AERATED IMPOUNDMENTS
Percent emitted for given Henry's
law constant. atm-nH/mole
Key emission model inputs
Base casea
50-percent increase from base caseb
Volatility
Air turbulence
Water turbulence
Retention time
Depth
Biomass concentration
No b i odeqradationd
10-7
2.7
3.9 (44)c
2.8 4)
3.6 33)
2.7 0)
1.8 (-33)
1.8 (-33)
20 (640)
10-5
79
85 (8)
80 (1)
85 (8)
80 (1)
73 (-8)
73 (-8)
94 (28)
10-3
99
99 (0)
99 (0)
99 (0)
99 (0)
98 (-1)
98 (-1)
100 (1)
corresponds to the model unit for mechanically aerated impoundments
used in the example calculation.
bEach parameter is increased individually by 50 percent from its base case
value.
cValues in parentheses are percent change from the base case.
dBase case with no biodegradation.
4-66
-------
TABLE 4-14.
RESULTS OF SENSITIVITY ANALYSIS FOR DISPOSAL
IMPOUNDMENTS
Percent emitted for given Henry's
Taw constant. atm«nH/mole
Key emission model inputs
Base casea
50-percent increase from base case
Volatility
Air turbulence
Retention time0
Depth
Biomass
No biodegradationd
10-7
13
10-5
93
10-3
92
18 (38) b
17 (31)
* * \ ** /
2.3 (-82)
9 (-31)
9 (-31)
94 (1)
96 (3)
55 (-41)
89 (-4)
89 (-4)
92 (0)
96 (4)
72 (-22)
88 (-4)
89 (-3)
84 (550) 100 (8) 100 (9)
aeased on the dimensions given in the example calculation 100 rag/L of the
constituent in 1,000 mg/L total organics, and a time since dnsposal of
12 months.
bValues in parentheses are percent change from base case.
CA retention time of 5 days was selected here to show the sensitivity to
retention time soon after disposal.
dBase case with no biodegradation.
4-67
-------
4.9 REFERENCES
1. Lunney, P. 0. Characterization of Wind and Depth Effects Upon Liquid
Phase Mass Transfer Coefficients: Simulation Studies. Master's
thesis, University of Arkansas, Fayetteville, AR. January 1983.
p. 119.
2. Springer, C., P. D. Lunney, and K. T. Valsaraj. Emission of Hazardous
Chemicals from Surface and Near Surface Impoundments to Air. U.S.
Environmental Protection Agency, Solid and Hazardous Waste Research
Division. Cincinnati, OH. Project Number 808161-02. December 1984.
p. 3-4 to 3-16.
3. Reference 2, p. 3-16 to 3-19.
4. Reference 2, p. 3-18.
5. Hwang, S. T. Toxic Emissions from Land Disposal Facilities. Environ-
mental Progress. 1^46-52. February 1982.
6. Mackay, D., and A. Yeun. Mass Transfer Coefficient Correlations for
Volatilization of Organic Solutes from Water. Environmental Science
and Technology. 17:211-217. 1983.
7. Reference 6, p. 214.
8. GCA Corporation. Air Emissions for Quiescent Surface Impoundments--
Emissions Data and Model Review. Draft Technical Note. Prepared for
U.S. Environmental Protection Agency. Contract No. 68-01-6871,
Assignment 49. August 1985. p. 5-1 and 5-2.
9. Reference 8, p. 4-4.
10. Reference 5, p. 47.
11. Thibodeaux, L. J. Air Stripping of Organics from Wastewater. A
Compendium. Air/Water, p. 373. (In publication.)
»
12. Westat Corporation. National Survey of Hazardous Waste Generators and
TSDF's Regulated Under RCRA in 1981. Prepared for the U.S. Environ-
mental Protection Agency. Contract No. 68-01-6861. April 1984.
13. Metcalf and Eddy, Inc. Wastewater Engineering. New York, McGraw-
Hill. 1972. p. 542-554.
14. Eckenfelder, W., M. Goronszy, and T. Quirk. The Activated Sludge
Process: State of the Art. CRC Critical Reviews in Environmental
Control. 15(2):148. 1984.
15. Beardsley, M., and J. Coffey. Bioaugmentation: Optimizing Biological
Wastewater Treatment. Pollution Engineering. December 1985. p. 32.
-------
16. Reference 13, p. 586.
17. Reference 15, p. 32.
18. Reference 13, p. 520-521.
19. U.S. Environmental Protection Agency. EPA Design Manual: Municipal
Wastewater Stabilization Ponds. Publication No. EPA-625/1-83-015.
October 1983. p. 3.
20. Reference 13, p. 557.
21. Englande, A. J. Performance "Evaluation of the Aerated Lagoon System
at North Gulfport, Mississippi. Prepared for U.S. Environmental
Protection Agency. Publication No. EPA-600/2-80-006. March 1980.
p. 39-41.
22. Allen, C. Project Summary: Site Visits of Aerated and Nonaerated
Surface Impoundments. Prepared for U.S. Environmental Protection
Agency. Contract No. 68-03-3253. Assignment 2-8. June 1987. p. 2.
23. Petrasek, A., B. Austern, and T. Neiheisel. Removal and Partitioning
of Volatile Organic Priority Pollutants in Wastewater Treatment.
Presented at the Ninth U.S.-Japan Conference on Sewage Treatment
Technology. Tokyo, Japan. September 1983. p. 16.
24. Bishop, D. The Role of Municipal Wastewater Treatment in Control of
Toxics. Presented at the NATO/CCMS Meeting. Bari, Italy. September
1982. p. 18.
25. Hannah, S., B. Austern, A. Eralp, and R. Wise. Comparative Removal of
Toxic Pollutants by Six Wastewater Treatment Processes. Journal WPCF.
58(1):30. 1986.
26. Kincannon, 0., and E. Stover. Fate of Organic Compounds During
Biological Treatment. Presented at ASCE Environmental Engineering
Conference. 1981. p. 6.
27. Melcer, H. Biological Removal of Organic Priority Pollutants.
Presented at Hazardous Substances in Wastewater Seminar. Toronto,
Canada. November 1982. p. 20.
28. Reference 19, p. 75-146.
29. Reference 13, p. 481-573.
30. Reference 14, p. 119.
31. Bailey, J. E., and 0. F. Ollis. Biochemical Engineering Fundamentals.
New York, McGraw-Hill. 1977. p. 343-349.
4-69
-------
32. Klncannon, 0., and E. Stover. Determination of Activated Sludge
Biokinetic Constants for Chemical and Plastic Industrial Wastewaters.
Prepared for U.S. Environmental Protection Agency. Publication
No. EPA-600/2-83-073a. August 1983. p. 18-20.
33 Pitter, P. Determination of Biological Degradability of Organic
Substances. Water Research. 10:231-235. 1976.
34. Kincannon, D.r A. Weinert, R. Padorr, and E. L. Stover. Predicting
Treatability of Multiple Organic Priority Pollutant Wastewater from
Single Treatability Studies.., Presented at the 37th Purdue Industrial
Waste Conference, West Lafayette, IN. May 1982.
35. Reference 23, p. 4-16.
36. Reference 25, p. 27-34.
37. Coburn, J..-C. Allen, D. Green, and K. Leese. Site Visits of Aerated
and Nonaerated Impoundments. Revised Draft Summary ^port Prepared
for U.S. Environmental Protection Agency. Contract No. 68-03-3Z5J,
Work Assignment No. 3-8. April 1988. p. A-l to A-34.
38. Matter-Muller, C., W. Gujer, W. Giger and W. Stumm. J«J;B1olog1cal
Elimination Mechanisms in a Biological Sewage Treatment Plant. Prog.
Water Tech. 12:305. 1980.
39. Dobbs, R., M. Jelus, and K. Chang. Partitioning of Toxic Organic
Compounds on Municipal Wastewater Treatment Plant Solids. Proceedings
of the International Conference on Innovative Biological Treatment of
Toxic Wastewaters. Scholze, R. J., Ed. Arlington, VA. June 1986.
p. 585-601.
40. Reference 5, p. 46.
41. Reinhardt, J. R. Gas-Side Mass-Transfer Coefficient and Interfacial
Phenomena of Flat-Bladed Surface Agitators. Ph.D. dissertation,
University of Arkansas. Fayetteville, AR. 1977. 96 p.
42. GCA Corporation. Emissions Data and Model Review for Wastewater
Treatment Operations. Draft Technical Note. Prepared for U.S.
Environmental Protection Agency. Contract No. 68-01-6871, Assign-
ment 49. August 1985. p. 4-3.
43. Reference 42, p. 4-2.
44. Reference 5, p. 47.
45. Reference 42, p. 4-3.
46. Reference 41, p. 48.
4-70
-------
47. Reference 19, p. 3.
48. Reference 13, p. 557.
49. Reference 13, p. 519.
50. GCA Corporation. Hazardous Waste TSDF Waste Process Sampling.
Prepared for U.S. Evironmental Protection Agency. Report No. EMB/85-
HNS-3. October 1985. p. 1-11.
51. GCA Corporation. Evaluation and Selection of Models for Estimating
Air Emissions from Hazardous Waste Treatment, Storage, and Disposal
Facilities. Prepared for U. 3. Environmental Protection Agency.
Publication No. EPA-450/3-84-020. December 1984. p. 69.
52. Thibodeaux, L., and D. Parker. Desorption Limits of Selected Gases
and Liquids from Aerated Basins. AIChE Sumposium Series.
72(156):424-434. 1976.
53. Reference 51, p. 67.
54. Reference 51, p. 67.
55. Reference 13, p. 519.
56. Allen, C. C. Prediction of Air Emissions from Surface Impoundments.
Paper 31a. (Presented at 1986 Summer Meeting of AIChE. Boston, MA.
August 1986.) 26 p.
57. Branscome, M., and A. Gitelman. Sensitivity Analysis: Emission Esti
mates for Surface Impoundments. Prepared for the U.S. Environmental
Protection Agency. March 1986. 67 p.
4-71
-------
5.0 LAND TREATMENT
This chapter presents the approach used to estimate air emissions from
'*«
land treatment operations. Analytical models to estimate emissions, repre-
sentative values of model input parameters, and example calculations are
included.
5.1 NARRATIVE DESCRIPTION OF LAND TREATMENT AIR EMISSIONS
Land treatment is one of several land disposal methods used for final
disposition of hazardous wastes. At land treatment facilities, wastes are
either spread onto or injected into the soil, after which they are normally
tilled into the soil. Other activities likely to occur at land treatment
facilities include storage of wastes in tanks or surface impoundments,
loading and unloading of wastes in vacuum trucks or dump trucks, and
dewatering of wastes using filtration devices. All of these activities
have emission points associated with them. The following paragraphs
describe analytical models used to estimate emissions from the application,
tilling, and final disposition of hazardous waste at a land treatment dis-
posal site. Emissions from other land treatment activities, such as truck
loading, storage tanks, and fugitive emissions from transfer and handling
operations, are estimated using procedures described in Chapter 7.0 of this
report.
Estimating emissions from land treatment may involve one to three
independent steps depending on operating practices at a land treatment
site. If waste is applied from a vacuum truck to the soil surface, allowed
to remain on the surface for a period of time, and then tilled into the
soil, emissions are estimated in three steps: (1) during application of
waste onto the soil from a vacuum truck, (2) after waste application and
before tilling, and (3) after tilling the waste into the soil. If waste is
applied to the soil surface and immediately tilled into the soil, emissions
5-1
-------
are estimated in only two steps: (1) during waste application, and (2)
after tilling. If waste is applied by subsurface injection and immediately
tilled, only one step is required to estimate emissions.
This section presents three separate analytical models that can be
used to estimate air emissions from separate land treatment activities.
Primary emphasis is given to the RTI land treatment model that is used to
calculate emissions from waste that is mixed with the soil. This condition
may exist when waste has been applied to the soil surface and has seeped
into the soil or when waste has been injected beneath the soil surface or
has been tilled into the soil. The RTI land treatment model is described
below in Subsection 5.2.1, which includes separate discussions of the
following topics:
Subsection Topic
5.2.1.2 Biodegradation
5.2.1.4 Effective diffusivity
5.2.1.5 Waste partitioning
5.2.1.10 Model selection rationale
After waste is applied to the surface of the soil by spray application, it
seeps into the soil. While the waste is on the surface, the concentrations
at the surface can be approximated by the concentration in the waste. Dur-
ing this short period that the waste covers the surface of the soil, the
maximum short-term emission rate is expected. For this special case, emis-
sions are estimated as the product of an overall mass transfer coefficient,
constituent concentration, and surface area of the land treatment site.
The model for estimating the mass transfer coefficient from the surface to
the wind was developed by McKay and Matsuga and is briefly discussed in
Subsection 5.2.3.
Emissions from a waste stream as it is applied onto the soil surface
from a vacuum truck, regardless of waste type, are also calculated as the
product of an overall mass transfer coefficient, the surface area of the
waste stream, and the concentration of a specific constituent. Preliminary
calculations indicate that emissions from the spraying waste application
are relatively small and can be ignored in most situations. Even so, a
brief discussion of a model for estimating these emissions is presented in
5-2
-------
Subsection 5.2.2, and the model can be used if desired. Also included in
this section are Subsection 5.2.4, which discusses representative values of
input parameters for the analytical models, and Subsection 5.2.6, which
presents example calculations using each of the three models presented.
At many existing land treatment sites, waste is applied onto the soil
from a vacuum truck and is allowed to remain for about 24 hours before
being tilled into the soil. Under these conditions, three separate calcu-
lations may be needed to estimate air emissions. Emissions during waste
application could be estimated using the waste application model described
in Subsection 5.2.2; emissions after application but before tilling would
be estimated using the RTI land treatment model as described in Subsection
5.2.1 (or, if a visible oil film exists on the soil surface, the oil film
surface model as presented in Subsection 5.2.3); and emissions after
tilling would be estimated using the RTI land treatment model. At other
existing sites, waste is injected into the soil using subsurface injection
and is immediately tilled. At these sites, only one calculation is needed
to estimate emissions. In this situation, the RTI land treatment model
would be used.
5.2 LAND TREATMENT
5.2.1 Land Treatment Emission Model Descriptions
5.2.1.1 Analytical Correlations. Emissions from land treatment after
waste is applied to the soil are estimated using a model developed by Clark
Allen of Research Triangle Institute (the RTI model). This model assumes
that emissions from the surface of the soil/waste mixture are limited by
the diffusion of vapors through the pore spaces in the soil/waste mixture
and further assumes that an equilibrium concentration of organic vapors
exists at all times within the pore spaces. The model is based on Pick's
second law of diffusion applied to a flat slab as described by Crank1 and
includes a term to estimate biological decay assuming a decay rate that is
first order with respect to waste loading in the soil.
The solution to the diffusion equation developed by Crank is for
diffusion out of a slab that initially has a uniform concentration of.
diffusing material throughout and that has equal concentrations of
diffusing material at each surface.
5-3
-------
The general solution to the diffusion equation for those conditions,
as presented by Crank, is:
F - • ~ 8 iDJ2n±nVt
F -
where
F * fraction of initially applied material that has diffused out of
the slab at time t
Mt * mass of material that has diffused out of the slab at time t
MO * initial mass of material present
0 * diffusion coefficient
1 * distance from center to surface of slab
t » time after initial distribution of diffusing material into *he
slab.
This series solution converges very slowly for small values of time (i.e.,
Dt/l2 < 0.213). Because of this slow convergence at short times (i.e.,
immediately after waste application or tilling), Crank presented an
alternative solution that is valid during this short time. The following
equation is obtained from the alternative solution:
M f 11/2
f * tr * -2- 2J [for Dt/12 < 0.213] . (5-2)
Mo jr IrJ l J
Equation (5-2) approximates the Crank .solution but excludes a small error
function correction used by Crank.
To verify the validity of Crank's solution for short times and to test
the accuracy of an approximation for use over longer times, the values
predicted by the solution for short times and the values obtained using the
first tens of the series solution [Equation (5-3)] are compared to the
values obtained using the first three terms of the series solution. Table
5-1 presents the results for a range of values of the dimensionless
parameter, Dt/l2;
5-4
-------
(5-3)
Table 5-1 shows that, for values of the dimensionless parameter
greater than 0.213, the first term of the series solution, Equation (5-3),
can be used to estimate total emissions. The table also shows that the
solution for short times, Equation (5-2), is valid for values of the dimen-
sionless parameter below 0.213. Equations (5-2) and (5-3) give identical
results for a parameter value of 0.213. This comparison indicates that
sufficient accuracy can be attained under all conditions if the equation
for short times is used for values of the dimensionless parameter below
0.213 and the first term of the general solution is used for values above
0.213. It is observed that the fraction of material that diffuses out of
the slab is linear with respect to the square root of time up to the point
where approximately 50 percent of the diffusing material is lost.2
The conditions defined for the above solutions by Crank are analogous
to diffusion of volatile organics out of a surface layer of a soil/waste
mixture as happens in land treatment operations. Because of the symmetry
of conditions on which the above solutions are based, an impenetrable plane
could, in theory, be inserted at the midpoint of the slab without changing
the solution. One-half of the slab with an impenetrable boundary layer on
the bottom would represent the surface layer of soil into which waste is
mixed during land treatment.
In a land treatment operation, the applied material partitions into
several phases including evaporation into a vapor phase, adsorption onto
soil particles, and absorption into oil and water in the soil/waste mix-
ture. Only the vapor phase is available for diffusion out of the soil/
waste mixture. Therefore, to apply the above equations to land treatment,
the amount of material in the vapor phase must be known. The amount of
material that partitions into the vapor phase can be estimated by calcu-
lating equilibrium conditions within the soil/waste mixture. This equilib-
rium is estimated by defining Keq, the ratio of the amount of organics in
the vapor phase to the total amount of organics in the soil/waste mixture.
The instantaneous emission rate, E, at any time, t, can be estimated by the
following Equations (5-4) and (5-5), which are obtained by differentiating
5-5
-------
TABLE 5-1. COMPARISON OF THE ESTIMATED FRACTION
EMITTED USING THREE DIFFERENT EQUATIONS
(INTEGRATED FLUX FROM SOIL)
Time
parameter
(Dt/l2)_
0.000
0.025
0.050
0.075
0.100
0.125
0.150
0.175
0.200
0.213
0.250
0.275
0.300
0.325
0.350
0.375
0.400
0.425
0.450
0.475
0.500
0.525
0.550
0.575
0.600
0.625
0.650
0.675
0.700
0.725
0.750
0.775
0.800
0.825
0.850
0.875
0.900
0.925
0.950
0.975
Short-term
solution
2?Dt 11/2
0.000
0.178
0.252
0.309
0.357
0.399
0.437
0.472
0.505
0.521
0.564
0.592
0.618
0.643
0.668
0.691
0.714
0.736
0.757
0.778
0.798
0.818
0.837
0.856
0.874
0.892
0.910
0.927
0.944
0.961
0.977
0.993
1.009
1.025
1.040
1.056
1.070
1.085
1.100
1.114
First term of
series solution
. _ 8 f Dty2]
" 7 x Ui2 J
0.189
0.238
0.284
0.326
0.367
0.405
0.440
0.474
0.505
0.521
0.562
0.589
0.613
0.636
0.658
0.679
0.698
0.716
0.733
0.749
0.764
0.778
0.791
0.804
0.816
0.827
0.837
0.847
0.856
0.864
0.873
0.880
0.887
0.894
0.900
0.906
0.912
0.917
0.922
0.927
First
three terras
of series
solution
0.067
0.179
0.252
0.309
0.357
0.399
0.437
0.472
0.504
0.520
0.562
0.589
0.613
0.636
0.658
0.679
0.698
0.716
0.733
0.749
0.764
0.778
0.791
0.803
0.816
0.827
0.837
0.847
0.856
0.865
0.873
0.880
0.887
0.894
0.900
0.906
0.912
0.917
0.922
0.927
5-6
-------
Equations (5-2) and (5-3) and adding the equilibrium constant, Keq, and a
terra to account for waste biodegradation, e^ D':
(shorttin.es) E - -f e'b , (5-4)
and
(longer times) E - M exp_ e'b (5-5)
where
tb * the biological degradation time constant.
The above equations account for the removal of organic material from
the soil/waste mixture both biological degradation and air emissions. In a
land treatment operation, the primary objective is to dispose of organic
materials by biodegradation; thus, significant quantities of waste would be
expected to be depleted from the soil by biological degradation. Other
mechanisms of removal such as leaching and photolysis also are possible but
are not accounted for in this model because of the estimated small amount
of materials lost by these processes.
5.2.1.2 Biodegradation. Biodegradation at land treatment sites is
generally considered to be a first-order process with respect to waste
concentration in the soil up to the point where saturation is achieved." 3
In addition to literature sources that make such statements, comments on a
draft of this document provided by Chevron Research Company offer further
evidence of the first-order nature of biodegradation at land treatment
sites.4 A first-order decay process ts defined in the literature as having
the following form;5
where
M * mass of organic material in the soil
Kb * biological decay constant.
5-7
-------
Integrating and using the boundary conditions M » M0 at t * 0 results in:
log M « -Kjjt + GI
or
-Kht
M -* C2C D
where Cj and C2 are constants of integration. Substituting the boundary
conditions gives:
-Kht
M D
Kb has units of s'1 and can be expressed as the reciprocal of the
biological decay time constant, l/tD. The exponential was introduced
directly into the rate relationship, Equations (5-4) and (5-5), to reduce
the amount of material available for air emissions by the fraction of
material removed by biooxidation.
5.2.1.3 Estimation of Equilibrium Coefficient. Keg. Partitioning of
volatile constituents in the waste is assumed to occur between the vapor
space in the soil/waste mixture, adsorbent solids in the soil, and
absorbent liquids in the soil and waste. Using 1 cm3 of the soil /waste
mixture as a basis for calculation, the total volume of gas (i.e., void
space) in the cubic centimeter is described by the air porosity, ea. Using
the ideal gas law, the number of moles of gas in 1 cm3 of the soil /waste
mixture is Pca/(RT). *h«re P is the pressure of a constituent in the gas
phase and is usually equal to XP* (X is the mole fraction of the constitu-
ent in the liquid phase and P* is the pure component vapor pressure). The
moles of constituent in the gas phase in 1 cm3 of the soil/waste mixture is
thus XP*ea/(RT). Oil loading in the soil/waste mixture in units of grams
of oil per cubic centimeter of mixture is L (goil/cm3 mixture), and the
total moles of constituent per cubic centimeter of the mixture is XL/MW0-ji.
The equilibrium coefficient, Keq. is defined as the moles of constituent in
the gas phase per unit volume of the soil /waste mixture divided by the
5-8
-------
total moles of constituent per unit volume of the soil /waste mixture.
Therefore, the following equation can be written:
XP*ea/(RT) P*MW0il£a
= RTL
This equation differs from the usual equation for equilibrium coefficient
by the factor ea, which is included to account for the limited air space
available within the soil/waste mixture. The ratio of moles per mole and
'»*
grams per gram can be used interchangeably in this equation. The value of
Keq can be calculated from measurements, if available, of constituent con-
centrations in the pore space and in the soil /waste mixture.
In a similar manner, it can be shown that Keq can be estimated for
aqueous wastes with an assumed value of the Henry's law constant, Hc:
Hc 1<)6 ea
R
Klcwaste
where
fiwaste s tne volume fraction of the soil /waste mixture that is
occupied by waste.
In the above equations, it is assumed that equilibrium is controlled by
Raoult's law for oily wastes and by Henry's law for aqueous wastes. An
aqueous waste is assumed to contain water and organic constituents that are
dissolved in water. An example of an aqueous waste is a sludge containing
10 percent solids, 5 percent acetone, 1 percent methanol , 500 ppmw benzene,
and the remainder water. If the waste contains oil mixed with the water,
or the waste contains volatile constituents at concentrations greater than
the solubility in water, it is modeled as an oily waste. It is important
to specify the molecular weight of this separate organic phase for this
Raoult's law calculation of Keq.
5.2.1.4 Estimation of Effective Diffusivity. The diffusivity of
specific compounds, as reported in the literature, assumes that the diffu-
sion occurs in free air. In a land treatment operation, diffusion of
vapors out of the soil must take place within the confines of the
5-9
-------
air-filled voids within the soil. This characteristic of soil is referred
to as the air porosity. The ratio of effective diffusivity of a
constituent in the soil to its diffusivity in air can be described by the
following equation-.6
0 e10/3
e _ a
a eT
where
Oe * effective diffusivity of constituent in soil
Oa » diffusivity of constituent in air
ea * air porosity of soil
CT » total porosity of soil.
When air porosity and total porosity are the same (i.e., for dry
soil), this equation reduces to:
°e . ,4/3
«;•'• •
Total porosity refers to the fraction of the land treatment medium that is
made up of nonsoil (or nonsolid) materials, i.e., the sum of the void
space, water-filled space, and space occupied by the oil in the applied
waste.
Soil air porosity undergoes substantial changes over time as soil
dries out and when moisture is added by rainfall or by watering. As a
result, accurately accounting for soil porosity in an analytical model is
difficult. The use of average or typical values of soil porosity may be
the most practical approach.
5.2.1.5 Waste Partitioning. A large percentage of wastes that are
disposed of by land treatment are refinery sludges. These wastes are
mostly sludge emulsions and consist of varying fractions of water, oil, and
inorganic solids, where oil represents the total organic portion of the
waste including volatile compounds. A much smaller amount of land-treated
wastes are dilute aqueous solutions of water and organic compounds. When
wastes are applied to a land treatment area, volatile materials in the soil
-------
have the potential for partitioning into four different phases—a vapor
phase, an oil phase where volatile material is dissolved in the oil, a
water phase where volatile material is absorbed in the soil moisture, and a
soil phase where volatile material is adsorbed by organic carbon within the
soil. For oily wastes, VO compounds will preferentially dissolve in oil
rather than water so that the fraction of volatile materials in the water
phase is estimated to be very small. Partitioning of volatile materials
into the soil phase by adsorption is a function of the amount of organic
carbon in the soil. It is also estimated to account for only a small
fraction of the applied organics because the surfaces in the soil are
expected to contain oil from the application and tilling of waste materials
that contain oil. This oil in the soil is expected to both absorb the
volatile constituents and to interfere with the relatively lower adsorption
rates on soil surfaces. For high molecular weight constituents present in
aqueous wastes, adsorption may be more important. An equilibrium equation
can be written that takes all four phases into account in the estimation of
equilibrium vapor concentration in the soil. However, as presented here,
the equilibrium equation in the RTI model includes only two phases. Calcu-
lations by one researcher looked at the difference in estimated emissions
using two-phase partitioning of waste into an oil phase and vapor phase and
using four-phase partitioning. The results of these comparisons are given
in Table 5-2 and show for the conditions considered that, for soils having
an organic carbon content of up to 10 percent, the estimated fraction of
applied organics emitted using four-phase partitioning is only about 10
percent less than the estimated fraction emitted using two-phase partition-
ing. In a given situation, the amount of material adsorbed by organic
carbon in the soil is relatively constant; thus, in soils with high organic
carbon content, adsorption of materials in the soil may become more signif-
icant if low loading rates of oil or aqueous wastes are used. One of the
products of biodegradation is organic carbon; thus, land treatment sites
that have been active for an extended time may have elevated concentrations
of organic carbon. Even so, with the normal oil loading used in land
treatment, it is likely that a large fraction of the available adsorption
sites would be occupied by the oil itself, thus limiting the effects of
adsorption on emissions of the lighter constituents.
5-11
-------
TABLE 5-2. EMISSION ESTIMATES USING TWO DIFFERENT EQUATIONS
FOR THE VAPOR-SOIL PARTITION COEFFICIENT/
Organic carbon
content of soil
(fraction)
Estimated emission
fraction--two-phase
partitioning
Estimated emission
fracti on—four-phase
partitioning
0
0.001
0.010
0.100
0.622
0.622
0.622
0.622
0.622
0.621
0.614
0.559
5-12
-------
For oily sludges, Keq is calculated using vapor pressure and waste
loading is calculated exclusive of water content. For dilute aqueous
waste, partitioning is estimated to be in a water phase and a vapor phase,
and the parameter Keq is calculated using Henry's law constant; waste
loading is calculated using the total waste applied. Keq may be calculated
from site-specific land treatment soil, vapor, and solids analyses if
available. Table 5-3 summarizes the equations that make up the RTI land
treatment model.
5.2.1.6 Emissions at Short Times. When a sludge containing volatile
organics is applied onto or tilled into the soil at a land treatment site,
the maximum rate of air emissions will occur immediately after application
or tilling. Volatile organics will leave the surface and enter the envi-
ronment through wind currents. Although the RTI model is based on the
premise that emissions from land treatment are limited by vapor diffusion
through the soil, the maximum rate of air emissions immediately after
application or tilling will be limited by the gas-phase mass transfer
coefficient, kg. Within a few hours after application or tilling, the rate
of air emissions from the volatile components will be substantially less
than the maximum rate because the volatiles at the surface have been
removed by the wind and the remaining volatiles must diffuse up through a
layer of porous solids, a relatively slow process.
The land treatment model can be used for short-term emissions. The
emissions from the short-term use of the land treatment model will be
somewhat less than the oil-film model, although the initial rate from both
of these models is equivalent. The oil film model is used to estimate
maximum emission rates and the land treatment model is used to account for
%
surface drying during short-term emission estimations. The equation for
the emission rate immediately after application or tilling is:
(5-6)
The basis of the above equation is a resistance in series model where the
resistance (inverse of the mass transfer coefficient) is the sum of the
5-13
'•=?
f l 1
6a ,
[kG Keq j
irt
De Keq
-------
TABLE 5-3. RTI MODEL FOR LAND TREATMENT EMISSIONS
Emission rate equations
Short-term solution (Kyt < 0.22)
'•=?
Keq k
Long-term solution (K t > 0.22)
E » M.
2Keq Oe
exp
Keq De f t t
41
Fraction air emissions
Short-term solution [K t < 0.22]
KeqO t
1/2
at
Long-terra solution (Kyt > 0.22)
-1
Long-term solution (Fa < 0.33 and Kytfa < 0.22)
F,* • F» [1 - exP (-M - t/tj]
at a a D
Very long-term solution (t * •) (K^ I 0.62)
0.811 K, tK
Very long-term solution (t * •) (Kdtfa < 0.62)
5-14
(continued)
-------
TABLE 5-3 (continued)
P*MW
Keq
Keq
RT
oil fa
L
(used for oily sludges)3
— (10 ) - — - — (used for dilute aqueous waste)
RT
waste
4.82 (10-3) U°-78 ScG-°'67 de"0-11
Sc
G >a°a
L - L1C
KeqD.
V
t>
L 10/3 1
a
leT
0.5
(volatilization constant)
(if both air porosity and total
porosity are known)
4.83
4/3
7,
B
(if only air porosity is known)
W f
L = — ry^- (for oily sludges); L = TT (for dilute aqueous
Al Al waste)
Variable Definition
Keq Equilibrium coefficient of constituent
in the soil (dimensionless)
^Equilibrium equations are adjusted to account for volume
fractions of air and waste within the soil.
Data source
Calculated
(continued)
5-15
-------
TABLE 5-3 (continued)
Variable
KG
c
°a
°e
E
Hc
1
L
Mt
Mo
•"oil
MM
P*
R
t
*b
Definition
Gas-phase mass transfer coefficient (cm/sec)
Concentration (weight fraction) of constituent
in the oil phase or (for dilute aqueous waste)
in water
Diffusion coefficient of constituent in air,
cm2/s
Effective diffusion coefficient of constituent
in the soil, cm2/s
Emission rate of constituent, g/cm2/*
Henry's law constant for constituent,
atm«ciii3/g mol
Depth to which waste is mixed in the soil, cm
Oil or aqueous waste loading in the soil,
g/cm3
Air emissions of constituent from the soil,
g/cm2
Initial loading of constituent on the
land treatment site, g/cm2
Average molecular weight of the oil, g/g mol
Molecular weight of constituent, g/g mol
Pure component vapor pressure of
constituent, atm
Ideal gas constant, 82.1 atm»cm3/g mo1»K
Time after waste application to the land
treatment site, s
Time constant for biological decay of
constituent, sb
Data source
Calculated
Definition
Data base
Calculated
Calculated
Data base
Facility
operation
Calculated
from annual
throughput
Calculated
Calculated
Estimated
Data base
Data base
Literature
Facility
operation
Literature,
or site
specific
bTime constant is the time required for biodegradation of
63.2 percent of a pollutant.
(continued)
5-16
-------
TABLE 5-3 (continued)
Vartable-
T
£a
CT
ewaste
B
*v
Kd
Kb
Fa
Fat
foil
U
w
A
SCQ
de
f*
pa
Definition
Temperature of vapor in soil, K
Volume fraction of air-filled voids in the
soil (soil air-filled porosity)
(dimensionless)
Total porosity of the soil (equivalent to
dry basis bulk density divided by soil
particle density) (dimensionless)
Volume fraction of waste in the soil
(dimensionless)
Biorate of constituent, mg V0/g«h
Volatilization constant for constituent, s'1
Modified volatilization constant, s~l
Biodegradation constant for constituent, s"1
Fraction of constituent emitted to the air
after a long time
Fraction of constituent emitted to the air
at time t
Fraction by weight of applied waste that is
oil (organic)
Windspeed, m/s
Total waste applied to land treatment site, g
Area of land treatment site to which waste
is applied, cm2 (ro2 in calculation of de)
Schmidt number (gas phase)
Effective diameter of land treatment area, m
Viscosity of air, g/cm»s
Density of air, g/cm^
Data source
Assumed
Estimated
from litera-
ture data
Estimated
Calculated
Data base
Calculated
Calculated
Calculated
Calculated
Calculated
Definition
Estimated
Definition
Definition
Calculated
Calculated
Literature
Literature
5-17
-------
resistance of the soil and the resistance at the wind-porous solid inter-
face. The mass transfer coefficient of the soil is defined in Equation
(5-4) in the term f^J^J • The resistance of the soil to mass transfer
is the inverse of the above or (^^J • The resistance at the air-soil
interface is defined by 35^. Because Keq has previously been defined as
containing a factor to account for soil porosity, this factor (soil poros-
ity) must be included in the above"equation to maintain a consistent
definition of Keq throughout this discussion. The revised resistance is
represented by ^\ . Summing the two resistances and substituting into
Equation (5-4) gives Equation (5-6). The gas-phase mass transfer
coefficient, kQ, is calculated as described in Table 4-1 for a surface
impoundment.
5.2.1.7 Estimating the Fraction Emitted at Short Times. The fraction
of a constituent emitted to the air after some time, t, can be estimated by
integrating the equation for air emissions from time 0 to time t:
at
The exponential term can be replaced by a series,
~t7tb t 1ft 1* 1ft 3
"
e
which can Ue substituted into the above integral, and each of the
individual terms integrated. The results of these integrations are:
'.-(=£ f ""* i ' - Kf ' Mtf - Mtf • • • -
This series solution converges with only a few terms for values of t/tb
less than 1. Therefore, the following simplification can be used to
estimate the fraction emitted (i.e., integrated emissions) at short times
5-18
-------
The resistance to emissions presented by gas-phase mass transfer at the
soil surface is only considered important for the estimation of the emis-
sion rates immediately after application or tilling. This resistance is
omitted in the above equation with little loss in accuracy.
The above equation is used to predict the fraction of a constituent
emitted to the air when Kvt is less than 0.22.
5.2.1.8 Estimating the Fraction Emitted for Longer Times. For longer
times, when most of the constituent is not present in the soil, the short-
term solution (Equations (5-4) and"(5-7)) will overestimate air emissions.
Under these conditions, Equation (5-5) can be integrated to estimate the
fraction removed by volatilization. Equation (5-5) can be simplified by
Keq D rt
defining the constant, K., as - * — :
a 4 r
M_8Kj t .
E * -°2-a exp [ - Kdt - t/tb ] . (5-8)
Integrating from time 0 to t gives:
' " exp " " t/t + °'1878 • <5"9)
In the above equation, terms after the first (n > 0) in the series solution
are replaced by the constant 0.1878. This equation is used for estimating
air emissions when Kvt is greater than or equal to 0.22.
When Kvt is less than 0.22, the following simplification can be used
to estimate air emissions at longer times. An exponential decay factor is
established to relate the fraction emitted at any time, t, to the fraction
emitted at very long times (i.e., t * •) as estimated using Equation
(5-12), which follows. The resulting equation is:
Fat a Fa Cl ' exp ("
-------
For very long times (i.e., t * •) , the fraction emitted can be
estimated using the following procedure. The integrated form of the
general solution without dropping terms is:
8 ^- 1 ' exPH2"*1*2 Kdt " tftJ
f""72-. (2n+l)Z + i-
n=0 Vd
This equation can be simplified using the following rationale: For large
values of t, the exponential terns'" are negligibly small, and for large
values of n, l/(tDK 0,- the simplified
equation is:
Fa s rrr * °-2317
a f I Vb L J
The value of 0.2317 was obtained by evaluating the first 125 terms of
the series for n > 0 with negligibly small values of
125
- *— » * 0.2317 .
Combining terras and simplifying, the equation becomes:
0.81057Krith
Fa , K t ^ j b. 0.1878 . ( 5-11)
The assumptions used in developing Equation (5-11) are not valid for
small values of K
-------
Kv, and tb and then using a curve-fitting routine to derive the relation-
ship in Equation (5-12) for
-------
TABLE 5-4. ESTIMATED AIR EMISSION FRACTION AT LONG TIMES
Value of T
(T « K 1 1
0.050
0.100
0.150
0.200
0.250
0.300
0.350
0.400
0.450
0.500
0.550
0.600
0.650
0.700
0.750
0.800
0.850
0.900
0.950
1.000
1.050
1.100
1.150
1.200
1.250
1.300
1.350
1.400
1.450
1.500
1.550
1.600
1.650
1.700
1.750
1.800
1.850
1.900
1.950
2.000
Estimated
fraction
(rigorous equation)
0.222
0.313
0.381
0.435
0.480
0.518
0.551
0.579
0.604
0.626
0.646
0.664
0.680
0.694
0.708
0.720
0.731
0.741
0.751
0.760
0.768
0.776
0.783
0.789
0.796
0.802
0.807
0.813
0.818
0.822
0.827
0.831
0.835
0.839
0.843
0.846
0.850
0.853
0.856
0.859
Estimated
fraction3
0.224
0.316
0.387
0.447
0.500
0.548
0.592
0.632
0.671
0.707
0.742
0.775
0.806
0.837
0.866
0.894
0.922
0.949
0.975
1.000
1.025
1.049
1.072
1.095
1.118
1.140
1.162
1.183
1.204
1.225
1.245
1.265
1.285
1.304
1.323
1.342
1.360
1.378
1.396
1.414
Estimated
fraction^
0.277
0.348
0.407
0.456
0.497
0.533
0.563
0.590
0.614
0.635
0.654
0.672
0.687
0.701
0.714
0.725 -
0.737
0.747
0.750
0.765
0.773
0.780
0.787
0.794
0.800
0.805
0.811
0.816
0.821
0.826
0.830
0.834
0.839
0.842
0.846
0.849
0.853
0.856
0.859
0.862
h 0.81057 KHtK
F, - „ ,. ^ ? D + 0.1878
5-22
-------
TABLE 5-5. RIGOROUS VS. APPROXIMATE ESTIMATES OF EMISSION FRACTIONS
t/tb
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.10
0.30
0.30
0.30
0.30
0.30
Kvt
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0,60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
0.05
0.10
0.15
0.20
0.25
Mb
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
45.00
50.00
55.00
60.00
65.00
70.00
75.00
80.00
85.00
90.00
95.00
100.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4.50
5.00
5.50
6.00
6.50
7.00
7.50
8.00
8.50
9.00
9.50
10.00
0.17
0.33
. 0.50
0.67
0.83
Estimated Estimated
fraction fraction
(rigorous) (approximated by)a
0.25
0.35
.0.43
0.50
0.56
0.61
0.65
0.69
0.73
0.76
0.79
0.81
0.83
0.85
0.87
0.88
0.90
0.91
0.92
0.93
0.24
0.34
0.42
0.49
0.54
0.59
0.64
0.67
0.71
0.74
0.77
0.79
0.81
0.83
0.85
0.86
0.87
0.89
0.90
0.91
0.23
0.32
0.39
0.46
0.51
0.25
0.36
0.44
0.50
0.56
0.62
0.67
0.71
0.75
0.79
0.24
0.34
0.42
0.49
0.55
0.60
0.65
0.69
0.73
0.77
0.23
0.32
0.39
0.45
0.51
Estimated
fraction
(approximated by)b
0,28
0.36
0.44
0.50
0.56
0.61
0.65
0.69
0.73
0.76
0.79
0.81
0.83
0.85
0.87
0.88
0.90
0.91
0.92
0.93
0.28
0.36
0.43
0.49
0.54
0.59
0.64
0.68
0.71
0.74
0.77
0.79
0.81
0.83
0.85
0.86
0.88
0.89
0.90
0.91
0.27
0.34
0.41
0.46
0.52
See notes at end of table.
(continued)
5-23
-------
TABLE 5-5 (continued)
t/tb
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
0.30
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
Kvt
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
Mb
1.00
1.17
1.33
1.50
1.67
1.83
2.00
2.17
2.33
2.50
2.67
2.83
3.00
3.17
3.33
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
0.90
0.95
1.00
Estimated Estimated
fraction fraction
(rigorous) (approximated by}a
0.56
0.60
0.63
0.67
0.70
0.72
0.75
0.77
0.79
0.80
0.82
0.83
0.84
0.85
0.86
0.19
0.26
0.32
0.37
0.42
0.46
0.49
0.52
0.55
0.58
0.60
0.62
0.64
0.66
0.68
0.69
0.71
0.72
0.73
0.74
0.56
0.60
0.64
0.68
0.72
0.17
0.24
0.29
0.34
0.38
0.41
0.44
0.48
0.50
0.53
Estimated
fraction
-{approximated by)b
0.56
0.60
0.64
0.67
0.70
0.73
0.75
0.77
0.79
0.80
0.82
0.83
0.84
0.86
0.87
0.25
0.30
0.35
0.40
0.44
0.47
0.51
0.54
0.56
0.59
0.61
0.63
0.65
0.67
0.68
0.70
0.71
0.72
0.74
0.75
Approximated by: F .
1.128 HCf (1-1/3 t/tb)
-1
Approximated by: F_,. » 4 fl * «ir-1 fl - exp f-K.t - t/tjl + 0.1878 .
at *z I VbJ I I d DJJ
5-24
-------
where
F5 » fraction of constituent remaining in the soil
F1 s fraction of material emitted to the air at time t assuming no
biodegradation (Fgt can be estimated by setting t/tfc = 0 in
Equation (5-7) or (5-9), whichever is appropriate).
To continue modeling emissions after retilling occurs, MQ is set equal to
F,-M and t is reset to zero. If a reapplication of waste occurs, the total
waste loading is the sum of the waste remaining in the soil and the newly
applied waste:
M0=FsMQ+Mn , ' (5-15)
where
M * amount of constituent newly applied to the land treatment
n site.
To continue the modeling after waste reapplication and tilling, t is reset
to zero.
5.2.1.10 Model Selection. The RTI model was selected for use in this
regulatory effort after a review of three models of land treatment emis-
sions. The models reviewed were the RTI model, the Thibodeaux-Hwang model,
and a model developed by EPA's Office of Research and Development located
in Ada, Oklahoma (the Ada model). The review considered three selection
criteria: technical basis, representativeness, and availability of inputs.
The Ada model is the most ambitious of the three in attempting to
account for mechanisms of pollutant removal other than air emissions and
biodegradation. However, that model requires detailed site-specific model
inputs that may not be available or reasonably estimated. Because of these
characteristics of the Ada model, it was not considered appropriate for use
in the current effort. Both the Thibodeaux-Hwang and the RTI models have
input requirements that are reasonably available, both have been compared
with available measured data, and both have shown reasonable agreement with
the measurements.8 Apparently, either of these two models is satisfactory
as a means of estimating emission rates at specific times for some organic
compounds. However, if the Thibodeaux-Hwang model is used to estimate
long-term steady-state emissions, it would predict that all of the applied
5-25
-------
volatile organics are emitted because it does not account for biodegrada-
tion. Such a prediction would contradict data obtained from laboratory and
field studies that indicate biodegradation of some organic compounds in
land treatment applications.9 The RTI model, in contrast, estimates
biodegradation of individual compounds based on constituent-specific
biodegradation rates. The RTI and the Thibodeaux Hwang models predict
similar emission rates for initial volatile losses in the absence of
biodegradation. Thus, the results.^of the RTI model show varying levels of
biodegradation when used to evaluate the fate of different organic com-
pounds .
In summary, the Ada model has had limited public review, accounts for
multiple waste removal mechanisms, requires numerous detailed model inputs,
and has no published comparisons of estimated and measured emissions. The
Thibodeaux-Hwang model has been publicly reviewed, accounts for one major
waste removal mechanism (volatilization), requires reasonably available
model inputs, and there are published comparisons of measured and estimated
emissions. The RTI model has had limited publ-c review, accounts for the
two major waste removal mechanisms (volatilization and biodegradation),
requires reasonably available model inputs, and there are published com-
parisons of measured and estimated emissions. The peer review, emission
comparisons, and data availability are compelling reasons for using the
Thibodeaux-Hwang model. However, the absence of biodegradation in that
model is considered a major shortcoming because of the importance attached
to this removal mechanism by industry personnel and many independent
researchers. Emission comparisons and data availability for the RTI model
are roughly equivalent to those of the Thibodeaux-Hwang model, and the RTI
model includes terms that account for biodegradation. The RTI and the
Thibodeaux models predict similar emission rates for initial volatile
losses in the absence of biodegradation. Thus, the RTI model was selected
for use in the standards development process.
5.2.2 Waste Application Model
At land treatment facilities that do not use subsurface injection with
immediate tilling, emissions may occur during the time that waste is being
applied to the soil surface and while the waste lies on the soil before it
is tilled into the soil. No existing models were identified that predict
5-26
-------
emissions during application of an oily sludge to the soil surface. The
approach selected for this case was to calculate an overall mass transfer
coefficient of volatile material from the surface of the stream of sludge
as it falls from the end of a hose to the soil surface. The mass transfer
coefficients were calculated using an equation presented in Section 4.0
(Table 4-1). The constant in the equation for gas-phase resistance was
increased by a factor of two in an attempt to account for,an increase in
mass transfer caused by the motion of the waste stream through the air.
The equations for making this calculation are presented in Table 5-6 along
with the definitions of the variables used and the sources of input data.
5.2.3 Oil Film Model
Emissions from waste lying on the soil surface are estimated in either
of two ways depending on the form of the waste as it lays on the soil
surface. In typical situations where the applied waste is spread over the
surface of soil, the RTI land treatment model can be used to estimate
emissions. The equation for short-term emissions given above as Equations
(5-4) and (5-7) would be used for this situation. If the applied waste has
a visible oil film on top, emissions immediately after spreading are
estimated by calculating an overall mass transfer coefficient as described
in Chapter 4.0 for an oil film on a surface impoundment. The mass transfer
equation was developed by McKay and Matsuga and is based on data obtained
from liquid hydrocarbon spills on land and water.10 The equations used to
calculate emissions under this situation are given in Table 5-7 along with
definitions of the variables used.
5.2.4 Model Inputs
Typical values of input parameters for the RTI model are based
primarily on a data base developed by EPA11 from site visits and contacts
with State, regional, and industry sources supplemented by information from
recent literature. These values were chosen as reasonably representative
of average or typical practices currently used at land treatment opera-
tions. Oil loading in the soil is a model input that is calculated from
several other parameters that might change independently. Varying the
value of the oil loading rate, thus, has the same effect as varying any one
or any combination of the other parameters. Oil loading is defined by
waste throughput, the percent oil in the waste, area of the land treatment
5-27
-------
TABLE 5-6. WASTE APPLICATION EMISSION MODEL
Emission equations
E » KCA
K * Keq kg (used for oily sludges); 4 *
kg - 9.64(10-3)U°-78ScG-°-67de-0-11;
(used for dilute
aqueous waste).
u
Keq » (used for dilute aqueous waste)
(used for oi
c. - 1 (10"6) + 144 (10*4) U*2'2 Sc •°*5; Sc, - A
^ L. L P. ,U,
°*5
* - (?) •• *G - ^
U * 0.01U(6.1 * 0.63U)
0'5
Variable
E
K
Keq
H
R
T
P*
PO
Definition
Emission rate for constituent, g/s
Overall mass transfer coefficient, m/s
Equilibrium coefficient, dimensionless
Henry's law constant for constituent,
atm oiH/g mol
Universal gas constant, atm cm3/
g mol K
Temperature, K
Vapor pressure of constituent, mm Hg
System pressure (atmospheric pressure),
mm Hg
Data source
Calculated
Calculated
Calculated
Literature
Literature
Measured
Literature
Definition
(continued)
3-28
-------
TABLE 5-6 (continued)
-Variable
U
r
1
C
de
w
U*
Definition
Gas-phase mass transfer coefficient,
m/s
Windspeed, m/s
Schmidt number on gas side
Schmidt number on-liquid side
Viscosity of air, g/cm»s
Density of water, g/cm3
Density of air, g/cm3
Density of oil, g/cnP
Diffusivity of constituent in air,
cm2/s
Surface area of cylindrical waste
stream, m2
Radius of cylindrical waste stream, m
Length of cylindrical waste stream, m
Concentration of constituent-in the
waste, g/cm3
Effective diameter of waste stream
surface area, m
Molecular weight of air, g/g mol
Molecular weight of oil, g/g mol
Viscosity of water, g/cm«s
Diffusivity of component in water,
cm2/s
Friction velocity, m/s
Liquid-phase mass transfer coefficient,
m/s
Data source
Calculated
Definition
Calculated
Calculated
Literature
Literature
Literature
Estimated
Literature
Calculated
Measured
Measured
Measured
Calculated
Literature
Estimated
Literature
Data base
Calculated
Calculated
5-29
-------
TABLE 5-7. OIL FILM SURFACE EMISSION MODEL
Emission rate equation
E = KCtA
Ct • CQ [exp (-Kt/D)]
K » k- Keq (used for oily sludges)
ks-4.82 (ID'3) U^Sc^de
Sc
(used for oily sludges);
P* ^a^oi 1
Keq » jy- ....
o ^L air
* • ra
0.5
Variable
E
K
Ct
Co
D
A
kG
Definition
Emission rate for constituent, g/s
Overall mass transfer coefficient, m/s
Concentration of constituent in the
oil phase at time t
Initial concentration of constituent
in the waste *
Oil film thickness, m
Area of land treatment, m2
Gas-phase mass transfer coefficient,
m/s
Data source
Calculated
Calculated
Calculated
Definition
Measured
Measured
Calculated
(cpntinued)
5-30
-------
TABLE 5-7 (continued)
Variable
U
ScG
"a
'a
Da
de
Keq
P*
PO
MWoil
MWa
PL
R
T
Definition
Windspeed, m/s
Schmidt number— gas phase
Viscosity of air, g/cm»s
Density of air, g/cm^
Diffusion coefficient of constituent
in air, cm^/s
Effective diameter of land treatment
area, m
Equilibrium coefficient of constituent
Vapor pressure of constituent, mm Hg
Atmospheric pressure, mm Hg
Molecular weight of the oil, g/g mol
Molecular weight of air, g/g mol
Density of oil, g/cm3
Universal gas constant, atm cm^/g mol K
Temperature, K
Data source
Definition
Calculated
Literature
Literature
Literature
Calculated
Calculated
Literature
Definition
Definition
Literature
Estimated
Literature
Measured
5=31
-------
site, and the depth to which the waste is mixed in the_soi][Jtilling
depth). Typical values of oil loading are defined fromjnedian values for
those parameters by which it is defined. The data basTThows annual
throughput varying from about 2 Mg/yr to about 400,000 Mg/yr with a median
value of about 1,800 Mg/yr. The area of land treatment sites ranges from
less than 1 hectare (ha) to about 250 ha with a median value of 5 ha. The
data base shows tilling depth varying from 15 cm to one case of 65 cm, with
most being in the range of 15 to 3Q cm. The single most frequently
reported tilling depth is 20 cm, which is selected as a typical value.
