&EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/3-80-025
December 1980
Air
Organic
Manufacturing
Volume 3: Storage,
Fugitive, and Secondary
Sources
-------
EPA-450/3-80-025
Organic Chemical Manufacturing
Volume 3: Storage, Fugitive,
and Secondary Sources
Emission Standards and Engineering Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
December 1980
-------
Ill
This report was furnished to the Environmental Protection Agency by IT Enviro-
science, 9041 Executive Park Drive, Knoxville, Tennessee 37923, in fulfillment
of Contract No. 68-02-2577. The contents of this report are reproduced herein
as received from IT Enviroscience. The opinions, findings, and conclusions
expressed are those of the authors and not necessarily those of the Environmen-
tal Protection Agency. Mention of trade names or commercial products is not
intended to constitute endorsement or recommendation for use. Copies of this
report are available, as supplies permit, through the Library Services Office
(MD-35), U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina 27711, or from National Technical Information Services, 5285 Port
Royal Road, Springfield, Virginia 22161.
D124R
-------
CONTENTS
Page
INTRODUCTION vii
Product Report Page
1. STORAGE AND HANDLING 1-i
2. FUGITIVE EMISSIONS 2-i
3. SECONDARY EMISSIONS REPORT 3-i
-------
Vll
INTRODUCTION
A. SOCMI PROGRAM
Concern over widespread violation of the national ambient air quality standard
for ozone (formerly photochemical oxidants) and over the presence of a number
of toxic and potentially toxic chemicals in the atmosphere led the Environ-
mental Protection Agency to initiate standards development programs for the
control of volatile organic compound (VOC) emissions. The program goals were
to reduce emissions through three mechanisms: (1) publication of Control Tech-
niques Guidelines to be used by state and local air pollution control agencies
in developing and revising regulations for existing sources; (2) promulgation
of New Source Performance Standards according to Section lll(b) of the Clean
Air Act; and (3) promulgation, as appropriate, of National Emission Standards
for Hazardous Air Pollutants under Section 112 of the Clean Air Act. Most of
the effort was to center on the development of New Source Performance Stan-
dards .
One program in particular focused on the synthetic organic chemical manufactur-
ing industry (SOCMI), that is, the industry consisting of those facilities
primarily producing basic and intermediate organics from petroleum feedstock
meterials. The potentially broad program scope was reduced by concentrating on
the production of the nearly 400 higher volume, higher volatility chemicals
estimated to account for a great majority of overall industry emissions. EPA
anticipated developing generic regulations, applicable across chemical and
process lines, since it would be practically impossible to develop separate
regulations for 400 chemicals within a reasonable time frame.
To handle the considerable task of gathering, assembling, and analyzing data to
support standards for this diverse and complex industry, EPA solicited the
technical assistance of IT Enviroscience, Inc., of Knoxville, Tennessee (EPA
Contract No. 68-02-2577). IT Enviroscience was asked to investigate emissions
and emission controls for a wide range of important organic chemicals. Their
efforts focused on the four major chemical plant emission areas: process
vents, storage tanks, fugitive sources, and secondary sources (i.e., liquid,
solid, and aqueous waste treatment facilities that can emit VOC).
121H
-------
ix
B. REPORTS
To develop reasonable support for regulations, IT Enviroscience gathered data
on about 150 major chemicals and studied in-depth the manufacture of about
40 chemical products and product families. These chemicals were chosen consid-
ering their total VOC emissions from production, the potential toxicity of
emissions, and to encompass the significant unit processes and operations used
by the industry. From the in-depth studies and related investigations, IT
Enviroscience prepared 53 individual reports that were assembled into 10 vol-
umes. These ten volumes are listed below:
Volume 1
Volume 2
Volume 3
Volume 4
Volume 5
Volume 6-10
Study Summary
Process Sources
Storage, Fugitive, and Secondary Sources
Combustion Control Devices
Adsorption, Condensation, and Absorption Devices
Selected Processes
This volume covers the generic VOC emission studies for storage and handling
emissions, fugitive emissions and secondary emissions. The focus of the reports
is on control of new sources rather than on existing sources in keeping with
the main program objective of developing new source performance standards for
the industry. The reports do not outline regulations and are not intended for
that purpose, but they do provide a data base for regulation development by the
EPA.