This value is in line with values of 15 to 30 cm reported in another
study.12 The data base shows oil content of the waste streams varying from
about 2 to 50 percent, with a median value of about 12 percent and a mode
value of 10 percent. The 10-percent figure is selected as typical.
Average molecular weight of the oil from which a particular
constituent evaporates is one of the determining factors in the rate of
evaporation and thus must be specified. Little data are available as
guidance for selecting a value for this parameter. The distribution of
constituents by molecular weight in land-treated wastes is not well known.
In one field measurement study of land treatment emissions,^ a value of
282 g/g mol was used as the average molecular weight of the oil. This
value was based on distillation of oil from a refinery sludge and identi-
fication of the constituent corresponding to the midpoint distillation
temperature (i.e., the temperature at which 50 percent of the oil was
distilled). The value 282 g/g mol is selected for use. A sensitivity
analysis using the RTI model shows that emissions are not highly sensitive
to this parameter.
Soil air porosity and total porosity impact the effective diffusivity
of a constituent in the soil. Very little soil porosity information has
been identified. One study reported measured values of soil porosity in a
land treatment plot as ranging from 43.3 to 65.1 percent,1* with an average
value of about 50 percent. The literature values are assumed to represent
air porosity. Total soil porosity would include the air porosity and the
space occupied by oil and water within the soil. One field study reported
measured values of both total porosity and air-filled porositv.15 Measured
values of total porosity ranged from 54.7 to 64.8 percent, with an average
5-32
-------
value of 60.7 percent. Measured values of air-filled porosity ranged from
27.4 to 46.9 percent, with an average of 37.2 percent. The value of 61
percent for total porosity is assumed to be a representative value. A
value of 0.5 is used in the model as a default for'air porosity.
Biorate data used in the RTI model data base (CHEMDAT7) represent
measured rates in aqueous systems. In order to use the aqueous biorate
data in a land treatment process, a factor was established for converting
aqueous data to land treatment values using measured data for benzene. A
re ent publication by the American"Petrol euro Institute (API) reported
experimentally determined values of biological decay constants for land
treatment studies using two different soil types.15 Decay constants were
measured for six compounds including two compounds, benzene and toluene,
that have aqueous biorates in the land treatment model data base
(CHEMDAT7). For benzene, the ratio of the API data, measured in units of
day"1, and the aqueous data, measured in units of mg V0/gbiomass*nr- was
calculated as 0.00179. This value is also a close approximation of the
ratio of the two data points for toluene, the other compound for which data
from both sources were available. The other compounds for which data were
reported by API did not have referenced aqueous data in the date base. The
above calculated relationship was used to calculate equivalent aqueous data
values for those compounds. Reported and calculated values of aqueous
biorates and land treatment biological decay constants are presented in
Table 5-8. The ratio of 0.00179 is used for all compounds to convert from
aqueous biorates to decay constants that can be used in the land treatment
model. The input parameter for the land treatment model is a biological
decay time constant, tfa, in units of seconds. The equation for calculating
tfc from the aqueous biorate is derived as follows.
The biological decay time constant is, by definition, equal to the
reciprocol of the biological decay constant, or
th = -J- (5-16)
b Kb
where Kb * biological decay constant. The ratio, r, of decay constant to
aqueous biorate is:
5-33
-------
TABLE 5-8. MEASURED AND ESTIMATED BIORATES AND DECAY CONSTANTS
FOR SELECTED ORGANIC CONSTITUENTS
Organic
.constituent n
Benzene
Ethyl benzene
Xylene(-o)
Naphthalene
Tol uene
Methyl naphthalene
Aqueous
bi orate,
19.0
46. 4d
40.8d
42. 5d
73.5
24. Od
Calculated
decay
constant ,
i day'1
0.034
0.083
0.073
0.076
0.132
0.043
Measured decay
constant, day"*3
Nunnb
0.034
0.083
0.073
0.076
0.106
0.043
Kidmanc
0.013
0.076
0.026
0.050
0.119
0.059
Reference 17.
bOata obtained using a clay loan soil (Nunn soil).
cData obtained using a sandy loam soil (Kidman soil)
dValues calculated from API experimental data.
5-34
-------
Substituting into Equation (5-16) gives
To obtain a result in seconds, this equation must be multiplied by 86,400
s/day. Making this conversion and inserting the value of r (i.e., 0.00179)
gives:
, 86.400 . 4.83 (107)
^b " 0.00179 B " B
For situations in which petroleum wastes are landfarmed and no
information is known about the nature of the volatile materials, it is
possible to estimate a default biorate from the average decay constant
values reported in the API investigation, 0.07 day"1, which corresponds to
a biorate, B, of 40 mg/g-h. This value is between the values for benzene
and toluene in the data base. The average value of the biological rate
constant in the two soils investigated by API was not significantly
different.
In summary, parameters and selected typical values for use in the RTI
model are as follows:
Annual waste throughput s 1,800 Mg
Area of land treatment = 5 ha
Oil content of waste = 10 percent
•
Average molecular weight of the oil = 282 g/g mol
Soil air porosity - 0.5
Soil total porosity « 0.61
Tilling depth « 20 cm
Temperature s 25 °C.
5-35
-------
5.2.5 Estimation of Total VQ Emissions
The preceding discussion has been limited to estimating emissions of a
single constituent in a hazardous waste. Using the models presented here
to estimate total VO emissions can be accomplished using any of several
approaches. The most obvious approach, and the one that should give the
most accurate results, would be to obtain a detailed analysis of the
constituents in the waste being land treated. The emission equations could
be used to calculate emissions of each constituent, and total emissions
could be calculated by summing the"emissions of individual constituents.
In many cases, a detailed analysis of the applied waste may not be avail-
able, and other, less accurate methods may be needed to estimate total VO
emissions. An alternative to the constituent approach could make use of a
boiling curve or steam stripping test of a sample of the waste. Experi-
mental data developed by Chevron Research Company18 indicate that a large
fraction of the constituents that boil at temperatures of 400 °F or lower
will be emitted to the atmosphere and that those constituents with higher
boiling points will tend to remain in the soil for a sufficient time to
undergo biodegradation. Similar results are obtained by applying the RTI
land treatment model to the constituents in the CHEMDAT7 data base. If a
sample of waste were subjected to a laboratory boiling test or other
equivalent test at a temperature of 400 «F,. the fraction of oil evaporated
would approximate the fraction that potentially would be emitted to the air
in a land treatment operation.
A third approach to estimating total VO emissions would again make use
of the experimental results generated by Chevron Research. The test
results showed that approximately 25 percent of the applied oil in the land
treatment test was emitted to the air. In the absence of a detailed
constituent analysis and with no boiling or steam stripping test of the
waste, a crude estimate of total VO emissions could be made by assuming
that emissions are equal to 25 percent of the applied oil. This approxi-
mating alternative would only apply to raw oily refinery wastes that have
not undergone any pretreatment to remove VO.
5.2.6 Example Calculations
5.2.6.1 Emissions from Land Treatment Soil. The following calcula-
tion demonstrates the use of the RTI model to calculate the long-term
5-36
-------
fraction of applied material emitted to the atmosphere and to calculate the
short-term and long-term emission rates and emission fractions. The calcu-
lations are made for benzene at a concentration of 2,000 ppm by weight in a
waste stream that is 10 percent oil.
Input values are:
Land area s 2.5 ha (half of total area of 5 ha
assumed active)
Annual throughput = 1,800 Mg
Oil content of waste = 10% (by weight)
Benzene concentration in oil s 2,000 ppm (by weight) (2 mg/g oil)
Tilling depth a 20 cm
Soil air porosity * 0.5
Soil total porosity * 0.61
Average molecular weight of oil * 282 g/g mol
a. Calculate oil loading (Equation from Table 5-3):
, (1'800x 1Q WeXQ-i goil/gwaste) 3
L « - 3 - * - s 0.036 gQ,-i/cm
(2.5 x 10® cnf)(20 cm) 01
b. Identify constituent properties:
Benzene properties:
B = 19.00 mg V0/gbiomass*n
Da = 0.088 cn>2/s
P* = 95.2 mm Hg = 0.125 atm.
»
c. Calculate the equilibrium coefficient (Equation from Table 5-3):
K - P* ""oil 6a = (0.125 atm) (282 q/q mol) (0.5 cm3/cm3)
6q " RT L " (82.05 atm-cm3/g mol -K) (298 K) (0.036 g/cm3)
Keq = 0.02002 .
d. Calculate the biological degradation time constant (Equation from
Table 5-3):
5-37
-------
e. Calculate the effective diffusivity of constituent in the soil (Equa-
tion from Table 5-3):
f 10/3 10/3
€ O /A C \ ***/ J o
0 » 0 a A * 0.088 cnr/s Viii2-i « 0.0235 cm /s .
e a CTZ o.er
T2KeqD
f. Calculate the value of K, » 3-= :
d 4 lz
K » (9.87) (0.0235) cm2/s (0.02002) , 2 g
d 4 (400) cm2
g. Calculate the fraction of constituent emitted to the air after'a long
time (Equation (5-11)):
0.81 K^ti,
F , a D > 0.1878 = 0.90 .
a Kdtb + i
h. Calculate the long-terra emission rate after 60 h (216,000 s).
Keq ^ . Q-02002 X4g0°235 x 216'000 = 0.254 .
Use Equation (5-5) (long-term equation):
* •«. -^ < i—n^-j
E m 2 x 0.72 x 2 * 0.02002 x 0.0235 x e f-0.02002 x Q.Q235 x 9.87 x 216.000J
x e[-216,000/2.54(106)]
• 3.38(10-6) e(-0.627) e(-0.085) - 3.38(10'6) e(-0.712)
E * 3.38(10-6) (0.491) « 1.7(10-6)
cm s
5-38
-------
c. Calculate the short-term emission rate after 15 min (900 s)
where
Keq . 0.02002 x^O.0235 (900)
kQ = 4.82 (ID'3) U°-78 Sc-°-67
U = windspeed = 4.47 m/s
e s effective diameter of land treatment area
0.5
de »
178 m
Sc
•G - D
a a
where
pg = viscosity of air = 1.81(10'*) g/cm»s
Pa = density of air = 1.2 (10'3) g/cm3
Da = 0.088 cm2/s
s . 1-81 (to"4) . 1>71
5 1.2 (10"-3) (0.088)
k
4.82(10'3) (4.47)0'78 (1.71)'0'67 (178)•°*11 - 0.0061 m/s - 0.61 cm/s
it
E -
_ 0.72 g 2 mq
1 cm 20 cm g
kQ Keq JDe Keq
1
,-t/tb
3.14 x 900
0.61 x 0.02002 Jo.0235 x 0.02002.
"5
•(-900/2.54(10.6))
0.072 (0.0004) e(-0.0004) - 2.87 (10")
cm s
5-39
-------
Table 5-9 shows estimated emission rates and emission fractions for
various times up to 40 days (960 hours).
5.2.6.2 Emissions from Haste Application. The following is an
example calculation for the application of an oily waste to a land treat-
ment plot using the equations in Table 5-6. For benzene in waste oil, the
calculation is:
Input values:
r » 0.038 m
L • 0.46 m
/*a * 1.81 (10-4) g/cm»s
/»a -1.2 (10-3) g/cm3
U - 4.47 m/s
R * 82.05 atm»cm3/K«g mol
T « 298 K
C * 200 ppm - 200 pg/g - 0.0002 g/cm3 » 200 g/m3
(assuming a density of 1 g/cm3)
A - 2 irL • 2(3.14)(0.038 m)(0.457 m) - 0.11 m2-
p. * 1 g/cm3
MWa = 29 g/g mol
MW0ii * 282 g/g mol.
a. Calculate the effective diameter of the waste stream surface
(Equation from Table 5-6):
r..,0.5
de > pjj . 0.37 m .
b. Calculate the Schmidt number (Equation from Table 5-6):
m _£a . 1.81 (IP"4) Q/cm*s
°G * /»a °a * [1.2 (10'3) g/cm3] (0.088 cm*
c. Calculate the equilibrium coefficient (Equation from Table 5-6):.
. _?1 'a ""oi 1 . (95.2 mm Ha)n.2 (IP"3) Q/cm31(282 q/q mol) . 0<0015
po ^L ^a (760 mm Hg)(l g/cm3)(29 g/g mol)
5-40
-------
TABLE 5-9. ESTIMATED EMISSION RATES AND FRACTIONS EMITTED
VERSUS TIME FOR EXAMPLE LAND TREATMENT CALCULATION
Time after
application/tilling,
h
Emission
rate,
'6 mg/cm2»s
Equation
used
Fraction
emitted,
fraction
Equation
used
1
2
4
8
24
48
96
192
480
960
14.4
10.3
7.30
5.12
2.90
1.98
1.08
0.348
0.011
0.00004
Short term
Short term
Short term
Short term
Short term
Short term
Long term
Long term
Long term
Long term
0.073
0.104
0.147
0.207
0.356
0.497
0.674
0.827
0.899
0.901
Short term
Short term
Short term
Short term
Short term
Short term
Long term
Long terra
Long term
Long term
5-41
-------
d. Calculate the gas-phase mass transfer coefficient (Equation from
Table 5-6):
kG = 9.64(10-3) U°'78 ScG -°'67 de-0-11
= 9.64(10'3)(4.47)0-78(1.71)-°'67(0.37)"0-11
= 0.024 m/s
e. Calculate the overall mass transfer coefficient (Equation from
Table 5-6):
K » kg Keq = (0.0015)(0.024) m/s » 0.000036 m/s
f. Calculate emissions from a unit volume of waste (Equation from
Table 5-6):
E » KCA » 0.000036 m/s (200 g/m3) (0.11) m2
= 7.92 (10"4) g/s .
Using a calculated fall time of 0.305 s:
E » (7.92»10~4 g/s)(0.30 s) = 2.38 (10"4) g .
Stream volume = Tr2L = (3.14)(0.038 m)2(0.46 m) = 0.002 m3
2,073 cm3
.J\ linn — /_Ji
Mass of constituent = (0.002 m"5) (200 g/nr) = 0.4 g .
g. Calculate the fraction of constituent emitted to the air:
Fraction emitted » 2'3QJi°g ^q - 0.00059 - 0.06% .
5.2.6.3 Emissions from an Oil Layer on Soil Prior to Tilling. An
example calculation for butanol-1 in an oil layer on the soil surface of a
land treatment site is given below using the equation from Table 5-7.
Input values:
1.81
1.2 (10-3) g/cm3
4.47 m/s
5-42
-------
MW0ii - 282 g/g mol
/>L * 1 g/cm3
MWa * 29 g/g mol
C « 0.0002 g/cm3 = 200 g/m3
A * 25,000 m2
t = 24 h = 86,400 s
d * 0.072 m
a. Calculate the effective diameter of the soil surface (Equation from
Table 5-7):
de .
b. Calculate the Schmidt number (Equation from Table 5-7):
Scr -
5
&- - - 1.81 do"4) q/cm.s
a°a [1.2 (10~j) g/cmj] (0.080 cnT/s)
c. Calculate the equilibrium coefficient (Equation from Table 5-7):
Van P* Pq ""oil . (6.5 mm Hg)[1.2 (10"3) q/cm3](282 q/q mol)
K6q s n ' SET" ~ ^ ^"f ^
po ^L MWa (760 mm Hg) (1 g/cm^) (29 g/g mol)
* 1.0 (10"4) .
d. Calculate the gas-phase mass transfer coefficient (Equation from
Table 5-7):
kg = 4.82 (ID'3) U°'78 SCg0-67 de'0'11 » 4.82(10'3)(4.47)°-78(1.89)'0-67
(178)-0'11
= 5.7 (10"3) .
e. Calculate the overall mass transfer coefficient (Equation from
Table 5-7):
K -
Keq = [5.7 (10"3) m/s][1.0 (10~4)] - 5.70 (10"7) m/s
f. Calculate the fraction of constituent emitted to the air at time t
(Equation from Table 5-7):
5-43
-------
f , i . t , l _ e-Kt/0 » ! . e [.5.70 (10-7)(86,400)70.072]
o
= 1 - 0.50 = 0.50 .
5.2.6.4 Examples of the Use of the Land Treatment Model for Specific
Cases. To illustrate how the land treatment model is used to estimate
long-term and short-term emissions from various waste management options,
the following case studies are presented. Each of these examples repre-
sents the land treatment of API separator sludge/DAF float, with the
exception of example 5. The waste contains 10 percent organics and is
applied at a rate of 2,500 Mg of waste per year. The loading, benzene
concentration, and porosity are identical to the example presented in
Section 5.2.6.1. The land area is 35,000 square meters.
Example 1. Waste is applied monthly for 9 months of the year. No
waste is applied during December, January, or February because the soil is
frozen or saturated with water during those months. The waste is applied
from a vacuum truck by spraying onto the soil surface with a nozzle. The
soil is tilled 24 hours after application and is tilled again (no waste
application) after 2 weeks.
Two calculations of air emissions are required: after application and
after tilling. The first time period is 1 day and the second time period
is 14 days. The amount of waste applied is 2,520 Mg/9 or 280 Mg/applica-
tion. It is assumed that the oil content in the soil is 0.036 grams of
oil/cm3. The amount of oil applied each application is 0.036/9 or 0.004
g/cm3. The concentration of benzene in the waste is 2,000 ppmw and the
concentration of benzene added to the*oil in the soil is 2,000/9 or 222
ppnw each application. The land area is 3,500 square meters. After appli-
cation, the liquid is assumed to seep into the soil to a depth of 5 cm, and
the oil loading in the waste on the soil surface was assumed to be the same
as the oil content of the soil.
To estimate the amount of air emissions between application and
tilling, the following parameters are used in CHEMDAT7:
5-44
-------
• Concentration of benzene: 2,000 ppmw
• Tilling depth: 5 cm
• Time of calculation: 1 day
• Loading (10 percent oil): 0.036 g/cm^.
The fraction lost during the first day is 98 percent with 0.007 percent
lost to biological decay.
The fraction lost after the first tilling is estimated by the use of
the following parameters:
• Concentration of benzene: 222 ppmw
• Tilling depth: 20 cm
• Time of calculation: 14 days
• Loading: 0.036 g oil/cm3.
The calculated fraction lost during the first tilling period is 0.89 to air
emissions and 0.095 to biodegradation. This fraction is independent of the
concentration of benzene and is expected to also apply to the second
tilling period.
The loss of benzene during the month is 97.8 percent during the
spreading period and (1-0.978-0.0067) x 0.89 x 100 or 1.4 percent during
the tilling periods, for a total air emission loss of 99.2 percent.
Example 2. Waste is applied weekly except when the ground is satur-
ated with water. Waste is applied from a vacuum truck, and the waste is
spread over the surface of the soil. The soil is tilled on the day
following the application and weekly between applications. The waste is
applied monthly to the land treatment plot throughout the year.
As in example 1, two calculation^ of emissions are required: after
application and after tilling. The first time period is 1 day and the
second time period is 7 days. The amount of waste applied is 2,520 Mg/12
or 210 Mg/application. It is assumed that the oil content in the soil is
0.036 grams of oil/cm3. The amount of oil applied each application is
0.036/12 or 0.003 g/cm3. The concentration of benzene in tne waste is
2,000 ppmw and the concentration of benzene added to the oil in the soil is
2,000/12 or 167 ppmw each application. The land area is 3,500 square
meters. After application, the liquid is assumed to seep into the soil to
5-45
-------
a depth of 5 cm, and the oil loading in the waste on the soil surface was
assumed to be the same as the oil content of the soil.
To estimate the amount of air emissions between application and
tilling, the following parameters are used in CHEMOAT7:
• Concentration of benzene: 2,000 ppmw
• Tilling depth: 5 cm
• Time of calculation: 1 day
• Loading (10 percent 6T1): 0.036 g/cm3.
The fraction lost during the first day is 98 percent with 0.007 percent
lost to biological decay.
The fraction lost after the first tilling is estimated by the use of
the following parameters:
• Concentration of benzene: 167 ppmw
• Tilling depth: 20 cm
• Time of calculation: 7 days
• Loading: 0.036 g oil/cm3.
The calculated fraction lost during the first tilling period is 0.80 to air
emissions and 0.083 to biodegradation. This fraction is independent of the
concentration of benzene and is expected to also apply to the second
tilling period.
The fraction loss of benzene during four tilling periods is (0.805) +
(0.0995)(0.805) + (0.805)(0.0995)2 + 0.805 (0.0995)3 or 0.89.
The loss of benzene during the month is 97.8 percent during the
spreading period and (1-0.978-0.0067).x 0.89 x 100 or 1.4 percent during
the tilling periods, for a total air emission loss of 99.2 percent.
Example 3. The waste is dewatered prior to land treatment. The
parameters are the same as those used in example 1, except the waste is
dewatered and the filter cake is land-treated. The oil content of the
filter cake is 20 percent. The waste is applied from a dump truck and is
spread by a bulldozer. The waste is tilled into the soil on the day
following spreading. It is assumed that the dewatering process removes
60 percent of the oil from the waste.
5-46
-------
As in the preceeding examples, two calculations of air emissions are
required: after spreading and after tilling. The first time period is
1 day and the second time period is 14 days. The amount of waste applied
is 2,520 Mg/9 or 280 Mg/application. It is assumed that the oil content in
the soil is 0.036 grams of oil/cm^. The amount of oil applied each
application is 0.036/9 or 0.004 g/cm3. The concentration of benzene in the
waste is 2,000 ppmw and the concentration of benzene added to the oil in
the soil is 2,000/9 or 222 ppmw each application. The land area is 3,500
square meters. After application;"*the liquid is assumed to be retained in
the waste, and the oil loading in the waste on the soil surface is assumed
to be the same as the oil content of the waste.
To estimate the amount of air emissions between application and
tilling, the following parameters are used in CHEMDAT7:
• Concentration of benzene: 2,000 ppmw
• Tilling depth: 2 cm
• Time of calculation: 1 day
• Loading (20 percent oil): 0.2 g/cm^.
The fraction lost during the first day is 98 percent with 0.006 percent
lost to biological decay.
The fraction lost after the first tilling is estimated by the use of
the following parameters:
• Concentration of benzene: 222 ppmw
• Tilling depth: 20 cm
• Time of calculation: 14 days
• Loading: 0.036 g oil/cm3.
The calculated fraction lost during the first tilling period is 0.89 to air
emissions and 0.095 to biodegradation. This fraction is independent of the
concentration of benzene and is expected to also apply to the second
tilling period.
The air emission loss of benzene during the application period is 98.3
percent during the spreading period and (1-0.983-0.006) x 0.89 x 100 .or 1.2
percent during the tilling periods, for a total air emission loss of 99.5
percent. The air emissions on the basis of the untreated waste would
5-47
-------
depend on the recovery of cil in the dewatering process and the air
emissions from the dewatering process.
Example 4. The waste is tilled as it is applied to the soil. The
tilling depth is 20 cm. The period between tillings is 3 days. Waste is
applied monthly.
Only one calculation of air emissions is required: after tilling.
The amount of waste applied is 2,520 Mg/12 or 210 Mg/application. It is
assumed that the oil content in the soil is 0.036 grams of oil/cm^. The
amount of oil applied each application is 0.036/12 or 0.003 g/cm3. The
concentration of benzene in the waste is 2,000 ppmw, and the concentration
of benzene added to the oil in the soil is 2,000/12 or 167 ppmw each
application. The land area is 3,500 square meters.
The fraction lost after the first tilling is estimated by the use of
the following parameters:
• Concentration of benzene: 167 ppmw
• Tilling depth: 20 cm
• Time of calculation: 3 days
• Loading: 0.036 g oil/cm^.
The calculated fraction lost during the first tilling period is 0.60 to air
emissions and 0.055 to biodegradation. This fraction is independent of the
concentration of benzene and is expected to also apply to the following
tilling periods.
The loss of benzene during the first tilling period is 60 percent with
(100-60-5.5) or 34.5 percent remaining. The loss of benzene during the
second tilling period is 0.60 (34.5) 91* 20.7 percent, with a fraction of
(0.345)2 or 0.119 of benzene remaining. The total loss of benzene for the
month is 92 percent.
Example 5. Waste is applied monthly for 9 months of the year. No
waste is app'ied during December, January, or February because the soil is
frozen or saturated with water during those months. The waste is applied
from a vacuum truck by spraying onto the soil surface with a nozzle. The
soil is tilled 24 hours after application and is tilled again (no waste
application) after 2 weeks. The waste is aqueous containing 10 percent
organics and 2,000 ppmw benzene. The waste contains 10 percent solids by
weight.
5-48
-------
Two calculations of air emissions are required: after application and
after tilling. The first time period is 1 day and the second time period
is 14 days. The amount of waste applied is 2,520 Mg/9 or 280 Mg/applica-
tion. It is assumed that the water content in the soil is 0.10 grams/cm^.
The concentration of benzene in the waste is 2,000 ppmw and the concentra-
tion of benzene added to the water in the soil is 300 ppmw each applica-
tion. The land area is 3,500 square meters. After application, the liquid
is assumed to seep into the soil to a depth of 5 cm, and the water loading
in the waste on the soil surface was assumed to be the same as the water
content of the soil.
To estimate the amount of air emissions between application and
tilling, the following parameters are used in CHEMDAT7:
• Concentration of benzene: 300 ppmw
• Tilling depth: 5 cm
• Time of calculation: 1 day
.• Loading (10 percent): 0.10 g/cnP.
The fraction lost to air emissions during the first day is 99.8 percent
with less than 0.01 percent lost to biological decay.
5.2.7 Assumptions and Sensitivity Analyses
The RTI model incorporates the following assumptions to simplify
development and use of the model:
• Volatilization and biodegradation are the predominant waste
removal mechanisms (i.e., other mechanisms can be ignored).
• Waste is mixed uniformly within a surface layer of the soil.
• Waste does not flow as a liquid within the soil.
• The adsorption isotherm of a constituent is linear within
the application surface layer and does not change with time.
• No bulk flow of gas is induced within the soil.
The rate of biological decay/chemical reaction is a first-
order process.
The diffusion coefficient does not vary with either concen-
. tration or time.
5-49
-------
• The concentration of a constituent in the gas phase at the
surface of the soil is much lower than the concentration of
that constituent in the gas phase within the soil.
• There is no diffusion of waste into the soil beneath the
zone of incorporation.
• Liquid-vapor equilibrium is established at all times within
the soil.
The RTI model was evaluated for sensitivity to each of the input
parameters. In the analysis, each input parameter was varied over the
entire range of reasonable values. The effect on emissions of parameter
variations was noted, and the parameters showing the highest sensitivity
were identified.
Individual constituent properties were found to have the most
significant impact on land treatment emissions. These properties include
vapor pressure, diffusivity in air, and biodegradation rate. The more
volatile compounds are mostly emitted to the atmosphere unless a volatile
compound also has a high biodegradation rate or low diffusivity. Compounds
with low vapor pressures tend also to have lower diffusivities; thus, even
if such compounds also have moderate or low biorates, they are mostly
biodegraded rather than emitted to the air.
Operating and field parameters also have an impact on emissions but to
a lesser extent than compound properties. Tilling depth, soil porosity,
and waste loading all have an impact on air emissions, with the largest
impact on the more volatile compounds. Tilling depth can have a sub-
stantial impact on air emissions of volatile compounds, especially if a
compound also has a relatively high btorate. As tilling depth increases,
materials remain in the soil for a longer time and thus have a greater
chance of being biodegraded.
Waste loading can be varied in two ways: by increasing the
concentration of a compound in the waste or by increasing the amount of
material applied to the soil per unit area. If the concentration of a
compound is changed, air emissions change in direct proportion to the
change in concentration (i.e.. the fraction of the compound that is emitted
to the air remains constant). If total waste loading is changed, air emis-
sions change in the same direction as the change in loading but not in
5-50
-------
direct proportion (i.e., the fraction emitted is lower for higher loading
rates). These results assume that a treatment site is not overloaded to
the point where biodegradation ceases to be a first-order process.
Average molecular weight of the oil has an effect on air emissions,
but the magnitude of the effect is less than that of the other parameters
studied. As average molecular weight goes up, the fraction emitted for a
specific constituent increases; the fraction emitted decreases if the aver-
age molecular weight is reduced.
5.3 REFERENCES
1. Crank, J. The Mathematics of Diffusion. London, Oxford University
Press. 1970. p. 45-47.
2. Crank, J. Diffusion in Polymers. New York, Academic Press. 1968.
p. 16.
3. Huddleston, R. I., C. A. Bleckman, and J. R. Wolfe. Land Treatment
Biological Decay Processes. In: Land Treatment, a Hazardous Waste
Management Alternative. Water. Resources Symposium Number Thirteen.
Austin, The University of Texas at Austin. 1985. p. 44.
4. Letter and attachment from Klett, R. J., Chevron Corporation, to
Wyatt, Susan R., U.S. Environmental Protection Agency, July 8, 1987,
p. 1-2. Comments on Draft Emission Models Report.
5. Levenspiel, 0., Chemical Reaction Engineering. John Wiley and Sons,
Inc., New York, New York. May 1967. p. 47.
6. Millington, R. J, and J. P. Quirk. Permeability of Porous Solids.
Trans. Faraday Society. 57:1200-1207. 1961.
7. Letter and attachments from Sonenville, G. F.. Chevron Research Com-
pany, to Thorneloe, S. A., U.S. EPA. May 22, 1986. p. 19. Comments
on preliminary draft BID for land treatment.
»
8 GCA Corporation. Air Emissions from Land Treatment—Emissions Data
and Model Review. Draft Technical Note. Prepared for U.S. Environ-
mental Protection Agency. Research Triangle Park, NC. September
1985. Chapter 4.
9. Radiation Technologies, Inc. Treatability Data in Support of a Treat-
ment Zone Model for Petroleum Refining Land Treatment Fa<£nties.
Prepared for American Petroleum Institute. Washington, DC. Marcn
1986. p. 154.
10. McKay, Donald, and Ronald S. Matsuga. Evaporation Rates of Liquid
Hydrocarbon Spills on Land and Water. The Canadian Journal of Chemi-
cal Engineering. 51:434-439. August 1973.
5-51
-------
11. Memorandum from Thorneloe, S., EPA/OAQPS, to Durham, J., EPA/OAQPS,
January 31, 1986. Land treatment data base.
12. Environmental Research and Technology. Land Treatment Practices in
the Petroleum Industry. Prepared for American Petroleum Institute.
Washington, DC. June 1983. p. 1-2.
13. Radian Corporation. Field Assessment of Air Emissions and Their Con-
trol at a Refinery Land Treatment Facility. Volume I. Prepared for
U.S. Environmental Protection Agency. Cincinnati, OH. September
1986. p. 154 and 179.
14. Reference 13. p. 43.
15. Utah Water Research Laboratory. Evaluation of Volatilization of
Hazardous Constituents at Hazardous Waste Land Treatment Sites. For
U.S. Environmental Protection Agency, Office of Research and Develop-
ment, Ada, Oklahoma. Undated, p. 55.
16. American Petroleum Institute. Land Treatability of Appendix VIII
Constituents Present in Petroleum Refinery Waste: Laboratory and
Modeling Svudies. API Publication No. 4455. American Petroleum
Institute, Washington, DC. April 1987. P. 3-10 and 3-12.
17. Reference 16.
18. Ricciardelli, A. J., et al. 1986. Landfarm Simulator Program.
Summary Report. Chevron Corporation, Richmond, California. July
1987. p. 18-24.
5-52
-------
6.0 LANDFILLS AND WASTEPILES
6.1 INTRODUCTION
The main facilities in this category that constitute the model plants
are waste fixation units, open landfills, closed landfills, and wastepiles.
All wastes that contain free liquids and that are destined for
disposal in a landfill must be treated to eliminate the free liquids. This
Is often accomplished by adding a "fixative" to the waste, such as portland
cement, cement kiln dust, or lime flue dust. These materials react with
water in the waste and set up to form a dry material that encapsulates or
binds the organic constituents of the waste. This fixation process is most
often conducted in lined open pits or open tanks Into which the liquid
waste Is poured. The fixative then is added and the materials are
thoroughly mixed, most often with a backhoe. Alternatively, mechanical
mixers such as pugmills can be used to blend the waste and fixative. Emis-
sions are generated for as long as the waste remains in the pit. Emissions
from this process may be estimated by using the open dump model.
A landfill is a facility, usually an excavated, lined pit, into which
wastes are placed for permanent disposal. Emissions from open landfills,
those still receiving wastes, can be estimated by applying the Research
Triangle Institute (RTI) land treatment model.1 Emissions from closed
landfills, those filled to design capacity and with a cap (final cover)
-installed, can be estimated with the RTI closed landfill model.
Wastepiles are temporary accumulations of waste. They serve a storage
function and have a limited life span. Emissions from wastepiles can be
estimated by applying the RTI land treatment model.2
6.2 CLOSED LANDFILLS
6.2.1 Emission Model Equations
The RTI closed landfill model 1s used to estimate the time-dependent
behavior of emissions from waste placed In a closed (capped) landfill that
6-1
-------
Is vented*to the atmosphere and (as a special case) open-landfill waste
covered with dally earth covers. This model accounts for escape of the
constituent of Interest via two primary, Independent mechanisms: diffusion
through the cap and convective loss from barometric pumping through the
vent(s). It 1s the purpose of this section to describe the model and its
evolution in a general way and to present all model equations and major
assumptions.
The model is based primarily upon the work of Farmer et al.,3 who
'*•
applied Pick's first law for steady-state diffusion to closed landfills.
Farmer's equation utilizes an effective diffusion coefficient for the soil
cap based on the work of Millington and Quirk.4 A previous EPA study5 was
dedicated to the evaluation of available models for estimating emissions
from hazardous waste treatment, storage, and disposal facilities (TSDF),
Including closed landfills. This study endorsed the models of Farmer
et al.6 and Thibodeaux7 for closed landfills, apparently because of their
treatment of soil-pore diffusion. Of the two, the Farmer et al.8 model
alone has received experimental verification (although to a limited degree)
via a laboratory experiment using hexachlorobenzene-containing waste in a
simulated landfill.
The diffusion model of Farmer et al.9 was subsequently modified by RTI
to allow for convective losses of the constituent of interest from the
landfill, which can occur from barometric pumping. Furthermore, the
decline In the emission rate from closed landfills over the long term was
accounted for via the Incorporation of a time-dependent decay function.
The barometric pumping emission mechanism results from changes in
atmospheric pressure—as the pressure*1s lowered, gas flows out of the
landfill through the vent(s) to equalize Internal pressure. The contri-
bution to total emissions resulting from barometric pumping equals the
concentration of the constituent of Interest in the gas within the landfill
multiplied by the total flow of gas from the landfill. It 1s recognized
that under certain conditions (e.g., the presence of significant biomass)
blogas could be generated in a landfill. Biogas consists of methane and
carbon dioxide, which is produced from the action of bacteria on organic
material. Because of the convective or purging action of blogas in remov-
ing the constituent of Interest In vapor form, biological decay (if it
6-2
-------
occurs) results in a net increase in the emission rate. However, it should
be noted that there is no evidence that there is significant biomass
(necessary for biogas generation) in any chemical waste landfill. There-
fore, in this analysis it is assumed, as suggested in the literature, that
the toxic property of the waste will inhibit biological processes and thus
prevent biogas generation. *0 Hence, closed landfill model equations
presented in this document account for diffusion through the cap and baro-
metric pumping only.
The equations inherent in the-RTI closed landfill model are as
follows: Pick's first law for steady-state diffusion, based on the work of
Farmer et al.,1* for a landfill is given as:
Ji ''-Dei (C2i - Cs1)/l (6-1)
where
Ji » vapor flux of the constituent through the soil surface,
g/cm2»s
Oei * effective diffusion coefficient, cm^/s
C2- * concentration of constituent in the air above the cap,
g/cm3 air
C • 3 concentration of the constituent in the vapor space beneath
51
the cap,
1 - cap thickness, cm.
(Because the concentration of the constituent at the surface is negligible,
C2i * 0.)
Emissions associated with diffusion alone (Eji, g/s) are obtained from
the above equation by multiplying by the landfill surface area (A) in cm2:
En - Ji x A . (6-2)
The effective diffusion coefficient of the constituent in soil, Dei,
is computed (using the expression developed by Millington and Quirk*2 an(j
applied by Farmer et al.13) from the diffusion coefficient of the constitu-
ent in air, Daj, as:
6-3
-------
where
D^ a vapor diffusion coefficient in air, cm2/s
Ql
e * soil cap air-filled porosity, cm3/cm3 (the actual
air-filled porosity of the moist soil)
CT « total porosity of the soil cap.
The concentration of the constituent of interest in the vapor space
beneath the cap is computed using'the ideal gas law as follows:
Csi * ^MV^' * P^i/RCT * 273) (6-4)
where
PI * equilibrium partial pressure of constituent, atra
MW-j * molecular weight of constituent, g/g rool
R * gas constant, 82.05
T1 * absolute temperature in the landfill, K
T * temperature in the landfill, *C.
Calculation of the equilibrium partial pressure, PJ, depends on the type of
waste liquid as follows:
a. For dilute aqueous solutions (where Henry's law applies), the equilib-
rium partial pressure of constituent within the landfill (Pi, atm) is
computed as:
Hcl'Tiquid»Xi x 1Q6 cm* (6.5)
1 ""liquid HT
where
Hci » Henry's law constant, m3«atm/mol
^Tn -A * density of waste liquid, g/cnP (1 g/cm3 is generally a
nquia good estimate for this parameter)
X. * mole fraction of constituent i in waste liquid
where
6-4
-------
g/18 +
where
C^ - weight fraction of constituent i in the
original waste liquid
CM n s weight fraction of water in the original
M2° waste liquid
^liquid * avera9e nwlecular weight of waste liquid, 9 • .
b. For two-phase (water + organic liquid) or organic liquid waste, the
equilibrium partial pressure of the constituent of interest within the
landfill (P^, atra) is computed using Raoult's law:
P. - X.Pt (6-6)
where
X. * mole fraction of constituent in the organic liquid phase
where
X1 - (Ci^.)/[Ci/MW. * Coil/MWoi1]
where
C- - weight fraction of constituent i in the
original waste liquid
C .1 * weight fraction of oil carrier-liquid in
the original waste liquid
MW ., * molecular weight of oil carrier-liquid,
01' g/g mol
P* * pure component vapor pressure of the constituent of
1 interest, atm. ,
Emissions from barometric pumping are computed as:
E2i • Q x Csi x A (6-7)
where
£21 * emissions from barometric pumping, g/s
Q » flow rate of gas through the vent, expressed as a flux,
cm3/cm2 landfill area»s
6-5
-------
CSf • concentration of constituent In the gas within the landfill,
g/cm3 gas (see Equation (6-4))
A » surface area of the landfill, cm2.
The gas flow rate, Q, is estimated using the following procedure.
a. Compute volume of gas available for barometric pumping, assuming the
entire void-volume of the waste Is available:
(6-8)
b.
Vc .J> x A x Cfw
where
V * volume of void space, era3
0 « thickness of waste bed within landfill, cm
A » surface area of the landfill, cm2.
ef « air porosity fraction of fixed waste (dlmensionless).
Compute the total volume of gas (cm3) exiting the vent of the landfill
due to changes 1n barometric pressure and/or temperature within the
landfill:
'B
frefl
lPlJ
' Tj + 273 '
kef * 273l
-1
(6-9)
where
VB « total volume of gas exiting landfill, cm3
P - » Initial (reference) barometric pressure, mm Hg
ret
P. » final barometric pressure, mm Hg
T. - final landfill temperature, *C
T f » initial (reference) landfill temperature, *C.
ret
For cases in which PI > Pref and/or T! < Tref, the computed value of
VB may be negative (indicating a condition of gas flow into the land-
fill). Because this condition results 1n no emissions associated with
barometric pumping, VB should be set equal to zero to avoid calcula-
tlonal errors in the following steps.
6-6
-------
c. Compute the average flow rate of gas from the landfill over the time
Interval of Interest:
VB
ITT <6-10>
where
Q = average flow rate of gas from the vent due to
barometric pumping, cm^/cm^ landfill area«s
At * time interval over-which the change in pres-
sure and/or temperature occurred, s
A » landfill area, cm2.
In an average day, barometric pressure drops 4 mbar from a typical
• value of 1,013 mbar. Landfill temperature is assumed to remain
constant. Hence, under these conditions, Pref --1,013 mbar, PI = 1,009
mbar, Tref = Tj * 15 *C, and At * 8.64 x 104 s.
Having computed the instantaneous emissions associated with diffusion
through the cap and barometric pumping, the total initial emission rate at
the time of landfill closure, E* (g/s), is computed as the sum:
The total instantaneous emission rate at any time then is computed via
an exponential decay function:
(3,600 s/h}(24 h/d)(365.25 d/-yr)E?
EI (t) » - —6 - exp (-Xt)
•
- 31.56 E} exp (-Xt) (6-12)
6-7
-------
where
Ej(t) * total time-dependent emission rate, Mg/yr
E? « initial emission rate, at time of landfill
cloture, g/s
t » time since landfill closure, mo
X - "decay" constant, mo'1.
The "decay" constant, X, is computed as follows:
(3,600 s/h) x (24 h/d) x 365.25 d/yr) x E?
x m 12 mo/yr x MQi
X « 2.63 x 106 E|/Moi (6-13)
where MO, is the total mass of the constituent of interest in the landfill
(g). This parameter can be computed from the weight fraction of the
constituent in the original waste liquid (CO, the mass of original waste
liquid in a unit volume of fixed waste (W), the landfill surface area (A),
and the thickness of the waste layer within the landfill (D):
M . = C. W A D . (6-14)
01 i
The average emission rate from a closed, vented landfill over the time
since landfill closure is equal to the integral of the emission rate equa-
tion over the time period divided by the time period, which yields the
following expression:
(3,600 s/h) (24 h/d) (365.25 d/yr) E* _u
-
where
E..(t) - average emission rate over the time since landfill
Al closure, Mg/yr
t » time since landfill closure, mo.
6-8
-------
Table 6-1 summarizes the equations necessary to apply the RTI closed
landfill model.
The model is highly sensitive to the air porosity of the clay cap
(ea), which largely determines the diffusion rate through the cap. The
model is sensitive to the properties of the constituent of interest,
particularly the vapor pressure (P*), Henry's law constant (HCi), and mole
fraction in the waste liquid (Xi). Because temperature affects volatility,
the model is sensitive to temperature. Other parameters to which the model
is sensitive include the depth of-the fixed waste (0), the air porosity of
the fixed waste (cfw), the landfill surface area (A), and the barometric
pressure change (Pref - PI). This latter group of parameters is
significant in that it impacts the barometric pumping rate or the volume of
gas available for pumping. In contrast to these parameters, the model
exhibits relatively low sensitivity to the diffusivity of the constituent
in air (Oai), the cap thickness (1), and the total mass of the constituent
in the landfill (M0j).
The major assumptions associated with the RTI closed landfill model
are as follows:
• The liquid waste containing the constituent of interest is
assumed to be bound in the fixed waste within the landfill.
• The constituent of interest in the gas within the landfill
is assumed to be in equilibrium with the liquid in the
waste.
• Adsorption of the constituent of interest onto the soil of
the cap is assumed to be negligible.
• The fraction of air-filled space in the landfill cap (air
porosity) is assumed to remain relatively constant over the
long term.
• The effective diffusion coefficient of the cap is assumed
not to vary with either the concentration of the constituent
of interest or time.
• The concentration of the constituent of interest in air at
the top of the landfill cap is assumed to be effectively 0.
No biodegradation (with concurrent production of biogas) is
assumed to occur due to the suppression of biological proc-
esses by the toxic waste.
6-9
-------
TABLE 6-1. RTI CLOSED LANDFILL MODEL
E.(t) - 31.56 E* exp (-Xt)
31.56 E*
X » 2.63 x 106 Et/MQi
i » C. W A D
Ei • Eli + E2i
Ji x A
C2. - 0
Csi * Pi ^i^ (T * 273)
H o X 3
p. , ci liquid 1 x 1Q6 cm (dilute aqueous waste liquids)
1 "Hi quid «J
X. = (C,/MW.)/(CU n/18 * C,/MW.) (dilute aqueous waste liquids)
I 11 fl^U 1 I
P. » X. P* (two-phase liquid or organic liquid waste)
X, - (C./MHj/tC./MW. + C-n/MW,,.,) (two-phase liquid or
1 1 1 ' ' on on organic liquid waste)
E2i - Q Csi
Q » VB/(At A)
f P
VB'VC
VC - D A cfw
Pi J I Tref *
273
- 1
(continued)
6-10
-------
TABLE 6-1 (continued)
Variable
A
Csi
Definition
Landfill surface area, cn»2
Concentration of constituent i
Data source
Westat survey3
Calculated
Coil
0
Dai
Dei
Ef
Eli
in the gas within the landfill,
g/cm-3 gas
Concentration of constituent i
in air above the cap, g/cm3
Weight fraction of constituent i
in the original waste liquid
(dimensionless)
Weight fraction of water in the
original waste liquid (dimension-
less)
Weight fraction of oil carrier-
liquid in the original waste
liquid (dimensionless)
Thickness of waste bed within
landfill, cm
Diffusivity of constituent i in
air, cm2/s
Effective diffusion coefficient
of constituent i in clay cap,
Average emission rate of con-
stituent i over time t since
landfill closure, Mg/yr
Total instantaneous ^emission rate
of constituent i at time t since
landfill closure, Mg/yr
Total initial emission rate of
constituent i at time of landfill
closure, g/s
Initial emission rate of constit-
uent i at landfill closure due to
diffusion alone, g/s
Assumed
Definition
Definition-
Definition
Westat survey13
Literature
Calculated
Calculated
Calculated
Calculated
Calculated
^Reference 14.
^Reference 15.
(continued)
6-11
-------
TABLE 6-1 (continued)
Variable
ci
Pi
PI
Pref
Pi
Q
R
t
Definition
Initial emission rate of constit-
uent i at landfill closure due to
barometric pumping alone, g/s
Henry's law constant for constit-
uent i, iiH«atm/mol
Initial diffusion flux of con-
stituent i, g/cm2»s
Landfill cap thickness, cm
Initial mass of constituent i
in the landfill, g
Molecular weight of constituent i,
g/g mol
Average molecular weight of the
dilute aoueous waste liquid,
g/g mol (assumed to be 18 g/g mol)
Molecular weight of the oil carrier-
liquid, g/g mol
Pure component vapor pressure of
constituent i, atm
Equilibrium partial pressure of
constituent i in the vapor space,
atm
Initial (reference) barometric
pressure, mm Hg .
Final barometric pressure after
At, mm Hg
Average flow rate of gas from
landfill vent(s) due to baro-
metric pumping, cm^/cm2 landfill
area«s
Ideal gas constant, 82.05 cm3-atm/
g mol*K
Time since landfill closure, mo
Data source
Calculated
Literature
Calculated
Literature
Definition or
calculated
Literature
Estimated
Definition
or estimated
Literature
Calculated
Meteorological
information
Meteorological
information
Calculated
Literature
Definition
(continued)
6-12
-------
. TABLE 6-1 (continued)
Variable
At
T
Tref
VB
w
liquid
Definition
Time interval used to determine
average barometric pumping rate, s
Temperature within landfill, °C
Initial (reference) landfill
temperature, *C
Final landfill temperature after
At, °C
Total volume of gas exiting land-
fill in At, cm3
Total volume of void space within
waste, cm3
Mass of original waste liquid in
a unit volume of fixed waste,
Mole fraction of constituent i in
the aqueous liquid (for dilute
aqueous waste) or in the organic
phase (for two-phase or organic
liquid waste) (dimensionless)
Air porosity of the clay cap
(dimensionless)
Total porosity of the clay cap
(dimensionless)
Air porosity of the fixed waste
(dimensionless)
Data source
Definition
Estimated from
literature data
Estimated from
literature data
Estimated
Calculated
Calculated
Definition or
estimated
Definition
Estimated from
clay property
data
Estimated from
clay property
data
Estimated from
fixed waste
property data
Definition or
estimated
Density of dilute aqueous waste
liquid. g/cm3 (generally equals
1 g/cnn)
Exponential decay constant, mo'1 Calculated
6-13
-------
• The landfill is assumed to be vented to the atmosphere. The
volume of gas available for barometric pumping is assumed to
consist of the total void-volume of the waste bed.
• No transport of the constituent of interest in moving water
is assumed to occur.
6.2.2 Model Plant Parameters for Closed Landfills
The characteristics of a model closed landfill facility are discussed
here. This model facility is used as the basis for an example calculation
in Section 6.2.3.