-------
REPORT 1
STORAGE AND HANDLING
D. G. Erikson
IT Enviroscience
9041 Executive Park Drive
Knoxville, Tennessee 37923
Prepared for
Emission Standards and Engineering Division
Office of Air Quality Planning and Standards
ENVIRONMENTAL PROTECTION AGENCY
Research Triangle Park, North Carolina
September 1980
D20A
-------
CONTENTS OF REPORT 1
I- ABBREVIATIONS AND CONVERSION FACTORS
II. CHARACTERIZATION AND DESCRIPTION
A. Introduction
B. Storage Tanks
C. Loading and Handling
D. References
III. EMISSIONS
A. Sources
B. Calculations
C. Industry Emissions
D. References
IV. CONTROL TECHNOLOGY
A. Storage Tanks
B. Loading and Handling
C. References
V. IMPACT ANALYSIS
A. Environmental and Energy Impacts
B. Control Cost Impact
C. References
VI. ASSESSMENT
A. Summary
B. Supplemental Information
Page
1-1
II-l
II-l
II-2
11-10
11-12
III-l
III-l
III-3
III-7
111-17
IV-1
IV-1
IV-18
IV-20
V-l
V-l
V-12
V-25
VI-1
VI-1
VI-2
APPENDICES OF REPORT 1
A. EMISSION CALCULATION SUPPLEMENTS
B. ALTERNATIVE FLOATING-ROOF-TANK EMISSION ESTIMATION EQUATIONS
C. COST ESTIMATE DETAILS AND CALCULATIONS
A-l
B-l
C-l
-------
1-v
TABLES OF REPORT 1
Number Page
II-l Number of Tanks and Storage Volume by Tank Size II-3
II-2 Number of Tanks and Storage Volume by VOC Vapor Pressure II-6
11-3 Vapor Pressure and Volume of Synthetic Organic Chemicals
Shipped 11-11
I1-4 Time Required for Shipments 11-11
III-l 1977 Storage-Tank VOC Emissions by Tank Size and Type
(Assuming No Controls) III-8
III-2 1977 Storage Tank VOC Emissions by Material Vapor Pressure
and Tank Type (Assuming no Controls) III-9
III-3 1977 Loading and Transportation VOC Emissions by Material
Vapor Pressure (Assuming no Controls) III-ll
III-4 Current 1977 Storage-Tank VOC Emissions by Tank Size and Type
(with Controls) 111-13
III-5 Current 1977 Storage-Tank VOC Emissions by Vapor Pressure and
Tank Type (with Controls) 111-14
III-6 Current 1977 Loading and Transportation VOC Emissions
by Material Vapor Pressure (with Controls) 111-15
V-l Model Parameters Used for Assessing Impacts of Industry
Storage Tanks v~2
V-2 Uncontrolled VOC Emissions vs Absolute Vapor Pressures of
Material Stored for Model Fixed-Roof Tanks V-2
V-3 VOC Emission Reduction from Replacing Fixed-Roof Tanks with
Internal Floating-Roof Tanks vs Absolute Vapor Pressures of
Materials v~3
V-4 Model Parameters and Operating Conditions for Assessing Impacts
of Refrigerated Vent Condensers V-5
V-5 VOC Emission Reduction from Fixed-Roof Tanks Before and After
Refrigerated Condenser Add-on V-6
V-6 Energy Impact from Refrigerated Vent Condenser V-7
V-7 VOC Emissions from Model Floating-Roof Tank vs Absolute Vapor
Pressures of Material Stored v~a
-------
1-vii
TABLES (continued)
Number Page
V-8 VOC Emission Reduction from Installation of New Pressure Vessels
Instead of New Floating-Roof Tanks as a Function of Stored-Material
Vapor Pressures V-10
V-9 Control Parameters of Model Loading Terminal for Refrigeration
Systems V-ll
V-10 VOC Emissions from Model Loading Terminal Before and After
Addition of Controls V-ll
V-ll Cost Factors for Computing Annual Costs V-13
V-12 Installed Capital and Operating Costs for Internal-Floating-Roofs V-13
V-13 Cost Effectiveness of Installing an Internal Floating Roof in a
Fixed Roof-Tank V-15
V-14 Cost Effectiveness for Add-on Refrigerated Vent Condenser to Fixed-
Roof Tank (New and Retrofitted Sources) V-17
V-15 Installed Capital and Operating Costs for Pressure Vessels V-20
V-16 Cost Effectiveness for Installing a New Pressure Vessel Instead
of a New Floating-Roof Tank (Incremental Cost Difference) as a
Function of Stored-Material Vapor Pressure V-21
V-17 Installed Capital and Operating Cost for Model-Loading-Terminal