The model facility for closed landfills has an area of 1.417 x 108 cm2
(3.5 acres); This value represents an approximately midrange value from
the Westat survey.16 A reasonable value of landfill depth, also selected
from the Westat survey,!7 is 458 cm (15 ft). The landfill cap is assumed
to be composed of compacted clay. The cap thickness value of 107 cm
(3.5 ft) represents the average of extremes in thickness of clay caps
reported in site studies (2 ft to 6 ft).18 The value used for air porosity
of the clay cap is 0.08 (8 percent), while the total porosity is 0.41
(41 percent). These values were computed based on reasonable physical
properties and level of compaction for compacted clay.19 The landfill is
assumed to be vented to the atmosphere. The temperature beneath the
landfill cap is estimated at 15 *C, which represents the temperature of
shallow ground water at a midlatitude U.S. location.20 This temperature is
assumed to remain constant. The landfill is assumed to be exposed to a
nominal barometric pressure of 1,013 mbar, which represents an estimate of
the annual average atmospheric pressure in the United States.21 Barometric
pumping Is estimated for the landfill using a dally pressure drop from the
nominal value of 4 mbar. The 4 mbar value represents an estimate of the
annual average diurnal pressure drop.22
The model closed landfill facility is assumed to contain fixed waste.
The waste liquid (before fixation) selected for the facility is assumed to
be a two-phase aqueous/organic containing 20 percent chloroform, 20 percent
low-volatility organic,* and 60 percent water (by weight). This liquid has
an average density of 1.16 g/cm3. The fixation Industry Indicates that
*For modeling purposes, this component of the waste liquid represents
the oil carrier-11quid.
6-14
-------
waste liquid, when combined with fixative, may in actuality increase in
volume by as much as 50 percent.23,24 jne volume change, which is a
function of the specific waste being fixed and the specific formulation of
the fixative, carr only be determined experimentally. In view of the
inherent variability in the fixation process and the lack of real data, for
the purpose of this calculation the assumption is made that the fixation
process does not change the waste volume. This assumption is environmen-
tally conservative and may result in an overestimation of the landfill
emissions. Actual volume changes that may take place as a result of
fixation can easily be accounted for because the change in the calculated
emissions is inversely proportional to the change in waste volume. One
industry contact indicated that, for the purposes of estimating emissions,
the assumption of no volume change during fixation was reasonable.25 Based
on the waste liquid density and the assumption of no volume increase from
fixation,, the mass of waste liquid in a unit volume of fixed waste is
1.16 g/cm3. The air porosity of the fixed waste (used to estimate the
total volume of gas available for barometric pumping) is taken to be 0.25
(25 percent). This value was inferred from measurements of total porosity
and moisture content of various fixed wastes,26 and, for the purposes of
this analysis, is assumed to pertain to waste within the landfill as
opposed to waste immediately following fixation. As discussed previously,
there is no evidence for significant biomass in any chemical waste land-
fill. Therefore, in this analysis it is assumed, as suggested in the
literature, that the toxic property of the waste will inhibit biological
processes and thus prevent biogas generation.27 Hence, the waste biomass
concentration is taken to be 0 g/cm3.
The properties of chloroform that are pertinent to this analysis
include the molecular weight (119.4 g/g mol), pure component vapor pressure
at 15 "C (0.162 atm), diffusivity in air at 15 °C (0.10 cm2/s) and density
(1.49 g/cm3). The low-volatility organic liquid present in the waste has a
molecular weight of 147 g/g mol and a density of 1.31 g/cm3.
Table 6-2 summarizes the model facility parameters for closed land-
fills used in the example calculation in Section 6.2.3. For facilities
that accept more than one type of waste, the weighted average constituent
concentrations may be used.
6-15
-------
TABLE 6-2. INPUT PARAMETERS—CLOSED LANDFILL
Parameter
Value
Area
Waste bed thickness
Cap thickness
Cap air porosity
Cap total porosity
Type landfill
Temperature beneath cap
Typical barometric pressure
Dally barometric pressure drop
1.417 x 108 cm2 (3.5 acres)
457 cm (15 ft)
107 cm (3.5 ft)
0.08 (8%)
0.41 (41%)
Vented
15 *C
1,013 mbar
4 mbar
Waste liquid (before fixation)
Liquid composition
Two-phase aqueous/organic
20% chloroform, 20% low-volatility
organic (oil), 60% water (by weight)
Liquid/fixative
Liquid 1n fixed waste
Air porosity fixed waste
Blomass concentration
Chloroform properties
Molecular weight
Vapor pressure (15 *C)
D1ffus1v1ty In air (15 »C)
Density
Low-volatility organic3 properties
Molecular weight
Density
1 unit volume liquid + dry fixative
» 1 unit volume fixed waste
1.16 g/cra3
0.25 (25%)
0 g/cm3
119.4 g/g mol
0.162 atm (123 mm Hg)
0.10 cm?/s
1.49 g/cm3
147 g/g mol
1.31 g/cm3
aAlso referred to as oil "carrier-liquid.1
6-16
-------
6.2.3 Example Calculation for Closed Landfill
This section presents a step-by-step calculation of emissions from a
closed landfill that is vented to the atmosphere. The equations discussed
in Section 6.2.1 and summarized in Table 6-1 are used with the model unit
parameters in Section 6.2.2 to estimate emissions from a fixed, two-phase
aqueous /organic waste containing chloroform:
• Waste liquid (before fixation): 20 percent chloroform, 20 per-
percent low-volatility organic
liquid, 60 percent water (by
weight)
• Liquid/fixative: 1 unit volume liquid + dry fixative = 1 unit
volume fixed waste
• Waste biomass concentration: 0 g/cm3
• Landfill area: 1.417 x 108 cm2 (3.5 acres)
• Waste bed thickness: 457 cm (15 ft)
Cap thickness: 107 cm (3.5 ft)
• Type landfill: vented
• Temperature beneath cap: 15 *C
• Time period for emission calculation: 1 yr.
a. Compute the effective diffusion coefficient, De-j (cni2/s) (Equation
(6-3)):
De1 * (0.10 cm2/s) (0.08)3'33/(0.41)2
De1 « 1.32 x 10"4 cm2/s
b. Compute the equilibrium partial pressure of chloroform in the vapor
space, Pi (atm):
The waste before fixation was a two-phase liquid. Hence, Raoult's law
applies. The mole fraction for this case 1s computed as:
X1 * (C./MW.mc./M^ + C011/MWQ11)
6-17
-------
X. = (0.20/119.4 g/g mol) * [0.20/119.4 g/g mol + 0.20/147 g/g mol]
X. « 0.55 .
From Raoult's law (Equation (6-6)):
Pi - Xi P|
Pi = (0.55)(0.162 atm)
Pi = 8.91 x ID'2 atm .
c. Compute the concentration of chloroform in the vapor space beneath the
cap, Csi (g/cm3 void space) (Equation (6-4)):
Csi - P^/Rd + 273)
m (8.91 x 10"2 atroU119.4 o/q mol)
si " (82.05 cm3-atm/g mol-K) (15 + 273)
C . * 4.50 x 10"4 g/cm3 .
d. Compute initial chloroform emission flux resulting from diffus
through the cap only, Ji (g/cn^-s) (Equation (6-1)):
J. = -(1.32 x 10'4 cm2/s)(0 g/cm3 - 4.50 x 10"4 g/cm3)/107 cm
in 9
J1 = 5.55 x 10 iu g/cm'«s .
e. Compute initial chloroform emissions resulting from diffusion through
the cap only, Eii (g/s) (Equation (6-2)):
- Ji x A
- (5.55 x 10-10 g/c«2.s) (1.417 x 108 cm2)
EX, * 7.86 x lO-2 g/s .
f. Estimate the barometric pumping-induced gas flow rate through the
vent(s):
6-18
-------
1. Compute the volume of gas available for barometric pumping,
Vc (cm3) (Equation (6-8)):
V = D x A x e,
C TW
Vc = (457 cm)(1.417 x 108 cm2)(0.25)
Vc * 1.62 x 1010 on3 .
2. Compute volume of gas exiting the vent due to barometric pressure
change, VB (cm3) (Equation (6-9)):
'B
M
lpiJ
Tj + 273 '
lTref + 273I
- 1
For this case, TI = Tref * 15 *C, and barometric pressure drops by
4 mbar from the nominal value of 1,013 mbar:
B
1.62 x 1010 cm3
[ f 1.013 mbarl f!5 *C + 273 iq , 1
L U.009 mbarJ U5 'C + 273 KJ " J
VD = 6.42 x 107 cm3
D
3. Compute the average flow rate of gas over the time interval, Q
(cm3/cm2 landfill area • s) (Equation (6-10)):
The average diurnal pressure drop of 4 mbar occurs within a 24-h
period. For convenience, the gas flow from this pressure change
is averaged over 24 h (equals 8.64 x 104 s).
Q * VB
AtA
6.42 x 107cm3
(8.64 x 104 s)(1.417 x 108 cm2)
Q = 5.25 x 10"6 cm3/cm2»s
6-19
-------
4. Compute the barometric pumping-induced emission rate, £31 (g/s)
(Equation (6-7)):
E21 - Q x CS1 x A
E21 « (5.25 x 10"6 cra3/cm2«s)(4.50 x 10"4 g/cm3)(1.417 x 108 cm2)
E21 * 0.335 g/s .
g. Compute the total Initial emission rate, E* (g/s) (Equation (6-11):
F* 3 F + F
Ei En 4i
Ef - 7.86 x 10"2 + 0.335
Ef » 0.413 g/s .
h. Compute the time-dependent instantaneous emission rate:
1. Compute total mass of constituent 1 in landfill, M01-:
First compute W, the mass of original waste liquid in a unit
volume of fixed waste. Assuming one unit volume of waste liquid
results in one unit volume of fixed waste, this parameter can be
computed using the densities of the waste liquid components and
their weight fractions as follows:
W * [(1.49 g/cm3)(0.2) + (1.31 g/cm3)(0.2) + (1 g/cm3)(0.6)]
x 1 cm liquid/cm3 fixed waste
« 1.16 g/cra3 .
MQJ 1s then computed as:
MQi » C1 W A D
M = 20 q chloroform 1.16 q liquid . 417 In8 __2
Ho1 1«6 g liquid * CJ,3 fUed ^ x 1-«7 * 10 »
x 457 cm » 1.50 x 10 g chloroform
6-20
-------
2. Compute the decay constant, X (mo'1) (Equation (6-13)):
X = 2.63 x 106 E*/MQl
X « (2.63 x 106) (0.413 g/s)/1.50 x 1010 g
X - 7.25 x 10"5 mo'1
3. Compute the instantaneous emission rate, Ei, in Mg/yr, after 1 yr
(Equation (6-12)):
E.(t) - 31.56 E* exp(-Xt)
E1 = (31. 56) (0.413 g/s) exp(-7.25 x 10"5 mo"1 x 12 mo)
E. - 13.0 Mg/yr
i. Compute the average emission rate in the first year, EAJ, in Mg/yr
(Equation (6-15)):
31.56 Et
EA. = _ (31.56)(0.413 q/s) - L f 12 mo x 7<25 x 1Q-5
Al (7.25 x ID'5 mo'1) (12 mo) l l
EA1 = 13.0 Mg/yr .
6.3 FIXATION PITS
This section is currently under review
for potential revisions.
6-21
-------
6-22 through 6-32
-------
6.4 OPEN LANDFILLS AND WASTEPILES
6.4.1 Emission Model Equations
The RTI land treatment model (also known as the Allen model,34
discussed in Section 5.2) is used to estimate the air emission rate of the
constituent of interest from open (active) landfills and wastepiles. This
6-33
-------
model is based on the theory of diffusion out of an infinite flat slab and
was intended originally for use in estimating emissions from land treatment
operations. The intent of this section is to discuss use of the model with
regard to the estimation of emissions from open landfills and wastepiles? a
detailed description of the model relevant to land treatment operations and
the theoretical basis for the model are presented in Section 5.0 of this
report and will not be repeated here.
A land-treatment-type model was selected for estimating emissions from
open landfills and wastepiles because (1) no adequate models exist for
these sources, and (2) there are a number of similarities in physical char-
acteristics of open landfills, wastepiles, and land treatment operations.
A previous EPA study35 dedicated to the evaluation of models for estimating
emissions from hazardous waste TSDF identified only one model for open
waste dumps such as landfills and wastepiles—the open dump model. A
serious limitation of the model for this application, however, is that it
does not account for depletion of the volatilizing chemical from the waste
surface. Hence, the open dump model is judged unsuitable for the estima-
tion of emissions from landfills and wastepiles over the time period of
interest (months or longer).
The similarity in physical characteristics among open landfills,
wastepiles, and land treatment operations is apparent upon close examina-
tion—in all three, the waste liquid is ultimately mixed homogeneously with
a "carrier" matrix (soil in the case of land treatment; dry fixative in the
case of active landfills; and soil, fixative, or some other solid matrix in
the case of wastepiles). In all cases, the matrix is porous and permeable,
allowing the diffusion of the constituent of interest through the matrix
and into the air. Hence, in all cases, diffusion theory can be used to
model the emission rate. The notable difference between land treatment
operations and open landfills/wastepiles is the presence of an additional
mechanism affecting emissions in the case of land treatment—biological
decay of the constituent. Because biodegradation is not thought to occur*
*There is no evidence that there is significant biomass (necessary for
biological decay) in any chemical waste landfill. It is assumed that the
toxic property of the waste will inhibit biological processes.36
6-34
-------
in open landfills/wastepiles, however, its effect is not accounted for in
the modeling of air emissions.
The RTI land treatment model, which was selected for estimating emis-
sions from open landfills and wastepiles, has the following character-
istics: a sound basis in scientific theory, limited validation against
measured emissions from land treatment operations, and reasonably available
input data.37 The model considers effects such as evaporation of the con-
stituent of interest from interstitial surfaces of the carrier matrix and
diffusion of material through air-filled pore spaces.
The equations necessary to apply the land treatment model to open
landfills and wastepiles are summarized in Table 6-6. These equations,
extracted from Chapter 5.0, can be used to estimate the fraction of the
constituent emitted (Ft) and the instantaneous emission rate (E). It
should be noted that the absence of biodegradation represents a special
case that allows some simplification of several of the Chapter 5.0 equa-
tions, e.g., Equations (5-4) and (5-5). (The absence of biomass implies
that biomass concentration equals 0. Hence, t^, the time constant for
biological decay, equals infinity. Consequently, the exponential term
e-t/tfa becomes unity.) Also, the absence of biodegradation implies that
the fraction of the constituent emitted after a long time, Fa, would equal
unity.
Because the land treatment model was derived originally for land
treatment operations, model input parameters'are not necessarily in the
most convenient units and terminology for open landfills and wastepiles.
Hence, several points should be noted:
• Fixed waste is analogous (for modeling purposes) to the
waste-laden soil in land treatment.
• M0l the area-loading of the constituent in g/cm2, is geared
toward land treatment operations. For open landfills and
wastepiles, it should be computed as indicated in Table 6-6.
• No "tilling" (as discussed in Chapter 5.0) is performed in
open landfills or wastepiles.
6-35
-------
• Waste liquid is "applied" or mixed with fixative only once.
Hence, waste "reapplication" (used in the sense discussed in
Chapter 5.0) does not occur in open landfills and waste-
piles.
• The waste bed depth in open landfills and wastepiles is
analogous to the "depth to which waste is mixed" in land
treatment, as discussed in Chapter 5.0.
The approach required to estimate emissions from open landfills or waste-
piles is as follows, based on equations in Table 6-6:
1. Compute the loading of waste liquid (L) in the fixative or
soil, using the known waste composition. (For two-phase
aqueous organics or organic liquid wastes, L should be computed
as grams organic phase per cubic centimeter solid material.
For dilute aqueous waste liquids, L equals grams aqueous liquid
per cubic centimeter solid material.)
2. Compute the effective diffusion coefficient (Oe).
3. Compute the partition coefficient (Keq).
4. Use the appropriate emission equation to compute the fraction
of constituent emitted (Ft) and/or the instantaneous emission
rate (E). For wastepile calculations, the time input to these
equations should be no greater than the life of the wastepile
(retention time).
The sensitivity of the land treatment model to some parameters differs
in its application to open landfills and wastepiles from that in land
treatment operations because of the difference (in some cases) in the
expected range of the parameters. In general, it can be stated that, for
application to open landfills and wastepiles, the model is sensitive to the
air porosity of the solid waste, the liquid loading in the solid waste, the
waste depth, the concentration of the constituent in the waste, and the
volatility of the constituent under consideration. In contrast, the model
exhibits a relatively low sensitivity to the diffusion coefficient of the
constituent in air.
The following major assumptions are associated with the RTI land
treatment model and its application to open landfills and wastepiles:
• The waste liquid is mixed uniformly with the carrier matrix.
(either fixative, soil, or some other granular solid mate-
rial) before placement in an open landfill or wastepile.
6-36
-------
TABLE 6-6. RTI LAND TREATMENT MODEL APPLIED TO OPEN
LANDFILLS AND WASTEPILES (NO BIODEGRADATION)
,1/2 ...^V
Emission fraction3
F + - 0.72 (K.t)1'* for —-*-2- < 0.25 (valid for no biodegradation
at a r only)
f . 8
1 - exp (-
Keq DAt
+ 0.1878 (for *£- I 0.25 - no biodegra-
r dation)
F = 1 (t + • - no biodegradation).
a
Keq
41
Keq
Hc (106cm3/m3)
RT
A
x —=— (for aqueous waste)
ewaste
" ™*oil ea (for two-phase aqueous/organic or organic liquid waste)
Keq - —^ r
Emission rate
Keq D
,
exp (-T) \ for Keq Det/l2 > 0.213
E = ^
172
(for all other times)
r -
DeKeq
1 L C
de - (4A/T)1/2
kg = 4.82 (10-3) uO-78 SC(,-0.67
See notes at end of table.
(continued)
6-37
-------
TABLE 6-6 (continued)
Variable Definition
C Weight fraction of constituent in the oil
(organic) phase (for two-phase or organic
liquid waste), or weight fraction of
constituent in the water (for aqueous waste)
Da Diffusion coefficient of constituent in air,
cm2/s
De Effective diffusion coefficient of constituent
in the solid waste, cm^/s
Data source
Definition
Data base
Calculated
E
Fa
Ft
He
Kd
kG
Keq
1
L
A
MO
See notes
Emission rate of constituent, g/cmZ/s
Fraction of constituent emitted to the atmos*
phere at infinity (equals unity for no biodeg-
radation)
Fraction of constituent emitted to the
atmosphere after time t
Henry's law constant for constituent,
atm»m3/g mol
Volatilization constant for constituent, s"1
Gas-phase mass transfer coefficient, m/s
Ratio of gas-phase constituent to total con-
stituent in the solid waste
Depth of waste in open landfill or wastepile,
cm
Waste loading in fixative or soil. For two-
phase aqueous/organ ics or organic liquids,
L = g organic (oil) phase/cnn solid material.
For dilute aqueous waste liquids, L = g aqueous
liquid/cnH solid material
Area of open landfill, m2
Area loading of constituent, g/cm2
at end of table.
Calculated
Definition
Calculated
Data base
Calculated
Calculated
Calculated
Literature
Definition
Definition
Calculated
(continued)
6-38
-------
TABLE 6-6" {continued)
Variable
P*
R
T
U
t
waste
Definition
Average molecular weight of the oil (less
constituent), g/g mol
Pure component vapor pressure of constituent,
atm
Ideal gas constant, 82.05 atm»cm3/g mol»K
Temperature of vapor in solid waste, K
Windspeed (m/s)
Time variable for emission calculation, s
(represents time lapse from initial waste
composition)
Dimensionless parameter used in the instan-
taneous emission rate expression
Void fraction (air porosity) of solid waste
(dimensionless)
Effective diameter of land treatment area, m
Schmidt number
Total porosity of solid waste (dimensionless)
Volume fraction of waste liquid in solid waste
(dimensionless) (can be computed as L/[density
aqueous liquid in
Data source
Estimated
Data base
Literature
Assumed
Assumed
Definition
Calculated
Estimated
from fixed
waste prop-
erty data
Calculated
Calculated
Industry
personnel
Calculated
aThe first equation presented represents a special case of Equation (5-7)
for no biodegradation.
bThis equation represents the first term of the series in Equation (5-5),
for the special case of no biodegradation. The exponential terms are
expressed, for convenience, in terms of the dimensionless parameter "T".
6-39
-------
• The liquid waste containing the constituent of interest is
assumed to be bound 1n the waste after fixation amrplace-
ment 1n the open 1andf111 or wastepi1e.
• The waste liquid does not flow within the carrier matrix.
• The adsorption Isotherm of the constituent of Interest 1s
linear within the depth of the waste and does not change
with tine.
• No bulk flow of gas is Induced within the waste matrix.
• The diffusion coefficient does not vary with either concen-
tration or time.
• The concentration of the constituent of Interest 1n the gas
phase at the surface of the open landflll/wasteplle is much
lower than the concentration of the constituent of Interest
1n the gas phase within the waste matrix.
No diffusion of the waste liquid into depths below the waste
layer 1s assumed.
• Liquid-vapor equilibrium is established at all times within
the waste matrix.
For the case of fixed waste in the landfill or wastepile,
the fixed waste.mixture behaves as a soil with regard to
diffusion of the constituent of Interest.
• No biodegradatlon of the constituent of Interest occurs in
open landfills or wastepiles.
6.4.2 Model Plant Parameters for Open Landfills and Wastepiles
The characteristics of model facilities for open landfills and waste-
piles are discussed here. The model open landfill facility is used as the
basis for an example calculation using the model.
6.4.2.1 Parameters for Open Landfills. The model facility for open
landfills has a surface area of 1.42 x lO* cm* (3.5 acres). This value
represents an approximately mldrange value from the Westat survey.38 A
reasonable value of landfill depth from the Westat survey39 was 458 cm
(15 ft). The model open landfill is assumed to be half full, and hence has
a waste depth of 229 cm (7.5 ft). The landfill is assumed to contain fixed
waste. A standard temperature of 25 »C is assumed to apply.
The waste liquid (before fixation) selected for this model facility is
assumed to be a two-phase aqueous/organic containing 20 percent chloroform,
6-40
-------
20 percent low-volatility organic, and 60 percent water (by weight). This
liquid has an average density of 1.16 g/cm3. The fixation industry indi-
cates that waste liquid, when combined with fixative, may increase in
volume by up to-50 percent,40 depending on the specific combination of
waste and fixative. In view of the inherent variability 1n the fixation
process and the lack of real data on volume changes, for purposes of this
report, the assumption is made that the waste volume does not change during
fixation. Measurements41 performed on various types of fixed waste yielded
a broad range of total porosities. Fifty percent,* as used in this study,
1s a reasonable estimate of this parameter. A 25-percent* air porosity
appears to be a reasonable value; this value was inferred from measurements
of total porosity and moisture content.42 As discussed previously, there
1s no evidence of significant biomass in any chemical waste landfill.
Therefore, in this analysis 1t 1s assumed, as suggested in the literature,
that the toxic property of the waste will inhibit biological processes and
thus prevent biogas generation.43 Hence, the waste biomass concentration
1s_ taken to be 0 g/cm3.
The properties of chloroform that are pertinent to this analysis
include the molecular weight (119.4 g/g mol), pure component vapor pressure
(208 mm Hg), and diffusivity in air (0.104 cm2/s). The low-volatility
organic liquid present in the waste has a molecular weight of 147 g/g mol.
Table 6-7 summarizes the model facility parameters for open landfills
used in the example calculation in Section 6.4.3.
6.4.2.2 Parameters for Wastepiles. A review of information in the
Westat survey44 led to the selection of an approximately midrange value for
basal area of 4.65 x 106 cm2. For modeling purposes, the pile is assumed
to be flat. A uniform height of 100 cm was inferred, using the Westat
Information and engineering judgment. All waste ultimately disposed of in
the landfill is assumed to be stored initially in the wastepile. The open
landfill described previously (Section 6.4.2.1) is assumed to be filled to
capacity in 1 yr. Based on the filled landfill volume (1.42 x 108 cm2 x
458 cm depth = 6.50 x lO™ cm3), the wastepile volume (4.65 x 10& cm2 x
*These porosity values are assumed to be representative of waste in an
open landfill, rather than waste that has recently undergone fixation.
6-41
-------
TABLE 6-7. INPUT PARAMETERS—OPEN LANDFILL
Area
Waste depth
Volume
Temperature
Waste liquid (before fixation)
Liquid composition
Liquid density (average)
Liquid/fixative
Air porosity fixed waste
Total -porosity fixed waste
B1oma£s concentration
Chloroform properties
Molecular weight
Vapor pressure
Dlffuslvlty In air
1.42 x 108 cm2 (3.5 acres)
229 c» (7.5 ft)a
3.25 x 1010 cn|3
25'C
Two-phase aqueous/organic
20X chloroform, 20X low-volatility
organic, 60X water (by weight)
1.16 g/cm3
1 unit volume liquid + dry fixative
* 1 unit volume fixed waste
0.25 (25X)
0.50 (SOX)
0 g/cm3
119.4 g/g mol
208 mm Hg
0.104 cm2/s
Low-volatility organic properties
Molecular weight 147 g/g mol
Represents half full.
6-42
-------
100 cm * 4.65 x 108 cm3), and the filling time of 1 yr, it can be concluded
that the wastepile undergoes a turnover rate of 140 turnovers/yr. Hence,
the wastepile retention time is 2.6 d/turnover. The properties of the
waste liquid and the resulting fixed waste accommodated by the model waste-
pile are identical to those for the open landfill (Section 6.4.2.1) and
will not be repeated here. Table 6-8 summarizes the model facility param-
eters used for wastepiles.
6.4.3 Example Calculation for Open Landfill
This section presents a step-by-step calculation of emissions from an
open landfill. The equations Identified in Table 6-6 are used with the
model unit parameters in Section 6.4.2 to estimate emissions from a fixed,
two-phase aqueous/organic waste containing chloroform; the same equations
would be applied to the estimation of emissions from wastepiles:
Waste liquid (before fixation): 20 percent chloroform,
20 percent low-volatility
organic liquid, 60 percent .
water
-:*• * Liquid/fixative: 1 unit volume liquid + dry fixative « 1 unit
volume fixed waste
• Waste biomass concentration: 0 g/cm3
• Landfill area: 1.42 x 108 cm2 (3.5 acres)
Landfill depth: 229 cm (7.5 ft)
• Temperature: 25 °C
• Time period for emission calculation: 3.15 x 107 s (1 yr).
a. Compute waste loading, L:
Liquid density before fixation =1.16 g/cm3
1 cm3 liquid waste + fixative = 1 cm3 fixed waste
• L = g organic phase/cm3 fixed waste
= (0.20 + 0.20) x 1.16 g/cm3 * 0.46 g/cm3 .
(Note that weight fraction of chloroform in the oil phase [C] =
0.2/(0.2 -•• 0.2) = 0.50.)
6-43
-------
TABLE 6-8. INPUT PARAMETERS—WASTEPILES
Surface area
Average height
Turnover rate
Retention time
Temperature
Wlndspeed
Waste type
Waste liquid (before fixation)
Liquid composition
Liquid density (average)
L1qu1d/f1xat1ve
Air porosity fixed waste
Total porosity fixed waste
Biomass concentration
Chloroform properties
Molecular weight
Vapor pressure (25 *C)
D1ffus1v1ty In air (25 °C)
4.65 x 106 cm2
100 cm
139/yr
2.6 d
25 'C
4.92 m/s
Fixed waste
Two-phase aqueous/organic
20% chloroform, 20% low-volatility
organic, 60% water (by weight)
1.16 g/cra3
1 unit volume liquid + dry fixative
» 1 unit volume fixed waste
0.25 (25%)
0.50 (50%)
0 g/cm3
119.4 g/g mol
208 mm Hg
0.104 cm2/s
Low-volatility organic properties
Molecular weight 147 g/g mol
6-44
-------
b. Compute effective diffusion coefficient for fixed waste:
3.33
D. • 0 ^
where
ea a air porosity fixed waste * 0.25
ej - total porosity fixed waste « 0.50.
Then
Da * diffusivity of chloroform in air - 0.104 cnj2/s
(0.104 cm2/s) <°'
(0.50)
De = 4.11 x 10-3 cm2/s . (Note: Oe/Da = 3.96 x 10'2.)
~c.~ Compute "partition" coefficient, Keq (ratio of gas-phase
chloroform to total chloroform in the waste):
For oily waste,
P*MW ., g
where
P* = pure component vapor pressure of chloroform = (208 mm Hg)/
(760 mm Hg/atm) * 0.274 atm
MW0ii = molecular weight low-volatility organic = 147 g/g mol
R = ideal gas constant * 82.05 cm3»atm/g mol«K
T * temperature within solid waste, K
T = 273 K + 25 °C = 298 K
Keq . (0.274 atm) (147 q/q mol) (0.25)
(82.05 cm3«atm/g mol«K)(298 K)(0.46 g/cm3)
Keq =8.95 x 10"4 .
6-45
-------
d. Compute fraction of total chloroform emitted, Ft, after 1 year:
Keq Det
First, determine which solution applies by computing —-^
(Table 6-6):
8.95 x IP"4 x 4.11 x IP"3 cm2/s
(229 cm)2
» 7.01 x 10"11 s'1
Therefore,
Keq. e. » 7.01 x 10"11 s'1 x 3.15 x 107 s
» 2.21 x lO3 .
K t a *** e Ij « 5.45 (10"3)
lecause Keq Det/l2 is less than 0.25, the first equation of Table 6-6
applies, and
Ft « 0.72 (Kdt)1/Z
Ft = 0.72 (5.45 x 10'3)1/2
Ft * 0.053 .
e. Compute instantaneous emission rate, E, after 1 yr:
1. Compute initial mass of chloroform in landfill:
Moi - 1 I C
where
1 - waste depth = 229 cm
L s g organic/cm3 fixed waste « 0.46 g/cm3
C • weight fraction chloroform in oil - 0.50.
6-46
-------
Then
M0 -- (229 cm) (0.46 g/cm3) (0.50)
M0 = 52.7 g/cm2
3. Compute instantaneous emission rate, E,. Because Keq Det/l2 <
0.213, use the following equation to compute the emission rate:
1
Keq
Keq D
4.82 do'3) U°'78 Scg-0'67 de'0-11
U - windspeed = 4.92 m/s
0.5
[4A1 "
—I = 134 m
viscosity of air * 1.8 (10~4) g/cm/s.
Density of air = 1.2 (10'3) g/cm^.
0.104 cm2/s.
where: /*g
pa
Oa
Sc
1.81 (10"4)
G 1.2 (10"3) (0.104)
= 1.45
= 4.82 (ID"3) (4.92)0'78 (1.45)"0'67 (134)'0-11
= 0.0076 m/s = 0.76 cm/s
E »
52.7
229
1
0.25 t
0.76 x 8.95(10"4)
3.14 x
8.95(10"4)
3.15(107)
.x4.11(10-3)
= 0.230
1
368 + 5.18(10)
E = 4.43 (lO-8) g/cm2/s.
6.5 REFERENCES
1. GCA Corporation. Air Emissions from Land Treatment-Emissions Data and
Model Review. Draft Technical Note. Prepared for U.S. Environmental
Protection Agency. Research Triangle Park, NC. August 1985. 120 pp.
6-47
-------
2. Reference 1.
3 Farmer, W. J., M. S. Yang. J. Letey, W. F. Spencer and M. H. Router.
Land Disposal of HexachlSrobenzene Wastes: Controlling Vapor Movement
in Soils. Fourth Annual Research Symposium. U.S. Environmental Pro-
jection Agency. Publication No. EPA-600/9-78-016. August 1978.
4 Millington, R. J., and J. P. Quirk. Permeability of Porous Solids.
Trans. Faraday Society. 57:1200-1207. 1961.
5. U.S. Environmental Protection Agency. Evaluation and Selection of
Models for Estimating Emissions from Hazardous Waste Treatment, Stor-
age, and Disposal Facilities. Office of Air Quality Panning and
Standards, Research Triangle Park, NC. Publication No. EPA-450/3-84-
020. December 1984.
6. Reference 3.
7 Thibodeaux, L. J. Estimating the Air Emissions of Chemicals from
Hazardous Waste Landfills. Journal of Hazardous Materials.
4:235-244. 1981.
8. Reference 3.
9. Reference 3.
10 ShenT T. T. Estimating Hazardous Air Emissions from Disposal Sites.
Pollution Engineering. 31-34. August 1981.
11. Reference 3.
12. Reference 4.
13. Reference 3.
14. Westat, Inc. National Survey of Hazardous Waste Generators and TSD
Facilities Regulated Under RCRA in 1981. Prepared for U.S. Environ-
mental Protection Agency. Contract No. 68-61-6861. 1981.
15. Reference 14.
16. Reference 14.
17. Reference 14.
18. Ely, R. L., G. L. Kingsbury, M. R. Branscome, L. J. Goldman C. M.
Northeim, J. H. Turner, and F. 0. Mixon, Jr. Performance of Clay Caps
and Liners for Disposal Facilities. Research Triangle Institute,
Research Triangle Park, NC. Prepared for U.S. Environmental Protec-
tion Agency; Cincinnati, OH. EPA Contract No. 68-03-3149. March
1983.
6-48
-------
19. Telecon. Goldman, Leonard, Research Triangle Institute, with Borden,
Roy, Department of Civil Engineering, North Carolina State University,
Raleigh, NC. August 13, 1986.
20. Gerachty, J. J., 0. W. Miller, F. Vander Leeden, and F. L. Troise.
Water Atlas of the United States. Water Information Center, Inc.,
Port Washington, NY. 1973. Plate 30.
21. Telecon. Goldman, Leonard, Research Triangle Institute, with Hughes,
John, National Climatic Center, Asheville, NC. May 15, 1986.
22. Reference 21.
23. Telecon. Goldman, Leonard, Research Triangle Institute, with Boyenga,
Dave, MBI Corporation, Dayton, OH, November 20, 1985.
24. Telecon. Goldman, Leonard, Research Triangle Institute, with Webber,
Emlyn, VFL Technology Corporation, Malvern, PA. November 12, 1985.
25. Telecon. Massoglia, Martin, Research Triangle Institute, with Webber,
Emlyn, VFL Technology Corporation, Malvern, PA. January 13, 1987.
26. Telecon. Goldman, Leonard, Research Triangle Institute, with Hannak,
J»eter, Alberta Environmental Center. April 4, 1986.
27^W*eterence 10.
28. Reference 5.
29. Arnold, J. H. Studies in Diffusion: III. Unsteady-State Vaporiza-
tion and Absorption. Transactions of the American Institute of Chemi-
cal Engineers. 40:361-379. 1944.
30. Reference 29.
31. Reference 29.
32. Letter and attachment from Shen, T., New York State Department of
Environmental Conservation, to McDonald, R., EPA/OAQPS. Modification
of Arnold's equation. February 16, 1986. (See also Reference 10.)
33. Trip Report. Goldman, Leonard, Research Triangle Institute, with
Chemical Waste Management, Sulphur, Louisiana. February 25, 1986.
34. Reference 1.
35. Reference 5.
36. Reference 10.
6-49
-------
37. Memorandum and attachment from M. Wright, Research Triangle Institute,
to S. Thorneloe, EPA/OAQPS. May 30, 1986. Selection of an emissions
model for land treatment.
38. Reference 14.
39. Reference 14.
40. References 23, 24, and 25.
41. Reference 26.
42. Reference 26.
43. Reference 10.
44. Reference 14.
6-50
-------
7.0 TRANSFER, STORAGE, AND HANDLING OPERATIONS
7.1 NARRATIVE DESCRIPTION OF MODEL PLANTS AND EMISSIONS
This chapter presents models for estimating VO emissions of hazardous
wastes from container loading, storage, and cleaning; stationary tank load-
ing and storage; spills; fugitive sources; and vacuum truck loading.
7.2 CONTAINER LOADING
This section addresses emission-estimating practices for hazardous
waste loading into tank trucks, railroad tank cars, marine vessels, and
55-gal (0.208-m3) drums.
7.2.1 Emission Model for Container Loading
*The%rocess of loading containers with volatile hazardous wastes
generates emissions. If containers were assumed to be clean when loaded,
only those vapors generated by the loaded waste could be emitted. If
containers hold residues of previous volatile wastes, additional emissions
from the residues will also be generated.
To calculate container loading losses, the AP-421 equation for loading
petroleum liquids is applied. This equation was derived for tanks, cars,
and marine vessels. It is also applied to 55-gal drums in this case
because the loading principles are similar and because no equation has been
developed exclusively for small containers such as drums. The loading
equation is as follows:
k . 12'46T SMP* (7-1)
where
L « loading loss, lb/103 gal of liquid loaded
M = molecular weight of vapors, Ib/lb mol
7-1
-------
P* * true vapor pressure of liquid loaded, psia
T * bulk temperature of liquid loaded, °R (*F + 460).
S 3 saturation factor, dimensionless (see Table 7-1).
Equation (7-1) for estimating emissions from containers is not applic-
able to open dumpsters because they are designed with no tops, unlike drums
that have limited venting through bungs.
7.2.2 Model Parameters
Containers are considered to be splash-loaded (as opposed to
submerged-loaded) for emission-estimating purposes. This loading method
creates larger quantities of VO vapors and increases the saturation factor,
S, of each volatile compound within the container. A saturation factor is
a dimensionless quantity that "represents the expelled vapors' fractional
approach to saturation and accounts for the variations observed in emission
rates from the different unloading and loading methods."2 A saturation
factor of 1.45 was selected for these emission estimates, based on previous
documentation of splash-loading petroleum liquids.3,4
Typical capacities for containers are assumed to be as follows:
• Drums: 55 gal (0.208 m3)
Tank trucks: 7,000 gal (26.5 m3)
• Railroad tank cars: 30,000 gal (114 ra3)
• Marine vessels: 20,000 tons.
It is assumed that 55-gal drums and tank trucks are the predominant
containers used in the waste management industry. Bulk liquid hazardous
waste is shipped predominantly by highway; therefore, hazardous waste tank
truck models are used for estimating emissions.
Molecular weight and vapor pressure are functions of the waste loaded,
and 25 °C is considered an annual average ambient operating temperature.
7.2.3 Sample Calculation for Tank Loading
The following sample calculation may be used to estimate VO emissions
from tank truck loading of an organic liquid. Waste stream compounds and
properties for the sample calculation are as follows. The same waste
stream is employed in each sample calculation in this section; only the
type of emission source is varied.
7-2
-------
TABLE 7-1. S FACTORS FOR CALCULATING PETROLEUM
LOADING LOSSES
Cargo carrier
Mode of operation
S factor
Tank trucks and tank cars
Submerged loading of a clean 0.50
cargo tank
Splash loading of a clean 1.45
cargo tank
Submerged loading: normal 0.60
dedicated service
Splash loading: normal, dedl- 1.45
cated service
Submerged loading: dedicated 1.00
vapor balance service
Splash loading: dedicated 1.00
vapor balance service
Marine vessels3
Submerged loading: ships
Submerged loading: barges
aTo be used for products other than gasoline.
0.2
0.5
7-3
-------
Molecular Vapor
Weight weight pressure Mole
Constituent fraction (Ib/mol) (psia) fraction
Benzene 0.3 78 1.84 0.368
Naphthalene 0.3 128 0.0044 0.224
Phenol 0.4 94 0.0066 0.408
The input parameters for the truck loading model are as follows:
• True vapor pressure of loading liquid, psia: 0.68 (calcu-
lated in a., below)
• Molecular weight of vapor, Ib/mol: 78.23 (calculated in b.,
below)
• Saturation factor, dimensionless: 1.45 (splash loading)
• Bulk temperature of liquid loaded, °R: 537
•- -Annual throughput, gal/yr: 28,000
a. §alc«late P*. true vapor pressure of liquid, by Raoult's Law:
P* = (Pj - Xj) + (P2 • X2) + (P3 • X3)
where
P* s true vapor pressure, psia
PI, ?2, and ?3 « vapor pressures of pure components
Xj, X2, and X3 - mole fractions of VO components in liquid
P* - (1.84 psia x 0.368) + (0.0044 psia x 0.224) + (0.0066
psia x 0.408)
• 0.68 (psia).
b. Calculate M, molecular weight of vapors:
(Pi-Xi) (P2»X,) (P,.X)
M - 1 l • M.+ i "
p* 1 p-
7-4
-------
where
M = molecular weight of vapor
Mj, M2, and M3 = molecular weight of each component
M fl.84 x 0.3681 ¥ ,« . f0.0044 x 0.2241
M ' I 5768 J x 78 + I OS J
x 128 + p.0066^0.40B) x 94
= 78.23 (Ib/mol).
c. Calculate emissions from truck loading:
I s 12.46 SMP*
12.46 x 1.45 x 78.23 x 0.68
537 °R
- 1.79 lb/1,000 gal
. . .ce,rtnc , 1.79 x IP"3 Ib/oal x 28.000 qal/yr
Annual emissions, LL = 2,205 1b/Mg
= 0.023 Mg/yr
7.3 CONTAINER STORAGE
This section addresses storage emissions from tank trucks, railroad
tank cars, 55-gal drums, marine vessels, and open dumpsters.
7.3.1 Emission Model for 55-Gal Drums. Tank Trucks, and Railroad Tank Cars
With regard to 55-gal drums, container storage is considered a loca-
tion where multiple drums are most likely to accumulate and be stored for
more than 90 days. Because drums are designed to be stored with a sealed
lid and bung, the potential for breathing losses is minimal. Therefore,
breathing loss is assumed to be negligible. However, the potential does
exist for a drum to rupture or become damaged and leak during storage.
Thus, the emissions from drum storage may be estimated using the same spill
fraction used for drum handling—10'4 (to be discussed in more detail in
Section 7.7, Spills). The following equation is used to estimate emissions
from drum storage:5
7-5
-------
E = 10~4 x I x Wi x V. (7-2)
where
E * emission from drum storage, -Mg/yr
I * throughput, Mg/yr
Wi = VO weight fraction
Vj = volatilization fraction.
Emission-estimating methodologies have not been developed for storage
in tank trucks and railroad tank cars. Only loading information was avail-
able in the literature for these containers. The assumed same emission
estimates principle for drum storage is applied with an emission factor of
ID"5 (to be discussed in more detail in Section 7.7, Spills).6
7.3.2 Model Parameters for Drum Storage
It i.s assumed that 50 percent of the VO storage loss from drum spUl
or rupture will be volatilized. The remaining volatiles will be captured
with RCRA spill response measures taken at the facility.
7.3.3 Sample Calculations for Drum Storage
Input parameters:
Waste stream: organic liquid
(See Section 7.2.3 for constituents.)
Waste density: 1.04 Mg/m3
Drum storage capacity: 182 drums (0.208 m3/drum)
Turnovers per year: 12
Spill fraction: 10'4
Weight fraction: 1
Volatilization fraction: 0.5.
a. Calculate annual throughput, I, Mg/yr:
I » 182 x 0.208 m3 x 12 x 1.04 Mg/m3
* 472 Mg/yr
7-6
-------
b. Calculate air emissions:
E * 10'4 x I x Wi x Vi
= 10"* x 472 Mg/yr x 1 x 0.5
= 0.024 Mg/yr .
7.3.4 Emission Model for Open Dumpsters
No information was found in readily available literature to estimate
VO emissions from the storage of hazardous wastes in open-top dumpsters.
The wastes held in dumpsters may range from sludges to contaminated
filters. An emission algorithm?,8 was developed for losses from an open
dump. Because an open dumpster is similar to an open dump, this algorithm
was used to calculate storage emissions. (See Section 6.3 for a detailed
derivation of the open dump model.)
2 "o 1*1 V • |°i '" (7-3)
"^^«J^—^^^^^^^^ ^ —
Ei RT ' ~F7
v
The open dump model is valid for short-term emission estimates only.
For long-term emission estimates, to avoid overestimates, it is important
to subtract the emissions from the waste's VO content on a daily, weekly,
or monthly basis depending on the waste volatility; this is done to iden-
tify the point in time when no VO remain in the waste.
7.3.5 Model Parameters for Open Dumpster Storage
This section is currently under review for potential revisions.
7
v
-------
itael
rH
7-8
-------
7.4 CONTAINER CLEANING
7.4.1 Emission Model for Container Cleaning
An AP-42 document? on tank truck cleaning is used as the primary
source for container-cleaning.emission estimates. AP-42 states that tank
truck cleaning typically involves washing the truck interior with agents
such as water, steam, detergents, or other chemicals. The document also
7-9
-------
provides emission factors that are a function of vapor pressure and viscos-
ity. These factors have been applied to emission estimates for cleaning
all types of containers, as follows:
Tank truck Emission factor
residue to be removed g/truck (Ib/truck)
High vapor pressure, low viscosity 215 (0.474)
Medium vapor pressure, medium viscosity 32.4 (0.071)
Low vapor pressure, low viscosity 5.5 (0.012)
The following equation is used to estimate emissions for container
cleaning:
E » Fc x N x Wj x 10'6 (7-3)
where
E, = cleaning loss, Mg/yr
FG»« ^mission factor for cleaning, g/container
N * number of cleanings per year
W^ = VO weight fraction.
7.4.2 Model Parameters
In all containers, the primary input parameter for estimating cleaning
emissions is the cleaning emission factor, which is determined from
(1) residue vapor pressure and viscosity (functions of waste handled), and
(2) container volume.
Based on AP-42,*0 a typical tank truck volume of 26.5 m3 (7,000 gal)
is assumed.
Because no data are currently available for drum cleaning, the emis-
sion factors for tank trucks were used to calculate cleaning emissions from
drums by comparing the proportion of drum volume (55 gal) to that of the
tank truck (7,000 gal).
7-10
-------
55-Gal drum
residue to be removed
High vapor pressure, low viscosity
Medium vapor pressure, medium viscosity
Low vapor pressure, low viscosity
Emission factor
g/drum
1.69
0.25
0.04
(Ib/drum)
(0.0037)
(0.0006)
(0.00009)
Emissions from marine vessels have not been addressed because of the
low usage of such vessels in the waste management industry.
7.4.3 Sample Calculation for Tank Truck Cleaning
The general assumptions for truck cleaning are as follows:
• Residue: pure organic liquid (benzene).
• Number of truck cleanings per year: 4
• Truck capacity: typical truck
• Weight fraction: 1.
a. Determine the cleaning emission factor, Fc:
(215 g/truck was used because of high vapor pressure and low
viscosity of pure benzene residue).
b. CalcuVate cleaning emissions:
E » Fc x N x Wi x 10-6
* 215 g x 4 x 1 x 10-6 Mg/g
« 8.6 x 10-4 Mg/yr .
7.5 STATIONARY TANK LOADING
7.5.1 Emission Model for Stationary Tank Model
Stationary tank working losses are those emissions from waste loading
and unloading operations. AP-42's "Storage of Organic Liquids"11 provides
an equation to estimate loading and unloading emissions from storage tanks.
The equation was developed for handling VO liquid in the following
industries:
7-11
-------
• Petroleum producing/refining
• Petrochemical and chemical manufacturing
• Bulk storage and transfer operations
• Other industries consuming or producing organic liquids.
Because hazardous wastes have the potential to contain VO compounds, as do
organic liquids, and because they are most commonly stored in the same
fashion as these liquid products, the following equation was selected from
AP-42:
L* « 2.40 x 10-5 My • P* • V • N • Kn • KC (7-4)
where
LH » working losses, Ib/yr
My » molecular weight of vapor in storage tank, Ib/lb mol
p* *.true vapor pressure at bulk liquid conditions, psia
N **number of turnovers per year (dimensionless)
M total throughput per year (gal)
n * tank capacity, V (gal)
V * tank capacity, gal
!(„ - turnover factor, dimensionless (for turnovers $36, Kn * 1;
for turnovers > 36, K,, » 18{?N* N)
*
KC « product factor, dimensionless (for crude oil, KC • 0.84; for
all other organic liquids, KC » 1).
7.5.2 Model Parameters
It is assumed that all stationary tanks are fixed-roof. According to
responses to the 1982 Westat Mail Survey,12 which were examined by the 6CA
Corporation,13 there are four sizes of tanks that best represent the waste
management industry:
7-12
-------
5.3 m3 (1,500 gal)
30.3 m3 (8,000 gal)
75.0 n»3 (20,000 gal)
795 m3 (210,000 gal).
Table 7-2 lists typical input parameters for these model tanks. Turnovers
per year were selected based on volume of waste processed in waste manage-
ment scenarios recorded in various documents. Molecular weight and vapor
pressure are a function of the waste loaded.
7.5.3 Sample Calculation for Tank Loading Emission Model
Input parameters:
Waste stream: organic liquid (see Section 7.2.3 for constituents)
Mv, molecular weight of vapor, Ib/lb mol: 78.23
P*, true vapor pressure of loading liquid, psia: 0.68
Kc, product factor for working loss: 1
V, "fixed-roof tank capacity, gal: 20,000
^N, turnovers per year: 44
Kn, turnover factor, dimensionless: 0.848.
a. Calculate Mv, molecular weight of vapor:
(see Section 7.2.3 for calculation).
b. Calculate P*, true vapor pressure of loading liquid:
(see Section 7.2.3 for calculation).
c. Calculate Kn, turnover factor: because N • 44, Kn = 18° ^ = 0.848
d. Calculate air emissions:
Lw = 2.40 x ID'5 x Mv • P* • V • N • Kn • Kc
= 2.40 x ID'5 x 78.23 x 0.68 x 20,000 x 44 x 0.848 x 1
"- 953 Ib/yr
=0.43 Mg/yr .
7-13
-------
TABLF 7-2. PERTINENT FIXED-ROOF TANK
Specifications
Capacity, m^
(gai)
Tank height, m
Tank diameter, m
Average vapor space
height, m
Adjustment for small
diameter .
(dlmenslonless)
Average diurnal temp.
change, *C (*F)
Paint factor
(dime/is ton less)
Relation of tank to
ground
Product factor
Model
A
5.3
(1,500)
2.4
1.7
1.2
0.26
11
(20)
1
Above
1
Model
B
30.3
(8,000)
2.4
4
1.2
0.65
11
(20)
1
Above
1
Model
C
75.7
(20,000)
2.7
5.8
1.4
0.86
11
(20)
1
Above
1
Model
D
795
(210,000)
12.2
9.1
6.1
1
11
(20)
1
Above
1
7-14
-------
7.6 STATIONARY TANK STORAGE
7.6.1 Model Description
Fixed-roof tank storage of hazardous wastes results in VO "breathing"
emissions through vents as ambient temperature and barometric pressure
fluctuate. Emissions occur in the absence of any liquid level change in
the tank. An existing AP-42*7 equation was used to estimate VO breathing
losses from hazardous waste storage tanks as follows:
0 CD
L = 2.26 x Hf2 My f-Ep.1 ' • D1'73 • H°'51 • AT0'5 • F • C • KC (7-5)
" "
where
l_b * fixed-roof breathing loss, Ib/yr
Mv * molecular weight of vapor in tank, Ib/lb mol
P* * true vapor pressure at bulk liquid conditions, psia
PA* average atmospheric pressure at tank location, psia
D* tank diameter, ft
H = average vapor space height, ft (assumed to be one-half of
tank height)
AT * average ambient diurnal temperature change, °F (20 °F
assumed as a typical value)
Fp = paint factor, dimensionless (see Table 7.3)
C - adjustment factor for small diameter tanks, dimensionless
(for diameter i 30 ft, c - 1; for diameter < 30 ft,
c * 0.0771 D - 0.0013 D2 - 0.1334)
Kc » product factor, dimensionless (for crude oil, 1^=0.65, for
all other organic liquids, Kc - 1.0).
The above equation requires an estimation of the true vapor pressure
using the liquid concentration. For very volatile constituents, the liquid
concentration depends on the amount lost as air emissions. To correct for
the loss to the air in estimating the liquid concentration, the following
equation may be used:
fraction lost to air * ^ + Lt
where Lt is the tank input of the volatile constituent in pounds per year.
7-15
-------
TABLE 7-3. PAINT FACTORS FOR FIXED-ROOF TANKS*8
Paint factors (Fp)
Tank color Paint condition
Roof Shell Good Poor
White
Aluminum (specular)
White
Aluminum (specular)
White
Aluminum (diffuse)
White
Light gray
Medium gray
White
White
Aluminum (specular)
Aluminum (specular)
Aluminum (diffuse)
Aluminum (diffuse)
Gray
Light gray
Medium gray
1.00
1.04
1.16
1.20
1.30
1.39
1.30
1.33
1.40
1.15
1.18
1.24
1.29
1.38
1.46
1.38
1.44a
1.58a
^Estimated from the ratios of the seven preceding paint factors.
7-16
-------
7.6.2 Model Parameters
Table 7-3 identifies the model parameters for estimating tank breath-
ing losses. Molecular weight and vapor pressure are functions of the waste
stored.
7.6.3 Sample Calculation for Tank Storage Emission Model
Input parameters:
Waste stream, organic liquid (see Section 7.2.3 for constituents)
Mv, molecular weight of vapor, Ib/lb mol: 78.23
P*, true vapor pressure of loading liquid, psia: 0.68
KC( product factor for breathing loss: 1
v, fixed-roof tank capacity, gal: 20,000
D, tank diameter, ft: 19
H, average vapor space height, ft: 4.5
AT, diurnal temperature change, °F: 20
Fp, paint factor, dimensionless: 1
.C, .adjustment factor for small tanks: 0.86 (calculate in c., below)
a. aCakulate molecular weight of vapor:
(see Section 7.2.3 for calculation).
b. Calculate true vapor pressure of loading liquid:
(see Section 7.2.3 for calculation).
c. Calculate adjustment factor for small tanks:
C = 0.0771 x 19 - 0.0013(19)2 . Q.1334
= 0.86 .
d. Calculate air emissions:
L = 2.26 x 10'2 Mtf f-^-1 ' • D1'73 • H°'51 • AT0'5 • F • C • K
D v p_p*l K
- 2.26 x 10-2 x 78.23 x [vS^'6* « («)»•" x (4.5)°-51
x (20)0'5 x Ix 0.86 xl
- 300 Ib/yr
* 0.14 Mg/yr
7-17
-------
7.7 SPILLS
7.7.1 Model Description
An ICF study19 of truck transport to and from TSDF and truck emissions
at TSDF terminals provided the background information necessary to estimate
spillage losses during TSDF and trucking operations. As a result of this
study, spill fractions of 10'4 and 10'5 were assumed for drum movement of
wastes and all other remaining waste movement, respectively. Thus, for
every 10,000 Mg of drummed hazardous waste moved, 1 Mg is assumed to be
spilled. The following equation is used to estimate spill emissions:
E - Fs x I x U{ x V, (7-6).
where
E » spill emissions, Mg/yr
Fs « emission fraction, 10'4 or 10'5
I - annual throughput, Mg/yr
Wj » VO weight fraction
VT » fraction for volatilization.
7.7.2 Model Parameters
In both cases of spills, it is assumed that 50 percent of the vola-
tiles in the waste are lost. The remaining 50 percent are recovered by
RCRA spill plan response. Therefore, most spills would be mitigated before
100 percent of VO is lost to the atmosphere.
It is assumed that spills do not occur during the transfer of waste
into a stationary tank if loading is automated through fixed piping.
7.7.3 Sample Calculation for Drum Storage Model
Input parameters:
Waste stream: organic liquid (see Section 7.2.3 for constituents)
Waste density: 1.04 Mg/m3
Emission fraction: 10'4
Weight fraction: 1
Volatilization fraction: 0.5
Number of drums handled: 2,184 (0.208 m3/drum).
7-18
-------
a. Calculate annual throughput, I, Mg/yr:
I = 2,184 x 0.208 m3 x 1.04 Mg/m3
= 472 Mg/yr .
b. Calculate air emissions:
E = 10-4 x 472 Mg/yr x 1 x 0.5
= 0.024 Mg/yr .
7.8 FUGITIVE EMISSIONS
7.8.1 Emission Model for Fugitives
Waste transfer operations often involve pumping wastes through pipe-
lines into a variety of containment units. Such pumping creates the poten-
tial for fugitive emission losses from pumps, valves, and flanges. Table
7-4 presents the Synthetic Organic Chemical Manufacturing Industries
(SOCMI) emission factors20 that had been developed to estimate VO that leak
from,pump seals, valves, and flanges. These factors are independent of the
throughput, type, or size of the process unit.
-~^» »
TABLE 7-4. SOCMI EMISSION FACTORS FOR FUGITIVE LOSSES
Equipment
Pump seals
Valves
Flanges
Type of
service
Light liquid
Light liquid
Emission factor
(kg/h-source)
4.94 E-2
7.10 E-3
8.30 E-4
The following equation is used to estimate fugitive emissions:
E = Z (Ff x Ni) x h x 10-3 (7-7)
where
E - fugitive emissions, Mg/yr
Ff = emission factor per source-type, kg/h-source (see Table 7-4)
NT = number of sources per source-type
h = residence time in the equipment (assume = 8,760 h/yr).
7-19
-------
7.8.2 Model Parameters
The major input parameters required for the emission model are emis-
sion factor, number of sources, and residence time. It is assumed that
waste remains in the transfer equipment 24 h/d, 365 d/yr; therefore, VO are
continuously being leaked to the atmosphere.
Minimal information has been compiled on typical quantities of pumps,
valves, and flanges at waste management facilities. Therefore, previous
contractors have turned to data collected from the petroleum refining
industry and SOCMI. 6CA recommended that "for any hazardous waste filling
operation, transfer operation, or handling operation involving pumps, the
estimate of two pumps, 35 valves, and 80 flanges be used. This includes
tank filling, tank truck or car filling, and drum filling."21 Because the
relationship 2:35:80 appears to be too high for pumping waste into a single
drum, one pump, three valves, and eight flanges are used for estimating
emissions. GCA recommended that smaller quantities of pumps, valves, and
flanges identified by SOCMI be applied for transfer operations to injection
wells* ana* incinerators, i.e., 1 pump, 18 valves, and 40 flanges.22
7.8.3 Sample Calculation for Fugitive Emission Model
Estimate the annual fugitive emissions from a set of piping lines that
connect to a storage tank, given the following information.
Input parameters:
Assume 2 pumps, 35 valves, and 80 flanges associated with the piping
equipment.
Assume the waste stream is organic liquid.
Assume waste remains in piping line 24 h/d, 365 d/yr.
a. The emission factor for light liquids was used because of the high VO
content.
b. Calculate fugitive emissions:
E = (0.0494 kg/h x 2 + 0.0071 kg/h x 35 + 8.3 x 10'4 kg/h x 8)
x 8,760 h/yr x 10'3 Mg/kg = 3.62 Mg/yr
7.9 VACUUM TRUCK LOADING
7.9.1 Emission Model for Vacuum Truck Loading
Emissions from vacuum truck loading are estimated by calculating an
equilibrium concentration of organic vapors in the vacuum truck at its
7-20
-------
operating conditions and assuming that a total volume of gas equal to the
vacuum truck volume is emitted to the atmosphere for each loading episode.
Equations for making the calculations are presented as follows:
Kt
Nv,
Nv x YJ x MWi
X. P*
(for oily waste)
[P0 VG (T/273)J/Pt
where
E-j = air emissions of compound i, g/truckload
Nv * total moles of vapor discharged, g mol
Y-J = mole fraction of compound i in vapor phase
X^ » mole fraction of compound i in liquid phase
MWi * molecular weight of compound i, g/g mol
P* = vapor pressure of compound i, mm Hg
-a «r
Pt = total system operating pressure, mm Hg
P0 = atmospheric pressure, mm Hg
V = vacuum truck volume, m3
VG = volume of 1 g mol of gas at STP, 0.0224 m3/g mol
T * operating temperature, K.
7.9.2 Model Parameters
Based on information obtained during site visits to refineries using
land treatment, vacuum trucks have a capacity of about 21 m3 (5,500 gal)
and operate at a pressure of approximately 303 mm Hg. These values are
used in estimating vacuum truck emissions.
Molecular weight and vapor pressure are functions of waste loaded, and
25 °C is considered a standard operating temperature.
7.9.3 Sample Calculation
The following is a sample calculation of benzene emissions during
loading of a vacuum truck with organic liquid.
7-21
-------
Input parameters:
Waste stream: organic liquid (see Section 7.2.3 for constituents)
VO constituent: benzene
MHi, molecular weight, g/g mol: 78
P*, pure compound vapor pressure: 95.2
Pt, system operating pressure, mm Hg: 303
P0, atmospheric pressure, mm Hg: 760
XT, mole fraction in liquid: 0.368
V, vacuum truck volume, o£: 21
V6, volume of 1 g mol of gas at STP, m3/g mol: 0.0224
T, operating temperature, K: 298
N, turnovers per year, truckload/yr: 10.
Calculate total moles of vapor discharged, g mol:
a.
V
[Pa VG
21
3 V(760 mm Hg x 0.0224 mj/g mol x 298 K/273 K)/303 m Hg
» 342.41 g mol/truckload .
b. Calculate mole fraction of benzene in vapor phase, Y,:
Y,
0.11* .
c. Calculate air emissions per truckload, g/truckload:
EI - Hv x Yi x MM,
• (342.41 g mol/truckload) (0.1156) (78 g/g mol)
» 3,087 g/truckload .
d. Calculate annual emissions for benzene, Mg/yr:
Annual emission « Ei x M
* 3,087 g/truckload x 10 truckload/yr
- 30,870 g/yr
= 0.031 Mg/yr .
7-22
-------
e. Repeat the above procedures, a through d., to compute emissions for
each constituent as follows:
Constituent Ei, g/truckload Annual emissions, Mg/yr
Benzene
Naphthalene
Phenol
3,087
7
14
0.031
0.00007
0.00014
Total emissions 3,108 0.0312
7.10 REFERENCES
1. U.S. Environmental Protection Agency. Transportation and Marketing of
Petroleum Liquids. In: AP-42. Compilation of Air Pollutant Emission
Factors. Third Edition, Supplement 12, Section 4.4. Research
Triangle Park, NC. Office of Air Quality Planning and Standards.
July 1979. 13 pp.
2. GCA Corporation. Air Emission Estimation Methods for Transfer, Stor-
age, and Handling Operations. Draft Technical Note. Prepared for
U.S. Environmental Protection Agency, Office of Air Quality Planning
•and- Standards. Research Triangle Park, NC. Contract No. 68-01-6371.
August 1985.
3. Reference 1.
4. Reference 2.
5. U.S. Environmental Protection Agency. Assessing the Release and Costs
Associated with Truck Transport of Hazardous Wastes. PB 84-224-468
(Contract No. 68-01-0021). Washington, DC. January 1984. 151 p.
6. Reference 5.
7. Shen, T. T. Estimating Hazardous Air Emissions from Disposal Sites.
Pollution Engineering. 31-34. August 1981.
8. GCA Corporation. Air Emissions of VOC from Waste Piles at Hazardous
Waste Treatment, Storage, and Disposal Facilities. Prepared for U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards. Research Triangle Park, NC. Contract No. 68-01-6871.
August 1985.
9. U.S. Environmental Protection Agency. Tank and Drum Cleaning. In:
AP-42. Compilation of Air Pollutant Emission Factors. Fourth
Edition, Section 4.8. Research Triangle Park, NC. Office of Air
Quality Planning and Standards. September 1985. 4 pp.
10. Reference 9.
7-23
-------
11. U.S. Environmental Protection Agency. Storage of Organic Liquids.
In: AP-42. Compilation of Air Pollutant Emission Factors. Fourth
Edition, Section 4.3. Research Triangle Park, NC. Office of Air
Quality Planning and Standards. September 1985. 25 pp.
12. Westat, Inc. National Survey of Hazardous Waste Generators and Treat-
ment, Storage, and Disposal Facilities Regulated Under RCRA in 1981.
Prepared for U.S. Environmental Protection Agency. Office of Solid
Waste, Washington, DC. April 1984.
13. Addendum to Memorandum dated September 6 1985 from Eichinger
Jeanne, GCA Corporation, to Hustvedt, K. C., EPA/OAQPS. September 12,
1985. TSDF model source parameters and operating practices data base.
14. Reference 11.
15. Reference 13.
16. Graver Standard Cone-Roof, Flat-Bottom Tanks. Sizes and Capacities.
17. Reference 11.
18. TRW-Environmental, Inc. Background Documentation for Storage of
Orqanic Liquids. Prepared for U.S. Environmental Protection Agency.
Research Triangle Park, NC. Contract No. 68-02-3174. May 1981.
19. Reference 5.
20. U.S. Environmental Protection Agency. Control of Volatile Organic
Compound Leaks from Synthetic Organic Chemical and Polymer Manufactur-
ing Equipment. Research Triangle Park, NC. Publication No. EPA-450/
3-83-006. March 1984.
21. Reference 2.
22. Reference 2.
7-24
-------
8.0 COMPARISON OF MODEL RESULTS WITH FIELD TEST DATA
8.1 INTRODUCTION
Predictions from TSDF emission models are compared with field test
data in this chapter. In general, considering the uncertainty of field
emission measurements, agreement between measured and predicted values is
considered reasonable. Measured and predicted emissions generally agree
within an order of magnitude.
The following caveats must be considered in any evaluation of the
comparison results presented in the following sections:
The field test data did not always include all of the input
parameters required to use the emission models. In such
cases, parameter values representative of field operations
were used as defaults.
2. The emission models use average influent and effluent con-
centrations to estimate annual emissions. Variations in
concentrations and constituents are not reflected.
i Field test data provide information on a limited number of
hazardous constituents. Extrapolation of comparisons on
limited constituents to all constituents of interest may
not always be possible.
4 The method of measuring emissions (e.g., flux chambers and
othefenclosure systems) could alter the real-world system
being tested and affect the representativeness of the
measured emissions.
8.2 SURFACE IMPOUNDMENTS AND OPEN TANKS
8.2.1 Summary
Emission test data were available from tests of five quiescent surface
impoundments. The overall mass transfer coefficients determined in these
tests agreed generally within an order of magnitude with the overall
coefficient predicted by the mass transfer correlations described in
8-1
-------
Chapter 4.0. Predicted values were higher than measured values in some
cases and lower in others.
The emission models used for impoundments also were applied to open
tanks. The comparison of measured and predicted values for the overall
mass transfer coefficient for open wastewater treatment tanks yielded mixed
results. For tanks with quiescent surfaces (e.g., clarifiers and equaliza-
tion basins), the model predictions were generally lower than measured
values but agreed within an order of magnitude. For the aerated systems,
the model predictions agreed well with material balance and ambient air
measurements for an open aerated system.
8.2.2 Details of Comparisons
The approach to the comparison of predicted and measured values is to
estimate the overall mass transfer coefficient from the correlations given
in Chapter 4.0 and to compare this value to the overall mass transfer
coefficient from the test data. The overall mass transfer coefficient from
the test-data is calculated from a measured emission rate and a measured or
estimated^ bulk concentration in the liquid phase. Note that errors in
either the measured emission rate or liquid-phase concentration have a
direct effect on the errors in the calculated mass transfer coefficient.
Most of the measured emission data were obtained by flux chamber
measurements. At a few sources, ambient air monitoring and material
balances were used to determine the emission rate for calculation of the
overall mass transfer coefficient.
6CA Corporation performed an analysis of data from impoundments. The
results are given in Tables 8-1 through 8-4 for four ponds at two different
sites. Site 5 is a commercial hazardous waste facility with a wastewater
treatment system onsite. The reducing lagoon receives wastes classified as
reducing agents from tank trucks. The lagoon is operated on a batch basis
and was observed to contain a zone of solids and a surface with a floating
oil film- The holding pond receives aqueous wastes from the water treat-
ment system and is filled (and discharged) on a monthly basis. The oxidiz-
ing lagoon receives oxidizing agents including halogens and other organic
compounds. The accumulation of solids and oil film also was observed on
this lagoon. Site 4 also is a commercial hazardous waste facility, and its
8-2
-------
TABLE 8-1. COMPARISON OF RESULTS FOR REDUCING LAGOON 1
AT SITE §1.2
Mass transfer coefficient (x 106 m/s)
Model predictions
Average flux (for 5 to 10 m/s
Constituent
Benzene
Toluene
Ethyl benzene
Naphthalene
Methylene chloride
Chloroform
1,1, 1-Trichloroethane
Carbon tetrachloride
p-01chlorobenzene
Styrene
chamber measurement3
4.9
5.0
5.5
2.6
12
5.7
7.6
11
2.6
5.7
windspeed)"
4.2-17
3.9-15
3.6-14
3.5-14
4.7-19
4.3-17
3.9-15
3.9-16
3.6-14
3.7-15
Calculated from reported emission rate and corresponding liquid-phase con-
. cerftration.
bW1ndspeed during the test ranged from 5 to 10 m/s.
8-3
-------
TABLE 8-2. COMPARISON OF RESULTS FOR HOLDING POND 6
AT SITE 53-4
Mass transfer coefficient fx 106 m/s)
Model predictions
Average flux (for 5 to 10 ra/s
Constituent chamber measurement0 winaspeea;-
Benzene
Toluene
Ethyl benzene
Naphthalene
Methyl ene chloride
Chloroform
1 , 1 , 1 -Trl ch 1 oroethane
Chlorobenzene
p-DI chl orobenzene
Ac et aldehyde
2.7
2.3
2.6
1.6
3.1
2.2
3.9
<0.039
4.3
3.4
5.3-21
4.9-19
4.6-18
4.4-18
6.0-24
5.4-21
4.9-19
4.9-20
4.6-18
5.7-19
•Caltulfted from reported emission rate and corresponding liquid-phase con-
centration.
&W1ndspeed during the test ranged from 5 to 10 ra/s.
8-4
-------
TABLE 8-3. COMPARISON OF RESULTS FOR OXIDIZING LAGOON 2
AT SITE 55«6
Mass transfer coefficient (x 106 m/s)
Average flux
Model predictions
(for 5 to 10 m/s
Constituent
Tol uene
Ethyl benzene
1,1,1-Trichloroethane
chamber measurement3
0.38
0.037
35
windspeed)b
3.8-15
3.6-14
3.9-15
Calculated from reported emission rate and corresponding liquid-phase con-
centration.
"Windspeed during the test ranged from 5 to 10 m/s.
.TABLE'8-4. COMPARISON OF RESULTS FOR SURFACE IMPOUNDMENT AT SITE 47,8
Mass transfer coefficient (x 1(?6 m/s)
Flux
chamber measurement3
Model predictions
Constituent
Toluene
Ethyl benzene
Methyl ene chloride
1,1, 1-Tri chloroethane
Chloroform
p-Dichlorobenzene
Average
2.4
1.0
8.4
2.6
12.0
0.44
Range
1.9-2.7
0.46-1.4
5.6-10.0
1.1-3.6
5.4-15.0
0.079-0.75
Viur 3 to iu m/s
windspeed)*3
6.3-25.1
5.9-23.5
7.7-30.5
6.3-24.7
7.0-27.6
5.9-23.1
Results for June 22, 1984.
bWindspeed during the test ranged from 5 to 10 m/s.
8-5
-------
impoundment is used to contain aqueous wastes. Table 8-5 presents a
comparison of results for Site 3, which is a chemical manufacturing plant
that produces primarily nitrated aromatics and aromatic amines. This
impoundment is a wastewater holding pond for the wastewater treatment
system at the plant. Two wastewater streams that enter the treatment
system are distillation bottoms from aniline production (K083) and the
nitrobenzene production wastewater (K104).
The results in Tables 8-1 through 8-5 show a reasonable agreement
between measured and predicted values of the overall mass transfer coeffi-
cient. The measured results for the impoundment in Table 8-3 may have been
affected by an oil film observed on the surface or from poor mixing in the
impoundment, which can complicate representative sampling of the liquid-
phase concentration. Table 8-5 shows good agreement of results for toluene
and benzene, which were present in the liquid phase at 2.6 and 17 mg/L,
respectively. The liquid-phase concentrations of the other four compounds
in Table 8-5 ranged from 0.029 to 0.15 mg/L. The differences in measured
and predicted values for these four compounds may have been affected by the
accuracy*of sampling and analysis of the liquid. The compounds listed in
Tables 8-1 through 8-5 are controlled by the liquid-phase mass transfer.
Consequently, the results are most dependent on Springer's correlation for
!CL (the liquid-phase mass transfer coefficient) and suggest that Springer's
model is probably accurate within an order of magnitude.
GCA, in a separate document, examined measured and predicted mass
transfer coefficients for open tanks that are part of wastewater treatment
systems.W The results for Site 8, which is an industrial wastewater
treatment operation, included a primary clarifier, an equalization basin,
and aerated stabilization basins. The various influent and effluent liquid
streams were analyzed, and air emissions around the unit were monitored.
Overall mass transfer coefficients were calculated from material balance
data and from ambient air monitoring. These values are listed in Tables
8-6 through 8-8 along with the predicted values from the mass transfer
correlations given in Chapter 4.0. The primary clarifier, equalization
basin, and the quiescent portion of the stabilization basin were modeled as
quiescent surfaces, and the correlations of Springer and MacKay/Matasugu
8-6
-------
TABLE 8-5. COMPARISON OF RESULTS FOR WASTEWATER HOLDING LAGOON
AT SITE 3»
Mass transfer coefficient (x 106 ra/s)
Constl tuent
Cyclohexane
Tetrachl oroethy 1 ene
Toluene
Benzene
n-Undecane
Methyl chloride
Flux chamber
measurement
0.39
0.10
9.0
3.7
0.21
35.0
Predicted3
3.8
3.7
3.8
4.1
2.8
3.1
aBased on an average measured wlndspeed of 3.7 m/s and an average
temperature of 22 •C.
TABLE 8-6. COMPARISON OF RESULTS FOR PRIMARY CLARIFIERS
AT SITE
Constituent
Tetralln
2-Ethyl hexanol
2-Ethyl hexyl aery late
Naphthalene
1,2-Olchloroethane
Benzene
Tol uene
Ethyl benzene
Mass transfer
Material
balance
—
96.0
•ntt
179.0
58.0
5.4
35.0
156.0
coefficient
Ambient
monitors
227.0
42.0
123.0
92.0
2.9
18.0
50.0
39.0
(x 106 m/s)
Model
prediction
2.0
2.7
3.4
4.0
4.1
3.8
3.5
8-7
-------
TABLE 8-7. COMPARISON OF RESULTS FOR EQUALIZATION BASIN
AT SITE 812
_
Constituent
1,2-Dichloroethane
Benzene
Toluene
Ethyl benzene
Mass transfer coefficient
Material Ambient
balance monitors
20 19.0
20 8.9
25 42.0
25 5.4
(x 106 m/s)
Model
prediction
5.0
5.1
4.7
4.4
TABLE 8-8. COMPARISON OF RESULTS FOR AERATED STABILIZATION BASINS
AT SITE 813
Constituent
2-Ethyl hexanol
2-Ethyl hexyl acrylate
1.2-Oichloroethane
Benzene
Toluene
Ethyl benzene
Mass transfer
Material
balance
0.05
4.8
2.0
12.4
5.0
2.9
coefficient
Ambient
monitors
0.01
8.3
0.52
1.1
5.8
0.55
(x 104 m/s)
Model
prediction
0.17
2.9
5.7
10.6
10.1
9.9
3-3
-------
were used. The turbulent portion of the stabilization basins was modeled
using the correlations of Thibodeaux and Reinhardt.
Useful conclusions from the comparison of measured and predicted
values are difficult because of the lack of consistent results from air
monitoring, probably due to very short sampling periods, changes in the
windspeed and direction, and the contribution to emissions from sources
near the mentioned source. In addition, material balance calculations are
subject to error from changes in influent/effluent flow rates and concen-
trations of specific compounds. In general, the model predictions for the
primary clarifier and equalization basin are lower than the measured
values. For the aerated stabilization basin, most of the predicted mass
transfer coefficients are higher than the measured values; however, the
agreement is within an order of magnitude. The measured values for the
primary clarifier may have been higher than the predicted values because of
the contribution from nearby sources to measured air concentrations or
because of the contribution from the falling film at the clarifier.
GCA also performed an analysis on an aerated lagoon at Site 7.*4 This
lagoon was covered and was purged with air at a rate of 1.4 m^/s (3,000
ft3/min). Predicted and calculated values for the mass transfer
coefficients are given in Table 8-9 and show that predicted values are 1 to
2 orders of magnitude higher. The basis of the predicted' values includes
Thibodeaux and Reinhardt's correlations for aerated systems. No strong
conclusions on the model's validity can be drawn from these data because
the dome enclosure over the system and its evacuation rate probably have a
direct effect on emissions. In addition, difficulties with air measure-
ments and determination of the appropriate driving-force concentration
(assumed to be the bulk liquid concentration) can lead to errors in the
calculated values of the overall mass transfer coefficient.
The results of the biodegradation model were also compared to avail-
able data from biodegradation measurements. The most desirable comparison
would be for a system in which the air emission rate and biodegradation
rate were measured directly. However, the extent of biodegradation from
studies of real systems has usually been determined by difference from a
material balance (what enters the system minus what leaves with the ef-
fluent and with air emissions).
8-9
-------
TABLE 8-9. COMPARISON OF RESULTS FOR COVERED AERATED LAGOON
AT SITE 715,16
Constituent
1,2-Olchloroethane
Benzene
Toluene
Mass transfer
coefficient (x 104 n/s)
Vent rate
•easurenent Predicted3
0.05
0.30
0.95
7.2
8.9
8.8
*Based on an estimated wlndspeed (not aeasured) of 5 «/s17 and an
estimated turbulent area of about 50 percent.19
8-10
-------
Petrasek et al. performed such a study on a large pilot-scale acti-
vated sludge system with diffused a-ir aeration. 19 The activated sludge
unit was enclosed, and the diffused air that was removed was sampled (for
flow rate and concentration) to determine air emissions. This system was
designed for a flow rate of 2.2 L/s (35 gal/min) with an air purge rate of
57 L/s. A summary of the operating parameters is given in Table 8-10. The
study used a synthetic wastewater that contained individual volatile com-
pounds at levels of 32 to 300 ppb. The biomass concentration was 2 g/L,
and the resultant food-to-microorganisra (F/M) ratio of 0.5 is well within
the recommended design range of 0.2 to 0.6.
Petrasek reported the percent of each compound entering the activated
sludge unit that was emitted with the diffused air; the fraction.bio-
degraded could be determined by difference-assuming all unrecovered
material was biodegraded. The results are summarized in Table 8-11 and
show a range of measured values from 5 percent for chlorobenzene to 62
percent for 1,1,1-trichloroethane. The predictions of the biodegradation
model discussed in Chapter 4.0 are also presented in Table 8-11 for compar-
isoa. Jhe comparison shows that the model predictions agree well with the
Petrasek measurements for nearly every compound.
Another type of comparison between measurements and predictions
involves effluent concentrations for well-defined wastewater treatment
systems. Namkung and Rittman22 reported influent and effluent concen-
trations of volatile organics for two Chicago wastewater treatment plants
that receive large shares of industrial discharges. The measurements were
made for two large activated sludge units aerated by diffused air. In
addition, the system's operational parameters were defined (Table 8-12) and
provided the necessary inputs for the mathematical model that includes air
emissions (diffused air system) and biodegradation.
The results of measured and predicted effluent concentrations are
summarized in Table 8-13. The most convincing comparison is the close
match for both plants for tetrachloroethylene, which the authors stated was
not biodegradable in these systems. Therefore, a biorate equal to zero was
used in the model for this compound. The close agreement between measured
and predicted effluent concentrations suggests that this compound is almost
8-11
-------
TABLE 8-10. DESCRIPTION OF PEJRASEK'S ACTIVATED
SLUDGE SYSTEM20
Parameter
Flow rate (L/s) 2.2
Volume (m3) 59.8
Residence time (h) 7.5
Air rate (L/s) 57
Fraction of surface agitated 0
Biomass concentration (g/L) 2.0
Concentration range for organics (ppm) 0.032 - 0.30
F/Ma . 0.5
Food to microorganism ratio (Ib/lb biomass • day)
based on chemical oxygen demand.
8-12
-------
TABLE 8-11. COMPARISON OF PETRASEK'S MEASUREMENTS AND MODEL PREDICTIONS
Compound
Benzene
Carbon tetrathloride
Chlorobenzene
Chloroform
Dichloropropane (1,2)
Ethyl benzene
Tertachl oroethane
and -ethene
Toluene
Trichloroethane (1,1,1)
Trichforoethane (1,1,2)
Trichloroethene
Reported
fraction
emitted3
0.15
0.59
0.05
0.34
0.32
0.21
0.27
0.20
0.62
0.25
0.41
Predicted
fraction
emittedb
0.19
0.54
0.02
0.20
0.09
0.15
0.37C
0.15
0.57
0.06d
0.37
Fraction Predicted
assumed fraction
biodegradeda biodegradedb
0.80
0.41
0.95
0.66
0.68
0.79
0.73
0.80.
0.38
0.75
0.59
0.78
0.44
0.97
0.75
0.88
0.82
0.58C
0.84
0.40
0.87d
0.59
aOata from Petrasek et al.;21 the fraction biodegraded is assumed to be the
fraction unaccounted for based on the analyses of the sludge, the air, and
the effluent.
bMode1 predictions based on the equations presented in Chapter 4.0 assuming
influent VO concentrations of 0.10 mg/L and operating parameters as provided
in Table 8-10.
cArithmetic average for the removal fractions calculated for 1,1,2,2-tetra-
chloroethane and tetrachloroethene.
dEmployed 1,1,1-trichloroethane's biodegradation rate constants.
8-13
-------
TABLE 8-12. DESCRIPTION OF TWO CHICAGO ACTIVATED SLUDGE UNITS23
— .--
Operating parameters
Volume (m3)
Depth (m)
Wastewater flow (mVs)
Air rate («3/s)
Residence tine (h)
Total organics (mg/L)
Biomass (g/L)
Fraction surface agitated
Concentrations (ppb)
Chloroform, in
out
Ethyl benzene, in
out
Methylene chloride, in
out
Tetrachloroethylene, in
out
Toluene, in
out
1,1,1-Trichloroethane, in
out
Trichloroethylene, in
out
Calumet
184.500
1.8
10.0
55
5.1
115
2.2
0
4.0
7.1
18
0.5
9.8
11
16
2.1
85
6.2
13
2.9
9.7
0.5
West-southwest
802,300
1.8
36.6
193
6.1
180
2.0
0
4.4
2.4
10
BDL
.48
11
12
1.6
22
BDL
15
2.2
22
2.1
BDL * Below detection limit.
8-14
-------
TABLE 8-13. COMPARISON OF MEASURED AND PREDICTED EFFLUENT
CONCENTRATIONS FOR CHICAGO WASTEWATER TREATMENT PLANTS24
Compound
Chloroform
Ethyl benzene
Methyl ene chloride
Tetrachloroethylene
Toluene
1,1, 1-Trichloroethane
Trichloroethylene
Calumet effluent
concentrations, ppb
Measured Predicted*
b b
0.5 0.68
b b
2.1 1.0
6.2 2.9
2.9 1.0
0.5 0.75
West-southwest effluent
concentrations, oob
Measured
2.4
c
11
1.6
c
2.2
2.1
Predicted9
2.3
0.16
7.1
0.77
0.69
1.1
1.6
aBased on the equations presented in Chapter 4.0.
bNo comparison possible because measured concentration in effluent was
greater than measured concentration in influent.
cMeasured effluent concentration was below detection limit.
8-15
-------
entirely removed by air stripping, and the quantity predicted to be air
stripped by the model is reasonably accurate.
The results in Table 8-13 also indicate that 1,1,1-trichloroethane and
tricfiToroethylene are biodegraded. The model predictions used a biorate
for these two compounds that was derived from Petrasek's data in Table
8-11. Both Petrasek's data and the comparison in Table 8-13 indicate that
these compounds are biodegraded to some extent; otherwise, the measured
effluent concentrations in Table 8-13 would have been higher than those
predicted by the model with biodegradation included.
Tabak et al.25 conducted an extensive study of the biodegradability of
numerous toxic compounds. They found that, when the microbial culture is
properly acclimated, almost all nonpesticide compounds could be, at least
partially, biodegraded. Although biodegradation rate constants were not
determined, the percent of compound biodegraded was shown to be dependent
on the acclimation of the culture, and (although to a lesser extent)
dependent on the concentration of the compound used. For every compound
tested, the percent biodegraded by the third subculture (presumably the
most*acclimated) always decreased when the concentration was doubled
(unless both cultures were either 100 percent or 0 percent biodegraded),
and this decrease was rarely a decrease of a factor of two or more.26 if
biodegradation were strictly a first-order process, the percent biodegraded
would be independent of the concentration. If biodegradation were strictly
a zero-order process, the percent biodegraded would decrease by a factor of
two (for those compounds not biodegraded, 100 percent) when the concen-
tration was doubled. Because an intermediate effect was generally
realized, Tabak's results suggest Monod-type biodegradation rate kinetics
are appropriate.
Another comparison that can be made is based on a series of field
studies, as reported by Coburn et al.,27 in which batch, biodegradation
rate studies were performed while controlling air emissions. The experi-
mental first-order biodegradation rate constant and the predicted, apparent
first order rate constant based on the Monod model can be compared in the
last two columns of Table 8-14. Note for compounds whose log mean
concentrations are near or are greater than the appropriate half-saturation
8-16
-------
constant (e.g., formaldehyde or methanol), the predicted first-order rate
constant according to the Monod model provides a better estimate of the
observed biodegradation rate than would be provided assuming simple first-
order kinetics (i.e., using KI straight from the data base as provided, for
comparison, in Table 8-14). Additionally, using the recommended biodegra-
dation rate constants and modeling approach, the predicted biodegradation
rates presented in Table 8-14 agree well with the reported biodegradation
rates for nearly every experimental run.
A separate study was conducted for EPA to evaluate measured and
predicted emissions for aerated waste treatment systems.29 The correla-
tions of Thibodeaux and Reinhardt were used (as recommended in Chapter 4.0)
to estimate the mass transfer coefficients of the turbulent zone. The
results showed an agreement between measured and predicted values that were
within an order of magnitude. The report concluded that, when adequate
descriptions of plant operating parameters are available, reliable emission
estimates can be obtained from the models (within the accuracy that results
from variations in sampling and chemical analysis). When plant operating
parameters are known, the major limitations in the models result from a
lack or accurate biooxidation rates and vapor/liquid equilibrium data for
specific compounds.30
8.2.3 Recommendations for Additional Data
The estimate of emissions from open liquid surfaces is provided by the
product of the mass transfer coefficient, driving-force concentration, and
surface area. Surface area can be determined with reasonable accuracy.
The previous comparison of mass transfer coefficients indicated that they
can be estimated within an order of magnitude. Probably the greatest
source of uncertainty is in the estimate of the appropriate driving force
for mass transfer. The concentration is likely to vary with time and loca-
tion in the impoundment. The type of flow system and extent of mixing in
the liquid also will affect this concentration.
For the less volatile compounds that may be controlled by gas-phase
mass transfer, the collection of equilibrium data may be useful to compare
with the estimated values used in the models. The comparisons presented in
this section primarily address compounds with high volatility in water
8-17
-------
TABLE 8-14. COMPARISON OF MEASURED AND PREDICTED BIODEGRADATION RATES
••^•••^•^—^•••••MMMMi^^MMM^^M
Compound
Acetone
Benzene
Chloroform
Dimethylphthalate
Ethanol
Ethyl benzene
Ethyl ene oride
(oxlrane)
Formaldehyde
Methanol
Methyl ethyl ketone
Methyl ene chloride
Log mean
cone.,**
«g/L
1.35
2.56
0.005
0.10
0.008
0.002
4.2
4.9
0.005
1.7
3.2
3.9
8.0
62
250
480
490
495
0.10
0.27
0.37
0.80
0.028
0.031
0.053
0.15
0.23
Rate
w
mg/g/h
1.3
19
2.94
2.2
8.8
6.8
4.2
5.0
18
2.0
22
constants3
Ke K.
Si 1.
mg/L L/g/h
1.1 1.15
13.6 1.4
3.7 0.79
0.71 3.1
9.8 0.90
3.2 2.1
4.6 0.91
20 0.25
90 0.20
10 0.20
55 0.40
First-order rate
constant, L/g/h
Experiment^
1.15
0.34
0.36
2.1
0.36
0.29
0.36
0.70
0.36
0.81
0.75
0.37
0.13
0.077
0.067
0.018
0.040
0.023
0.24
0.18
0.19
0.16
0.11
0.11
0.36
0.20
0.57
Predictedc
0.53
0.36
1.4
1.4
0.79
0.79
0.45
0.60
2.1
0.67
0.54
0.49
0.17
0.057
0.053
0.032
0.031
0.031
0.20
0»19
0.19
0.18
0.40
0.40
0.40
0.40
0.40
(continued)
8-18
-------
TABLE 8-14 (continued)
Log mea
cone . . b
Compound
2-Propanol
Thiobismethane
Toluene
1,1, 1-Trichloroethane
Trichloroethene
Total xylenes
mg/L
2.9
6.2
1.07
0.014
0.016
0.081
0.14
0.040
0.16
0.004
0.097
Rate
max,
mg/g/h
15
0.16
73.5
3.5
3.9
40.8.
constants3
"s,
mg/L
200
0.17
30.6
4.73
4.43
22.7
Kl
L/g/h
0.75
0.93
2.4
0.74
0.88
1.8
••^•••••^•^•MH
First-order rate
constant, L/g/h
Experiment1*
0.069
0.085
0.13
0.28
0.34
0.63
1.9
0.38
0.88
0.41
>2.2d
Predicted0
0.074
0.073
0.11
2.4
2.4
2.4
2.4
0.73
0.72
0.88
1.8
aRecommended rate constants from Appendix 0, fable D-l.
bFrom data reported by Coburn et al.;28 the log mean concentration was
calculated as follows:
CLM » (Ci - cf)/in(Ci/Cf)
where
Cf
initial concentration, mg/L
final concentration, mg/U.
cCalculated as the apparent first-order rate constant using the Monod model
(Equation 4-13) based upon the log mean concentration as follows:
Kl ' "max/Us + CLM) .
dFinal concentration was below detection limit. The final concentration was
assumed to be at the detection limit to calculate the first-order rate cons-
tant. The actual rate constant should be greater than the reported value.
8-19
-------
(high Henry's law constant). Because semi volatile compounds also can be
emitted to a significant extent, air emission measurements for these less
volatile compounds would be useful for comparison with model predictions.
8.3 LAND TREATMENT
Field data from four test sites and one laboratory simulation were
used as a basis for comparing measured emissions with estimated emissions
using the RTI land treatment model. Two other laboratory simulations of
land treatment are under way or near completion, but the documentation on
those tests is not yet sufficiently complete so that comparisons of meas-
ured and estimated emissions can be made. These include an EPA-sponsored
study being conducted by Radian Corporation and a simulation study by
Chevron Research Company in Richmond, California. Table 8-15 summarizes
the tests evaluated. Generally, estimated emissions are within an order of
magnitude of measured values. Values of estimated emissions varied both
above and below measured values.
.Comparisons of estimated and measured emission flgx rates are pre-
sented graphically in this section. Comparisons of estimated and measured
emissions by weight percent of applied material are presented in the next
section.
Considering the potential for error in measuring or estimating values
for parameters that are input to the model and the potential for error in
measuring emissions, differences in the range of an order of magnitude are
not unexpected. In making the comparisons, values for all model inputs
sometimes were not available in the emission test reports. In these cases,
values were estimated using averages of field data or values identified
previously as typical or representative of actual land treatment practices.
In the 1985 test at a Midwest petroleum refinery (Case I),31 emission
measurements were made at sample locations in six test plots. For each
plot, emission measurements were made after waste application but before
the plot was tilled, again after the waste was tilled, and for another
period after a second tilling. AH measurements were made using a flux
chamber and tenax traps. Emission rates were measured for six specific
organic constituents: benzene, toluene, p-xylene, o-xylene, m-xylene, and
naphthalene. Benzene and toluene were selected as a basis for comparing
8-20
-------
TABLE 8-16. SUMMARY OF LAND TREATMENT TESTING AND TEST RESULTS
ro
Sit* Test
No. T*at sit* location description
12 West Coast corporate Laboratory
research facility simulation
13 Southwest research Laboratory
fseillty slmulstlon
14 Midwestern r*fin*ry Flux chamber
Tost Tost Tost
yosr sponsor procedures
1980 - Private u Run 1
1987 corporation (raw wast*)
* Run 2
(raw wast*)
Run 2
(tr*at*d
wast*)
1980 EPA Run 1*
(API separ-
ator sludge)
Box fl
Box |2
Box |3
Box |4
Run 2*
(IAF flost)
Box |1
Box |2
Box |3
Box |4
1986 ORO Plot A
Plot B
Test
durstton
2.6
months
22 days
22 days
31 days
31 days
'
6 dsys
8 days
Tost result*
West* constituent
on
Oil
Oil
Oil
Oil
Oil
Oil
OH
Oil
on
Oil
Benzene
Toluene
Ethyl b*nz*ne
p-Xyl*n*
m-Xy 1 one
o-Xylen*
Naphtha l*n*
Benzene
To 1 uene
Ethyl benzene
p-Xylene
m-Xyl*ne
o-Xyl*n*
Naphthalene
Emissions,
wt. 91
36
11
1
6.2
NA
a. 6
0.7
16
NA
18
19
81
41
196
16
39
28
1
lie
06
402
21
83
38
2
See notes at end of table.
(continued)
-------
TABLE 6-IS (continued)
00
k
Site Teat Teat Teat Teat Teat
No. Teat site location description year sponsor procedures duration
14 (con.)
Plot C 8 daya
Plot 0 8 daya
Plot E 8 daya
Plot F 6 daya
16 West Coaat refinery Flux chamber 1984 ORO Surface 6 weeks
appl 1 cation
i _^____ _
Tf|t results
Waste constituent
Benzene
Toluene
Ethyl benzene
p-Xylene
•-Xylene
o-Xy 1 ene
Naphthalene
Benzene
Toluene
Ethyl benzene
p-Xylene
•-Xylene
o-Xy 1 ene
Naphthalene
Benzene
Toluene
Ethyl benzene
p-Xylene
•-Xylene
o-Xy lone
Naphtha 1 ene
Benzene
Toluene
Ethy (benzene
p-Xylene
•-Xylene
o-Xy! ene
Naphthalene
n-Heptane
Uethy 1 eye 1 ohexane
3-Me thy 1 -heptane
n-Nonane
1-Methylcyelohexene
1-Oetene
A-Plnene
Llmonene
Emissions,
wt. X
39
17
140
26
26
17
142
86
363
66
79
62
2
107
63
346
43
62
39
1
84
47
208
13
28
24
1
60
61
62
66
49
60
17
22
See notes at end of table.
(continued)
-------
TABLE 8-16 (continued)
CD
ro
Site Teat Test Test Test
No. Test sit* location description year sponsor^ procedures
16 (con.)
Subsurface
16 Southwest research Laboratory 19BS API/EPA Run no.
facility simulation Run no.
Run no.
Run no.
Run no.
Run no.
Run no.
Run no.
Run no.
Run no.
Run no.
Run no.
Run no.
Run no.
Run no..
Run no.
Run no.
Run no.
Run no.
Run no.
Run no.
IS
21
24
27
28
32
as
S4
as
aa
a?
40
41
44
46
40
47
48
49
60
61
Test reaulta
Test Eailaalena,
duration Waste constituent wt. K
Toluene
p-,av-Xylene
1 , 1 , 6-Tr 1 nethy 1 ben lene
o-Ethyl- toluene
Total VO
Total oil
6 weeks n-Heptsne
Methy Icyc lohexane
a-Methyl -heptane
n-Nonane
1 -Uethy 1 cy c 1 ohexane
1-Octene
A-P Inane
Llawnene
Toluene
p-,M-Xylene
1 ,i, 6-Tr I me thy Ibeniene
o-Ethyl -toluene
Total VO
Total oil
• hours" 01 1
Oil
on
Oil
Oil
Oil
Oil
Oil
Oil
Oil
Oil
on
Oil
Oil
Oil
on
on
on
on
on
on
a7
16
21
82
a0
1.2
94
88
77
80
78
74
21
28
68
48
27
42
88
1.4
91
4.
0.0
0.
f.
a.
2.
0.0
0.
70.
9.
0.
2.
4.
49.
7:
0.
6.
9.7
1.1
0.47
See notes at end of table.
(continued)
-------
TABLE 8-16 (continued)
Site
No.
Test site locstion
Test
description
Test
year
10
Oulf Coaat e
TSDF
clal Flu* chaster
IMS
Tost
sponsor
OHO
Test results
Test
procedures
Test
duration
Waste constituent
Emissions,
wt. *
Single test« 09 hours Total VO
' ' ' 60 hours Boniene
0.77
1.91
17 Midwestern refinery Flux chamber 1979 API Centrifuge
sludge
Test no. 6
Test no. 6
API separa-
tor sludge"
Test no. 7
Test no. 9
Test no. 9
19.9 hours
•07 hours
619 hours
122 hours
620 hours
Oil
Oil
Oil
on
on
0.1
a. 6
13.6
1.1
13.6
00
ro
API « Anorlcsn Potrolou* Institute.
IAF * Induced air Ilotstion.
ORO • Office of Research and Development.
•Sludge applied to BOH |1 and Bon |9 as duplicate tests; sludge treated with Mercuric chloride to eliminate (or reduce)
bloactivlty applied to Box |4 and no aludge epplied to Bon |2, which served aa a control.
*Each run for which results sre reported waa 9 houra.
cTeat was conducted using aged wastes.
^Allowed to westher for 14 days in open 6-gal buckets in an outdoor open shelter prior to application.
-------
measured and estimated emissions in this test. The comparison was made for
test plot A after the waste was tilled for the first time. Estimated emis-
sions for each compound are higher than the measured values but generally
are within a factor of 10. Estimated and measured values are shown
graphically for benzene and toluene in Figures 8-1 and 8-2, respectively.
At the West Coast refinery (Case 2),32 emission tests were made using
three adjacent plots marked off in the land treatment site. The center
plot was used as a control and had no waste applied while waste was applied
to the other two plots. One plot had waste applied to the soil surface and
the other had waste applied by subsurface injection. Flux chambers were
situated on each test plot and emission measurements were made during three
different test periods each lasting 4 days. Canister air samples, sludge
samples, and liquid samples were analyzed by gas chromatography (GC).
Emissions of both total VO and selected specific constituents were measured
during the test. For comparing measured and estimated emissions, total VO
and toluene emissions from the surface application plot were used. Esti-
mated emission rates for both toluene and total VO agree reasonably well
with measured rates but range from higher to lower than measured rates at
different times. Estimated cumulative emissions over the entire test
period agree reasonably well with the measured values. For both toluene
and total VO, estimates covered a 4-day period with a till occurring after
2 days. Estimated and measured values over the 4-day period for which the
comparison is made are generally within an order of magnitude, as can be
seen in Figures 8-3 and 8-4. Measured values were reported as half-day
average emission rates.
For the test at the commercial hazardous waste site in 1983,
(Case 3),33 waste was applied to a single test plot and tilled into the
soil. Air emission measurements were made over a 3-day period using a flux
chamber and gas canisters. Sampling locations were selected randomly, with
a control point used to provide a common sampling position each day.
Sample analyses were made by GC. Emission comparisons of measured and
estimated emissions were made for total nonmethane hydrocarbon (NMHC) emis-
sions using data generated by GCA in a separate study of the data from this
test.34 As with previous tests, estimated emission flux rates were greater
8-25
-------
240
IN)
01
M
£
O»
3
X
3
0
*M
M
I
30
Time (hours)
-t- . Measured
i
40
Figure 8-1. Estimated vs. measured'benzene emission flux rates—Case 1.
-------
ro
««4
I/O-;
160 -
150 -
140 -
130 -
^» 120 -
f 110 -
N. 100 -
o>
3 90 -
3 80 -
c 70 -
o
8 60 ~
1 50-
40 -
30 -
20 -
10 -
0 -
C
1
I
\
N^s
XB — - . m
i i i i " i i i i '
> 10 20 30 40
D Estimated
Time (hours)
Measured
S-2. Es85raaii©d vs. measyreti ioiuen© ®mlssi®n
raSes— Case 1.
-------
26
M
80
100
D Estimated
Time (hours)
Measured
Figure 8-3. Estimated m, measured
emission WKX rafles— Case 2 (idale foer 4 days
-------
(A
\
CM
E
0>
3
ro
UJ
c
0
*M
IA
500
450
400
350
300
250
200
150
100
50
0
20
1
40
Time (hours)
T"
60
80
D Estimated + Measured
Figur® 8-4. Estimated vs. measured iotai ¥O emission iiux rafles—Case 2.
-------
than Measured values but mostly were within a factor of 10 or less of the
measurements. Estimated cumulative emissions also were substantially
higher than measured values. Estimated and measured values of instantan-
eous emission flux rates are shown in Figure 8-5.
In the 1979 test at the Midwest petroleum refinery (Case 4),35 three
test plots were laid out. One plot was used as a control and had no waste
applied, one plot had an API separator sludge applied, and the other plot
had a centrifuge sludge applied. A 1-ft2 collector box was placed on the
test plot and continuously purged with fresh air. The outlet from the box
was analyzed for total VO (as methane and NMHC) using a continuous hydro-
carbon analyzer. For one test run, total VO emissions were estimated with
the land treatment model for comparison with the measured values. Measured
and estimated values are shown graphically in Figure 8-6. As can be seen,
the estimated and measured values agree quite well for this test. Total
cumulative emissions for each test were also estimated using the model and
compared with measured values. The estimated values were generally higher
than measured values for these emissions.
8.3.1 Midwest Refinery—1985 (Case 1)
Table 8-16 presents the model input values used to compare estimated
and measured emissions for plot A of the Case 1 test data. The information
in Table 8-16 represents data for plot A as reported in the test report.
Similar information was reported for plots B through F and those data were
used as appropriate for input to the model. Table 8-17 shows measured
emissions of six constituents made during the test. In this test, the
waste was allowed to stay on top of the soil for 24 hours before it was
tilled into the soil. Measured emissions during the first 24 hours were
combined with measured emissions after tilling to get total emissions.
Table 8-17 shows variations in measured emissions among the different test
plots and shows emissions greater than applied material for some plots and
some waste constituents. In Table 8-17, weight fraction represents the
fraction of applied material that is emitted to the air. For ethylbenzene,
all plots have measured emissions in excess of the amount applied. To
compare measured and estimated emissions, the RTI land treatment model can
be used for estimating emissions both before and after tilling. Estimated
8-30
-------
00
CJ
M
\
O»t>
3 C
•^ O
o
*• r
n
M
5 -
4 -
3 -
2 -
1 -
0
Measured
liux rates—Case 3.
-------
CO
u>
ro
u
OJ
E
X
3
C
0
8
I
800
700 -
600 -
500 -
400 -
300 -
200 -
100 -
0
160
O Estimated
Time (hours)
Measured
i-6. Esitosfied vs. measured eircteion flux rales—Case 4.
-------
TABLE 8-16. INPUT PARAMETERS FOR RTI LAND TREATMENT MOOEL^
Parameter
Value
Source
Organic loading
Tilling depth
Soil air porosity
Soil total porosity
Benzene concentration^
Toluene concentration**
Benzene diffusivity
Toluene diffusivity
Benzene vapor pressure
Toluene vapor pressure
Benzene biorate
Toluene biorate
Molecular weight of oil
aSource of field data:
bWeight fraction of oil
0.0236 g/cm3
20 on
0.40
0.61
0.000249
0.000632
8.80 E-02 cm2/s
8.70 E-02 cm2/s
95.2 mm Hg
30.0 mm Hg
19.0 mg V0/g»h
73.0 mg V0/g«h
282 g/g mol
Calculated from field data
Field data
Field data
Field data
Calculated from field data
Calculated from field data
Data base
Data base
Data base
Data base
Data base
Data base
Assumed
Reference 36. Data represent conditions in plot A.
3-33
-------
TMLC •-!». ICASURED MO HtlNAICD IMUSIONS--CAM I
If (ft
Ul
oo
lOCCtlM Hfl/C**
A »fl.«l
• IM.M
C IM.M
• 4M.4I
I ttl.lt
r 1*4. M
Ir.c.
• •!
I.M
• .M
1.4*
t 91
• ••4
M/CB*
14*. I|
414. M
Mf.M
MI.M
IM.M
4*4. •!
Ir.e.
• .41
4).M
• .II
• .M
• ••1
• .41
Jta/cn*
II. M
M.4>
M.ll
MI.M
IM.II
II. M
«1.
Ir.e.
I.M
4.M
1.4*
l.ll
B . 4ft
I.M
Jto/ca*
I.I*
I.M
II. M
•I.M
M.I4
• .•1
•i.
Inc.
• .!•
«.tl
• .*•
rt»
• .41
• .II
•i.
Jlfl/c* lr«c.
M.4* •.!•
WI.M *.M
•l.ll •.*•
IM.M •.!•
IM.M •••*
M.«4 «.M
M/<.*
11.11
M If
tt.tt
M.M
tl.M
II. M
•».
lr.e.
• .M
•.M
• .II
• .I*
•.M
«.*4
„ »
t.lf
l.ll
I.M
l.ll
*.4«
1.44
•i.
»r.c.
• ••1
•.M
•••1
• .«
•••1
•••1
Ust
•k.
^/«.' «rM.
All
•••1
•.M
-------
cumulative emissions for benzene and toluene for all plots are shown in
Table 8-17 and show reasonable agreement with measured values.
8.3.2 West Coast Refinery (Case 2)
The data .in Table 8-18 were used to estimate emissions of toluene and
total VO from the surface application plot at the Case 2 land treatment
facility. Estimated and measured cumulative emissions are compared in
Table 8-19. The comparisons were made for total VO (as determined by purge
and trap) and for toluene.
8.3.3 Commercial Waste Disposal Test (Case 3)
Table 8-20 shows the inputs used to estimate emissions from the Case 3
land treatment operation. Mo specific constituent data were available so
emissions were estimated using average characteristics of the total organic
phase. Results are shown in Table 8-21. The comparison is made for the
estimated versus measured cumulative weight percent of applied oil that is
emitted after 24 hours and after 68 hours, which is the duration of the
entire test.
8.3.4 Midwest Refinery—1979 (Case 4)
The information in Table 8-22 was used to estimate emissions from the
CaSfe 4*facility test. No specific constituent data were available; emis-
sions were estimated for total organics using average parameter values.
Results are presented in Table 8-23. The comparisons are for the cumula-
tive weight percent of applied oil that was emitted over the entire period
of each test.
8.4 LANDFILLS AND WASTEPILES
Emission testing has been performed on at least one active (open)
landfill at each of five sites. Only three of these sites have closed or
inactive landfills at which emission measurements were performed. No emis-
sion test data are available for wastepiles.
Meaningful comparisons can be performed of emission test data with
mathematical model predictions provided that all key model input parameters
are available from the tests. A review of documentation from the emission
tests indicates that generally more than half of the needed model input
parameters (other than chemical property data) are unknown, despite the
fact that several emission tests were performed with the stated intention
8-35
-------
TABLE 8-18. INPUT PARAMETERS FOR RTI LAND TREATMENT MODEL*
Source
Organic (oil) loading
Tilling depth
Soil porosity
Molecular weight of oil
Toluene concentration
Toluene diffusivity
Toluene vapor pressure
Toluene biorate
VO concentration
VO diffusivity
VO vapor pressure
VO biorate
0.0328 g/cm3
20 CB
0.5
282 g/g ml
0.00157 (wt. frac-
tion of oil)
8.70 E-02 c«2/s
30.0 m Hg
73.0 ng V0/g»h
0.04 (wt. fraction
of oil)
6.60E-02 cm2/*
14.6 mm Hg
23.68 mg V0/g«h
Estimated from field data
Field data
Field data
Field data
Calculated from field data
Data base
Data base
Data base
Calculated from field data
•
Average from field data
Average from field data
Average from field data
a$our%e df field data: Reference 37.
TABLE 8-19. ESTIMATED VS. MEASURED EMISSIONS-CASE 2
Tol uene
Total VO
Total oil
Tine
after
tilling, day/h
33/793
33/793
33/793
Estimated
emissions,
wt. %
31
32
1.3
Measured
emissions,
wt. %
37
30
1.2.
8-36
-------
TABLE 8-20. INPUT PARAMETERS FOR RTI LAND TREATMENT MODEL*
Parameter
Value
Source
Organic loading
Tilling depth
Soil porosity
Molecular weight of oil
Vapor pressure
Diffusivity in air
Biorate
0.0406 g/cm3
19.6 cm
0.5
282 g/g mol
0.57 mm Hg
2.70 E-02 cra2-/s
23.68 mg V0/g«h
Calculated from field data
Field data
Assumed
Assumed
Calculated by GCAb
Average from field data
Average from data base
aSource of field data:
^Reference 39.
Reference 38.
TABLE 8-21. ESTIMATED VS. MEASURED TOTAL VO EMISSIONS—CASE 3
Time
after
tilling, h
68.00
Estimated
emissions,
wt. % total
applied oil
4.5
Measured
emissions,
wt. % total
applied oil
0.77
8-37
-------
TABLE 8-22. INPUT PARAMETERS FOR RTI UNO TREATMENT MODEL3
Parameter
Value
"Source
Organic loading
Tilling depth
Soil porosity
Molecular weight of oil
Diffusivity in air
Vapor pressure
Biorate
0.002125 g/cm3
20 en
0.5
282 g/g mol
9.12 E-02
g.76 an Hg
23.68 mg V0/g«h
Estimated from field data
Assumed
Assumed
Assumed
Average fron data base
Calculated by GCA&
Average from data base
^Source of field data: Reference 40.
^Reference 41.
TABLE 8-23. ESTIMATED VS. MEASURED EMISSIONS—CASE 4
Test
5
6
7
8
9
Elapsed
time, day/h
1/20
13/307
26/619
5/122
22/520
Estimated
emissions,
wt. % total
applied oil
5.0
14.0
16.0
14.0
28.0
Measured
emissions,
wt. % total
applied oil
0.14
2.5
13.5
1.1
13.4
8-38
-------
of validating emission models. Examples of key model input parameters that
are generally unknown or poorly defined include waste porosities (air and
total), average waste bed temperature (for active and closed landfills),
waste composition at depths greater than the surface layer, barometric
pressures, clay cap porosities (air and total), clay cap thickness, waste
bed depth, and (for active landfills particularly) time between core sam-
pling and air emissions determination. To apply the models, representative
default values have been used where necessary. Because of the necessity to
estimate key input parameters, the comparisons that follow are of extremely
limited value for model validation. To achieve validation of emission
models, additional field tests or laboratory experiments are needed for
active and closed landfills and wastepiles.
Field data from two sites were used for comparison with the land
treatment model as applied to active landfills. These sites (5 and 8) were
chosen because of similarity in constituency of selected chemicals and
relative availability of model input parameters. However, it should be
noted that at each of the sites more than half of the needed model input
parameters were not available from the tests and thus required estimation.
~* Information on the waste composition within closed landfills was
insufficient to allow use of the closed landfill model. At two of the
three closed/inactive landfill sites (4 and 5), no solid samples of waste
were taken; at the remaining site (Site 9), a single soil core was appar-
ently extracted from the 3-ft clay cover, providing no information about
the composition of the waste below the cover. However, it should be noted
that Farmer et a1.42 (who developed the precursor* to the RTI closed land-
fill model) mentioned that their model has received experimental verifica-
tion via a laboratory experiment using hexachlorobenzene-containing waste
in a simulated landfill.
Following are the results of the comparison for active landfills at
Sites 5^3,44 an(j g45. Table 8-24 presents model input parameters used in
*The Farmer et al. model accounts for diffusion through the clay cap only
(not barometric pumping).
8-39
-------
TABLE 8-24. MODEL INPUT PARAMETERS USED
RTI LAND TREATMENT MODEL TO AN ACTIVE
Parameter
Value '
IN APPLICATION OF THE
LANDFILL AT SITE 5*
Data source
L, total organic loading 2.65 x 10'3 g/c«3
in soil
Ct, weight fraction of Xylene: 0.178
constituent i in Methylene chloride:
organic phase
T, temperature of
constituent vapor
in soil
1, depth of waste in
landfill
ey, total porosity of
waste
:- --.» fe
ea, air porosity of
waste
Sb, soil biomass
concentration
8.48 x lO'4
Tetrachloroethy1ene:
1.37 x 10-3
25 »C
229 on (7.5 ft)
0.50 (50%)
0.25 (25%)
0 g/on3
MH0ii, molecular weight 150 g/g mol
of organic carrier
liquid
t, tine between soil
sampling and air
emission measurement
900 s (15 min)
Inferred from field data
(solid sample analysis)
assuming soil density »
2.3 g/cm3
Computed from field data
(solid sample analysis)
Default value
Default value
Default value
Default value
Default value
Default value
Engineering judgment
8-40
-------
the application of the RTI land treatment model to an active landfill at
Site 5. Table 8-25 presents a comparison of measured and predicted
emission rates for the Site 5 landfill.
Model predictions for the chemicals assessed here are higher than
field data values by a factor ranging from 13 to 441. This discrepancy may
be largely a result of the presence of daily earth covers (6 in. thickness)
and layers of drums within the waste bed—neither of which are accounted
for by the model. Other contributing factors may include the estimation of
key model input parameters (e.g., air porosity of waste, temperature of the
constituent within soil) and the nonrepresentative nature of the waste
sample (obtained at the surface) for describing the *aste composition at
depth.
Table 8-26 presents model input parameters used in the application of
the RTI land treatment model to an active landfill at Site 8. Table 8-27
presents a comparison of measured and predicted emission rates for the
Site 8 landfill.
- Model predictions of the emissions at Site 8 are, in general, closer
to,fiey data than were the predictions made for Site 5. Better overall
agreement may be due to the absence of drums and daily earth covers in this
landfill.
8.5 TRANSFER, STORAGE, AND HANDLING OPERATIONS
8.5.1 Models Documented in AP-42
Emission methods for the following models were taken from AP-42; they
have been developed from the field data in the petroleum industry and
should be applicable to TSDF:
• Container loading (from AP-42, Section 4.4)52
• Container cleaning (from AP-42, Section 4.8)53
• Stationary tank loading (from AP-42, Section 4.3)54
• Stationary tank storage (from AP-42, Section 4.3).55
8.5.2 Fugitive Emissions
Fugitive emission sources have been studied extensively for the
petroleum and Synthetic Organic Chemical Manufacturing Industries
8-41
-------
TABLE 8-25. COMPARISON OF MEASURED AND PREDICTED EMISSION RATES
FOR SITE 5 ACTIVE LANDFILL^ _
-- ' Field data result, Model prediction,
Cheaical
Methylene chloride
Tetrachloroethylene
atandfill 10, General Organic Cell.48'49
8-42
-------
TABLE 8-26. MODEL INPUT PARAMETERS USED IN APPLICATION OF THE
RTI LAND TREATMENT MODEL TO AN ACTIVE LANDFILL AT SITE 8*0
Parameter
Value
Source
L, total organic loading
in soil
Ci, weight fraction of
VOj in organic phase
T, temperature of VO
vapor in soil
1, depth of waste in
landfill
•
eji total porosity of
waste
ea. air porosity of
waste
$5, soil biomass
concentration
MW0ji, molecular weight
of organic carrier
liquid
t, time between soil
sampling and air
emission measurement
1.71 x 10-6 .g/cm3 Field data
Xylene: 0.012
1,1,1-TCE:. 0.19
Tetrachloroethylene:
0.096
25 •€•
229 on (7.5 ft)
0.50 (50%)*>
0.25 (25%)b
0 g/cm3
150 g/g mol
900 s (15 min)
Computed from field data
(solid sample analysis)
Default value
Default value
Default value
Default value
Default value
Default value
Engineering judgment
aSoil surface temperatures at this site were reported at 26 to 36 *C. The
model unit default value of 25 *C is applied to the constituent within
the soil in this analysis.
bA single "porosity" value of 31.7 percent was reported for this site,
based on one soil sample. Because this value is not defined explicitly,
and may not be representative of typical waste in the landfill, the model
unit default values of air porosity and total porosity were applied in
this analysis.
8-43
-------
TABLE 8-27. COMPARISON OF MEASURED AND PREDICTED EMISSION RATES
FOR THE SITE 8 ACTIVE LANDFILLS1
Field da$a result. Model prediction,
Cheaical
Total xylene 6.21 0.23
1,1,1-Trichloroethane 3.57 3.8
Tetrachloroethylene 6.31 !•»
8-44
-------
(SOCMI) facilities.56 These SOCMI emission factors are assumed to be
applicable to similar operations at TSDF.
8.5.3 Spillage
An ICF57 study of truck transport to and from TSDF and truck
emissions at TSOF terminals provided the information necessary to
estimate spillage losses during TSDF and trucking operation. However,
none of the reports listed in Appendix F of the BID contains field
test data for comparison.
8.5.4 Open Dumpster Storage Emissions
No field data were available for comparison.
8.6 REFERENCES
1. 6CA Corporation. Air Emissions for Quiescent Surface Impoundments-
Emissions Data and Model Review, Draft Technical Note. Prepared for
U.S. Environmental Protection Agency. Contract No. 68-01-6871,
Assignment 49. August 1985.
2. Radian Corporation. Hazardous Waste Treatment, Storage, and Disposal
Facility Area Sources: VOC Air Emissions. Prepared for U.S. Environ-
mental Protection Agency, Emissions Standards and Engineering Divi-
sion. "Research Triangle Park, NC. Contract No. 68-02-3850. Janu-
ary, 25, 1985.
3. Reference 1.
4. Reference 2.
5. Reference 1.
6. Reference 2. .
7. Reference 1.
8. Radian Corporation. Hazardous Waste Treatment, Storage, and Disposal
Facility Area Sources: VOC Air Emissions at Chemical Waste Manage-
ment, Inc., Kettleman Hills Facility. Volume 1. Prepared for U.S.
Environmental Protection Agency, Emissions Standards and Engineering
Division. Research Triangle Park, NC. EMB Report 85-HW5-2. December
1984.
9. GCA Corporation. First Chemical Corporation Wastewater Holding Lagoon
Field Study. Prepared for U.S. Environmental Protection Agency.
Contract No. 68-02-3851 (WA10). 143 p. August 1986.
10. GCA Corporation. Emissions Data and Model Review for Wastewater
Treatment Operations. Draft Technical Note. Prepared for U.S.
8-45
-------
Environmental Protection Agency. Contract No. 68-01-5871, Assign-
went 49. August 1985.
11. Reference 10.
12. Reference 10.
13. Reference 10.
14. Reference 10.
IS. Reference 10.
16. Nelson, Thoms P., et al. (Radian). Field Assessment of Air Emissions
and Their Control at Hazardous Waste Facilities (Draft). Prepared for
U.S. Environmental Protection Agency, Industrial Environmental
Research Laboratory. Cincinnati, Ohio. December 1984. 77 p.
17. Reference 10.
18. Reference 16.
19. Petrasek, A., B. Austern, and T. Neiheisel. Removal and Partitioning
of Volatile Organic Priority Pollutants in Wastewater Treatment.
.Presented at the Ninth U.S.-Japan Conference on Sewage Treatment
Technology. Tokyo, Japan. September 1983. 31 p.
20. Reference 19, p. 2-4.
21. Reference 19, p. 16.
22. Namkung, E., and B. Rittman. Estimating Volatile Organic Compound
Emissions from Publicly Owned Treatment Works. Journal WPCF.
59(7):677.
23. Reference 22, p. 671-672.
24. Reference 22, p. 672.
•
25. Tabak, H.. S. Quave, C. Mashni, and E. Barth. Biodegradability
Studies with Priority Pollutant Organic Compounds. Staff Report.
Wastewater Research Division. U.S. Environmental Protection Agency.
Cincinnati. Ohio. 1980.
26. Reference 25.
27. Cobum, J.. C. Allen, D. Green and K. Leese. Site Visits of Aerated
and Nonaerated Impoundments Revised Draft Summary Report. Prepared
for U.S. Environmental Protection Agency. Contract No. 68-03-3253,
Work Assignment No. 3-8. April 1988. p. A-l to A-34.
8-46
-------
28. Reference 27.
29. Allen, C. C., et al. Preliminary Assessment of Aerated Waste Treat-
ment Systems at Hazardous Waste TSDFs (Draft). Prepared for U.S.
Environmental Protection Agency. Contract No. 68-02-3992. December
1985. 108 p.
30. Reference 29.
31. Utah State University. Evaluation of Volatilization of Hazardous
Constituents at Hazardous Waste Land Treatment Sites. Prepared for
U.S. Environmental Protection Agency. Ada, Oklahoma. Undated.
157 p.
/
32. Radian Corporation. Field Assessment of Air Emissions and Their Con-
trol at a Refinery Land Treatment Facility. Volumes I and II. Pre-
pared for U.S. Environmental Protection Agency. Cincinnati, OH.
September 1986. 699 p.
33. Radian Corporation. Hazardous Waste Treatment, Storage, and Disposal
Facility Area Sources—VOC Air Emissions. Prepared for U.S. Environ-
mental Protection Agency. Research Triangle Park, NC. January 1985.
141 p.
34. GCA Corporation. Air Emissions from Land Treatment—Emissions Data
" and Model Review. Draft Technical Note. Prepared for U.S. Environ-
mental Protection Agency. August 1985. p. 4-36.
35. Suntech, Inc. Atmospheric Emissions from Oily Waste Land Spreading.
Final Report SWM-8(563). Prepared for American Petroleum Institute.
Washington, DC. Undated. 63 p~
36. Reference 31.
37. Reference 32.
38. Reference 33.
39. Reference 34.
40. Reference 35.
41. Reference 34.
42.. Farmer, W. J., M. S. Yang, J. Letey, W. F. Spencer, and M. H. Roulier.
Land Disposal of hexachlorobenzene Wastes: Controlling Vapor Movement
in Soils. Fourth Annual Research Symposium. U.S. Environmental
Protection Agency. Publication No. EPA-600/9-78-016. August 1978.
43. Radian Corporation. Hazardous Waste Treatment, Storage and Disposal
Facility Area Sources: VOC Air Emissions. Prepared for U.S. Environ-
8-47
-------
mental Protection Agency. Research triangle Park, NC. DCN 85-222-
078-17-09. January 25, 1985. 141 p.
44, Radian Corporation. Evaluation of Air Missions from Hazardous Waste
Treatment/Storage and Disposal Facilities in Support of the RCRA Air
Emission Regulatory Impact Analysis (RIA): Data Volume for Site 4 and
Site 5 Prepared for U.S. Environmental Protection Agency. Cincin-
nati, OH. DCN 83-203-001-63-19. January 11, 1984. 454 p.
45. Radian Corporation. Hazardous Waste Treatment, Storage and Disposal
Facility Area Sources—VOC Air Emissions. Prepared for U.S. Environ-
mental Protection Agency. Research Triangle Park, NC. EMB Report
85-HWS-l. May 1985. 54 p.
46. Reference 43.
47. Reference 44.
48. Reference 43.
49. Reference 44.
50. Reference 45.
51. .Reference 45.
52. JJ.5^ Environmental Protection Agency. Transportation and Marketing or
Petroleum Liquids. In: AP-42. Compilation of Air Pollutant Emission
Factors. Third Edition, Supplement 12, Section 4.4. Office of Air
Quality Planning and Standards. Research Triangle Park, NC. July
1979. 13 p.
53. U.S. Environmental Protection Agency. Tank and Drum Cleaning. In:
AP-42 Compilation of Air Pollutant Emission Factors. Third Edition,
Supplement 12, Section 4.8. Office of Air Quality Planning and Stand-
ards. Research Triangle Park, NC. February 1980. 4 p.
54. U.S. Environmental Protection Agency. Storage of Organic Liquids.
In: AP-42. Compilation of Air Pollutant Emission Factors. Third
Edition. Supplement 12, Section 4.3. Office of Air Quality Planning
and Standards. Research Triangle Park, NC. April 1981. 25 p.
55. Reference 47.
56. U.S. Environmental Protection Agency.. Control of Volatile Organic
Compound Leaks from Synthetic Organic Chemical and Polymer Manufactur-
ing Equipment. Research Triangle Park, NC. Publication Nc. EPA-450/
3-86-006. March 1984.
57. U.S. Environmental Protection Agency. Assessing the Release and Costs
Associated with Truck Transport of Hazardous Wastes. PB 84-224-468
(Contract No. 68-01-0021). Washington, DC. January 1984. 151 p.
8-48
-------
APPENDIX A
CHEMDAT7 USER'S GUIDE
-------
APPENDIX A
CHEMDAT7 USER'S GUIDE
A.I INTRODUCTION
CHEMDAT7 is a Lotus 1,2,3 spreadsheet that includes analytical models
for estimating volatile organic (VO) compound emissions from treatment,
storage, and disposal facility (TSDF) processes under user-specified input
parameters. The available models include disposal impoundments, closed
T-andfills, land treatment facilities, and aeration and nonaeration impound-
ment processes. Predicted emissions can be viewed on the screen or
printed. A graphical presentation of the relationships between emission
prediction and vapor pressure and between emission prediction and the par-
tition coefficient is also available to the user. The resulting scatter
diagrams can be printed via PrintGraph, another Lotus procedure.
VO emission models from hazardous waste TSDF were described in
Chapters 4.0, 5.0, 6.0, and 7.0 of this report. The emission rates from
some TSOF can be estimated via CHEMDAT7. In this regard, Exhibit A-l
specifies the appropriate CHEMDAT7 model to estimate emissions from partic-
ular TSDF. For example, the nonaerated impoundment model in CHEMDAT7 can
estimate emissions from storage impoundments. The CHEMDAT7 model for pre-
dicting emissions from treatment impoundments is the aerated impoundment
model. Furthermore, the land treatment model in CHEMDAT7 can estimate
emissions from land treatment soil, open landfills, and wastepiles. Emis-
sions from an oil film surface in a land treatment facility or an oil film
on a surface impoundment are predicted via the oil film model in CHEMOAT7.
When a CHEMDAT7 model is not available to predict emissions, the reader
must resort to the prediction equations in the previous sections of this
report to estimate VO emissions from TSDF.
CHEMDAT7 calculates the fractions of waste constituents of interest
that are distributed among the pathways (partition fractions) applicable to
the facility under analysis. Estimated annual emissions from many of the
A-3
-------
TSOF are calculated by this spreadsheet. Otherwise, estimated annual
emissions can be calculated offline by multiplying the throughput of the
constituent of interest by the partition fraction.
the seventh version of the CHEMOAT spreadsheet contains several major
operational modifications. In CHEMDAT7, the user can select a subset of
target compounds for investigation. The user can also specify which TSDF
processes are to be considered during a session. These two selections
improve the efficiency of CHEMDAT7 relative to some of the earlier versions
by minimizing storage requirements as well as the actual loading and
execution time.
Default input parameters in the accompanying CHEMOAT7 diskette
demonstrate example calculations in Chapters 4.0. 5.0, and 6.0 of this
report. However, the user can readily change the input parameters to
reflect different hazardous waste TSDF characteristics and then recalculate
emissions under these modified conditions. The diskette with this user's
guide is wrHe-protected. It is suggested that a copy be used in estimat-
ing emissions. Furthermore, the list of compounds in CHEMDAT7 can be aug-
mented by any of the 700 chemicals in Appendix 0. Procedures for introduc-
ing additional compounds into CHEWAT7 are described in Appendix 0.
Instructions on the use of CHEMOAT7 appear in this appendix. (We have
assumed throughout this appendix that the user has some minimal knowledge
of Lotus 1,2,3.) CHEMJAT7 contains a data base of component-specific prop-
erties used to generate internally the inputs for the environmental fate
models of waste disposal practices. Section A.2 specifies how to get
started in Lotus 1,2,3. Section A.3 details proper utilization and
modification of CHEMDAT7 via the alternative command menu. The data base
and model parameters are described in Section A.4. The steps required to
obtain a printout of the graphs are described in Section A.5. The
referenced exhibits complete this appendix.
A.2 SETTING STARTED IN LOTUS 1,2,3
The applications program, Lotus 1,2,3, Version 2.0, should be in-
stalled on the computer system for use. Place the diskette containing*
CHEHDAT7 in a low density drive.
Access the 1,2,3 mode of operation in Lotus. A skeleton of a
spreadsheet will appear on the screen, and the message READY will appear in
A-4
-------
the upper right corner of the screen. (Consider this and other messages as
road signs that guide you through CHEMOAT7.) Either copy the contents of
the CHEMOAT7 diskette to your default directory or set the directory in
Lotus 1,2,3 to the disk drive containing the CHEMDAT7 diskette. To
retrieve CHEMOAT7, press the following sequence of keys, i.e., those keys
that appear between brackets { }:
{/> in w •
Note that {/} calls the active command mode. (The active command mode is
indicated by the message MENU in the upper right corner of the screen.)
The F represents file and the R represents retrieve. In other words, these
are the commands to retrieve a file. We indicate the file to retrieve by
typing the drive label, followed by a colon, followed by CHEMDAT7,
{B-.CHEMDAT7} ,
and then pressing {RETURN}.
An equivalent approach is to position the cursor over CHEMDAT7, which
appears below the command line, and then press {RETURN}. The cursor con-
trols^(keys wi'th arrows on top, i.e., t, 4, __>, and <__) position the
cursor. After CHEMDAT7 loads for a minute or two (note the message WAIT
flashing at the bottom of the screen), the user will be in the help mode of
the alternative command menu developed specifically for CHEMDAT7. The
message CMD at the bottom of the screen indicates that the alternative
command menu is active. After reviewing the appropriate help screens (see
Section A.6 for further details), press {Q} for quit to return to the main
alternative command menu. A discussion of this menu follows.
A.3 ALTERNATIVE -COMMAND MENU
The alternative command menu is a menu-driven set of instructions
specific to CHEMDAT7. Choose an option (DATA-FORMS, VIEW, SORT, PRINT,
SELECT, HELP, or QUIT) by positioning the cursor over the selection and
pressing {RETURN}. Alternatively, the first letter or set of letters that
uniquely identify the selection can be entered: e.g., {P} can be pressed
to choose the PRINT option. Alternative command menu options and subop-
tions are listed in Exhibit A-2; please study this guide carefully. If you
get lost after accessing the alternative command of CHEMDAT7, press {ESC}
several times to completely exit the macro. Then the first level of
A-5
-------
options (DATA-FORMS, VIEW, etc.) can be accessed via {ALT M}. "ALT M"
stands for the alternative command menu.
A.3.1 DATA-FORMS Option
The DATA-FORMS option in the alternative command menu is the crux of
this software package. It is the vehicle by which the user can change
model input parameters in CHEMOAT7.
Once the user has specified {DATA-FORMS} and has selected the desired
•odel, the corresponding input parameters will appear highlighted on the
computer screen. Move the cursor to the cell location containing the model
input parameter to be modified. After the proper cell has been located in
the spreadsheet with the cursor, type the new input parameter, then press
{RETURN}. The new input parameter will appear on the screen. At this
point, the message CALC will appear in the lower right side of the screen
indicating that the model predictions must be recalculated to reflect the
new input parameter. Once all the input parameters have been modified,
press (F9}~to recalculate the emission predictions under the new input
parameters-. The WAIT message will flash as the model predictions are
recalculated. The CALC message will disappear once the model predictions
have been revised. The user must perform this latter step so that the
emission results will reflect the modified input parameters. Furthermore,
this revised spreadsheet must be saved if updated graphs are desired.
To save the revised spreadsheet, access the active command mode (press
{ESC}). Then type
B: / F S {ENTER} R ,
where F represents file, S represents save, the file name automatically
appears^ and R represents replace. Consequently, the old version of the
CHEMDAT7 spreadsheet is replaced by the new, modified version.
The DEFAULT suboption for DATA-FORMS is used to replace model input
parameters with the default values, i.e., those values corresponding to
example calculations described in Section A.4.3. The model default values
for impoundments (non-aerated, diffused, oil film, or disposal), aerated
wastewater treatment, land treatment soil, open landfills, and closed •
landfills appear in Exhibits A-3, A-4, A-5, A-6, A-7, and A-8, respec-
tively. After input parameters are replaced with default values, remember
to recalculate model predictions.
A-6
-------
The CONC suboption for DATA-FORMS is used when individual concen-
trations are to be specified for each VO in the data set of interest. If
you select this option, you nay specify the individual concentrations for
each of the units that support this feature, but you must enter a zero as
the concentration in the standard data forms. Failure to do this will
result in the concentration used for each of the compounds defaulting to
the concentration on the standard data form.
The IMPORT suboption is used with a specially designed program, CHEM7,
that complements CHEMOAT7. One of the uses of the {IMPORT} {RESTORE}
option is to replace the data in CHEMOAT7 with the original data supplied
with the diskette. The {IMPORT} {GO} option loads a set of compounds into
the spreadsheet under the compounds selected with the {SORT{ {SELECT}
option.
A.3.2 VIEW Option
A.3.2.1 Viewing Model Emission Results. The VIEW option enables the
user to view on the computer screen emission results from disposal impound-
ments; land treatment facilities, open landfills/wastepiles, closed land-
fills* and nonaerated and aerated models. Exhibit A-2 describes the
options and suboptions available. To exit the model results, press
{ALT M}.
A.3.2.2 Viewing Graphs. The user is able to view graphs on the
computer screen by pressing {VIEW} {GRAPH} and then selecting the desired
model. The user must exit from this procedure via {QUIT} in order to
reformat the spreadsheet.
A.3.2.3 Viewing Mass Transfer Coefficients. The user is able to view
mass transfer coefficients on the computer screen by pressing {VIEW} {K}
and selecting the desired facility. A list of the facilities that use mass
transfer coefficients is presented on screen, preceded by a number. After
the user presses the desired number, the mass transfer coefficients and
Henry's law constants appear on the screen for each compound. The units of
the mass transfer coefficients are M/sec.
A.3.3 SORT Option
This option in the alternative command menu enables the user to
rearrange the order of the compounds in the CHEMDAT7 spreadsheet. Subop-
tions include sorting by biorate, selected compound, alphabetically, by
A-7
-------
compound type (class), and by vapor pressure. Model predictions must be
recalculated after sorting.
The {SELECT} suboption under {SORT} is a very powerful feature of
CHEMDAT7. Analogous to a "cut and paste" procedure, {SORT} {SELECT}
enables the user to select a subset of the 61 compounds for entry into the
CHENDAT7 spreadsheet. Model calculations are only performed for the
selected compounds, which improves the performance of the spreadsheet.
Consequently, emission results are limited to the selected compounds. This
{SORT} {SELECT} option was employed to choose the compound of interest,
benzene, in previous exhibits.
Two approaches are available for selecting compounds. To select a set
of chemicals, use a "1" to flag those compounds of interest. Those com-
pounds preceded with a "0* will be ignored. The user may also order the
selected compounds so that compound-specific emission results are printed
in a desired sequence. If there are n compounds of interest, place an
integer from 1 to n in front of each compound. Once {ALT} {1} is typed,
CHEMOAT7 will rearrange the compounds in descending order according to this
integer. As before, emission results will not be estimated for those com-
pounds preceded by a "0".
{SORT} {DATA} is an option used to add additional compounds to the
{SORT} {SELECT} compounds that are selected. An extended data base of
approximately 300 compounds is added to the spreadsheet, starting at Cell
{A 255}. The compounds in this extended data base are selected in an
identical fashion to the compound selection process under {SORT} {SELECT}.
After the compounds of interest are selected, press {ALT A} to add the
selected compound to your CHEMOAT7 compound list. CAUTIONt This option
will destroy data for compounds that are not selected in the {SORT}
{SELECT} option, so perform the {SORT} {SELECT} step first. If you change
your mind later, you can always restore the original data set by {DATA-
FROM} {IMPORT} {RESTORE} and start over.
A.3.4 PRINT Option
The PRINT option allows the user to obtain a hard copy of model-
specific emission results (via the suboptions LAND-TREAT, DISPOSAL,
AERATED, LANDFILL, and NON-AERATED) and the corresponding set of input
parameters, as well as mass transfer coefficients (via the K suboption) and
the data summaries (via the DATA suboption). The data summaries include
A-8
-------
biorate data (refer to columns N through P in Exhibit A-9), land treatment
data (see columns AM and AN in Exhibit A-9), and selected data tables (see
columns F, G, H, and I in Exhibit A-9) that are printed via the commands
{PRINT} {DATA} {BIORATE}, {PRINT} {DATA} {UNO-TREAT}, or {PRINT} {DATA}
{DATA}, respectively. Examples of printed model input parameters and
emission results appear in the exhibits.
The PRINT option has another important function: it enables the user
to store the graph file on the diskette. Enter the commands {PRINT}
{GRAPH} followed by the selected model. This procedure is required to
obtain a hard copy of the model-specific graph. Further details are
provided in Section A.5.
A.3.5 SELECT Option
The SELECT option in the main menu, another "cut and paste" feature of
CHEMDAT7, allows the user to activate only a subset of the TSDF models for
entry into the spreadsheet. Models are selected through the 0-1 (off-on)
flag system described prevously. Note that, if the user attempts to use a
model that is not active, error messages will be printed instead of model
emission results.
A.3.6* HELP Option
The HELP option provides the user with information regarding the
proper application and various assumptions of the specific models in
CHEMOAT7. A general help screen is also available. The help screens are
presented in Section A.6.
A.3.7 QUIT Option
The active command mode can be accessed from the alternative command
menu by locating the cursor over {QUIT} and pressing {RETURN}. In this
mode, all the features of Lotus 1,2,3 are available by pressing {/}.
A.3.8 Adding Additional Compounds
There are three ways to add new compounds to the CHEMDAT data base:
overwriting existing data, selecting compounds from the supplemental list,
and using a separate compound data program (CHEM7).
A.3.8.1 Overwriting Existing Data. To overwrite existing data, use
the cursor to locate a compound that you do not need and wish to overwrite:
• Use the {F5} key to locate the cursor in the cell containing the
name of the compound to be overwritten. The data for this
compound appears in the corresponding row.
A-9
-------
• Enter the data for the new compound in that row. Data sources
include:
Reference books
— CHEM7
Default values.
• Use the alternative command menu {ALTUM} option of {SortJ
{Select} to select the compound and reform the worksheet by
pressing {ALT}{Z}.
A.3.8.2 Selecting from a Supplemental List. To select compounds from
a supplemental list, use the command menu {ALT}{M} option of {Sort}{Data}.
Select the compounds in the supplemental list by tagging each selected
compound with a number greater than 1 in column A of the row containing the
compound name. After the selection is complete, press {ALT}{A} to reform
the spreadsheet.
The compounds selected from the supplemental list will be sorted from
the list and transferred to the bottom of the working list in rows 14
through 82. The spreadsheet will then be sorted with all selected
compounds (both from the working list and the supplemental list) trans-
ferred to the top of the list, in decending order of the tag number in
column A.
There is a limit of 20 compounds when individual concentrations are
specified for each compound. Row (DP43..0Y43) can be copied in lower rows
to expand the compound number limit for individual concentration
specifications.
A.3.S.3 Loading Compounds from a Master List. To load a compound
from a master list, a separate program (CHEM7) is available. To use CHEM7,
you must quit CHEMDAT7 by the {/}{Q}{Y} option. Change the directory to
the same directory that you are using for the CHBCAT7 program, then type
CHEM7 and press {ENTER}.
For the users of a hard disk system, copy the files on the CHEM7
diskette to your hard disk directory used by CHEMDAT7. Be sure to save a
backup copy of all your original diskettes.
Users of a floppy disk system will need to use the CHEM7 diskette.
Again, be sure to make a backup copy before using your diskette. When you
load the compound data, you need to put a diskette containing the compound
data (MASTERCL.CH7) in the disk drive. When you print your data to
diskette, you need to use a diskette that you will use with the CHEMDAT7
program. You may place the CHEMDAT7 diskette (CHEMDAT7.WK1) in the disk
drive.
A Ifl
-------
A separate manual is available to describe the program CHEM7. The
user oust load compound data from a master data base, select compounds, and
print the data to disk. The user can optionally append compound data to
the master data base. If the user only wants to select additional com-
pounds from an extended list and let the computer automatically fill in
missing data, the command CHEM7 1 can be entered, with a space between the
7 and the 1.
After using CHEM7, return to CHEMOAT7 and load the files you have
created. In CHEMOAT7 you will use the command menu {ALT}{M} option of
{Data-form}{Import}{Go}.
A.4 ANATOMY OF THE CHEMOAT7 SPREADSHEET
This section describes the structure of the CHEMDAT7 spreadsheet; a
generalized layout of the spreadsheet appears in Attachment A. In general,
rows 1 through 5 of this spreadsheet contain column labels or names. The
compound-specific data base and calculation results appear in rows 14
through 83. The locations of model-specific input parameters are specified
throughout Section A.4.3.
A.4.1 Data Base
Vhe data base in the CHEMDAT7 spreadsheet is a matrix of component-
specific properties or calculations. Sixty-one chemicals or compounds
appear in rows 14 through 83 of the spreadsheet. These compounds are
listed in Exhibit A-10.
Compound properties and model computational results appear in columns
B through BU of the spreadsheet. For example, molecular weight appears in
column D. Thus, each compound-specific data item has a unique cell address
"cr" where c (alphabetic) represents the appropriate column and r (numeric)
the row. Suppose data on toluene appear in row 16 of the CHEMDAT7 spread-
sheet. Then, the molecular weight of toluene would appear in cell D16.
A.4.2 Column Labels
In general, column labels or names appear in rows 1 through 5 of the
spreadsheet. Exhibit A-9 lists the column labels in CHEMDAT7.
A.4.3 Model Input Parameters and Predicted Emissions
Brief descriptions and locations of the CHEMDAT7 model-specific
parameters are presented in the following subsections. Additionally, use
of CHEMDAT7 is demonstrated via example calculations presented previously
in this report. VO emissions estimated from CHEMDAT7 models are compared
to the example calculation results.
A-ll
-------
Missing data in the CHEMOAT7 spreadsheet (e.g., vapor pressure for
dioxin) frequently preclude estimating emission rates for the affected
compound. In such cases, "ERR" is printed in place of the estimated emis-
sion. The user is encouraged to insert these missing data as values become
available. The procedure for modifying input parameters is described in
Section A.3.1.
A.4.3.1 Nonaerated Model. CHEMDAT7 nonaerated model input parameters
are located in cells C06 through C015 of the spreadsheet and are illus-
trated in Exhibit A-ll. (This and subsequent exhibits depict what the user
will see on the computer screen.) The windspeed in meters/second is placed
in C06. C07 contains the depth of the nonaerated impoundment (in meters).
The surface area of the impoundment (in meters squared) and the flow rate
(in cubic meters/second) appear in COS and C09, respectively. The VO inlet
concentration expressed in milligrams/liter is placed in C010. Cell C011
contains the sum of the organics (in milligrams/liter) entering the
facility. The conversion of these organics to biomass is used to estimate
the rate of adsorption as a pathway for the removal of organics from the
system." Note that the VO inlet concentration in C010 should be less than
the overall concentration of organics in C011. Because biodegradation is
presumed to be nonexistent in a storage impoundment, the amount of active
biomass (C012) has been set at zero. If the uniqueness of the TSDF permits
biodegradation as a pathway and the user desires to consider its impact on
'emissions, C012 can be changed to reflect active biomass using standard
Lotus procedures (see Section A.3.1). The input parameter, biomass solids
in (C013), is appropriate for municipal facilities only. An approximation
of the solids input (C014) for adsorption in municipal facilities is 200
ppmw. Of these solids, approximately 50 percent are removed in the primary
clarifier, leaving 0.1 g/L solids input to the treatment basin. The
ambient air temperature for the facility in degrees Celsius is placed in
C015.
The nonaerated model input parameters in Exhibit A-ll have the same
values as those used in the example calculation for storage impoundments in
Section 4.2.3 where the compound of interest is benzene. The predicted
emissions for benzene from CHEMDAT7 (see Exhibit A-ll) are to those model
results presented in the latter portion of Section 4.2.3. The predicted
fraction of benzene that will be emitted to the air is calculated as 0.801.
A-12
-------
The estimated annual -air emissions of benzene from the example nonaerated
surface impoundment total 0.39 Mg/yr. CHEMDAT7 model input parameters and
results are printed using the PRINT option as discussed below in Section
A.3.4.
In contrast, the example calculation in Section 4.3.3 considers
biodegradation as a pathway. The amount of active biomass is 0.05 g/L; see
Exhibit A-3. The remaining model input parameters are identical to those
in Section 4.2.3. Almost 11 percent of the benzene is emitted to the air
so that the annual emission rate is 0.052 Mg/yr.
A.4.3.2 Aerated Model. CHEMOAT7 parameters for the aerated
impoundment model are located in column CO, rows 78-88, and column CS, rows
79-85 (see Exhibit A-4) The equations for estimating relative pathways
include those for the quiescent surface of nonaerated systems, but these
must be supplemented to account for a turbulent zone.
Because biodegradation is assumed to occur in aerated impoundments,
the amount of active biomass (grams/liter) must be specified in C082. Cell
C087 contains the fraction of the impoundment's surface area that is agi-
tated. xThe submerged air flow (m3/s) in C088 accommodates a mechanical air
source located below the surface of.the impoundment. Consequently, emis-
sions from a diffused air flow-through system, i.e., one that emphasizes
biodegradation, are estimated by placing an appropriate value in C088. The
remaining input parameters in column CO are the same as those explained in
the previous section.
The number of impellers is specified in CS79. The oxygen transfer
rating of the aerator (pounds 0£/horsepower/hour) appears in CS80. The
total power of the mechanical aeration system expressed in horsepower is
placed in cell CS81. Cell CS82 contains the power efficiency factor (no
units). The remaining parameter inputs include the water temperature
(CS83) in degrees Celsius, the impeller diameter (CS84) in centimeters, and
the impeller speed (CS85) in rads/second. If the impoundment's surface is
not agitated (C087=0), the final seven model input parameters (CS79-CS85)
are ignored in the emission calculations.
The aerated model input parameter values in Exhibit A-4 are the same
as those used in the example for a mechanically aerated treatment
impoundment discussed in Section 4.4.3. The compound of interest is
benzene. CHEMDAT7 reproduces the materials balance results presented in
A-13
-------
Section 4.4.3; see Exhibit A-4. The predicted fraction of benzene that
will be emitted to the air is 0.83. The estimated annual air emissions of
benzene fro* such aerated impoundments are 0.81 Mg.
Emission results for the activated sludge unit in Section 4.4.4 and
the diffused air activated sludge unit in Section 4.6.3 are also available
fro* the CHEMDAT7 aerated model. See Exhibits A-5 and A-12, respectively,
for the model input parameters, the materials balance results, and the
annual air emission rate.
Diffused aerated systems retain the input parameters for mechanical
aeration. It is possible to have both in an activated sludge system. When
not physically present, the mechanical aeration parameters can be used to
estimate the mass transfer coefficients of exposed surfaces agitated by
diffusing air bubbles.
The diffused air option in the aerated impoundment model can be used
with or without surface aeration. If the surface is not agitated by the
churning action of submerged aeration, the fraction agitated can be
specified as zero. If the surface is visibly agitated, the fraction
agitated can be specified as the fraction of the surface that is estimated
to be agitated by visual .inspection or calculations. The mass transfer
coefficients of the agitated zone are estimated the same as for surface
aeration, so the default parameters of the surface-agitated units are used
for the diffused air model.
A.4.3.3 Land Treatment Model. The CHEMDAT7 land treatment model can
predict emissions from land treatment soil, open landfills, and wastepiles.
A general description of the land treatment model parameters is followed by
specific guidelines for each application.
The land treatment model parameters for CHEWDAT7 are located in cells
CV7 to CV18 as illustrated in Exhibit A-6. The oil loading must be
obtained from the site operator or manually calculated offline as indicated
in the report. It then is entered in CV7. The weight concentration of the
VO (ppm by weight) (see CV8) is in the oil phase (for a two-phased liquid),
in the water (for a dilute aqueous liquid), or in the liquid (for an
organic liquid waste). The depth (CV9) is the depth of tilling in the land
treatment facility in centimeters. The total porosity and the air porosity
appear in CV10 and CV11, respectively. CV12 contains the average molecular
weight of oil. In cell CV13, the value "1" is entered if the waste liquid
A-14
-------
is a dilute aqueous solution or a "0" is entered for an (Raoult's law)
organic waste or a two-phase (water + organic liquid) waste model. An
intermediate time period over which emissions are to be calculated is
specified in CV14 in units of days. Cell CV15 is an indicator for biolog-
ical activity in the TSDF. When the user selects the land treatment model
to predict emissions from land treatment soil, this flag is automaticallv
set to 1. When the land treatment model is used to predict emissions from
open landfills or wastepiles, biodegradation is assumed not to occur, and
this indicator is automatically set to zero. The user may change this
value if biodegradation is to be considered in the open landfill or waste-
pile. The remaining three parameters have been described previously.
The values of these parameters in Exhibit A-6 estimate relative
emissions from land treatment soil for benzene under the scenario detailed
in Section 5.2.6.1. The estimated emissions for benzene presented in this
section correspond to the CHEMDAT7 predictions in Exhibit A-6. Because an
intermediate time period of 365.25 days was specified in CV14, the interme-
diate time emission fractions reflect annual estimates. The predicted
annual fraction of benzene that will be emitted to the air is calculated as
0/90. This is equivalent to the example calculation result for land treat-
ment facilities in Section 5.2.£.l.g. The long-term emission fraction for
this example is equal to the intermediate result. Annual air emissions of
benzene from the land treatment soil can be estimated from these results as
follows:
Calculation
From Section 5.2.6.1:
Annual throughput: 1.800 Mg
Oil content of waste: 10 percent
Benzene concentration in oil: 2,000 ppm
E » 0.903 x 1,800 Mg x 0.10 x 0.002 * 0.33 Mg/yr
Benzene emission from the land treatment facility is 0.33 Mg/yr.
Short-term emission results for the land treatment facility are also
available from CHEMOAT7. Exhibit A-7 contains the estimated emission- rates
for benzene at 1/4, 1, 4, 12 and 24 hours after application to the land
treatment soi1.
Different tilling frequencies can also be accommodated by using the
CHEMDAT7 land treatment model to predict emissions for each time period
A-15
-------
defined by the tilling frequencies. This approach additionally requires
that initial weight concentration of the VO (CVS) for each time period be
adjusted for the estimated mass of the VO emitted or biodegraded during the
previous time period. For each run of the model, the time period (in days)
over which emissions are to be calculated is placed in CV14.
As indicated previously, the CHEMDAT7 land treatment model can be used
to estimate emissions from open landfills. Exhibit A-13 contains the
CHEMDAT7 land treatment model parameters and emission results that corre-
spond to the example calculation in Section 6.4.3 for predicting the emis-
sion fraction of chloroform from open landfills. Because biodegradation is
presumed not to occur in an open landfill, CV15 has been set to zero.
One year post-application, the fraction of chloroform lost to the
atmosphere from open landfills is 0.053 from both CHEMOAT7 and the example
calculation in Section 6.4.3.d. The following calculation demonstrates how
to convert this emission fraction into an annual emission rate for an open
landfill.
Calculation
From Section 6.4.3:
Landfill depth: 229 cm
Landfill area: 1.42 x 1()8 cm2
Loading: 0.46 g/cm3
Weight fraction of chloroform in oil: 0.5
E « 229 cm x 1.42 x l
-------
the open landfill. Multiply this value by 140, the estimated number of
turnovers per year, to obtain the annual emission rate of chloroform for
the wastepile. The user is reminded that, for open landfills and
wastepiles, the biomass concentration in the land treatment model defaults
to zero (CV15).
A.4.3.4 Disposal Impoundment Model. CHEMDAT7 disposal impoundment
model input parameters are located in C046 through C056 of the spreadsheet.
Please refer to Exhibit A-14. The concentration of VO (C051) is the
initial concentration in the waste expressed in milligrams/liter. The
adsorbing biomass concentration (in C050) is the concentration available to
remove the VO. This and the active biomass concentration (grams/liter in
C049) are set to zero when adsorption and biodegradation are assumed to be
nonexistent in a disposal impoundment. The remaining input parameters were
previously defined.
The parameters specified in Exhibit A-14 reflect the example
calculation for benzene emissions from a disposal impoundment described in
Section. 4.5.3. The CHEMOAT7 emission results also shown in this exhibit
suggest that 14 percent of the benzene in the disposa.1 impoundment will
^ *""
have been emitted to the air in the specified period of 6 months. Section
4.5.3 contains the same fractional result. With two turnovers per year,
the corresponding emission rate is 0.04 Mg/yr.
The CHEMDAT7 disposal impoundment model can also estimate emissions
from a diffused air system. This modification is achieved by defining a
positive (nonzero) submerged air flow (mVsecond) in C055. If the user
wishes to consider biodegradation as a pathway in the diffused air system,
the concentration of active biomass (g/L) should be placed in C049.
A.4.3.5 Closed Landfill Model. Cells CV30 through CV52 in the
CHEMDAT7 spreadsheet contain the closed landfill model input parameters;
see Exhibit A-8. The user must first select the appropriate model. .The
options include the aqueous model (CV50=1) and the (RaouH's law) two-
phased or organic liquid model (CV50*0). Additional CHEMDAT7
specifications are required for each of these models.
With the dilute aqueous model:
• The weight fraction oil (CV41) is set at zero (consequently,
the sum of weight fraction water (CV42) and weight fraction
VO (CV43) must be one).
A-17
-------
MW-liquid (CV52) and rho-liquid (CV51) are used in the esti-
mation procedure. Default values of 18 g/g mol and 1 g/cnP,
respectively, appear in the software as information to the
user. These values cannot be changed.
• mt oil (CV45) does not contribute to the emission predic-
tion.
• Liquid in waste (CV44) is set at 1 g/ca3 in accordance with
dilute .aqueous waste.
With the (Raoult's law) two-phased model:
• The values of HU-liquid (CV52) and rho-liquid (CV51) are
ignored in the emission estimation process.
Finally, the closed landfill model can accommodate biodegradation. The
input parameter CC/GVOC CONY (CV48) is the amount of gas produced in cubic
centimeters per gram of VO biodegraded and is only applicable when
biodegradation is considered a pathway. The user may wish to change the
given representative value of CC/GVOC CONV to investigate other scenarios.
The amount of active biomass (g/cc) is placed in CV49. The emissions then
are augmented by those resulting from this additional pathway. Please note
that the gas conversion input is ignored if biodegratation is presumed not
to occur (i.e., when biomass in CV49 is set to zero). The remaining closed
landfill model parameters are self-explanatory.
The values of the closed landfill model parameters in Exhibit A-8
correspond to those in the example calculations of chloroform emissions
from a closed landfill (see Section 6.2.3). A two-phased model was used,
and both the instantaneous emission rate after 1 year and the average emis-
sion rate in the first year were equivalent (13.0 Mg/yr of chloroform is
the value for both rates from the sample calculations) to the estimated
emission values from the CHEMDAT7 landfill model as shown in Exhibit A-8.
A.4.3.6 Oil Film Model. The CHEMDAT7 oil film model predicts
emissions from an oil film surface impoundment or from an oil layer on land
treatment soil. Input parameters for the oil film model are located in
cells CV69 to CV78; see Exhibit A-15. Depth in meters (CV70) refers to the
oil film thickness measured in meters. Area (CV71) is the surface area of
the facility in square meters. Flow (CV72) is the rate of flow through a
surface impoundment and, consequently, is not applicable when predicting
emissions from an oil layer on land treatment soil. When there is no flow,
A-18
-------
set CV72 to zero and enter the number of months for disposal in CV77. This
latter Input is needed to calculate residence time.
The Input parameters specified in Exhibit A-15 reflect the example
calculation for butanol-1 emissions from an oil layer on the soil surface
of a land treatment site as described in Section 5.2.3.3. This exhibit
shows the equivalent CHEMDAT7 emission result.
A.5 PRINTING A GRAPH
A.5.1 Graph Selection
Two types of graphs are available in CHEW)AT7. The first is a plot of
emission prediction versus the partition coefficient; the second is a plot
of emission prediction versus vapor pressure. Both graphs are available
for all models except the oil film. Select a graph in the active command
mode via
{/} {6} W {U} ,
where G represents graph,- N represents name, and U represents use. (Recall
that you can also position the cursor over the desired option and press
{RETURN}.) Now select the desired graph. GRAPH-PC'is the plot of emission
prediction versus the partition coefficient, and GRAPH-VP is the plot of
emission prediction versus vapor pressure. You can ignore the plot that
appears on your screen. To return to the alternative command menu, press
{ESC}, {QUIT}, {ALT M}.
A.5.2 Storing the Graph File
The sequence {PRINT} {GRAPH} followed by selection of the desired
model will create the plot chosen via Section A.5.1 for the selected model
and simultaneously store it in a specific file on the diskette (see Section
A.3.1). CHEMOAT7 will store only one plot for each model. Furthermore, it
stores the most recently created plot. Consequently, the user must first
create the plot for which he wants a hard copy using the procedures
described in this and the previous section.
A.5.3 The Hard Copy
To obtain a hard copy of a plot, access the active command mode;
follow the commands to exit Lotus 1,2,3; and then access Lotus
{PRINTGRAPH}. Once in this mode, {SELECT} plots for printing using the
space bar to flag (#) the desired plots. The most recently created plot
for closed landfill emissions is Ql; for land treatment emissions, Q2; for
A-19
-------
nonaerated process emissions. Q3; for aerated process emissions, Q4; and
for disposal impoundment emissions, Q5. Press {RETURN} and then {60} to
print the selected plots.
A.6 HELP SCREENS
Exhibits A-16 to A-22 present help screens that are available in
CHEMOAT7. These help screens are provided to assist the user when other
documentation may not be available.
A-20
-------
TSDF (Section)
CHEMDAT7 Model
SURFACE IMPOUNDMENTS AND OPEN TANKS
Storage impoundments (4.2)
Treatment impoundments, mechanically
aerated systems (4.3)
Disposal impoundments (4.4)
Diffused air systems (4.5)
Oil film surfaces (4.6)
Diffused air-activated sludge
LAND TREATMENT
Land treatment soil (5.2)
Waste application (5.2)
Oil film surface (5.2)
LANDFILLS AND WASTEPILES
Closed landfills (6.2)
Fixation pits (6.3)
Open landfills (6.4)
Wastepiles (6.4)
TRANSFER, STORAGE, AND HANDLING OPERATIONS
Container loading (7.2)
Container storage (7.3)
Container cleaning (7.4)
Stationary tank loading (7.5)
Stationary tank storage (7.6)
Spills (7.7)
Fugitive emissions (7.8)
Vacuum truck loading (7.9)
Nonaerated impoundment
Aerated impoundment
Disposal impoundment
Disposal impoundment
Disposal or aerated
impoundments
Oil film
Aerated impoundment
Land treatment
NA
Oil film
Closed landfill
NA
Land treatment
Land treatment
NA
NA
NA
NA
NA
NA
NA
NA
(modified)
(modified)
NA - Not available in CHEMDAT7.
EXHIBIT A-l. RELATIONSHIP BETWEEN HAZARDOUS WASTE TSDF AND CHEMDAT7 MODELS
A-21
-------
Selection
DATA-FORMS
Potions (explanation)
(Go to data-entry fonts)
IMPOUND (Go to the data entry fom MENU for impound-
ments)
NON-AERATED (Go to the data entry form for
flowthrough impoundments)
DIFFUSED (Go to the data entry fom for diffused-
air disposal)
FILM (Go to the data entry form for oil films)
DISPOSAL IMPOUNDMENTS (Go to the data entry form
for disposal impoundments)
QUIT
AERATED (Go to the data entry form for activated sludge
processes)
CLOSED-LANDFILL (Go to the data entry form for capped
landfill)
OPEN-LF/WP (Go to the data entry form for open land-
fill/wastepile)
LAND-TREAT (Go to the data entry form for land applica-
tion of wastes)
CONC (Enter concentration for each compound, permits
different concentration for different compounds)
A IMPORT (Import compound data selected from the Master
Data Set)
DEFAULT (Replace parameters with default parameters)
Municipal (municipal waste aerated impoundment
parameters)
Aerated (default aerated model parameters)
Non-aerated (default non-aerated model parameters)
Landfill (default landfill model parameters)
Land-tr. (default land treatment model parameters)
EXHIBIT A-2. ALTERNATIVE COMMAND MENU
A 31
-------
Open If (set open landfill/wastepile to default)
Disposal (default disposal Impoundment model
parameters)
Quit (Return to data forms menu)
QUIT (Return to main menu)
(Go to a portion of the worksheet)
IMPOUND. (Go to the MENU for impoundments)
NON-AERATED (Go to the calculations for
flowthrough impoundments)
DIFFUSED (Go to the calculations for diffused air
disposal impoundments)
FILM (Go to the calculations for oil film
impoundments)
DISPOSAL (Go to the calculations for disposal
impoundments)
QUIT
AERATED (View results)
OPEN LF/WP (Use land treatment model to simulate an
open landfill)
CLOSED-LF (View calculations for a closed landfill)
LAND-TREAT-(View results)
TOTAL-EMISSIONS (View the results for long-term
integrated emissions)
EMISSION-RATE (View the results for initial
emission rates)
RETURN (Main menu)
GRAPH (VIEW GRAPH)
LANDFILL (View a graph)
LAND-TREAT (View a graph)
EXHIBIT A-2 (continued)
A-23
-------
NON-AERATED (View a graph)
AERATED (View a graph)
DISPOSAL (View a graph)
QUIT
K (View the mass transfer coefficients)
SORT (Rearrange the order of the compound listings)
SELECT (Sort by selected compounds)
DATA (Select additional compounds from an extended data
base)
ALPHABETIC (Sort alphabetically)
BIOLOGICAL (Sort in descending biological rates)
CLASS (Sort by compound type)
V-PRESSURE (SORT BY VAPOR PRESSURE)
QUIT (RETURN TO MENU)
PRINT (Print a portion of the worksheet)
LAND-TREAT (Print land treatment, open landfill, or
wastepile results)
LONG-TERM (Print long-term environmental fate)
SHORT-TERM (Print short-term emission estimates)
DISPOSAL (Print disposal lagoon results)
AERATED (Print aerated process results)
LANDFILL (Print closed landfill results)
NON-AER. (Print non-aerated impoundment results)
K (Print mass transfer coefficients for selected
facilities
DATA (Print data summaries)
BIORATE (Print biorate data)
LAND-TREAT (Print land treatment data)
EXHIBIT A-Z. (continued)
A-24
-------
DATA (Print chemdat information)
GRAPH (Put-a graph file on the disk for later printing)
FILM (Print thin film lagoon results)
SELECT (Select which models are to be used)
HELP (Look at an instructional screen)
GENERAL (View general information)
MODEL (Help in selecting the data form for your unit)
DATA (Discussion of data entry)
LANDTREAT (Information for the use of the land
treatment models)
LANDFILL (Information for the use of the landfill
model)
BIO-RATE (Information for the use of the biological
reaction rates)
IMPOUND (Information for the use of the impoundment
models)
QUIT (Return to the main menu)
QUIT (Return to spreadsheet command mode)
EXHIBIT A-2. (continued)
A-25
-------
NON-AERATED WASTEWATER TREATMENT
WINDSPEED 4.47 •/$
depth 1.8 •
AREA 1500 «2
FLOW 0.00156 «3/s
VO Inlet cone. 100 ng/1
TOTAL ORGANIC; IN 250 «g/i
ACTIVE BIOMASS 0.05 g/1
BIOMASS SOLIDS IN 0 g/1
TEMPERATURE 25 deg. C
* TOTAL AIR EMISSIONS 0.61 Mg/yr
EFFLUENT EMISS. BIOL. PHOTOL.ADSORB, air
COMPOUND NAME & HYDRO. eralss.
.(Mg/yr)
BENZENE 0.031 6.124 0.845 o.ooo o.ooi o.eioi
EXHIBIT A-3. NONAERATED MODEL INPUT PARAMETERS AND EMISSION RESULTS
A-26
-------
AERATED WASTEWATER TREATMENT
WINDSPEED
DEPTH
AREA
FLOW
ACTIVE BIOMASS
BIOMASS SOLIDS
VO INLET CONC.
TOTAL ORGANICS
TOTAL BIORATE
FRACT. AGITATED
SUBMERGED AIR FLOW
Number Impellers
Oxygen trans, rat.
POWR (total)
Power efficiency
Temperature
IN
IN
Impeller speed
4.47
1.8
1500
0.0031
0.25
0
100
250
19
0.16
0
5
3
75
0.83
25
61
126
m/s
m
m2
m3/s
9/1
TOTAL AIR EMISSIONS
mg/1
mg/1
mg/g b1o-hr
m3/s
Ib02/h-hp
HP
deg C
cm
rad/s
7.984 Mg/yr
COMPOUND NAME
BENZENE
RELATIVE AERATED WASTEWATER VOC PATHWAYS
EFFLUENT EMISS. BIOL. PHOTOL.ADSORB, air
& HYDRO. emiss.
, (Mg/yr)
0.002 0.816 0.182 0.000 0.000 7.9842
EXHIBIT A-4. AERATED MODEL INPUT PARAMETERS AND EMISSION RESULTS
FOR MECHANICALLY AERATED TREATMENT IMPOUNDMENT
A-27
-------
AERATED WASTEWATER TREATMENT
WINDSPEED 4.47
DEPTH 4
AREA 27
FLOW 0.0075
ACTIVE BIOMASS 4
BIOMASS SOLIDS IN 0
VO INLET CONC. 100
TOTAL ORGANICS IN 250
TOTAL BIORATE 19
FRACT. AGITATED 0.7
SUBMERGED AIR FLOW 0
Number Impellers 1
Oxygen trans, rat. 3
POWR (total) 7.5
Power efficiency 0.83
Temperature 25
Impeller d1a 61
Impeller speed 126
TOTAL AIR EMISSIONS
ra/s
m
m2
m3/s
g/i
0/1
mg/1
mg/g b1o-hr
ra3/s
Ib02/h-hp
HP
deg C
cm
rad/s
9.257 Mg/yr
COMPOUND NAME
BTRZENT
RELATIVE AERATED WASTEWATER VOC PATHWAYS
EFFLUENT EMISS. BIOL. PHOTOL.ADSORB, air
& HYDRO. eraiss.
(Mg/yr)
6.032 0.391 6.576 0.000 0.001 9.2572
EXHIBIT A-5. AERATED MODEL INPUT PARAMETERS AND EMISSION RESULTS
FOR ACTIVATED SLUDGE UHIT
A-28
-------
LAND TREATMENT MODEL DATA
(land treatment)
L,Loading (g oll/cc soil) 0.036
Concentration 1n oll(ppmw) 2000
1,Depth of tilling (cm) 20
Total porosity 0.61
Air Porosity(01f unknown) 0.5
MW oil 282
For aqueous waste, enter 1 0
Time of calc. (days) 365.25
For blodegradatlon,enter 1 1
Temperature (Deg. C) 25
Wind Speed (m/s) 4.47
Area (m2) 25000
LANDTREATMENT INTERMEDIATE TIME
FRACTION LOST 365.25 days
COMPOUND NAME AIR BIOL. AIR BIOL.
BENZENE 57903 07097 O03 0.097
EXHIBIT A-6. LAND TREATMENT MODEL INPUT PARAMETERS AND EMISSIOH
RESULTS FOR LAND TREATMENT SOIL
A-29
-------
LAND TREATMENT MODEL DATA
(land treatment)
L,Loading (g oll/cc soil) 0.036
Concentration In oll(ppinw) 2000
1,Depth of tilling (cm) 20
Total porosity 0.61
Air Porosity(0 1f unknown) 0.5
MW oil 282
For aqueous waste, enter 1 0
T1«e of calc. (days) 365.25
For blbdegradatlon,enter 1 1
Temperature (Deg. C) 25
Wind Speed (a/s) 4.47
Area (m2) 25000
COMPOUND NAME
LANOTREATMENT EMISSION RATES (g/cm2-s)
TIME (hours)
0.25 1 4 12 48
fENZENE
2.89E-08 1.4OE-08 7.286-09 4.1ot-09 I.y8t-Uy
EXHIBIT A-7. LAKO TREATMEHT MOOa INPUT PARAHETEH
FOR OPEN LANDFILLS AND WASTERILES
A-30
-------
CLOSED LANDFILL
A, LF area (cm2)
1, cap thickness (cm)
ea, cap air porosity
et,cap total porosity
T, avg. LF temp. (C)
efw, waste porosity
Pref, ref. pressure (mbar)
PI, new pressure (mbar)
Tref, reference temp. (C)
Tl, new temp. (C)
dt,bar.pump time int.(s)
weight fraction oil
weight fraction water
weight, fraction vo
W, liquid in waste (g/cm3)
mwt.oil
t, time of calc. (mo.)
D, depth of waste (cm)
CC/GVOC CONV
active biomass (g/cc)
VO diss. in water, enter 1
rho-liquid,density (g/cm3)
MW-liquid, (g/g mol)
1.4170E+08
107
0.08
0.41
15
0.25
1013
1009
15
15
86400
0.2
0.6
0.2
1.16
147
12
457
1750
0.OOE+00
0
1
18
COMPOUND NAME
CHLOROFORM
CLOSED LANDFILL AIR EMISSIONS
EMISSION RATES
EMISS. BIOL. 12 months
.FRACTION FRACTION AVERAGE INST.
1.000 0.000 12.99431 12.98869
EXHIBIT A-8. CLOSED LANDFILL MODEL INPUT PARAMETERS AND EMISSION RESULTS
A-31
-------
Coluan label [explanation]
B COMPOUND NAME
C COMPOUND TYPE
D M.H. [molecular weight]
E DENSITY (g/cc)
F VAP. PRESS. (««Hg) [vapor pressure at 25 *C]
6 H LAN CONST (ata*«3/«ol) [Henry's law constant]
H DIFF. HAT («2/s) [diffusion coefficient in water]
I DIFF. AIR (c«2/s) [diffusion coefficient in air]
•
J BOILING POINT CO
K VAPOR PRESSURE COEFFICIENT - A
L VAPOR PRESSURE COEFFICIENT - B
M VAPOR PRESSURE COEFFICIENT • C
V K(Y/X) [Henry's law coefficient, «ol fraction] (calculated)
U UNO-TREATMENT BIORATE (DAY'l) (calculated)
N LOG OCT/WATER PARTITION
0 BIORATE (mg/g-hr)
EXHIBIT A-9. LIST OF COLUMN LABELS IN CHEMDAT7
A-32
-------
Column
~~ 1MMM
P
Q
X
Y-
Z
AA
AB
AC
AD
AE
AF
AG
AH
AI
AJ
AK
AL
AM
AN
AO
AP
AQ
Column label [explanation]
Kl L/g-hr.
HYOROL (s-1) [rate const, photolysis] + [rate const.
hydrolysis]
ADS. (n3/kg solids) [adsorption]
DISPOSAL IMPOUNDMENT - total fraction removed
OIL FILM - emissions fraction
VP in landfill (im Hg
ln(K)
temp correction factor
Ks (g/cro3 per g/cm3) [partition factor,
VO into sludge]
Biorate (hf-1)
KO (g mol/cn»2-s) * [overall mass transfer
coefficient (ra/s)] x 5.56
Co (Mg/yr)
Temperature correction CS83
biorate (s-1)
KO (g mol/an2-s) » [overall mass transfer
coefficient (m/s)] x 5.56
Co(mg/L)
ADS. COEF. (VAP/SOL) [adsorption coefficient]
BIO RATES (1/tb s-1) [biological rates]
DECAY PARAMETER
EMISS. FRACTION [emission fraction for
specified period]
BIOL. FRACTION [biological removal fraction
for specified period]
AVERAGE EMISSION RATE (Mg/specified time period)
EXHIBIT A-9 (continued)
NON-AERATED
I
AERATED
1
CLOSED
LANDFILL
A-33
-------
Column
AR
AS
AT
AU
AV
AU
AX
AY
AZ
BA
Coluan label [explanation]
BC
BO
BE
BF
BG
BH
BI
BJ
BK
NON-AERATED
INSTANTANEOUS EMISSION RATE (Mg/speclfled
tine period)
EMISS. [long-tern emission fraction] LAND TREATMENT
BIOL. [long-term biological removal fraction]
EMISS. [intermediate emission fraction]
BIOL. [intermediate biological removal fraction]
EFFLUENT [annual effluent fraction]
EMISS. [annual emission fraction]
BIOL. [annual biological removal fraction]
PHOTOL. & HYDROL. [annual fraction removed via
photolysis and hydrolysis]
ADSORB, [annual fraction removed via
adsorption]
AIR EMISSIONS (Mg/year)
EFFLUENT [annual effluent fraction]
EMISS. [annual emission fraction]
BIOL. [annual biological removal fraction]
PHOTOL. & HYDROL. [annual fraction removed via
photolysis and hydrolysis]
ADSORB, [annual fraction removed via adsorption]
AIR EMISSIONS (Mg/year)
AQUEOUS [fraction remaining]
EMISS. [emission fraction for specified period]
BIOL. [biological removal fraction for specified
period]
EXHIBIT A-9 (continued)
AERATED
DISPOSAL
IMPOUNDMENT
A-34
-------
€oUmnv Column label [explanation]
BL PHOTOL. & HYDROL. [fraction removed via photolysis
and hydrolysis for specified period]
BM ADSORB [fraction removed via adsorption for
specified period]
BN Air emission (Mg/yr)
BO VP aqueous landfill
BP K1 aerated
BQ KO (9 mol/cm2-s) » [overall mass transfer
coefficient (m/s)l x 5.56
BR SUM RATES DISPOSAL [sum of the various
rate processes] DISPOSAL
IMPOUNDMENT
BS lambda
BT CONC (g/on3)
BU K' non-aerated
EXHIBIT A-9 (continued)
A-35
-------
CONFOUND NAME
ACETMJKMYDE
ACETONE
ACROLEIN
ACRTLONITRILE
ALLYL ALCOHOL
BENZENE
BENZYL CHLORIDE
BUTANOL-1 .
CARBON OISULFIOE
CARBON TETRACHLORIOE
CHLOROFORM
CHLOROPRENE
CRESOL(-)
CRESOL(-O)
CRESOL(-P)
CRESYUC ACID
CUNENE (IsoprapylbMZvw)
ftTfHI JHHHffH/FHff 1 71 f-ff)
11 1 1 II OOlBf Hf f fir [1*11 f-n)
OICHLOROETHANE(1.2)
OICHLOROFROPANE(1.2)
OWETHYL MTROSANINE
OIOXIN
ENCNLOROMVORIN
ETHYLACETATE
ETHVLBENZENE
ETHYLENEOXIOE
ETHYLETHER
FORNALDEHVOE
FREONS
HEXACHLOR06UTAOIENE
HHHB
HHLB
HHHB
IS08UTVL ALCOHOL
LHHB
LHNB
MALEIC ASHYSRIOE
NETHANOL
METHYL ACETATE
METHYL ETHYL KETONE
METHYL IS08UTYL KETONE
METHVLENE CHLORIDE
NETHYLNAPHTHALENE (1)
HHHB
MHLB
MHNB
fUPHfHALEKE
NITROBENZENE
NITROSOMORFHOLINE
PHENOL
PHTHALIC ANHYDRIDE
POLYCHLORINATED BIFHENVLS
PROPYLENE OXIDE
PYRIDINE
TETRACHLOROETHANEd . 1 .2.2
Tt?TfiA/*UI rtflftamil VBTHC
1 C 1 KMUV^UNIJIl InLTCnt
TOLUENE
TRICHLOROETHANE(l.l.l)
TRICHLOROETHANE(1.1.2)
TRICHLOROETHYLENE
TRICHLOROFLUORONETHANE
TRICHLOROd. 1.2)TRIFLUORO
VINYL CHLORIDE
VINVLIOENE CHLORIDE
XYLENE(-o)
DENSITY
N.W. (o/CC)
44.0
58.0
56.1
53.1
58.1
78.1
126.6
74.1
76.1
153.8
112.6
119.4
88.5
108.1
10B.1
108(1
108.0
120.2
98.2
147.0
147.0
99.0
113.0
74.08
322.0
92.5
88.1
106.2
44.0
74.1
30.0
120.92
260.8
mm
•5
98.4
144.0
117
74.0
97.3
78.4
98.1
32.0
74.1
72.1
100.2
85.0
142.2
69.9
112.0
57
128.2
123.1
116.14
94.1
148.1
290
58.1
79.1
168.0
1 AC tt^
lea. 83
92.4
133.4
133.4
131.4
137.4
187.38
62.5
97.0
106.2
0.79
0.79
0.84
0.81
0.8S
0.87
1.10
0.81
1.26
1.59
1.11
1.49
0.958
1.03
1.05
1.03
1.05
0.88
0.95
U29
1.26
1.16
1.005
1.83
1.18
0.90
0.87
0.87
0.71
0.08
1.486
1.67
1»«
• /I
0.79
0.93
0.79
0.92
0.82
0.80
1.34
"1.14
1.20
1.07
1.33
1.45
0.83
0.98
1.59
ICVd
.624
0.87
1.33
1.3
1.40
1.49
1.58
0.91
1.21
0.88
VAP.PRESSH
(-HQ) (I
266
244.2
114
23.3
95.2
1.21
6.5
366
113
11.8
208
273
0.08
0.24
0.11
0.3
4.6
4.8
1.5
1.2
80
40
1.52E-09
17
100
10
1250
520
3500
5000
0.15
OMf
• UN
1
1
1
10
1
1
0.0001
114
235
100
15.7
438
1
1
1
oTB
0.3
0.341
0.0015
524.S
20
6.5
19
30
123
25
75
796
300
2660
591
7
LAW CONST. OIFF. WAT
iteHti/wI) (O£/SK)
9.50C-O9 1.41C-09
2.SOE-OS 1.14E-OS
S.66E-OS 1.22E-OS
8.80E-OS 1.34E-05
1.80E-03 1.14E-OS
S.50E-03 9.80E-06
4.33E-04 7.80E-08
8.90C-06 9.30E-06
1.68E-02 l.OOE-05
3.00E-02 8.80E-06
3.93E-03 8.70E-06
3.39E-03 l.OOE-OS
0.331 l.OOE-OS
4.43E-07 l.OOE-OS
2.60E-06 8.30E-06
4.43E-07 l.OOE-OS
1.70E-08 8.30E-08
1.46E-02 7.10E-06
4.13E-08 8.62E-08
1.94E-O3 7.90E-06
1.60E-03 7.90E-06
1.20E-03 9.90E-06
2.30E-03 8.73E-06
l.OOE-OS
8.12E-OS S.60E-06
3.23E-05 9.80E-06
1.28E-04 9.66E-06
6.44E-03 7.80E-06
1.42E-04 1.45E-OS
6.8OE-04 9.30E-06
S.76E-OS 1.98E-OS
4.01E-01 l.OOE-OS
2.56E-02 6.16E-06
IJEJtf _4M C 1 ** —*
.OOt-vc tt.lK-vO
5.38E-03 9.40E-06
3.00E-02 9.39E-06
1.18E-03 8.24E-06
2.20E-06 9.30E-06
1.58E-07 9.64E-06
1.58E-07 1.13E-OS
4.00E-08 1. HE-OS
2.70E-06 1.64E-OS
1.02E-04 l.OOE-05
4.35E-OS 9.80E-06
4.95E-OS 7.80E-06
3.19E-03 1.17E-05
6.BOE-OS 1.16E-05
2.22E-OS 8.ME-06
4.08E-OS 1.18E-OS
1.18E-03 7.50E-06
1.31E-OS 8.ME-06
l.OOE-OS
4.S4E-07 9.10E-06
9.00E-07 8.60E-06
2.94E-04 l.OOE-OS
1.34E-03 l.OOE-05
2.36E-OS 7.60E-06
3.SOE-04 7.90E-06
2.90E-02 8.20E-06
6.68E-03 8.60E-06
1.72E-02 8.30E-06
7.40E-04 8.80E-06
9.10E-03 9.10E-06
5.83E-02 9.70E-06
4.3SE-01 S.20E-06
8.60E-02 I.23E-OS
1.50E-02 1.04E-OS
S.27E-03 l.OOE-OS
EXHIBIT A-10. ALPHABETICAL LIST OF COMPOUNDS IN CHEMOAT7
A-36
-------
COMPOUND NAME
BOILING VAPOR PRESSURE COEFFICIENTS
OIFF. AIR POINT ~ —-—-
(ott/SK) (dtg.C) ABC
ACETALDEHYOE 1.Z4E-O1
ACETONE 1.246-01
ACROLEIN l.OSE-01
ACRYLONITRILE 1.22E-01
ALLYL ALCOHOL 1.146-01
BENZENE 8.806-02
BENZYL CHLORIDE 7.50E-02
BUTANOL-1 8.006-02
CARBON OISULFIDE 1.04E-01
CARBON TETRACHLORIOE 7.806-02
CHLOROB6NZENE 7.306-02
CHLOROFORM . 1.04E-01
CHLOROPRENE 1.04E-01
CRESOL(Hi) 7.406-02
CRESOL(-o) 7.406-02
CRESOL(-p) 7.406-02
CRESYLIC ACID 7.406-02
CUM6N6 (lMpropylbmztM)6.50E-02-
nfCLOHBWMHE 7 84E-W
OICHLOROBENZENEU.2) (-0)6.906-02
OICHLOROBENZENE(1.4) (-p)6.90E-02
OICHLOROETHANE(1.2) 1.04E-01
OICHLOftOPROPANE{1.2) 7.82E-02
DIMETHYL NITROSAHINE 1.046-01
OIOXIN 1.046-01
EPICHLOROHYORIN 8.606-02
ETHYLACETATE 7.32E-02
ETHYLBENZENE 7.50E-02
ETHVLENEOXIDE 1.04E-01
ETHYLETHER 7.406-02
FORMALDEHYDE 1.78E-01
FREONS 1.04E-01
HEXACHLOROBUTAOIENE 5.61E-02
HEXACMLOROCYaOPENTAOIENE5.61E-02
MM 8.73E-02
HHLB 8.76E-02
HHMB 7.40E-02
ISOBUTYL ALCOHOL 8.60E-02
LHHB S.27E-02
LHNB l.OOE-01
MALEIC ANHYDRIDE 9.SOE-02
H6THANOL 1.50E-01
METHYL ACETATE 1.04E-01
METHYL ETHYL KETONE 8.08E-02
METHYL ISOBUTYL KETONE 7.50E-02
HETHYLENE CHLORIDE 1.01E-01
METHYLNAPHTHALENE (1)
MHHB 9.50E-02
MHLB 7.606-02
MHM 1-lSe-Ol
NAPHTHALENE 5.90E-02
NITROBENZENE 7.60E-02
NITROSOMORPHOLINE 5.90E-02
PHENOL 8.20E-02
PHTHALIC ANHYDRIDE 7.10E-02
POLYCHLORINATEO BIPHENYLS1.04E-01
PROPYLENE OXIDE 1.04E-01
PYRIOINE 9.10E-02
TETRACHLOROETHANEU. 1.2.27. 10E-02
TETRACHLOROETHLYENE 7.20E-02
TOLUENE 8.70E-02
TRICHLOROETHANE(l.l.l) 7.80E-02
TRICHLOROETHANEU.1.2) 7.80E-02
TRICHLOROETHVLENE 7.90E-02
TRICHLOROFLUOROMETHANE 8.70E-02
TRICHLORO( 1 , 1 .2) TRIFLUOR07 .80E-02
VINYL CHLORIDE 1.06E-01
VINYLIOENE CHLORIDE 9.00E-02
XYLENE(-O) 8.70E-02
20.8 8.005 1600.017
56.2 7.117 1210.595
53.0 2.387833
77.4 7.038 1232.53
97.0 1.367408
80.1 6.905 1211.033
179.4 0.082788
117.7 7.476 1362.39
46.3 6.942 1169.11
76.8 6.9339 1242.43
132.0 6.978 1431.05
61.2 4.493 929.44
59.4 6.161 783.45
202.0 7.508 1856.36
190.8 6.911 1435.5
203.0 7.03S 1511.08
235.0 -0.52289
153.0 6.963 1460.793
157.0 0.681266
179.0 0.176097
173.4 0.079184
83.4 7.025 1272.3
96.8 6.98 1380.1
153 ERR
421.4 12.88 6465.5
117.0 1.230494
77.0 7.101 1244.95
136.2 6.975 1424.255
10.7 7.128 1054.54
34.5 6.92 1064.07
-14.0 7.195 970.6
-29.8 3.699108
215.0 -0.82393
234.0 -1.09155
107.9 1.000036
200.0 -4.00014
65.0 7.897 1474.08
54.0 7.065 1157.63
79.6 6.97421 1209.6
115.8 6.672 1168.4
39.8 7.409 1325.9
218.0 7.01 1733.71
210.8 7.115 1746.6
225 ERR
182.0 7.133 1516.79
284.0 8.022 2868.S
ERR
33.9 2.719846
115.S 7.041 1373.8
146.2 .631 1228.1
121.4 .976 1386.92
110.7 .954 1344.8
75.0 .643 2136.6
74 .951 1314.4
87.0 .518 1018.6
23.8 .884 1043.004
48.0 6.88 1099.9
-13.9 3.4 5008
31.9 .972 1099.4
144.4 .998 1474.679
291.809
229.664
222.47
220.79
178.77
241.59
230
217.55
196.03
179.7
199.07
165.16
161.85
207.78
222.9
22.8
273
217.88
213.21
237.76
228.8
244.1
229.13
219.73
216
191.9
252.6
201.86
201.8
174.95
0
214.98
179.9
217.53
219.48
302.8
209
192.7
236.88
227.5
237.2
213.69
EXHIBIT A-10 (continued)
A-37
-------
COMPOUND MME
BIOMTC Kl
•gVO/**r L/g-hr.
ACETONE
AOOLEIN
AcmoNrraiLE
ALLYLALCOHOL
BENZYL CHLORIDE
BUTANOL-1
CARBON OBULFIOE
CARBON TETRACHLOMOE
oiCHLOWETHAMEU.j)'
OICMUMOP«MNE(1.2)
OOC1MTL NITMSMaiK
OIOUM
ETHYLACETATE
ETHVLBEN2ENE
CTNYLENCOUDC
ETHVLETNER
nMMLOOIVOE
FREONS
HEXACHLOROBUTAOIENE
HEXACHLONOCTaOPOITAOIEME
HHLB
HMNB
ISOBUTTL M.CONOL
LHH8
UM
MALEIC
NtnMML
Ntnm.
HCTMTU ETHYL KETOMC
HETHTt ISOBUTTL KETONC
NETHTLEK CHUWIOE
NETMTUMmTHALEK (1)
HHHB
mu
iSffll
m TmoizpiC
NITROSOMORmOLINE
PHTHALIC ANHYDRIDE
POLTCNLORIMTED BIMENVLS
OXIDE
TETRACHLOROETHANE(1.1.2.2
TETRACMUMOETHLYENE
TOLUENE
TRICHLOROETHANE(l.l.l)
TRICHLOROETHANEd. 1.2)
TRICHLOROETHYLENE
TRICHLOROFUJONOMETMANE
TRICHLORO(1.1.2)TRIFUIONO
VINYL CHLORIDE
VINYUOENE CHLORIDE
XYLEME(-o)
R.4Z 0.190681
1.30 1.19
7.80 0.34
18.00 0.79
0.00 0.196699
19.00 1.4
17.79 1.010399
0.11
1
1.9
10
0.79
I
17
17
17
1
1
7.80
0.00
1.90
0.39
2.94
0.00
23.21
2Z.7t
23.21
19.00
0.00
11.49 0.274276
2.90 0.98
6.40 2.3
2.10 0.98
17.00 1.4
0.00 0.081977
0.00 1
0.00 0.138991
17.98 1
6.80 2.1
4.20 0.91
0.77 0.279119
9.00 0.29
0.00 1
0.00 3.962786
0.00 4.434119
1
1
1
0.11
1
1
1
0.2
.124779
0.2
29.2
0.30
2.71
7.80
23.2
3.99
4.08
18.00
19.87 0.
2.00
0.74 0.491677
22.00 0.403106
24.03 1
40.1 1
0.39 1
4&* 1
11.00 2.3
0.00 0.091899
97.00 13
0.00 0.078467
0.00 1
0.00 1
39.03 0.238439
6.20
6.20
73.48
3.90
3.90
3.90
0.00
0.00
0.68
0.68
2.4
0.74
0.74
0.88
.239682
1
0.00 0.142278
0.00 1
40.79 1.784978
EXHIBIT A-10 (continued)
A-3&
-------
NON-AERATED WASTEWATER TREATMENT
WINOSPEEO 4.47 m/s
depth 1.8 a
AREA 1500 ra2
FLOW 0.00156 m3/s
VO Inlet cone. 100 rag/I
TOTAL ORGANICS IN 250 mg/1
ACTIVE BIOMASS 0 g/1
BIOMASS SOLIDS IN 0 g/1
TEMPERATURE 25 deg. C
TOTAL AIR EMISSIONS 3.94 Mg/yr
EFFLUENT EMISS. BIOL. PHOTOL.ADSORB, air
COMPOUND NAME & HYDRO. emiss.
(Mg/yr)
BENZENE"0.1990.8010.0000.0000.000 3.9432"
EXHIBIT A-ll. NONAERATED MODEL INPUT PARAMETERS AND EMISSION RESULTS
A-39
-------
AERATED WASTEWATER TREATMENT
WINDSPEED 4.47 n/s
DEPTH .4 •
AREA 27 «2
FLOW 0.0075 m3/s
ACTIVE BIOMASS 4 g/1
BIOMASS SOLIDS IN 0 g/1
VO INLET CONC. 100 rog/1
TOTAL ORGANICS IN 250 mg/1
TOTAL BIORATE 19 mg/g blo-hr
FRACT. AGITATED 0.7
SUBMERGED AIR FLOW 0.04 m3/s
Number Impellers 1
Oxygen trans, rat. 3 Ib02/h-hp
POWR (total) 7.5 HP
Power efficiency 0.83
Temperature 25 deg C
Impeller dla 61 cm
Impeller speed 126 rad/s
TOTAL AIR EMISSIONS 9.745 Mg/yr
RELATIVE AERATED WASTEWATER VOC PATHWAYS
EFFLUENT EMISS. BIOL. PHOTOL.ADSORB, air
COMPOUND NAME & HYDRO. • emlss.
„ ,c c (Mg/yr)
BENZENE0.030 0.412 0.557 0.000 0.001 9.7448
EXHIBIT A-12. AERATED MODEL INPUT PARAMETERS AND EMISSION
RESULTS FOR DIFFUSED AIR-ACTIVATED SLUDGE UNIT
A-40
-------
LAND TREATMENT MODEL DATA
(open landfill, waste pile)
L,Loading (g o1l/cc soil) 6.464
Concentration 1n oil(ppmw) 500000
1,Depth of tilling (on) 229
Total porosity 0.5
A1r Porosity(0 1f unknown) 0.25
MW oil 147
For aqueous waste, enter 1 0
Time of calc. (days) 365.25
For blodegradatlon,enter 1 0
Temperature (Deg. C) 25
Wind Speed (m/s) 4.47
Area (m2) 14200
LANDTREATMENT INTERMEDIATE TIME
FRACTION LOST 365.25 days
COMPOUND NAME AIR BIOL. AIR BIOL.
CHLOROFORM TTooo oTooo OBIo.ooo
EXHIBIT A-13. LAND TREATMENT MODEL INPUT PARAMETERS AND EMISSION
RESULTS FOR OPEN LANDFILLS AND HASTEPILES
A-41
-------
DISPOSAL IMPOUNDMENT
(no outlet flow)
WINDSPEED 4.47 m/s
DEPTH 1.8 m
AREA 9000 m2
ACTIVE BIOMASS 0.05 g/1
BIOMASS SOLIDS IN 0 m3/s
VO INLET CONC. 100 mg/1
TOTAL ORGAN ICS IN 250 mg/1
TOTAL BIORATE 19 rag/g blo-hr
TIME FOR EMISSIONS 6 months
SUBMERGED AIR FLOW 0 m3/s
TEMPERATURE 25 deg. C
•
TOTAL AIR EMISSIONS 1.77E+00 Mg/yr
DISPOSAL IMPOUND. VO EMISSIONS 6 MONTHS
AQUEOUS EMISS. BIOL. PHOTOL.ADSORB, air
COMPOUND NAME * HYDRO. emiss.
(Mg/yr)
b.666 6.546 0.454 6.000 6.dob i.759547
EXHIBIT A-14. DISPOSAL IMPOUNDMENT INPUT PARAMETERS AND EMISSION RESULTS
A-42
-------
OIL FILM SURFACE
WINDSPEED (ra/s) 4.47
depth (m) 0.072
AREA (m2) 25000
FLOW (m3/s) 0
VO cone. 1n oil (mg/1) 200
oil (fraction of'waste) 1
molecular weight of oil 282
density of oil (g/cc) 1
Enter months for disposal 0.033
Temperature (deg C) 25
OIL FILM
COMPOUND NAME emissions
fraction
BUTANOt-10.494451
EXHIBIT A-15. OIL FILM SURFACE MODEL INPUT PARAMETERS AND
EMISSION RESULTS
A-43
-------
Y99: 'GENERAL HELP : IF THESE HELP SCREENS 00 NOT ANSWER YOUR QUEST MENU
GENERAL MODEL DATA LANOTREAT LANDFILL BIO-RATE IMPOUND QUIT
View general Information
Y Z AA AB AC AD AE
99 GENERAL HELP : IF THESE HELP SCREENS DO NOT ANSWER YOUR QUESTIONS,
100 ' CONSULT THE MANUAL.
101 for ...
102
103 all compounds sane concent. enter concentration on data fora.
104 compounds different concent, enter values using CONC on DATA-FORM
105 select compounds use the SORT menu option
106 which data fora to use? press N for model help screen
107 General use of spreadsheet Consult your Lotus 123 manual.
108
109 New compounds can be added by copying other data over the existing
110 compound data base. [b!4..wS2] See data help screen.
Ill Go to a cell for data entry by pressing [F5J and typing cell label
112 New compounds can be added by selecting [IMPORT]. See DATA-FORM menu.
113 New compounds can be added by selecting [SORT][DATA].
114 hold {ALT} down and press {0} to restore borders for screens.
115
116 calculate: press {F9} before you use the predictions!
117 last use:
118 20-Mar-89 VER 7.06 3-20-89 cca
20-Mar-89 02:51 PM CMC CALC
EXHIBIT A-16. GENERAL HELP SCREEN
A-44
-------
Y139: Model Assistance Help Screen MENU
GENERAL MODEL DATA LANDTREAT LANDFILL BIO-RATE IMPOUND QUIT
Help In selecting the data form for your unit
Y Z AA AB AC AQ AE
139 Model Assistance Help Screen
140
141 to model..
142
143 Treatment Tanks Use the non-aerated model.
144 Lagoon sludges Use the 1andtreatment model.
145 Open waste piles Use the 1andtreatment model.
146 For municipal waste treatment assume that approximately IX of flow
147 Is Inlet solids by volume.
148
149 Disposal for non flow system you specify time for calc.
150 Non-aerated and aerated Impoundments are flow through systems at
151 steady state conditions.
152 Plug flow lagoon Use disposal model: first order kinetics are
153 assumed for a batch. The equations for plug flow
154 can be simulated by non-flow, since there 1s
155 no backmlxing. Emission time 1s residence time.
156 Plug flow aeration Not supported since strong baekmixing Is assumed.
157 You can run multiple times for series reactors.
153 Use cone, out for Inlet cone, for next reactor.
20-Mar-89 02:52 PM CMO CALC
EXHIBIT A-17. MODEL ASSISTANCE HELP SCREEN
-------
AV138: 'DATA ENTRY CONCERNS: CONSULT MANUAL FOR ADDITIONAL DETAILS MENU
GENERAL MODEL DATA LANDTREAT LANDFILL BIO-RATE IMPOUND QUIT
V1«w general Information
AV AW AX AY AZ BA BB BC BD
138 DATA ENTRY CONCERNS: CONSULT MANUAL FOR ADDITIONAL DETAILS
139
140 New compounds can be added by copying other data over the existing
141 compound data base. [b!4..w82]
142 Sometimes data omissions cause errors. The following are needed:
143
144 COLUMNS OF DATA REQUIRED
145 Landfills E F 6 I K L M 0
146 landtreatnent F G IKLMNO
147 non aerated GHIKLMNOP
148 aerated r GHIKLMNOP
149 disposal GHIKLMNOP
150 oil f11« F I K L M
151 impoundment G H I OP
152
153 Where no data 1s available and the pathway 1s not evaluated, non-zero
154 values can be Inserted and the effect of the value 1s canceled In data
155 entry. Example: put any positive value for 0 and P and set b1omass-0.
156
157 column [C]: A Is aromatic,N nitrogen,0 oxygenate,C chlorinated,P phenol
20-Mar-89 02:53 PM CMD CALC
EXHIBIT A-18. DATA ENTRY HELP SCREEN
A-46
-------
AI118: 'LANDTREATMENT HELP SCREEN: CONSULT MANUAL FOR ADDITIONAL DETAILS. MENU
GENERAL MODEL DATA LANDTREAT LANDFILL BIO-RATE IMPOUND QUIT
View general Information
AI AJ AK AL AM AN AO
118 LANDTREATMENT HELP SCREEN: CONSULT MANUAL FOR ADDITIONAL DETAILS.
119
120 The model used simulates the diffusion of a gas out of a porous
121 solid. The rate of diffusion depends on the concentration 1n the
122 gas phase. The concentration 1n the gas phase Is reduced by
123 absorption 1n the oil which 1s mixed with the porous solid. The
124 wind resistance to diffusion 1s assumed to be unimportant for the
125 long term emissions.
126
127
128
129
130
131
132
133
134
135
136
137
20-Mar-39 02:53 PM CMD CALC
EXHIBIT A-19. LAND TREATMENT HELP SCREEN
A-47
-------
AP118: 'LANDFILL HELP SCREEN: CONSULT MANUAL FOR ADDITIONAL DETAILS MENU
GENERAL MODEL DATA LANOTREAT LANDFILL BIO-RATE IMPOUND QUIT
View general Information
AP AQ AR AS AT AU AV
118 LANDFILL HELP SCREEN: CONSULT MANUAL FOR ADDITIONAL DETAILS
119
120 The landfill model Is based upon air emissions fron these
121 pathways:
122
123 diffusion through cap
124 gas generation by blodegradatlon
125 barometric pumping
126
127 Open landfills can be modeled using the equations
128 In the landtreatment model.
129
130
131
132
133
134
135
136
137
20-Mar-89 02:54 PM CMD CALC
EXHIBIT A-20. LANDFILL HELP SCREEN
/V-48
-------
AW118: 'BIORATE HELP SCREEN FOR AQUEOUS SYSTEMS MENU
GENERAL MODEL DATA LANDTREAT LANDFILL BIO-RATE IMPOUND QUIT
View general Information
AW AX AY AZ BA BB BC BD BE
118 BIORATE HELP SCREEN FOR AQUEOUS SYSTEMS
119
120 The rate of b1oox1dat1on 1s estimated by the Monod kinetic model.
121 The value of Kl 1s specified 1n column V14...V84.
122 If Kl 1s not estimated empirically, there 1s an equation
123 based upon the octanol- water partition coefficient.
124 If the octanol-water partition coefficient 1s absent, Kl 1s
125 assumed to equal 1.
126 The value of Kv 1s determined empirically and 1s Identical to the
127 previously used blorate constants 1n CHEMOAT6.
128
129 , [blomass] [ Kv ] [concentration]
130 RATE «
131 [concentration] + [ Kv/Kl ]
132
133 The effect of temperature on the rate 1s as follows:
134 ( 1.046 )~(deg C-25deg)
135
136
20-Mar-89 02:55 PM CMD CALC
EXHIBIT A-21. BIORATE HELP SCREEN
A-49
-------
BF118: 'IMPOUNDMENTS HELP SCREEN: FOR ADDITIONAL DETAILS CONSULT THE MENU
GENERAL MODEL DATA LANOTREAT LANDFILL BIO-RATE IMPOUND QUIT
View general Information
BF BG BH BI BJ BK BL BM BN
118 IMPOUNDMENTS HELP SCREEN: FOR ADDITIONAL DETAILS CONSULT THE MANUAL.
119 The flow 1n the aerated and non-aerated Impoundments 1s assumed to be
120 well mixed. Plug flow can reduce effluent concentrations, but may not
121 significantly change fraction emitted.
122 If only the exit concentration Is known, the Inlet may be estimated
123 by trial and error.
124 Aerated processes can describe Impoundments as well as processes.
125 -
126 Both surface aeration and submerged aeration are available at the same
127 time. ^Submerged aeration 1s expected to agitate the surface 1n a
128 similar manner to surface aeration. If there Is no agitated surface
129 available for air transport, the fract. agitated can be set to 0.
130
131 The component properties of mass transfer are based on the ratio to
132 a reference compound. The following exponents are used with the ratios.
133
134 nonaerated • aerated
135 D temp, dlf.coef. dlf.coef.
136 liquid 1 0.67 0.5
137 gas 1.75 0.67 0.5
20-Mar-89 02:55 PM CMC CALC
EXHIBIT A-22. IMPOUNDMENTS HELP
A-5C
-------
tn
^
.1
A
14
74
B
8
M
P
O
N
D
S
66
127
C
14
74
»
«
iAA
C
GO*
X Y A»
99 "
H
P
MACROS 5
C
R
E
E
N
\2l
AZ.BA
LABELS
^POUND-SPECIFIC CHARACTERISTI
AND EMISSION RESULTS
i
BH
14S"ITkCI GDC/^ICIS^ I4CI D C/^dCCMC
MUUbL-brtlylrK> ntLr ol»ntkNo
BU
BU
CS
b
11
14
74
BZ.CA
BZ CL
MACROS
BV CJ
COMPOUND-
SPECIFIC
EMISSION
RESULTS
fi
Ifi
57
76
H8
CM CP
NON- AERATED
IP*
\
DISPOSAL
IMPOUNDMENT
IP'
>
CM CP
AERATED
IP*
C
1
fi
If
t/
52
\U
78
14
17
CU CV
LAND-
TREATMENT
IP*
CLOSED
LANDFILL
IP*
CU CV
OH.
FILM
IP*
CS CZ
DEFAULT
VALUES
CZ
\
- Input paiameleis.
ATTACHMENT A. GENERALIZED LAYOUT OF CHEMOAT7 SPREADSHEET
-------
APPENDIX B
A GUIDE THROUGH THE LITERATURE
-------
APPENDIX B
A GUIDE THROUGH THE LITERATURE
B.I INTRODUCTION
There 1s concern that volatilization of organic compounds (VO) from
hazardous waste treatment, storage, and disposal facilities (TSDF) poses a
public health problem. These VO emissions may adversely affect ambient air
quality 1n or around a hazardous waste TSDF. However, there are other
competing mechanisms or pathways through which VO can leave hazardous waste
facilities. These Include adsorption, migration, runoff, biological or
chemical degradation, hydrolysis, oxidation, and hydroxyl radical reaction.
Consequently, the potential hazard of volatilization should be assessed in
relation to the potential Importance of these other pathways.
Hazardous waste TSDF Include, among others, surface Impoundments,
landfarms, landfills, and wastewater treatment plants. The Important
competing pathways for each hazardous waste site have been Identified in
earlier sections of this report. This evaluation has been based on field
and laboratory measurements as well as predictive or mathematical models of
these pathways. This appendix supplements the body of the report; it
serves to direct the reader through the literature concerning VO pathways
from hazardous waste TSDF.
For the convenience of the reader, a comprehensive source list is
presented 1n Appendix C of this report. In addition to the references
cited in this appendix and in the individual sections of the report, this
bibliography lists other literature of interest for VO pathways and TSDF
emission models.
B.2 SURFACE IMPOUNDMENTS
B.2.1 Volatilization
Direct measurement of volatilization rates from surface impoundments
is extremely complicated. Hwang and Thibodeaux* reviewed the concentration
B-3
-------
profile and plume mapping technique and proposed a new method requiring
fewer concentration measurements. This latter technique has yet to gain
popularity. Thibodeaux et al.2 used the concentration profile technique to
measure the rate at which selected VO were emitted to the air from basins
in the pulp and paper industry. The ranges of the average flux for meth-
anol and acetone were 1.4 to 3.8 ng/cm2** and 0.028 to 0.10 ng/cni2»s,
respectively, which were higher than background. Radian3 obtained emission
rates from four different hazardous waste sites containing surface impound-
ments as well as landfills and landfarms. They used the concentration
profile, transect, materials balance, and vent sampling approaches. These
methods4*5 are also applicable to other nonpolnt source hazardous waste
facilities such as landfills and landfarms.
Volatilization rates from surface impoundments can be estimated via
mathematical models. Mackay and Leinonen6 predicted air emissions from
nonaerated surface impoundments receiving influent irregularly (unsteady
state). The liquid and gas mass transfer coefficients in this model were
modified by Mackay and Yeun.7 Thibodeaux et al.8 developed predictive
models for both aerated and nonaerated steady-state surface impoundments.
DeWolf and Wetherold9 critiqued these models and presented a protocol for
their proper use. Shen*0 modified the nonaerated model of Thibodeaux et
al.11 In an extensive review of these and other predictive models, GCA*2
judged the theoretical work of Thibodeaux et al.13 and Mackay and
Leinonen14 as most suitable for predicting air emissions from surface
Impoundments.
The use of these mathematical models for predicting volatile emissions
is less expensive and faster than actual field measurements. However, to
be cost-effective, these mathematical models must provide accurate esti-
mates of volatilization rates. It is disappointing to note that relatively
few validation studies are reported 1n the literature. A description of
these follows.
Hwang*5 compared predicted and measured emission rates of toluene and
1,1,1-trichloroethane from an evaporation pond. The transect method was
used for field measurements, and the models summarized in Hwang*6 provided
the predicted rates. For each organic compound, the predicted result was
B-4
-------
within the confidence Units of the average measured result. Balfour
et al..I? used the Thlbodeaux et al.18 model to predict emissions from these
surface Impoundments. Emission rates were measured via the flux chamber in
all three ponds as well as via the concentration profile In the third pond.
In this latter pond, the emission rate of most compounds as determined
using the flux chamber was statistically greater than that determined using
the concentration profile. Furthermore, results of a comparison of .
measured versus predicted emission rates were Inconclusive. Vaught19 used
the Springer et al.20 and Mackay and Yeun^l approaches to predict air
emissions from quiescent surface Impoundments. He concluded that volatil-
ization rates predicted from the Mackay and Yeun model were comparable to
rates measured via the flux chamber. In contrast, the predicted rates from
Springer et al. exceeded the measured rates.
B.2.2 Other Pathways
.The role of other pathways In the removal of VO from surface Impound-
ments has not been addressed extensively In the literature. However,
biological removal mechanisms associated with stabilization ponds and
lagoons?2 wm be applicable where conditions of pH, temperature, and
nutrient levels are suitable for biological growth.
B.3 LAND TREATMENT
For the past 25 years, the petroleum Industry has operated land
treatment, sludge farming, and land disposal facilities. The pharma-
ceutical and organic chemical manufacturers recently have considered this
method of hazardous waste disposal because of Its comparatively reasonable
cost, simplicity, and use of natural processes. How does a land treatment
effectively and safely treat and dispose of VO? The purpose of land
treatment 1s to exploit the microbiological actions of the upper soil layer
to degrade toxic organic material at a controlled rate. Although photo-
degradation takes place In a land treatment facility,23 the short time that
the materials are exposed to direct sunlight and the screening effect of
the oil in which hazardous materials are suspended make this pathway
negligible. Several other pathways may exist: volatilization, runoff, and
leaching.24,25 However, these latter mechanisms do not occur at a properly
sited, operated, and maintained RCRA-permltted land treatment facility.
8-5
-------
B.3.1 Degradation
The chemical structure of the hazardous waste, application and mixing
techniques, and soil characteristics (texture, temperature, moisture
content, oxygen level, nutrient level, pH, and the kind and number of
microbes) affect b1odegradat1on.26,27 Although blodegradatlon 1s purported
to be the principal mechanism for removal of organic compounds by land
treatment, only a few experiments have attempted to quantify the resulting
removal. A laboratory simulation of land treatment of oily sludge revealed
that 85 percent of the polynuclear aromatics degraded.28 Results from
Snyder et al.29 are comparable: oil removal on fertilized plots approached
80 percent at 1 year postappHcation.
Mathematical models for degradation could not be found in the
literature.
B.3.2 Volatilization
Techniques for direct measurement of volatilization at landfarms30'31
were discussed previously. Exogenous factors affecting volatilization in
land treatment include properties of the soil, waste application tech--
niques, mixing schedules, and atmospheric conditions.32'33'34
Farmer and Letey35 proposed five gradient models for pesticide vola-
tilization rates from the soil based on diffusion laws. The models
accommodate soil-incorporated pesticides with and without significant
mobility in flowing water. With nonincorporated pesticides, vapor density
relationships and air flow rate rather than movement in the soil control
the volatilization rate. These approaches do not, however, accommodate
subsurface injection. Thibodeaux and Hwang3^ developed a gradientless
model of air emissions from petroluem waste landfanns. Their approach
accurately predicted the volatilization of dieldrin reported in Farmer and
l,etey37 and is considered most suitable for estimating air emissions from
land treatment.
B.3.3 Migration and Runoff
Migration and runoff of VO from a landfarm may occur after improper
application or treatment of the hazardous waste. A description of factors
affecting these two pathways appeared 1n Reference 38. Results from a
laboratory study of refinery and petrochemical sludge39 suggested that the
B-6
-------
presence of hazardous waste 1n runoff decreases with time after applica-
tion. In addition, leachate water collected 1.5 meters below the subsur-
face was essentially free of toxic components.
However, as previously mentioned, migration and runoff do not occur at
a properly sited, operated, and maintained RCRA-perm1tted land treatment
facility. This paragraph 1s Included for the sake of completeness.
B.4 LANDFILLS
B.4.1 Volatilization
Direct measurements of VO emissions from landfills are possible.
During field tests conducted for EPA's Office of A1r Quality Planning and
Standards (OAQPS), Radian40 measured air emissions from landfills at three
of the four monitored hazardous waste TSOF. Markle et al.41 collected air
samples from three landfills representative of those used by the polyvlnyl
chloride Industry for health hazard evaluations. To compare the effici-
encies of water and soil coverings 1n reducing volatilization, Farmer et
al.42 measured emission rates from simulated landfills.
Numerous equations also have been developed to model VO emissions from
hazardous waste landfills. The procedure of Fanner et al.43 based on
Pick's law for steady-state diffusion, estimates emission from covered or
burled landfills. This was later modified by Shen.44 Thlbodeaux's45
emission models differentiate covered landfills by the presence or absence
of Internal gas generation. Another approach4^ Incorporates time-varying
atmospheric pressure Into the emission model. Volatilization rates from
landfills with no covering, I.e., open dumps, were modeled by Shen.47
DeWolf and Wetherold48 recommend Shen's49 emission model for covered land-
fills. GCA, 1n their excellent comprehensive review of these and other
emission models, prefers the work of Farmer et al.50 and Th1bodeaux.5l
Field validation of these mathematical models has not been reported In
the literature. Despite this, Baker and Mackay5* employed Shen's53 model
1n their protocol to evaluate toxic air pollution downwind of hazardous
waste landfills.
B.4.2 Migration
Several scientists have Investigated the potential problem of
migration of toxic contaminants from landfills. Rovers and Farquhar54
B-7
-------
suggested that the production of leachate within a landfill is not
uncommon. However, the migration of harmful compounds through adjacent
soils is not significant. Shen and Tofflemire^S reported that annual
losses of PCB to'migration from uncovered landfills in the Hudson River
Basin (New York) were substantially less than losses to volatilization.
B.4.3 Other Pathways
The impact of other pathways is not discussed quantitatively in the
literature.
B.5 WASTEWATER TREATMENT PUNT EFFLUENT
A description of the pathways operating in a wastewater treatment
plant is complicated by the number of different treatment systems. There
are closed tanks and open tanks (with and without mixing). Air emissions
from closed tanks occur during venting.
E. C. Jordan56 and Bums and Roe57,58 examined the fate of priority
toxic pollutants in publicly owned treatment plants. They observed a
decrease in VO concentrations across the activated sludge process and a
lack-of pollutant accumulation in the waste-activated, sludge. This
suggestsethat VO are substantially air-stripped or biodegraded during
secondary treatment. Results from the controlled laboratory experiments of
Roberts et al.59 imply that organic solutes more likely volatilize during
wastewater treatment with surface aeration than with bubble aeration.
Lurker et al.5^ examined how aeration rate, suspended particle concentra-
tion, and detergent concentration influence aerial organic chemical release
from an activated sewage treatment process.
The nonaerated open tank system is similar to the nonaerated surface
impoundment discussed previously; see Section B.2.1 for a discussion of the
corresponding emission rate models. Similarly, open tank wastewater treat-
ment processes with mixing can be estimated from Thibodeaux et al.61
Hwang62 went a step further in his activated sludge surface aeration model.
His approach was to estimate pollutant removal by degradation, adsorption,
and air stripping via a mass balance equation. Like Hwang and Thibodeaux
et al., Freeman^ considered air stripping losses at the air-water inter-
face. Unlike Hwang, however, he viewed the adsorption pathway as insignif-
icant and, thus, ignored it. In an entirely different approach,
B-8
-------
Freeman64*65 Modeled the mass transfer of a toxic compound Into the bubbles
of the aerated system (diffused air [subsurface] activated sludge model).
The structure of these and other models was critiqued 1n SCA.66
Allen et al.67 presented models of VO losses at each process encoun-
tered 1n wastewater treatment systems. The models Include a methodology
for estimating the relative Importance of competing pathways. Addition-
ally, these Investigators compared the loss of volatHes obtained from
field tests at several treatment facilities68 and from these mathematical
models. The models predict VO losses due to 51©degradation or volatiliza-
tion In close agreement with the field data. Results from other validation
studies are not as consistent. Predicted emission rates from aerated
surface Impoundments at two wastewater treatment plants were generally
larger than measured values.69 The difference between measured and
predicted emission rates 1n Cox et al.70 appears to be a function of the
type of compound and the presence of aerators.
B.6 SUMMARY
This appendix serves to guide the reader through the literature
concerning VO pathways from hazardous waste TSDF. The pathways of Interest
Include"volatilization, adsorption, migration, runoff, biological or
chemical degradation, hydrolysis, oxidation, and hydroxyl radical reaction.
The hazardous waste TSOF considered are surface Impoundments, landfills,
landfarms, and wastewater treatment plants. The body of this report
expands on the pathways and models most applicable.to current research.
B.7 REFERENCES
1. Hwang. S. T., and L. J. Thlbodeaux. Measuring Volatile Chemical
Emission Rates from Large Waste Disposal Facilities. Environmental
Progress. 2:81-86. 1983.
2. Thlbodeaux, L. J., D. G. Parker, and H. H. Heck (University of
Arkansas). Measurement of Volatile Chemical Emissions from Wastewater
Basins. Prepared for U.S. Environmental Protection Agency.
Washington, DC. Contract No. R805534. December 1981.
3. Radian Corporation. Hazardous Waste Treatment, Storage, and Disposal
Facility Area Sources: VOC Air Emissions. Prepared for U.S.
Environmental Protection Agency. EPA Contract No. 68-02-3850.
January 1985.
B-9
-------
4. Reference 1.
5. Reference 3.
6. Mackay, D., and P. J. Lelnonen. Rate of Evaporation of Low-Solubility
Contaminants from Water Bodies to Atmosphere. Environmental Science
and Technology. 13:1178-1180. 1975.
7. Mackay, D., and A. T. K. Yeun. Mass Transfer Coefficient Correlations
for Volatilization of Organic Solutes from Water. Environmental
Science and Technology. 17:211-217. 1983.
8. Reference 2.
9. DeWolf, G. B., and R. G. Wetherold (Radian Corporation). Protocols
for Calculating VOC Emissions from Surface Impoundments Using Emission
Models, Technical Note. Prepared for U.S. Environmental Protection
Agency. Washington, DC. Contract No. 68-02-3850. September 1984.
10. Shen, T. T. Estimation of Organic Compound Emissions from Waste
Lagoons. Journal of the Air Pollution Control Association. 32.
1982.
11. Reference 2.
12. GCA Corp. Evaluation and Selection of Models for Estimating Air
Emissions from Hazardous Waste Treatment, Storage, and Disposal
Facilities. Prepared for U.S. Environmental Protection Agency.
Publication No. EPA-450/8-84-020. December 1984.
13. Reference 2.
14. Reference 6.
15. Hwang, S. T. Model Prediction of Volatile Emissions. Environmental
Progress. 4:141-144. 1985.
16. Hwang, S. T. Toxic Emissions From Land Disposal Facilities. Environ-
mental Progress. 1:46-52. 1982.
17. Balfour, W. D., C. E. Schmidt, R. G. Wetherold, D. L. Lewis, J. I.
• * Steinmetz, and R. C. Hanish. Field Verification of Air Emission
Models for Hazardous Waste Disposal Facilities. In: Proceedings of
the Tenth Annual Research Symposium. Publication No. EPA-600/9-
84/007. Fort Mitchell, Kentucky. April 1984.
18. Reference 2.
19. Vaught, C. C. (GCA). Air Emissions for Quiescent Surface Impound-
ments—Emissions Data and Model Review, Draft Technical Note.
Prepared for U.S. Environmental Protection Agency. Washington, DC.
Contract No. 68-01-6871. August 1985.
B-10
-------
20. Springer, S., P. D. Lunney, K. T. Valsaraj, and L. J. Th^jaux
(University if Arkansas and Louisiana State University.) Emissions of
Hazardous Chemicals from Surface and Near Surface Impoundments to Air,
Draft Final Report. Prepared for U.S. Environmental Protection
Agency. Washington, DC. Project No. 808161-2. December 1984.
21. Reference 7.
22. Metcalf and Eddy, Inc. Wastewater Engineering: Collection, Treat-
ment, Disposal. New York, McGraw-Hill. 1972.
23. Kaufman, D. D. Fate of Organic Compounds 1n Land-Applied Wastes. In:
Land Treatment of Hazardous Wastes, Parr, J. F., P. B. Marsh, and J.
M. Kla (eds.). 1983. p. 77-151.
24. U.S. Environmental Protection Agency. Hazardous Waste Land Treatment,
Technical Resource Document. EPA Contract Nos. 68-03-2940 and 68-03-
2943. April 1983,
25. Brown, K. W. Chapter 36, Land Treatment of Hazardous Wastes. In:
Proceedings of the Fourth Life Sciences Symposium, Environment and
Solid Wastes. GatHnburg. TN. October 4-8, 1981. p. 449-482.
26. Reference 24.
27. Reference 23.
- -a e •
28. Bossert, K. W. M. Kachel, and R. Barth. Fate of Hydrocarbons During
01ly Sludge Disposal 1n Soil. Applied and Environmental Microbiology.
47:763-767. 1984.
29. Snyder. H. J., G. B. Rice, and J. J. Skujins. Residual Management by
Land Disposal. In: Proceedings of the Hazardous Waste Research
Symposium, Fuller, W. H., (ed.) Publication No. EPA-600/9-76-015.
July 1976.
30. Reference 1.
31. Reference 3.
32. Wetherold, R. G.. J. L. Randall, and K. R. Williams (Radian Corpora-
tion). Laboratory Assessment of Potential Hydrocarbon Emissions from
Land Treatment of Refinery 01ly Sludges. Prepared for U.S.. Environ-
mental Protection Agency. Washington, DC. Publication No. EPA-600/2-
84-108. June 1984.
33. Reference 24.
34. Farmer, W. J., and J. Letey. Volatilization Losses of Pesticides from
Soil. Prepared for U.S. Environmental Protection Agency, Publication
No. EPA-660/2-74-054. August 1974.
B-ll
-------
35. Reference 34.
36. Thlbodeaux, 1. J., and S. T. Hwang. Land Farming of Petroleum
Wastes—Modeling the Air Emissions Problem. Environmental Progress.
1:42-46. 1982.
37. Reference 34.
38. Reference 23.
39. Reference 25.
40. Reference 3.
41. Markle, R. A., R. B. Iden, and F. A. Sllemers (Battelle). A Prelimi-
nary Examination of Vinyl Chloride Emissions from Polymerization
Sludges during Handling and Land Disposal. Prepared for U.S. Environ-
mental Protection Agency. Washington, OC. Publication No. EPA-
660/2-74-054. February 1976.
42. Farmer, W. J., M. Yang, J. Letey, and W. F. Spencer. Problems Associ-
ated with the Land Disposal of an Organic Industrial Hazardous Waste
Containing HCB. In: Residual Management by Land Disposal, Proceed-
ings of the Hazardous Waste Research Symposium, Fuller, W. H. (ed.).
Publication No. EPA-600/9-76-015. July 1976.
43. farmer, W. J., M. S. Yang, and J. Letey. Land Disposal of Hazardous
Wastes: Controlling Vapor Movement 1n Soils. In: Fourth Annual
Research Symposium. Publication No. EPA-600/9-78-016. August 1978.
44. Shen, T. T. Estimating Hazardous Air Emissions from Disposal Sites.
Pollution Engineering. 31-34. August 1981.
45. Thlbodeaux, L. J. Estimating the A1r Emissions of Chemicals from
Hazardous Waste Landfills. Journal of Hazardous Materials.
4:235-244. 1981.
.46. Thlbodeaux, L. J., C. Springer, and L. M. RHey. Models of Mechanisms
for the Vapor Phase Emission of Hazardous Chemicals from Landfills.
Journal of Hazardous Materials. 7:63-74. 1982.
47. Reference 44.
48. DeWolf, G. B.r and R. S. Wetherold (Radian Corporation). Protocols
for Calculating VOC Emissions from Land Applications Using Emission
Models, Technical Note. Prepared for U.S. Environmental Protection
Agency. Washington, DC. Contract No. 68-02-3850. December 1984.
49. Reference 44.
50. Reference 43.
B-12
-------
51. Reference 2.
52. Baker. L. W., and K. P. Mackay. Hazardous Waste Management: Screen-
ing Models for Estimating Toxic A1r Pollution Near a Hazardous Waste
Landfill. Journal of the A1r Pollution Control Association. 35:1190-
1195. 1985.
53. Reference 44.
54 Rovers, F. A., and 6. J. Farquhar. Evaluating Contaminant Attenuation
in the Soil to Improve Landfill Selection and Design. In: Proceed-
ings of the International Conference on Land for Waste Management.
1974. p. 161-173.
55. Shen, T. T., and T. J. Tofflemlre. A1r Pollution Aspects of Land
Disposal of Toxic Wastes. Environmental Engineering Division Journal.
106:211-226. 1980.
56. E. C. Jordan Co. Fate of Priority Toxic Pollutants 1n Publicly Owned
Treatment Works, 30 Day Study. Prepared for U.S. Environmental
Protection Agency. Publication Mo. EPA-440/1-82-302. August 1982.
57. Bums and Roe Industrial Services Corp. Fate of Priority Pollutants
1n Publicly Owned Treatment Works, Final Report, Volume II. Prepared
for U.S. Environmental Protection Agency. Publication No. EPA-440/1-
82-303. July 1982.
58. Bu*ns and Roe Industrial Services Corp. Fate of Priority Pollutants
In Publicly Owned Treatment Works, Final Report, Volume I. Prepared
for U.S. Environmental Protection Agency. Publication No. EPA-440/1-
82-303. September 1982.
59. Roberts. P. V., C. Munz, P. Dandliker, and C. Matter-Muller.
Volatilization of Organic Pollutants in Wastewater Treatment-Model
Studies, Project Summary. Prepared for U.S. Environmental Protection
Agency. Publication No. EPA-600/52-84-047. April 1984..
60. Lurker, P. A., C. S. Clark, V. J. EHa, P. S. Gartslde, and R. N.
Klnman. Aerial Organic Chemical Release.from Activated Sludge.
Water Research. 18:489-494. 1984.
61. Reference 2.
62. Hwang, S. T. TreatabllHy and Pathways of Priority Pollutants in
Biological Wastewater Treatment. (Presented at the American Institute
of Chemical Engineers Symposium. Chicago, Illinois. November 1980.)
63. Freeman, R. A. Stripping of Hazardous Chemicals from Surface Aerated
Waste Treatment Basins. In: APCA/WPCF Speciality Conference on
Control of Specific Toxic Pollutants. Gainesville, Florida.
February 13-16, 1979.
B-13
-------
64. Freeman, R. A. Comparison of Secondary Emissions From Aerated Treat-
ment Systems. (Presented at AlChe Meeting. Paper 5c. Orlando.
Florida. February 1982.)
65. Freeman, R. A. Secondary Emissions from Subsurface Aerated Treatment
Systems. Environmental Progress. 1:117-119. 1982.
66. Reference 12.
67. Allen, C. C., S. Simpson, and G. Brant (RTI and Associated
Technologies, Inc.). Field Evaluations of Hazardous Waste
Pretreatment as an A1r Pollution Control Technique. Prepared for U.S.
Environmental Protection Agency. EPA Contract No. 68-02-3992. April
1985.
68. Alsop, G. M., R. 1. Berglund, T. W. Slegrist, G. M. Whipple, and B. E.
Wllkes. Fate of Specific Organics in an Industrial Biological Waste-
water Treatment Plant, Draft Report. Prepared for U.S. Environmental
Protection Agency, Industrial Environmental Research Laboratory.
Research Triangle Park, NC. June 29, 1984.
69. Melsenheimer, 0. C. (GCA). Emissions Data and Model Review for
Wastewater Treatment Operations, Draft Technical Note. Prepared for
U.S. Environmental Protection Agency. Washington, DC. Contract No.
68^01-6871. August 1985.
70. Cox, R. D., Lewis, R. G. Wetherold, and J. I. Stelnmetz (Radian
Cbtpo'ration). Evaluation of VOC Emissions from Wastewater Systems
(Secondary Emissions). Prepared for U.S. Environmental Protection
Agency. Washington, DC. Project No. 68-03-3038. July 1983.
B-14
-------
APPENDIX C
COMPREHENSIVE SOURCE LIST
-------
APPENDIX C
COMPREHENSIVE SOURCE LIST
Addendum to memorandum dated September 6, 1985, from Eichinger, Jeanne, GCA
Corporation, to Hustvedt, K. C., EPA/OAQPS. September 12, 1985. TSDF
model source parameters and operating practices data base.
Allen, C. C. Prediction of A1r Emissions from Surface Impoundments, Paper
31a. (Presented at 1986 Summer Meeting of AIChE. Boston, MA. August
1986.) 26 p.
Allen, C. C., and D. A. Green (Research Triangle Institute). Review of VOC
Pathway Models, Draft Report. Prepared for U.S. Environmental
Protection Agency. EPA Contract No. 68-01-6826. 1985.
• •
Allen, C. C., D. A. Green, and J. B. White (Research Triangle Insltute).
Preliminary Assessment of Aerated Waste Treatment Systems at TSDFs -
Phase I, Draft Final Report. Prepared for U.S. Environmental
Protection Agency. EPA Contract No. 68-03-3149. May 1985.
Allen, C. C., D. A. Green, and J. B. White (Research Triangle Institute).
Preliminary Assessment of Aerated Waste Treatment Systems at TSDFs—
Phase I. Draft. Prepared for U.S. Environmental Protection Agency.
Research Triangle Park, NC. EPA Contract No. 68-03-3149, Task 54-01F.
1985.
Allen, C. C., S. Simpson, and G. Brant (Research Triangle Institute and
Associated Technologies, Inc.). Field Evaluations of Hazardous Waste
Pretreatment as an A1r Pollution Control Technique. Prepared for U.S.
Environmental Protection Agency. EPA Contract No. 68-02-3992.- April
1985.
Alsop, G. M., R. L. Berglund, T. W. SlegHst, G. M. Whipple, and B. E.
Wllkes. Fate of Specific Organlcs In an Industrial Biological
Wastewater Treatment Plant, Draft Report. Prepared for U.S.
Environmental Protection Agency, Industrial Environmental Research
Laboratory. June 29, 1984.
Armstrong, N. E., E. F. Gloyna, and 0. Wyss. Biological Counter-measures
for the Control of Hazardous Material Spills, Project Summary.
Publication No. EPA-600/S2-84-071. March 1984.
C-3
-------
Arnold, J. H. Studies 1n Diffusion: III. Unsteady-State Vaporization and
Absorption. Transactions of the American Institute of Chemical
Engineers. 40:361-379. 1944.
Bailey, J. E., and D. F. 0111s. Biochemical Engineering Fundamentals. New
York, McGraw-Hill. 1977. p. 343-349.
Baker, L. W., and K. P. Mackay. Hazardous Waste Management: Screening
Models for Estimating Toxic A1r Pollution Near a Hazardous Waste
Landfill. Journal of the Air Pollution Control Association. 35:
1190-1195. 1985.
Balfour, W. 0., C. E. Schmidt, R. G. Wetherold, D. L. Lewis, J. I.
Steinmetz. and R. C. Hanlsh. Field Verification of Air Emission
Models for Hazardous Waste Disposal Facilities. In: Proceedings of
the Tenth Annual Research Symposium. Publication No. EPA-600/9-
84/007. Fort Mitchell, Kentucky. April 1984.
Bossert, I., W. M. Kachel, and R. Barth. Fate of Hydrocarbons during Oily
Sludge Disposal in Soil. Applied and Environmental Microbiology. £7:
763-767. 1984.
Branscome, M.f and A. Gitelman. Sensitivity Analysis: Emission Estimates
for Surface Impoundments. Prepared for the U.S. Environmental
Protection Agency. (Docket Number ). March 1986. 67 p.
Brown, K.. W. Chapter 36, Land Treatment of Hazardous Wastes. In:
Proceedings of the Fourth Life Sciences Symposium, Environment and
Solid Wastes. Gatlinburg, TN. October 4-8, 1981. p. 449-482.
Burns and Roe Industrial Services Corp. Fate of Priority Pollutants in
Publicly Owned Treatment Works, Final Report, Volume II. Prepared for
U.S. Environmental Protection Agency. Publication No. EPA-440/1-82-
303. July 1982.
Burns and Roe Industrial Services Corp. Fate of Priority Pollutants in
Publicly Owned Treatment Works, Final Report, Volume I. Prepared for
U.S. Environmental Protection Agency. Publication No. EPA-440/1-82-
303. September 1982.
Canter, L., and R. C. Knox. Evaluation of Septic Tank System Effects on
Ground Water Quality. .Publication No. EPA-600/S2-84-107. September
1984.
Clark, J. W., W. Viessman, Jr., and M. J. Hammer. Water Supply and
Pollution Control. Scranton, PA, International Textbook Company.
1971.
Cox, R. D., D. L. Lewis, R. G. Wetherold, and J. I. Steinmetz (Radian
Corporation). Evaluation of VOC Emissions from Wastewater Systems
(Secondary Emissions). Prepared for U.S. Environmental Protection
Agency. Washington, DC. EPA Project No. 68-03-3038. July 1983.
C-4
-------
Cox, R. D., J. L. Stelnmetz, and 0. L. Lewis (Radian Corporation).
Evaluation of VOC Emissions from Wastewater Systems (Secondary
Emission), Volme II, Appendices. Prepared for U.S. Environmental
Protection Agency. Washington, DC. EPA Project No. 68-03-3038. July
1983.
Crank, J. The Mathematics of Diffusion. London, England. Oxford
University Press. 1970. p. 45-47.
OeWolf, G. B.f and R. G. Wetherold (Radian Corporation). Protocols for
Calculating VOC Emissions fron Surface Impoundments Using Emission
Models, Technical Note. Prepared for U.S. Environmental Protection
Agency. Washington, DC. Contract No. 68-02-3850. September 1984.
DCWolf, G. B., and R. G. Wetherold (Radian Corporation). Protocols for
Calculating VOC Emissions from Land Applications Using Emission
Models, Technical Note. Prepared for U.S. Environmental Protection
Agency. Washington, DC. Contract No. 68-02-3850. December 1984.
E. C. Jordan Co. Fate of Priority Toxic Pollutants 1n Publicly Owned
Treatment Works, 30 Day Study. Prepared for U.S. Environmental
Protection Agency. Publication No. EPA-440/1-82-302. August 1982.
Ely, R. L., G. L. Klngsbury, M. R. Branscome, L. J. Goldman, C. M.
Northelm, J. H. Turner, and F. 0. M1xon, Jr. (Research Triangle
Institute). Performance of Clay Caps and Liners for Disposal
Facilities. Prepared for U.S. Environmental Protection Agency.
Cincinnati. OH. EPA Contract No. 68-03-3149. March 1983.
Farlno, W.. P. Spawn, M. Jaslnskl, and B. Murphy (GCA/Technology). Evalua-
tion and Selection of Models for Estimating A1r Emissions from
Hazardous Waste Treatment, Storage, and Disposal Facilities. Prepared
for U.S. Environmental Protection Agency. Publication No. EPA 450/3-
84-020. 1984.
Farmer, W. J., and J. Letey. Volatilization Losses of Pesticides from
Soil. Publication No. EPA-660/2-74-054. August 1974.
Farmer, W. d.t M. S. Yang, and J. Letey. Land Disposal of Hazardous
Wastes: Controlling Vapor Movement 1n Soils. In: Fourth Annual
Research Symposium. Prepared for U.S. Environmental Protection
Agency. Publication No. EPA-600/9-78-016. August 1978.
Farmer, W. J., M. S. Yang, and W. F. Spencer. Land Disposal of
Hexachlorobenzene: Controlling Vapor Movement 1n Soil. Publication
No. EPA-600/2-80-119. August 1980.
Farmer, W. J., M. S. Yang, J. Letey, W. F. Spencer, and M. H. Rouller.
Land Disposal of Hexachlorobenzene Wastes: Controlling Vapor Movement
In Soils. Fourth Annual Research Symposium. U.S. Environmental
Protection Agency. Publication No. EPA-600/9-78-016. August 1978.
C-5
-------
Fanner, W. J.. M. Yang, J. Letey, and W. F. Spencer. Problems Associated
with the Land Disposal of an Organic Industrial Hazardous Waste
Containing HCB. In: Residual Management by Land Disposal,
Proceedings of the Hazardous Waste Research Symposium, Fuller, W. H.
(ed.). U.S. Environmental Protection Agency. Publication No. EPA-
600/9-76-015. July 1976.
Freeman, R. A. Comparison of Secondary Emissions From Aerated Treatment
Systems. (Presented at AlChe Meeting. Paper 5c. Orlando, Florida.
February 1982.)
Freeman, R. A. Secondary Emissions from Subsurface Aerated Treatment
Systems. Environmental Progress. 1:117-119. 1982.
Freeman, R. A. Stripping of Hazardous Chemicals from Surface Aerated Waste
Treatment Basins. In: APCA/WPCF Speciality Conference on Control of
Specific Toxic Pollutants. Gainesville, Florida. February 13-16,
1979.
Galegar, J. (East Central Oklahoma State University). Annotated Litera-
ture Reference on Land Treatment of Hazardous Waste. Prepared for
U.S. Environmental Protection Agency. Publication No. EPA-600/2-84-
098. Hay 1984.
GCA Corporation. Evaluation and Selection of Models for Estimating A1r
Emissions from Hazardous Waste Treatment, Storage, and Disposal
Facilities. Prepared for U.S. Environmental Protection Agency.
Publication No. EPA-450/8-84-020. December 1984.
GCA/Tecnnology Division. Evaluation and Selection of Models for Estimating
A1r Emissions from Hazardous Waste Treatment, Storage, and Disposal
Facilities. Prepared for U.S. Environmental Protection Agency.
Research Triangle Park, NC. Publication No. EPA-45073-84-020.
December 1984. pp. 85-89.
GCA Corporation. Hazardous Waste TSDF Waste Process Sampling. 1985.
p. 1-11.
GCA Corporation. A1r Emission Estimation Methods for Transfer, Storage,
and Handling Operations, Draft Technical Note. Prepared for U.S.
Environmental Protection Agency, Office of A1r Quality Planning and
Standards. Research Triangle Park, NC. EPA Contract No. 68-01-6871.
August 1985.
GCA Corporation. Air Emissions for Quiescent Surface Impoundments—Emis-
sions Data and Model Review, Draft Technical Note. Prepared for U.S.
Environmental Protection Agency. Contract No. 68-01-6871, Assignment
49. August 1985.
GCA Corporation. Air Emissions from Land Treatment—Emissions Data and
Model Review, Draft Technical Note. Prepared for U.S. Environmental
Protection Agency. Research Triangle Park, NC. August 1985. 120 pp.
C-6
-------
GCA Corporation. Air Emissions of VOC front Waste Piles at Hazardous Waste
Treatment, Storage, and Disposal Facilities. Prepared for U.S.
Environmental Protection Agency, Office of Air Quality Planning and
Standards. Research Triangle Park. NC. EPA Contract No. 68-01-6871.
August 1985.
GCA Corporation. A1r Emissions from Land Treatment—Emissions Data and
Model Review. Draft Technical Note. Prepared for U.S. Environmental
Protection Agency. Research Triangle Park, NC. September 1985.
Chapter 4.
Gerachty, J. J., D. W. Miller, F. Vander Leeden, and F. L. Trolse. Water
Atlas of the United States. Port Washington, NY, Water Information
Center, Inc. 1973. Plate 30.
Ghasseml, M., M. Haro, and L. Fargo. Assessment of Hazardous Waste Surface
Impoundment Technology: Case Studies and Perspectives of Experts,
Project Summary. Publication No. EPA-600y$2-84-173. January 1985.
Graver Standard Cone-Roof, Flat-Bottom Tanks. Sizes and Capacities.
Hornlck, S. B. In: Land Treatment of Hazardous Waste, Parr, J. F. (ed).
Park Ridge, NJ, Noyes Data Company. 1982.
Muddieston, R. L. Solid-Waste Disposal: Land Farming. Chemical
Engineering. 119-124. February 26, 1979.
Hwang,.4. T. TreatabllUy and Pathways of Priority Pollutants 1n
Biological Wastewater Treatment. (Presented at the American Institute
of Chemical Engineers Symposium. Chicago, Illinois. November 1980.)
Hwang, S. T. Toxic Emissions from Land Disposal Facilities. Environmental
Progress. ,1:46-52. February 1982.
Hwang, S. T. Model Prediction of Volatile Emissions. Environmental
Progress. 4:141-144. 1985.
Hwang, S. T., and L. J. TMbodeaux. Measuring Volatile Chemfcal Emission
Rates from Large Waste Disposal Facilities. Environmental Progress.
2:81-86. 1983.
ICF, Inc. The RCRA Risk-Cost Analysis Model Phase III Report, Appendix E.
Chemical and Physical Processes Affecting Decay Rates of Chemicals in
Aquatic Environments, Draft. Prepared for U.S. Environmental
Protection Agency, Office of Solid Waste, Economic Analysis Branch.
Washington, DC. 1984.
Kaufman, D. D. Fate of Organic Compounds In Land-Applied Wastes. In:.
Land Treatment of Hazardous Wastes, Parr, J. F., P. B. Marsh, and J.
M. Kla (eds.). 1983. p.- 77-151.
C-7
-------
Klncannon, D. F. Evaluation of Biological Tower Design Methods.
(Presented at First International Conference on Fixed-Film Biological
Processes. Kings Island, OH. April 1982.)
Land Farming Fills on HPI Need. Hydrocarbon Processing 60:97-103. 1980.
Letter and attachment from Shen, T., New York State Department of
Environmental Conservation, to McDonald, R., EPA/OAQPS. Modification
of Arnold's equation. February 16, 1986.
Loehr, R. C., and J. F. Mallna, Jr. (eds.) Land Treatment, A Hazardous
Waste Management Alternative. In: Proceedings of the Water Resources
Symposium Number Thirteen. Austin, Texas, Center for Research in
Water Resources, The University of Texas at Austin. 1986.
Lunney, P. D. Characterization of Wind and Depth Effects Upon Liquid Phase
Mass Transfer Coefficients: Simulation Studies. Master's thesis,
University of Arkansas, Fayettevllle, AR. January 1983.
Lurker, P. A., C. S. Clark, V. J. Ella, P. S. Gartslde, and R. N. Kinroan.
Aerial Organic Chemical Release from Activated Sludge. Water
Research. 18:489-494. 1984.
Lyman, W. J., et al. Research and Development of Methods for Estimating
Physicochemlcal Properties of Organic Compounds of Environmental
Concern. Phase II, Part I. NTIS AO 11875A. 1981.
"•-•at-
Lyman, -Warren" J., William F. Reehl, and David H. Rosenblatt. Handbook of
Chemical Property Estimation Method—Environmental Behavior of Organic
Compounds. New York, NY, McGraw-Hill Book Company. 1982. Chapter 9.
Mabey, W. R., T. Mill, and D. G. Hendry. Photolysis in Water. In:
Laboratory Protocols for Evaluating the Fate of Organic Chemicals in
Air and Water, Draft. Prepared for U.S. Environmental Protection
Agency. EPA Contract 68-03-2227. 1980.
Mackay, D., and A. Yeun. Mass Transfer Coefficient Correlations for
Volatilization of Organic Solutes from Water. Environmental Science
and Technology. 17:211-217. 1983.
Mackay, D., and P. J. Leinonen. Rate of Evaporation of Low-Solubility
Contaminants from Water Bodies to Atmosphere. Environmental Science
and Technology. 12:1178-1180. 1975.
Mackay, D., W. Y. Shiu, A. Bobra, J. Blllington, E. Chau, A. Yeun, C. Ng,
and F. Szeto (University of Toronto). Volatilization of Organic
Pollutants from Water. Prepared for U.S. Environmental Protection
Agency. Publication No. EPA-600/3-82-019. April 1982.
C-8
-------
Markle, R. A., R. B. Iden. and F. A. Slletpcrs (Battelle Laboratories). A
Preliminary Examination of Vinyl Chloride Emissions fro* Polymeriza-
tion Sludges during Handling and Land Disposal. Prepared for U.S.
Environmental 'Protection Agency. Washington, DC. Publication No.
EPA-660/2-74-054. February 1976..
Matter-Huller, C., W. Gujer, W. Gelger, and W. Stum. The Prog. Wat. Tech.
(Toronto). 12:299-314. Great Britain. lAWPR/Pergamon Press, Ltd.
1980.
Melsenhelmer, D. C. (GCA). Emissions Data and Model Review for Wastewater
Treatment Operations, Draft Technical Note. Prepared for U.S.
Environmental Protection Agency. Washington, DC. Contract No. 68-01-
6871. August 1985.
Memorandum and attachment from Wright, M., Research Triangle Institute, to
Thomeloe, S., EPA/OAQPS. Selection of an emission model for land
treatment. Nay 30, 1986.
Memorandum from Thomeloe, S., EPA/OAQPS, to Durham, J., EPA/OAQPS.
January 31, 1986. Land treatment data base.
Metcalf and Eddy, Inc. Wastewater Engineering: Collection. Treatment,
Disposal. New York, McGraw-Hill. 1972. 782 p.
Meyers, J.-D.. and R. L. Huddleston. Treatment of 01 ly Refinery Wastes by
. Land Farming. In: Proceedings of the Industrial Waste Conference.
Lafayette, Indiana. May 8-10, 1979. p. 686-698.
M1l11ngton, R. J., and J. P. Quirk. Permeability of Porous Sol Ids. Trans.
Faraday Society. 57:1200-1207. 1961.
0111s, D. F. Contaminant Degradation 1n Water. ES&T. 19(6):480-484.
1985. ~~
Parr, J. F., P. B. Marsh, and J. M. Kla (eds.). Land Treatment of
Hazardous Wastes. Park Ridge, NJ, Noyes Data Corporation. 1983.
Perry, R. H., and C. H. Chllton. Chemical Engineers' Handbook, Fifth
Edition. New York, McGraw-Hill. 1973.
PUter, P. Determination of Biological DegradablHty of Organic
Substances. Water Research. 10:231-235. 1976.
Radian Corporation. Hazardous Waste Treatment, Storage, and Disposal
Facility Area Sources: VOC Air Emissions. Prepared for U.S.
Environmental Protection Agency. EPA Contract No. 68-02-3850.
January 1985.
Radiation Technologies, Inc. Treatablllty Data In Support of a Treatment
Zone Model for Petroleum Refining Land Treatment Facilities. Prepared
for American Petroleum Institute, Washington, DC. March 1986.
154 pp.
C-9
-------
Relhhardt, J. R. Gas-Side Mass-Transfer Coefficient and Interfadal
Phenomena of Flat-Bladed Surface Agitators. Ph.D. dissertation,
University of Arkansas, Fayettevllle, AR. 1977.
Roberts, P. V., C. Munz, P. Oandliker, and C. Matter-Muller.
Volatilization of Organic Pollutants 1n Wastewater Treatment-Model
Studies, Project Summary. Publication No. EPA-600752-84-047. April
1984.
Rovers, F. A., and G. J. Farquhar. Evaluating Contaminant Attenuation in
the Soil to Improve Landfill Selection and Design. In: Proceedings
of the International Conference on Land for Waste Management. 1974.
p. 161-173.
Research Triangle Institute. Exponential Biological Decomposition Model,
Draft Technical Note. Prepared for U.S. Environmental Protection
Agency. Contract No. 68-01-6826. August 15, 1985.
Research Triangle Institute. RTI Review of VOC Pathway Models, Draft
Report. Prepared for U.S. Environmental Protection Agency. Contract
No. 68-01-6826. October 31, 1985.
Savage, G. M., L. F. Diaz, and C. G. Golueke. Disposing of Organic
Hazardous Wastes by Composting. BloCycle. 26:31-34. 1985.
Shen T T. Emission Estimation of Hazardous Organic Compounds from Waste
^Disposal Sites. (Presented at the Air Pollution Contro Association
Annual Meeting. Montreal, Quebec. June 1980.)
Shen, T. T. Estimating Hazardous Air Emissions from Disposal Sites.
Pollution Engineering. 31-34. August 1981.
Shen T T. Estimation of Organic Compound Emissions from Waste Lagoons.
'journal of the A1r Pollution Control Association. 32. 1982.
Shen T T. and T. J. Tofflemire. Air Pollution Aspects of Land Disposal
'of Toxic Wastes. Environmental Engineering Division Journal. 106:
211-226. 1980.
Snyder. H. J., G. B. R1ce, and J. J. Skujins. Residual Management by Land
Disoosal In: Proceedings of the Hazardous Waste Research Symposium,
Fuller, W. H. (ed.). Publication No. EPA-600/9-76-015. July 1976.
Spivey, J. J.. C. C. Allen, D. A. Green, J. P. Wood, and R. L. Stallings
(Research Triangle Institute). Preliminary Assessment of Hazardous
Waste Pretreatment as an Air Pollution Control Technique. Prepared
for U.S. Environmental Protection Agency. EPA Contract No. 68-03-
3149, Task 12-5. 1984.
C-10
-------
Springer. S.. P. 0. Lunney, K. T. ValsaraJ, and L. J. Thlbodeaux
P ^University of Arkansas and Louisiana State University.) Emissions of
Hazardous Chemicals from Surface and Near Surface Impoundments to A1r,
Draft Final Report. Prepared for U.S. Environmental Prot«ct1on
Agency. Washington, DC. EPA Project No. 808161-2. December 1984.
Springer, C., P. D. Lunney, and K. T. ValsaraJ. Emission of Hazardous
Chemicals from Surface and Near Surface Impoundments to A1r. U.S.
Environmental Protection Agency, Solid and Hazardous Waste Research
Division. Cincinnati, OH. EPA Project Number 8081*1-02. December
1984.
StalUngs, R. L., T. N. Rogers, and M. E. Mulllns. A1r Stripping of
Volatile Organlcs. In: Proceedings of the Institute of Environmental
Sciences. Las Vegas, NV. April 30-May 2, 1985. p. 170-174.
Stiver, W., and 0. Mackay. Evaporation Rate of Spills of Hydrocarbons and
Petroleum Mixtures. Environmental Science Technology. 18:834-840.
1984.
Telecon. Goldman, Len, Research Triangle Institute, with Hannak, Peter,
Alberta Environmental Center. April 4, 1986.
Telecoir. Goldman, Len, Research Triangle Institute, with Hughes, John,-
National Climatic Center. Ashevllle, NC. May 15, 1986.
Telecon. Goldman, Len, Research Triangle Institute, with Wall 1s, Al,
National Climatic Center, Ashevllle, NC. May 30, 1986.
Telecon. Goldman, Len, Research Triangle Institute, with Borden, Roy,
Department of C1v1l Engineering, North Carolina State University,
Raleigh, NC. August 13. 1986.
Thlbodeaux. L. J. A1r Stripping of Organlcs from Wastewater. A
Compendium. Air/Water, p. 373. (no date yet).
Thlbodeaux. L. J. Air Stripping of Organlcs From Wastewater: A
Compendium. Proceedings of the Second National Conference on Complete
Water Use. Chicago. IL. May 4-8, 1978.
Thlbodeaux. L. J. Estimating the Air Emissions of Chemicals from Hazardous
Waste Landfills. Journal of Hazardous Materials. 4:235-244. 1981.
Thlbodeaux. L. J., and S. T. Hwang. Land Farming of Petroleum
Wastes—Modeling the Air Emissions Problem. Environmental Progress.
1:42-46. 1982.
Thlbodeaux, L. J., and S. T. Hwang. Toxic Emissions from Land Disposal
Facilities. Environmental Progress. 1(1):46. February 1982.
C-ll
-------
Thlbodeaux, L. J., C. Springer, and L. M. R1ley. Models of Mechanisms for
the Vapor Phase Emission of Hazardous Chemicals from Landfills.
Journal of Hazardous Materials. 7:63-74. 1982.
Thibodeaux, L. J., D. 6. Parker, and H. H. Heck (University of Arkansas).
Measurement of Volatile Chemical Emissions from Wastewater Basins.
Prepared for U.S. Environmental Protection Agency. Washington, DC.
Contract No. R805534. December 1981.
Trip Report. Goldman, Len, Research Triangle Institute, with Chemical
Waste Management, Sulphur, Louisiana. February 25, 1986.
Truong, K. N., and J. W.'Blackburn. The Stripping of Organic Compounds in
Biological Treatment Processes. Environmental Progress. 3:153-152.
1984.
Turner, J. H., C. C. Allen, and D. A. Green (Research Triangle Institute).
VOC Pathways from Certain Hazardous Waste Sites, Draft Interim Report.
Prepared for U.S. Environmental Protection Agency. EPA Contract No.
68-01-6826. September 1985.
U.S. Environmental Protection Agency. Transportation and Marketing of
Petroleum Liquids. In: AP-42, Compilation of Air Pollutant Emission
Factors, Third Edition, Supplement 12, Section 4.4. Office of Air
Quality Planning and Standards. Research Triangle Park, NC. July
1979. 13 p.
U.S. *Erwironmental Protection Agency. Tank and Drum Cleaning. In: AP-42,
Compilation of Air Pollutant Emission Factors, Third Edition,
Supplement 12, Section 4.8. Office of Air Quality Planning and
Standards. Research Triangle Park, NC. February 1980. 4 p.
U.S. Environmental Protection Agency. Storage of Organic Liquids. In:
AP-42, Compilation of Air Pollutant Emission Factors, Third Edition,
Supplement 12, Section 4.3. Office of Air Quality Planning and
Standards. Research Triangle Par.k, NC. April 1981. 25 p.
U.S. Environmental Protection Agency. Hazardous Waste Land Treatment,
Technical Resource Document. EPA Contract Nos. 68-03-2940 and 68-03-
2943. April 1983.
U.S. Environmental Protection Agency. EPA Design Manual: Municipal
Wastewater Stabilization Ponds. Publication No. EPA-625/1-83-015.
October 1983. 327 p.
U.S. Environmental Protection Agency. Assessing the Release and Costs
Associated with Truck Transport of Hazardous Wastes. Publication No.
PB 84-224-468 (Contract No. 68-01-0021). Washington, DC. January
1984. 151 p.
C-12
-------
U.S. Environmental Protection Agency. Control of Volatile Organic Compound
Leaks from Synthetic Organic Chemical and Polymer Manufacturing
Equipment. Research Triangle Park. NC. Publication No. EPA-450/3-
86-006. March 1984.
U.S. Environmental Protection Agency. Evaluation and Selection of Models
for Estimating Emissions from Hazardous Waste Treatment, Storage, and
Disposal Facilities. Office of Air Quality Planning and Standards.
Research Triangle Park, NC. Publication No. EPA-45073-84-020.
December 1984.
U.S. Environmental Protection Agency. Hazardous Waste Ranking - Assessment
of A1r Emissions from Hazardous Waste Treatment, Storage, and Disposal
Facilities. Publication No. EPA-450/3-85-006. February 1985.
Vaught, C.'C. (GCA). Air Emissions for Quiescent Surface Impoundments-
Emissions Data and Model Review, Draft Technical Note. Prepared for
U.S. Environmental Protection Agency. Washington, DC. Contract No.
68-01-6871. August 1985.
Westat Corporation. National Survey of Hazardous Waste Generators and
TSDF's Regulated Under RCRA 1n 1981. Prepared for the U.S.
Environmental Protection Agency. Contract No. 68-01-6861. April
1984.
Wetherold, R. S.t J. L. Randall, and K. R. Williams (Radian Corporation).
Laboratory Assessment of Potential Hydrocarbon Emissions from Land
Treatment of Refinery 01ly Sludges. Prepared for U.S. Environmental
Protection Agency. Washington, DC. Publication No. EPA-600/2-84-108.
June 1984.
Zepp, R. G. Quantum Yields for Reaction of Pollutants 1n Dilute Aqueous
Solution. Environ. Sc1. Technol. 12(3):327-329. 1979.
Zepp, R. G., and 0. M. Cllne. Rate of Direct Photolysis In Aquatic
Environment. Environ. Sc1. Technol. 11(4):359-366. 1977.
C-13
-------
APPENDIX D
PROPERTIES FOR COMPOUNDS OF INTEREST
-------
APPENDIX 0
PROPERTIES FOR COMPOUNDS OF INTEREST
This appendix contains compound-specific properties of about 700
chemicals, most of which are not included in CHEMDAT7. These data,
presented as a source of information, can be easily incorporated into
CHEMDAT7. Consequently, this appendix greatly increases the utility of
CHEMDAT7. The chenrical "universe" tested in this appendix represents those
chemicals that could be encountered in TSDF and that are useful for
calculating emission rates for the facilities modeled in the body of this
report. The list was extracted from the GCA Physical/Chemical Database,
WET model stream compositions, and the Industrial Studies Database.
'The"compounds listed in this appendix were not all included in
CHEMDAT7 because their inclusion would seriously slow the execution time of
the program, the memory requirements would prevent the program from being
run on many machines, and because the data are not complete for some of the
chemicals. Compounds included in the CHEMDAT7 program were selected on the
basis of the estimated frequency with which they are found in hazardous
wastes and on their position in prioritized lists of pollutants.
Due to the limited number of compounds for which adequate biodegra-
dation rate data were available to determirte appropriate biodegradation
rate constants, these constants are provided separately as Table D-l. It
is recognized that biodegradation rates can vary widely from site to site.
Therefore, the following priority schedule is provided as guidance in
determining the appropriate biodegradation rate constants to be employed in
the emission models:
• Use site-specific biodegradation rate data where available.
• Use the rate constants suggested in Table D-l as available.
D-3
-------
TABLE 0-1. SUMMARY OF RECOMMENDED BIODEGRADATION RATE CONSTANTS
Compound
Acenaphthene
Acetic acid
Acetone
Acrolein
Aery Ion it rile
Aniline
Anthracene
Arochlor (1254)
Benzene
bis (2-Ethy 1 hexy 1 ) phthal ate
Butanol (iso)
Butyl benzyl phthal ate
Carbon tetrachloride
Chloroaniline(s)
Chlorobenzene
Chloroform
Chlorophenol(o)
Chlorophenol(p)
Chrysene
Cresol (s)
Di bromochl oromethane
Di butyl phthal ate
Dichlorobenzene(l,2)
Dichlorobenzene(l,3)
Dichloroethane(l.l)
Dichloroethane(l,2)
Dichloroethene(l.l)
4*
14
1.3
7.8
18
7.1
19
0.77
7.8
1.5
0.27
0.39
2.94
15
6.5
23
0.40
2.5
6.4
2.1
L/g-h
2.7
0.98
1.15
0.36
0.75
21
2.2
0.95
1.4
0.35
0.11
2.2
1.5
0.86
10
0.79
1.4
17
0.035
1.0
0.58
2.3
2.3
0.98
1.8
mg?L
14
1.1
22
24
0.34
13.6
2.2
71
1.0
0.31
0.039
3.7
1.4
0.40
4.4
2.8
2.1
(continued)
D-4
-------
TABLE 0-1 (Continued)
Compound
Oichloropheno1(2,4)
01 ch 1 oropropane ( 1 , 2)
Olethylphthalate
DlMthy1phenol(2,4)
01aethy1pheno1(3.4)
Oloethylphthalate
D1n1trophenol(2,4)
Ethanol
Ethyl benzene
Ethylene oxide (oxirane)
Fluoranthene
Fuorene
Formaldehyde
Undane
Nethanol
Methyl ethyl ketone
Methyl ene chloride
Naphthalene
Nitrobenzene
Pentachl orophenol
Phenanthrene
Phenol
Propanol (2)
Styrene
Tet rach 1 oroethane
(& ethene)
Thiobisaethane
Toluene
****•
•g/g-h
25
17
10.7
5.5
2.2
8.0
8.8
6.8
4.2
5.0
18
2.0
22
42.5
11
130
1.5
97
15
6.2
0.16
73.5
L/g-h
3.3
1.4
2.7
4.7
1.05
3.1
0.62
0.9
2.1
0.91
1.5
3.5
0.25
0.031
0.20
0.20
0.40
1.0
2.3
3.4
2.2
13
0.075
0.11
0.68
0.93
2.4
*$L
7.5
12
2.3
5.2
0.71
13
9.8
3.2
4.6
20
90
10
55
42.5
4.8
38
0.67
7.6
200
9.1
0.17
30.6
(continued)
0-5
-------
TABLE D-l (Continued)
.u
Compound mg/g-h L/g-h mg/L
Trichloroethane(l,l,l) 3.5 0.74 4.73
Trichloroethene 3.9 0.88 4.43
Trichlorophenol(2,4,6) 0.26
Trinitrotoluene (alpha) 4.4 0.45 9.8
Xylenes (total) 40.8 1.8 22.7
D-6
-------
Estimate the biodegradation rate constants using the following
Methodology:
— Approximate K^x f ro« available data for K^ for compounds
. of similar structure and/or functional groups; and
— Approximate KI either by using the following correlation:
K^L/h/g) - 0.135 K^ 0.38
where K^ - octanol -water partitioning coefficient
ow
or by using the default (average) value for KI, which Is KI
« 1 L/h/g. and then calculate Ks as: Ks " Kmax/Kl«
The following properties are given for each chemical (listed by name
and Chemical Abstract Source [CAS] number) :
• Molecular weight
• Density
* r Vapor pressure at 25 *C
Solubility
• Henry's law constant
• Diffusion coefficient in water
• Diffusion coefficient in air
• Boiling point
• Coefficients for the Antoine equation for estimating vapor
pressure at temperatures other than 25 *C.
To estimate vapor pressures at temperatures other than 25 *C, the
Antoine equation coefficients are used with the following equation:
109 (10) Vapor Pressure fa* Hg) » A -
where
A, B, and C » the Antoine equation coefficient
T » temperature in *C.
Two approaches may be used to introduce a new compound and its
properties into CHEMDAT7. First, the data for one compound in
CHEMDAT7 may be replaced with data for the compound of interest in the
D-7
-------
columns specified above. With this approach, the list of compounds in
-CH0«AT7 remains constant at 62. The second approach involves append-
ing the new compound and its properties to the existing list of chemi-
cals in CHEMOAT7. All the equations/calculations must then be copied
from one of the existing rows via Lotus 1,2.3 into the appropriate
cells in the new row of the spreadsheet. As mentioned above, the
inclusion in CHEMOAT7 of all or a large part of the chemicals listed
in this appendix could result in increasing the time required to exer-
cise CHEMDAT7 and could prevent its use on some microcomputers.
The properties of interest listed above, with the exception of
the CAS number, mimic those in columns B, D-M, and Q of the CHEMOAT7
spreadsheet.
D-8
-------
2.4.« r
too
WA1OCM
AJOUR
4U3KNV
MOON
AJCIUUM CMOOMJC AOD
AUCOHOL
A
A
0
N
0
0
0
0
N
0
e
o
A
H
0
N
0
N
N
0
N
0
N
0
e
N
o
2BS.48 0
104.21 1^7
182^1 0
44 .78
0 8
0 0
1.0S
.79
.78
128.10 1^8
Tt^C 1.U
19*
223^7
U0.0S
238
12§
121.18
118.18
148
Ski
71.01
0
83.1
188 JOk
287.1
138.13
148.14
324J
.12-
.37
384.91
88,1
78.S3
138
238
UQ.8J
88
182
138.19
98.14
137.18
109.23
98.18
134.1
188.12
188.12
108.13
70.13,
94.12
140
X63.S
20S.C
1S2
182.17
128 J§
212
84^8
17 M
138.23
138.18 .88
8
1.02
8 9
S.2C-10
k
0
0
0
.124
0
0
.113
3*
.124
.128
0
.099
0
0
0
0
0
0
•
0
0
0
.108
.007
.889
.122
4BS9
114
•394
.0774
4774
.259
0
.484
8
.07
0
Ml pfc A
0 1
278 '
20S 1
20.8 1
•8.1 1
8 1
118 i
139.8 1
SB.2 1
tt.8 \
282.3 J
82 <
8
8
8
8
8
8
8
0
0
0
B
0 :
141.0 I
77.4 1
0
0
220
288
0
0
297
0
0
97
48
-38.8
8
8
8
8
8
8
0
302
0
0
174
104 -j
184.
0
273
-33.4
0
148.4
8
184
8 1
B
1 1
r.728 :
r .72819 5
I.00B 1
M5939S:
1 1
r.387 :
M49 1
r.u7 1
r.u9 :
».i3f :
1.949 :
a.29ns ;
5.8S2 e
r.038 j
.927094
90
.923
.0112
•
.387
CKJ4M •
•9O4QB
(
f
1
r.32
1 1
C
1 0
B34.234 241
S4.234 241
L0K.S.7 291
1388.448 22<
1 0
LS33oSJ 22!
1444.718 19!
L2I0C§» 22!
L314.4 23t
J878.8 27;
LU5<.i64 22!
0
0
0
0
0
0
0
0
0
0
0
(839, S/7 27
H8.63? IS
L232.SI 23
0
0
0
0
0
0
1908.751 It
0
a
0
e
496.§@ a
0
0
a
0
LS3g,:i 2
e
2
0
0
a
e
N0
-------
AXXSU
/N1MMCBC
MMOTE
ARSBOCAOD
xnwzae
AJRMOC
aaeueme
B6NZB4 3JLFOMTE
AOD
AOD
BBCETWNTLM
B9CDME
8E«ODICW»e-l,3
uQRwnvec.3,4.
BBCDTWAZDLE
BENZYL MEW*. E1»6R
BHC.b**-
A
0
Z
P
N
N
C
0
c
s
A
0
0
C
A
C
A
0
N
0
A
0
3
C
C
A
A
A
A
A
A
0
C
0
A
P
P
P
P
BIOfCLO[2.2.l3-2,
emem.
BB (2-otXR06T>oco icnwc
r>cn
ET>et
BXS(2-QUMHSOPRaPER
BRCMDB6NZYI. ALCOHOL -(•*
ALOMX. -(e)
BRCKOUROBENZYL AtOHL
BROMOOUOROdHM. ACETATE
emaaaianetme.
ACETATE
BROOCTHNC
BROme«L
BROAJHMJPUNLIKILE 3
8
C
0
C
c
c
c
0
c
0
0
0
c
c
c
c
c
c
c
c
c
c
c
c
100.1*
170.2*
•
3*4.86
141.98
21648
207.37
**>!*
43.1
127
108.13
192
0
70.1
20244
163.17
184.23
287.13
233.3
252.32
122.13
18047
182.23
262.32
108.1
135.1*
195.47
113.57
228.3
262.3
262.32
276.34
252.32
262.32
108.16
126.6
122.10
229.29
298.81
298.81
298.88
290.88
164.2
173.1
228.31
377.7
148
171.1
39147
US
138.9*
16742
18744
1*744
18744
191.48
205
129.39
187
163.8
107.02
106.98
252.77
94.96
249.11
133.98
in
.26
87
.11
.11
.2*
.43
36 88
.08741
.27
.97E-05
.1
987
476
18
,22
,11
.32
.4962
43.5
.636-06 4892
.416-08 4902
a06C~t3B *fl36X
.97
W
1.616
1.3959
25.9
1.876-07
13.4
Ail pt
M
nr
78.9
I
-
231
221
0
436
312
4
0
0
480
0
179.4
0
0
288
0
0
323.4
0
254
218.1
0
21.3
178
189
384
104
136
y
.91
.289244
-
.6424
.033
e
0
1781 273.18
MU.734 1
211. 033
S26.8
1333.3 S
167.0398
20.79
73
.74831 1438.72 181
.34998 2331.4 186
6 0
6 8
a a
a i
.9246 2372.1 !
8
??3=i8
.9824 2428.6 IS§,6
.246506 3724.363 273,16
.19817
.246
99
.86054
e
0
0
c
I
1632.593 172,79
0
0
(
f
g
i
a
<
9
a
a
993.725
(
S2.733
i
a
0
<
9
e
i
9
a
a
438.817 5
206.441
e
a
a
e
0
.49805 942.267 192,587
a a
0 0
0 0
0 I
S
49.6 0 8 0
.60 8. 0
310.1 0 a a
000
7.00782 1612.36 ;
208.38
-------
1904 •
100 0
240 0
241 0
148.42 0
98.U
74
74.14
74.1
U0
•
.79
0
0
0
90.10
78.14
48
'4
0'
10448
10448
72.U
72.12
•8.1
13740
13840
20142
1074
70.1
0
0
0
0
0
0
44
0
0
148
0
1834
U0.ll
88447
147.4
38*48
0
0
1.31
0
1.U
0
0
0
704
0
1274
12747
12747
210.7
1224
17842
3284
18847
18847
18847
21047
1374
11041
13441
1344
1204
0
0
1.11
0
0
0
0
0
0
0
0
0
0 •
0
0
0
0
0
0
0
0
hr
440
88
0
404
m
M
473
toil pt A C
0
0
0
2
-4.4
137.0
9
9
9
9
148
9
370
0
170 .
0
774
0
0
0
0
744
0
1084
0
189
484
-88.1
704
2404
974
9
ITS
9
f
0
98
9
140
9
0
232
0
0
0
(
0
0
0
330
0
237
0
0
0
0
0
0
0
0'
0
0
0
e
s
840 930448 238.86
47481 1314.19 188.56
4708 1382.38 178.77
4708 1382.39 178.77
127 1438.410 210.7<
9 9
98817 1877.988 281.37
9 0
74114 20S8.90* iae.es
988 1380.988 172. U
0 §
.8«0«8« 2028.130 573,16
^BflBaVA
.942
.9830
.86010
48288
48889
.970
436468
•
0
0
0
0
03.89 186.45
ff
g
ff
e
0
8
409.11 241.55
S
1242.43 36
S
0
6
e
e
0
0
0
723.386 i79.9E
9
L988.8 220
2073.78 215
0
0
431.0S 217.5s
0
0
0
0
0
0
e
2300.12$ 194.2
0
0
0
-------
THIOBWK P
OtXRGHiaOM
typ*M*
C 204.67
C 04.62
2C 80.62
C •
toll pt A
.9214
kenm. Beeoic AODC
ouRncnm. Acentae
OUROCnflL B94ZDA1E P
BIMKL ien>e
tenm. ETHER
nem. HETOC
p
OURQMITROeEMB€(-»)
OUROPHEKL
OUROPHEML-2
nem. crHER,4
OUROPHBALErHttO. 1.1
OtflROPWHUC JMADOBE 4
(-0
OUROIEnUHICRORJUN
ouononumou. P
119.4
2ae.c
T4.f
M.Sl
1S4.«
1S7
117 J§
168.0
104
M.K
7I.84
7».S4
87
78.B
113.9
OUOROIUCS..S-
OJOL
OijORO-l,3-OraflPB<»01B€ C
C
C
c
3
0«JQRO-2-teiHCKYGBaQZC ACID 4
ouRO-2-NnRaBeen. /UJOCKL, 4
ouRO-3-KnrnwaLae 4
QURO-VNTTRO < P>Oftl£VCU>e(¥C 2
OtCRO-4AlflMXnjMWN aOMOClUC AOD 2C
ALCOHOL 2 c
3 C
AOD 2 C
3 C
OtflO-4-NnROAWSOLE 2 C
OCORO tflOftLPYRPPC 2 C
CARBJdD AMICE 2C
AOD 4 C
3 c
3 C
OURO-t-PWOODBenM. PM1HALA1E 4 C
OtORO-N4CnftUB
-------
A
0
0
0
0
0
0
A
0
N
0
N
C
C
N
N
N
N
N
N
0
0
H
0
0
H
0
N
N
N
f
M
H
N
ounce*. «,«-onciH«L PHOSJUMDME »
N
A
CfCUMM. tCBXm.
CKUMOLN0C
croLoee
i.a
^
OZMCETII.
2,4
0
N
N
OOUHL EDA
. 1,3
OXOUCNO PWWML 2.3
9OHMIUCIHNC
OKHJOMXIHIL enai
c
c
c
ff
f
6.911
7.
6 199.1
143S.8 MS.:
1SU.08 Ifll.
.17B3S
24M.79I
207.
•41
tan.ss 222.
. 912.87 109.
2137.192 273.
0
Z 188
0
0
0
14184
-------
KHJOROBHVLB6(1,2)
KXJNOF»P*€(1,2)
KnjowpROpeea.3)
AOD<2,4)
HQHRSTYREhC 2,6
3,4
SDURo-2-euiac 1,2
5.5
•OEMIL Ei»e* AOD ounce
3ZJMKL
JEIHtL 94^9 *ffl_B€
JZIMlflBCB^P
BEIWU) We«.
BEIHtLSC CLYCO.
1,1
MI>C H.N
JDCIHn.-1-NnROBBCBC 2.4
typ*Mt dm vp hi 41 dv boil pt A B C
C 99.95 1.21 200 4319
C 98.9*
C 102.92
16341
in
221
us
1U
110.97
110.97
U0.97
172
141
134.5
134.0
125
99.94
269.99
390.99
73.14
74.14
139.5
292
271.39
90
N 149.23
A 124.22
N 208.39
0 10.14
•.2
N 110.47
0 101
A 162.28
299.27
0 78.1
0 244.32
N 40.69
N 135.23
212.21
S 94.2
N 71.69
60.1
S 119
N US
N 74.09
194.2
S 129.14
S 62.12
N •
A 212.99
H U2.21
A 266.33
N 73.19
N 129.15
N 60.1
0 192.16
0 122.17
0 138.19
N 151.19
N 169.1
N 104
N 192.14
N 182.1
N 199
f 364.4
P 240.24
0 99.2
C 322
B 170.21
B 188.28
331 *0B632
1360 921
.12 .0DBBB49
4M4 .6800940
290 .0321
.10 40 .0823
.2 4S .08233
375.941 0
0 0
0 .0129
0 .6886
0 1.41E-07
0 .6873
0 0
.19 2.97 .689389
409 .095
.ffffWBg 6.S4G-09
1.8E-07 .0880594
0 JB731
v •£00200$
0 0
.12 0 All
.008894 0
0 .45
tt .68283 5.746-08
0 .60571
0 .6134
1 .6886014
0 0
0 .0090059
0 0
1 .107
0 9.17E-07
400 .600121
0 .09244
1690 5.24G-0S
0 .69135
0 0
0 .6880016
95 4 .6088192
.9 157 408124
0 0
0 .088286
.605 0 0
.19 .688197 2.1EE-0S
.1 5.966-97
420 .03646
0 0
0 0
.24 .690481
1.62E-09 2.73E-10
.11 29 .080385
0 0
157 0
.04 .0573 .686921
0 0
0 0
0 .68759
•88 *08 •000002
.66 1.170479 1.S3E-07
6 9.11E-09
.31 .0951 4.07E-08
.018 .009914
.091 198099
1 .0812
.68 37 .6880231
.83 1.5ZE-49 .0000312
.075 .62 .60224
0 .00931
•^viT^BO
• /vlf^Bw
•l^^&^BB
.61E-68
.87E-4S
'
6000168
0088169
98081
.29E-0B
*9BB^0B
.64&«6
^6E-69
B00162
WWffiff
0 7J823 1205.4 230.6
0
9
210
290
.6688 47.2
4792 0
112
92.5
77
•
4725
4792
4513
4939
106
104
668
0461
779
203
-
229
104
159
46
t9B
.9
LS3
B3.8
77
8.9
ai.s
202
50
264.7
168.7
21.4
.9661 1141.9 231.9
0 0
.497976 1990.059 199.3598
.999339 1677.797 197 .4598
0 0
.99 1380.1 222.8
0 0
.946299 1246.569 221.6846
6 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
0 0
.8016 583.3 144.1
.92032 1054.07 228.8
0 0
0 0
0 0
0 0
.466 1993.57 218.5
.9962 1586.31 291.97
0 0
.6367 1939.4 162.7
0 0
0 0
0 0
.9933 1053.88 194.41
6 0
0 0
0 0
.08212 980.242 221.67
6 0
0 0
.1609 1196.59 242.68
.928 1400.87 196.43
.409 1305.91 225.53
0
0
0
.522 700.31 51.42
0
.1609 1196.58 242
0
0
0
0
.08212 980.242 221.67
0
.337 229.2 -137
.516903 2048.072 191.7S98
.372 386 -43.6
.799 1118 61.8
0 0
0 0
0 0
.431 1554.88 240.34
L2.88 9466.5 273
0 0
0 0
-------
100.2
1*3
1.1
nowum
t
•
t
•
t
M
N
0
H
0
1.10
0
2)0.24
0
UB 0
19004 •
104J8 0
104.24 0
0
70
390
220
0
•COUffN 3UMV
BM9.
tune 3,3-og»««*.
OHIUNI O-ICOL
03k03m.en0jt
0M4MI CLKQL tOOON*. CROR
N
0
N
0
0
0
C
N
0
0
OS
0
N
A
H
N
C
0
0
0
30 .
40.1
0
0
0
0
0
0
0
1.10
0
0
.70
.137
•5
120
101
6442
104
310.3
•5.1
0
.
0
0
0
0
107^0 2.7
0 1.U
00.22 0
T9.1 0
U0U0 0
14
fUWM
74,1 .n
0 0
U2J7 0
0
1.10
0
0
0
0
0
30
CBJICM3L)
0
0
N
N
N
0
C
0 -U22
12042 1.405
0
1.10
10241 0
• 0
74.1 0
15103 0
4V h
1 450 5
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1 451
0
0
.121
\ .107 1
0 :
0 1
477 1
t .271 1
0 1
1 474 3
0 1
0 1
1 4732 1
0 I
1 475 1
0 1
0 1
0 1
.100 1
0 1
0 1
0 1
0 1
.104 I
474 3
0 1
0 1
0 1
490 3
0 1
0 a
0 1
0 i
0 i
.170 -1
0 1
470 1
.104 -3
0 1
. .104 3
4072 1
0 1
0 J
0 3
0 1
0 1
0 1
0 1
0 1
0 1
0 1
til p* A
•B 0
0
0
0
0
0
120 0
' 0
6.291
0
0
m.
M0 7401507
0
0
0
0
0
LIT 0.22043
0
W*4012fl&0¥
0.321
172 7.400
05 7.074
1 0
mm «•> 4M44^K
2.4 0.900
17.3 0
144 0.92
1 0
1 0
F7 7.101
I 745413
L38.2 6.975
1 0
128 7.337200
1314 6.721
1 649
1 0 1
1 7.0490
1 0
,10.6 0
1 7.120
144 6.92
1 0
1 7.490
1 90
37 0
00.6 0
S00 6.373
1 7.701
I 0
1 0
A 7.195
1 2.602059
00.7 7.501
E9.0 0
1 0
tt.4 6.975
01.7 6.575
1 0
E70 0
M0 0.947832
1 6.105
» 0
1 0
1 0
162 0 '
L9.2 99
1 0
C
1
t
VI
2821.431
2055.813 <
0 t
583.72 S
1710.2S ?
1577.67 1
1043.5 2
9 «
1897.011 2
1030.01 ;
0 1
1054.07 S
0 t
0 1
1244.91 :
957.31 i
1424.251 2
9 8
1521.061 ?
S2B0.82 :
2656.9 3
1 fi
1793.9 S
9 g
9 e
1054.54 ;
1054.07 ;
9 t
1858 3
6 (
9 8
9 8
1758 3
as37.i ;
9 8
8 «
970.6 2
19.62512 :
1699.2 :
9 i
9 (
1059.87 I
1198.7
0 1
0 I
2993.110
1058
i
i
I
t
t
£
1
1
SB
46.1!
m.it
t
58.8:
ar.s:
73 J-
94.2
(
!73.1i
38.0
1
226.8
I
1
S7.8
SB
113.2
r
OS .3
B1.7
03 .5
36.S
f
5
37.7
28. £
>
L87
1
»
1
as
M.:
I
i
{44.:
S3*.:
380.*
>
9
ST.
162.
B
J
174.
20
1
I
I
1
I
1
-------
WWCBO
>epn»c(-«i)
teeiouw
H
M
tEXNMJ
HWZBC
IGOQUNGEIHN'C
100.21 0
100.02 49
272.77 0
294.9 2.04
269.9 1.67
298.98 1.97
272.9 l.n
237
276
116.21
89.2
HiB
H10MZ3NE
WlCROCWNIC AOD
sance
MME
HUCfOCf DXMETHA. ETVB*
^vraD
AOD
4
H
M
H
0
0 102.19
0 108.168
99.4
99.4
144
144
117
117
34.1
.82
947
AOD aa
WTMMJkTE 4
HVtMODMeiMM. tSCPHOPTL ICIOC
H*aOMCIM«Lne«l. CMBMuWTE N
HtCRODMCim., N-teiHA£1Hn. AMDC N
i
PHI>MLATE 4
OCQLE
BOPHOOC
UE»
LEW 9JBACTBMe
IBJC9C
UM
LMfl
0
0
0
N
N
N
N
U
U
7«
161.14
78.CC
114
zm
188 .
1M
131
89
100.5
89.7
91
71.49
•
276.34
117.16
249
138.21
114.1
325.29
897.89
UM
UhOMC
MLAIMZDN
MIL£KACID
MAL£XC AMftORDE
_
OODC
MEIHAOnUCAOD
SLJLPONIC AOD
MEIWKL
v€TH«Pm
VETHOMA.
0
0
N
M
0
0
H
0
S
0
N
97.3
97.3
78.4
78.4
299.89
339.38
118.97
98.1
138.2
98.18
88.1
18.04
98.11
48.11
32
281.38
182.2
.878
.88
16.1188
.087
l£-89
1
1
1
1
0
93
79
9
.1
9
0
1628
114
3
.0114
3.61E-07
.0000113
1.34
4.81E-07
.08419
0
8.11E-97
.0523
.0827
64E-48 .0827
BB0113 .1
0320113
628
.217
.614
.137
638
2S4
16
16
368.68
• C
1331.63 2X2.41
1284.9 216.64
9882922289.189
166.1698
1171.17
761.28
395.67
224.41
196.68
•69.48
1329.6 280.4
1289.37 27S.g?
895.319 2S0.2S
2481 18S
9
897
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
474.08 S9.13
-------
87.12
* M
184
228 •
.0888
1 pi A 8 €
000
0.83629 1121.3 229.887
\ 7.088 US7.83 219.73
r 8.818848 1198.779 224.825
000
7.3388 18U.8 233.3
0 0
0 0
0 0
0 0
« 7.088 949.S8 249.34
0 9
0 9
0 0
0 0
6.828 1270.78 221.42
000
9.0 0.97421 1209.6 ,218
3.027 3.02 -11.9
0.C782 1007.S 181.4
42.4 0
144 0
.078 11S.8 8.872
38.1 0
0 0
.077 101 8.408
MM m
9fcW 9
7.0888
C
8.88408
8
\
•
101 39.8 7.488
8 .098
8 .098
8 jm
8 .078
8 .US
a .us
0
g
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0 •
0
0
0
0
0
0 -40.0 0
0
®
168.4 191.9
9
0
050.5 274.4
826.9<£ 196
640.26* 198.4
&
ee.25 234. s«
0
0
0
.325.9 252.3
S
a
e
0
0
0
0
0
a
a
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
8
0) .091 0 7.71813 1746.8 23S
-------
ttfiBUOOS
(LYOOL
NEftON
NDdNHflDC
NrmoB3«n. ALOMJ. P
Nrrooawew.,4-
NT1KXBJXLOSC
NTT
NrTOFHMX.,2-
Nrmonea.,4-
ALjOMJL 4
NPFOSOMGRPHUC
NlDUUfRRCUDDC N
NTm050-N4CIH«UAEA N
(-p)
ocwte
aawa. i
OCTML2
OCT>M1.3
OCWN3.4
an. _
OCDIETHtUfiETH.TKIOCIM.
OdUCAOD
OOMZCAOD
MM1KION
KB 1010
PCS 1221
PCS 1232
PCB1242
PCS 1248
PCS 1264
PCS 1280
pa's
PBffACHJROETWC
PBfl'AOtflROffTHCBB'CBC
remozse 1,2
IblKANTTRATE
nBMCETIN
,1-OBnetMYLETKrt.)-
173.01
.0
.4
,14
,42
.23
0
10
.111
3.410910
0
0
0
.3
0
10
0
0
0
7
flgflj
1207
8210
8192
99
.20
27.9
0
2.2
0
0
8
0
0
0
0
0
1
0
0
17
.124
0
0
0
0
1.36-13
0
0
0
0
26.3
0
0
0
0
0
0
0
0
0
12.5
.07
2.216-14
4
0 '
.081
.0135
0
0
0
.11
.61
.67,
4.4
0
0
0
0
0
0
.00872
.0973
.021
0
0'
0
0
0
2.236-07
.10
.07
.341
0
0
4.54E-07
0
.0199
Iv
0
.069
.073
.076
r
.059
.067
096
.056
.057
082
bell f* A
217 1
210
0
0
100
300.9
308.1
208
0
0
0
0 •
0
0
0
0
340
0
0
MgtfJ
4009
0
210
279
0
0
164
0
208
226
0
0
0.
0
316
0
126.7
0
0
0
0
0
649.5
0
0
239
0
125
376
0
0
0
0
0
0
0
0
277
162
0
310
0
0
0
147
0
266
340
182
0
0
1
r.0i
r .28421
r.347
.219423
.6596
.868
.116
.28100
.9949
.918
12.0701
.3888
.2215
.7398
.74
.9182
.133
a c
0 i
1733.71 '
2077.68
2136
12170.788
4039.73 !
3338.6 !
1748.0 i
448.94
'
720.39
1361.99 '
4605.8 -
10S0.4
680.3
760.6
0
0
0
0
0
0
0
1378
0
0 1
1104.991
0 1
0 1
0 1
0 I
0 (
0 <
1516.79 '
0 1
0 <
1
201.86
184
183
99.7398
73.16
273.16
201.8
27 .6
84. 9
209. 5t
»9.§
122.6
S4.7
99.6
8
9
9
9
9
197
9
9
228,86
9
9
9
I
}
I
174.95
9
9
-------
hi
MMMJC
N
U
0
0
0
N
N
N
N
0
S
0
0
N
0
N
13043
U7
13844 0
IB 0
175 0
10344 1.34
103.10 0
138)4] 0
1824S 0
298.4 0
03.81 0
34 0
108.14 1.50
147.1 0
93.12 40
1384! 40
88.14 0
288 1.40
24.7
.171
ACEIXIE ISO
zso
PrtODVC
mODMUl
OJMLD9C
ACEI3W1
SOKIE-
SULFGNME
10
SUOONKAC9
U
M
M
M
N
f
A
0
0
S
S
A
C
12844 0
0 0
7440 0
8040 .70
72.1 0
8840 41
7440 0
101.13 47
5940 0
10240 0
17043 0
42.32 0
78.11 14H
58.1 .83
8740 0
10243 40
120.10 40
1524! 0
5840 0
2824 147
78.1 40
338 0
0 0
12840 0
888.7 0
0 0
340 0
133.10 0
102.10 0
28942 0
201.7 0
824M 0
28840 0
284.1 0
•849, 0
20542 0
394.45 0
334.42 0
10*4! 4>
11040 0
98.1 0
1394! 0
3440 0
480 0
240 0
.104
2.8
Ml
0
0
0
287
0
0
320
110.5
8.1
-07.78
724)
284
004*
128.8
0
M0
0
0
0
180
0
141.1
974!
82.4
0
494;
0
0
884!
0
88.4
188
33.9
88
101.8
0
0
113.0
404
115.5
0
0
2S7.7
234.5
4!
145
A B e
0 0
-.108 108.4 <
0 0
34075 3022.8
0 0 •
0.842 941.25
0 0
0 0
8462 2888.8
0 g
7403 1415.73
0.0525 1448.4
0 0
0 0
0 0
0.88838 804
0 0
0 0
0 f
7.84787 1499.21
0.117 1580.92
0 0
(0.6
18.1
230
0
0
273. 1C
0
211.82
268
0
0
9
347 .a*
0
0
g
204.6'
219.6:
0
10.2315 2859.02 -=J4.1S
0.403 951.2
5478- 665.52
8.843410 1231.175
0 0
6.834991 1386.131
0 0
6.77811 770.85
84082432085.9
8478775 1656.8SS
0 0
74U 1282.28
6.95142 1401.297
0
0
0
7441
0
0
0
0
6.924308
0
0
0
0
0
0
0
0
0
0
0
7.14
0
0
0
0
0
373.8
•4.547
74.51
190.3
1S9.1
222. S
0
28.2
0
245.5
233.5
273.1
e
208. e
207.:
14. £
ISS.i-
0
0
0
0
0
0
0
0
Z4.i
9
0
-------
TETWO*HWBBCB€C1,2,3,4)
TE!mOUROB9CB€(l,2,4,6)
1EIRAOlJORODXBB4ZmjWI,2l3X9-
TETRA«jCRCQlBBffl>-p-Oin«N,2,3,7,0-
tEIRAOUROEnwea.1,1.2)
TETRAOCCRCEnee
1ETRAQUaROFtea.(2.3,6,6)
TCTWEIH*. LBO
OLYOOL
TETWUN
naaACEnuzcE
TWceexM. AJCGML r
iwnc»N«E com. At »*»)
TWDWKK
•MmenwcL
typ*Mt
216.9
206.9
23S.9
305.98
321.98
109
108
106.83
106.03
231.9
231.9
179.85
323.46
322.31
322.34
194.28
290.2
72.12
132.2
0
19840
76.13
124.21
0
2X8.36
40.U
130.17
TWDLR6A
iHiajeA,i-(*-otXNcnemj-
MRAM
nrnae
icuee
lOJUBC OIAM&C(2,4)
ICUJB* onsoar*«TE(2.4)
N
N
N
A
N
N
Tauee>r*MiNEt2,4-
TOJJBOIMOh€,2,6-
iai8OZMoe,3,4-
TOUIC AOD (para-)
IQLUXC Aueme
TOlflDDC (-«)
TW»Pte€
TOBjm.
iiuirQOKTE SSS
erAiauBOZMiCDe HCIN
N
N
N
0
0
c
A
A
c
0
0
c
c
c
c
c
c
c
c
c
TOBUTILPHOSWE
IRICHjORQWrSOL£,2,3f»-
TOQLORCBBCB€ 1.2,3
TRIOUDRQBeCBC 1,2,4
TRKHJK8eC&€ 1.3,5
•tfOMJJKBJTtte, 1,2,3
nODUORCEIHIwe 1,1,1
TOOtCROEnwe 1.1,2
noDtavEiHnBC
iRXoureaLaaenwe
naouDROP»ecLC2,4,s-
nXOUXFWa.(2,4,6)
nODUROPR(P*E(l,l,l)
rawjORQprop*€a,i,2)
•RZOtOROPRQP*€(l,2,2)
•RKHJDRQPROP*ea,2,3)
•RXOUOROPROPeC (1,1,2)
•RKMIWa.l^TraHJUQRO
1QDURO(l,l,2)'miaUHCETm«a,2,2)
TOCHJRO-l,2,2-TRIRJUORCEim€,l,l,2-
lOOURO-l.S.^lRZAZDC 2,4,6
91.14
70.12
240.41
138.12
92.4 •
122
174.10
428
122.17
122.17
122.17
138.10
120.14
107.17
107.17
414
346.67
314.64
208.32
211.48
181.46
181.6
181.6
101.46
133.4
133.4
131.4
137.4
197.46
197.48
147.43
147.43
147.38
147.4
146.4
0
107.30
187.38
184.41
0
0
1.88
0
0
0
1.6
1.69
1.624
1.624
0
0
0
1.683
0
0
0
0
0
0
0
0
0
0
0
0
0
1.41
0
0
0
.07
1.11
1.2
0
0
0
0
0
1.03
1.048
.989
1.11
0
0
0
0
0
0
0
0
1.33
1.3
1.4
1.49
0
0
0
0
0
0
0
0
1.58
0
0
.010
72.1
Iv
.071
.071
.072
.072
498
107
487
481
468
478
478
479
487
0
0
471
0
0
471
0
0
.078
0
0
boil pt A B C
264
240
246
0
0
0
146.3
146.2
121.4
121.4
104
0
0
0-
0
0
' 0
0
67
207.6
0
128
0
0
0
316
6
108.7
0
320
182
323
0
0
110.7
0
261
0
140
0
286
275
204
242
209.6
200.4
0
0
0
209
227
221
213
208.5
0
76
74
87
23.8
263
244.6
107
140
124
168
0
0
48
47.7
0 0
0 0
0 0
0 0
0 0
0 0
.898 1386.88 209.74
.631 1226.1 179.9
.976 1388.92 217.63
.976 1386.92 217.53
.986
47056
.964
.427826
.77229
.825311
»
202.29
741.3
L344.8
632.941
.
2317.388
2815.078
8.25
08.26
19.48
11.1388
273.16
273.16
.643 2138.6 392.8
.961 1314.4 209
.518 1018.6 192.7
.884 1043.004 236.38
.903 788.2 43.23
.88 1099.9 227.5
0
00 0
-------
V99X.AOBMI
V9«LOURBf
N
N
0
0
e
o
N
A
H
N
A
N
A
M
0
N
C
M
A
N
A
N
N
0
0
C
M
C
f
X
A
A
A
A
0
e
C
M
199.2 •
199J2 •
94 •
.192.13 A
99.U 9
129^ •
U4^2 •
19* 9
213U1 9
227JI 9
244 JB
377
•
99
132.99 9
1 •
1^4
UKOS 9
97
199 ja
i
9
191.19 .9
122.17 9
149.9 9
191.49 9
toll p* A
m
339.4
•
•
-47.3
242.S
•
•
•
•
319
9
•
•
•
9
9
•
•
9
139
-49.3
9
73
•13.9
9
31.9
399
£
9
139
144.4
139.4
239
9
0
9
.421
.21
C
789.21 20S
9
1296. iS 228.6
9
9
.972 10B9.4 237.2
9 S
19.3999 3816.44 M6.I2
9 0
.099 142S.28S 215.:
.999 1474.87? 23. £
.99 14S3.43 SS.3
.07919 1821.4S 1S9.2
9 (?
9 S
9 9
-------
any u
.7
.24
.34
ACKLBN
QKN9OHUC MSB
/ceoc
MATOOM
NJOUK
AXIUUDC CMBOCIUC AOD N.SLB
JLLYLMJOHL
/ILtL OUQR3CC
>U>HA tenm. sTYiee
icnm. smee oaets
7
OflMfBlf IIIU Hill B 3
ALCDWL 2
2
jHflK) g (HjORCTYRlDJC 2
AMCHM
/ML
AdHXJE
ztnr
.1C
07-44.1
74-M-3
.•17
411188:1;«
7.1
-------
SO* Id
WO CAS
flffllUCOC
4.4»
AOD
AZOBBSfti
BBCNXBNDE
8B4WKONOJM
BBCS! SULFGNKIE
AKSOOC AOD
BBOBC SJJWK AOD
3,4
1.41
2.1S
.«
B&gDTHUZOE
6B40JIKXOLORSC
MJCOHI.
COQCE
BC.!
PTPM Ctflff
S.tt
7.23
2.3
3.M,
4.14
3.72
0
BO (2-OUNOE»
2.M
-.31
BROMUCETOC
ALOMX. -(n4
-(e)
ALCDWL -GO
P
BWUXHJDROBBCYL AU3ML
ACETATE
2.3
1.1
4.2t
BROMVIttPIONllKlIE 3
7.7»
.4
iffieofe
T5-27-* 200
543-a»-2 2
7S-2S-2 3190
m-fiB-3 0
2417-00-
UB-3S-7 10
l»l;l]«
30$
32:0;i
2: :2;<
32JJ
36:l;lf
11'
34:iai
-------
l*m
9
•
•
Mb IAO CM
in too*
4LCBO.2 •
•JMOM-ClfJi
.CS)
BJDML-1
•/TA.
ura. vciu rniwun
•
•
•
*
•
.71
.71
0
•
4.71
BUTtL-»-»«IWDCr
OT
MH)T
OD,2
aUHOBBOQXC ACffl.V
OURCBBOOK AOD,4>
(f«M)
ODt*»
2.*
2.M
CHJOMOBBCn. AJCQO. -(«)
OUROCCUINVN2
atamienM.
S&-TS-* 10
UV-tt-2 2MB
74-U-a 77
C2M-25-1 .2
•73-0-2 Set
17849-38- SOB
•73-78-7 881
O24:l;Znf0Bi
S.9
2
•73-32-S 2
S42-l»-7 2
.2
721
-------
BBV
RtKiO CM
aunenwc
aUROEIMK
OUKEIHlLaC
OROCIHWC
OUKHtOON) 2
BBOCQC AOD
oumeniH. BBOMTE P
otaocww. enw. too«
ouRoenm. tenrn. Ernst
oiooenm. nem. i
2.IS
CMJQROPHTWtJC
OLQROPnPMe.1
l aCAHBBOUC AC3D 2
MJCBCL 2
AOD 2
3
OtCRO-4-NTmWNISOLE 2
0*J3*>**aN33PYRJDB« CAW JkGD AMB6 2
OURO-fr-Cr/MPH1HNJC AOD 4
3
3
Qjjnn IT ptmnnoncTHYL PHTHMJCTE 4
P
P
(TOTAL)
omsoc
CUBIC AQD
COPPER
2.9*
11
•S
.1
10
ma «8 t .2
107-30-2 •
532-27-4
a]
91-69-7 9.74 37:lf2;<
430 1009
88-73-3 2
34:
34:181]
2710 34:lSlf
7085-72-3 3.3
•
•
7S-29-8 9
6C7-fl»-2 10
542-79-7
6S7-&-2 •
1331-2»-«2
•7 .2
31vl7U
31tl21]t
an]«
21S1]«
aa]s
34:ia
U2B-U-1C
-7
S41 Si
248 .2
2
428 20
57479-70-S
-18-
-22-3 108
•
115
2
81 68
2
485
598 50
1088
08 10
271 .2
479 2B0B
1887.11-42
104-82-6 .2
87-48-1 8
7440-47-38
218-01-8 .388
77-82-9 40080
147-14-8 10
28:<;«
-------
Ln
-------
aaemuMa.9
sm u,
OS
2.7«
onuNoonMHicmKMN 3,4
oran
1.2
OSBMLMONI
09&MI OVBI
OOIMI.
OODML
LUB1ML 1KQJOV9
BZTHIL 9*M 4NBJM
OOBMCBC GLYCOU
sw
(D»W»OML>O>€)
oneiMocr
Q&CIHIL OBOVLMOC N^l
B&CIMI. BB«(A)>Nr 7.12
oacnm. oaLLn&c
oocnm.
oocnm.
oacnm. ICIHILTMIDCWMMIC N.N
ooenm.
ODcnrn.
oncnw. SULFATE
oacnm. amnoc
0&CIHIL-1
2,4
2.42
1.S4
2J
2
2.
03MOIP
OBKWCa.4)
OXDCH
om«m.
.4
2.7
.*
.7
!»-• 2
i*
•
!•
•
3414:9
32
-------
BMUMOUiaL
BMLM OLVCEL UMEML RMR
•MUM O.YCQL tOOCMM.
i a.'wa. koocnm. oat ACECKTE
maoKine «B, nan M.T
2.4
.a
.2
TT
.1
2
a
•4-17-1
3l*USsi
2s *s
n-a-t •
23S2[
-------
MHID CM
3.9*
•.47
S.74
•
3.9*
MMKOCAOD
HDCMOL-1
ACS 2
MICMWVAUU ill 4CD
rniwune 4
MICNDCACIHfU
maoenciMnjf«M.
ICIOC
QHBt
maonienm., M4cnm£iHri. AIOM N
-
3
e«vo»€ <
3
3HJOLE
IODOCBAMM4 2
LOO/CCWH
LEWSJWCTEME
MDUIHSM
MDLBKAOD
tcnwanucAOD
MRWWC 9ULKMK AC3D
MEnwcmra.
MEMWO.
MEIHQUM.
^
.n
2.14
.2
737 ma
723 2
as UB
273 A
-73-4 UB
i47*-e*-e
•72-9
•1
U0»l»»7
141
.73-1 "am
7430-97-6 •
141-79-7 2MB
7»-tt-« IflBi
74-C2-C 24
78-T3-2 2BBB
7443-1 23BBI
'1 •
•
18782-77- 2BB
-------
mm.
rnirnrn T
mm.
mm.
mm.
MUML
mm,
mm.
mm.
AGO
ICIHVUVl
•
• MUV «^4* t«
MWLJTV N
74
U*-U-7 2
4M U
«
•
U
M-» •
473
4Ti
•1*.
tf
2
•
-------
ODV tt
MM!
n
.17t
rau
monwa.
k^MMk
ACID
ACETA1E 190
xso
MOD3ME
awmrac
UMUM
S4OC.CYOJ
SM04NQN
sn^ex
SDMZ2NE
9DDXUN AOEKKIE
SULMTE
sooam
sncpiozunxjN
S.18
•t
ETKBCf-
Z.3C
•1-0-4
-------
*MO Ctt
.*
.4
9:2;]
ant
UK74-U-
-------
TPMOI mm nacg,2,a,4)
nmxH, i • •• i »a€gt2>a.c
TCTOOll M'BH IPCCl.2,4,0
MklflO CM
.•
TWOBBCW. JbOQHL P
iMociMMK cnm. M aa«.)
THffiWMK
•navwa.
TWDUW
Twai«A,
HOMM
10LBC
10LOC OOMNIC2.4)
icuec
IRSBUTVL fH BH I H II Bill III K> 1 1
1.14
44
7».4t
.8
.S
.9
.4
74
74
3434*3
44:10ttQf
79-O-6
-------
-t.7*
.4
a.as
.1
.32
-------
TECHNICAL REPORT DATA
DA
b*f
NWOHT NO.
TITUI AND SUSTITIS
Hazardous Waste T-reatment, Storage, and Disposal
Facilities (TSCF) - Air Emission Modsls
MIPONT DAT!
- April 1989
•. PCHFOflMING ORGANIZATION COOC
AUTHOR!*)
10. PROGRAM CUEMINT NO.
PCRPORMIMO ORGANIZATION NAMC AND AOORI5S
Air Quall€y_P_lajffiing_and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
11. CONTHACT/BnANT NO.
68-02-4326
. .^w~~~.~~i AGtNCY-NAMC AND AOORSSS
DAA for Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
13. TVM Of MITOMT AND *««IOO COVIREO
1*. SPONSORING AOINCY COOC
EPA/200/004
NOTIS
i. Aesf RACT
Analytical models are presented for estimating air emissions form hazardous
waste treatment, storage, and disposal facilities (TSDF). Air emission models havt
been developed for aerated and nonaerated surface impoundments, land treatment ^
facilities, landfills, and wastepiles. Emission model predictions are compared.to
available field data. This report also includes emission factors for transfer,^.
storage, and handling operations at TSDF. " ' >^&
The models have been assembled into a spreadsheet that is included in this
report as floppy diskette for use on a microcomputer.
Appendices include a list of physical-chemical properties for approximately
700 compounds and a comprehensive source list of pertinent literature in addition
to that cited in the report.
7.
K«Y \*0«OS AND OOCUMKNT ANALYSIS
-OSSCftfTOMS
Air Pollution
Hazardous Waste Air Emissions
TSDF Air Emission*
Air Emission Models
.uiafRISUTION STATINS*,.
Release unlimited-^-available from MTIS,
5285 Port Royal Road, Springfield, VA
22161
fe.lOINTIIMCRS/OMN INOKO TKMMS
Air Pollution
M! SiCURlTV CLASS ITM* 9*9*1
Unclassified
c. COSATI Fieid/Gfoup
------- |