Control Systems (950 m3/day) V-24
V-18 Cost Effectiveness for Loading-Terminal (950 m3/day) Control
Systems vs Material Vapor Pressure V-24
A-l Paint Factors for Fixed-Roof Tanks A-3
A-2 Tank, Type, Seal, and Paint Factors for Floating-Roof Tanks A-4
A-3 "S" Factors for Loading Loss Calculations A-5
B-l Emission Factors K and and n B-5
D
B-2 Summary of Emission Factors K and m for Floating Roofs B-5
r
B-3 Fitting Multipliers B-6
C-l Cost of Building a Fixed Roof Tank C-4
-------
1-ix
TABLES (Continued)
Number Page
C-2 Cost of Installing a Contact Single Seal Internal Floating Roof C-5
C-3 Cost of Installing a non-Contact Single Seal Internal Floating Roof C-6
C-4 Cost of Building an External Floating Roof Tank C-7
C-5 Pressure Vessel vs Floating-Roof Tank Installed Capital and
Operating Cost Differential C-10
FIGURES OF REPORT 1
II-l Number of Tanks and Storage Volume vs Tank Size II-4
II-2 Number of Tanks and Storage Volume vs Tank Type and Use II-5
II-3 Average Number of Turnovers of VOC Storage Tanks vs
Tank Size II-8
IV-1 Rim-mounted Secondary Seals on External Floating Roofs IV-3
IV-2 Metallic Shoe Seal with Shoe-Mounted Secondary Seal IV-4
IV-3 Rim Mounting of a Secondary Seal on an Internal Floating Roof IV-7
IV-4 Activated-Carbon Adsorption System IV-8
IV-5 Thermal Oxidation Unit IV-10
IV-6 Refrigerated Vent Condenser System IV-11
V-l Cost Effectiveness of Controlling Emissions by Retrofitting
Existing Fixed-Roof Tanks with Floating Roofs V-16
V-2 Cost Effectiveness of Controlling Emissions by Refrigerated Vent
Condenser Added On to Fixed-Roof Tank V-18
V-3 Cost Effectiveness of Controlling Emissions (100%) by
Installing New Pressure Vessel vs New Floating-Roof Tank V-22
V-4 Cost Effectiveness of Controlling Emissions (98%) by
Installing New Pressure Vessel vs New Floating-Roof Tank V-23
C-l Installed Capital Cost vs Condenser Area for Various Materials of
Construction for a Complete Condenser Section C-8
-------
Number
1-xi
FIGURES (Continued)
C-2 Installed Capital Costs vs Refrigeration Capacity at Various
Coolant Temperatures for a Complete Refrigeration Section C-9
C-3 Installed Capital Cost of Carbon Adsorption Systems C-ll
C-4 Total Installed Capital Cost for Thermal Oxidation Systems with
Waste-Gas Heat Content = 10 Btu/scf, Residence Time = 0.5 sec, and
Combustion Temperature = 1600°F C-12
-------
1-1
ABBREVIATIONS AND CONVERSION FACTORS
EPA policy is to express all measurements used in agency documents in metric
units. Listed below are the International System of Units (SI) abbreviations
and conversion factors for this report.
To Convert From
Pascal (Pa)
Joule (J)
Degree Celsius (°C)
Meter (m)
Cubic meter (m3)
Cubic meter (m3)
Cubic meter (m3)
Cubic meter/second
(m3/s)
Watt (W)
Meter (m)
Pascal (Pa)
Kilogram (kg)
Joule (J)
To
Atmosphere (760 mm Hg)
British thermal unit (Btu)
Degree Fahrenheit (°F)
Feet (ft)
Cubic feet (ft3)
Barrel (oil) (bbl)
Gallon (U.S. liquid) (gal)
Gallon (U.S. liquid)/min
(gpm)
Horsepower (electric) (hp)
Inch (in.)
Pound-force/inch2 (psi)
Pound-mass (Ib)
Watt-hour (Wh)
Multiply By
9.870 X 10"6
9.480 X 10"4
(°C X 9/5) + 32
3.28
3.531 X IO1
6.290
2.643 X 102
1.585 X 104
1.340 X 10"3
3.937 X 101
1.450 X 10"4
2.205
2.778 X 10"4
Standard Conditions
68°F = 20°C
1 atmosphere = 101,325 Pascals
PREFIXES
Prefix
T
G
M
k
m
M
Symbol
tera
giga
mega
kilo
milli
micro
Multiplication
Factor
1012
109
106
103
10"3
io"6
Example
1 Tg =
1 Gg =
1 Hg =
1 km =
1 mV =
1 pg =
1 X IO12 grams
1 X IO9 grams
1 X 10G grams
1 X IO3 meters
1 X IO"3 volt
1 X IO"6 gram
-------
II-l
II. CHARACTERIZATION AND DESCRIPTION
A. INTRODUCTION
The synthetic organic chemical manufacturing industry (SOCMI) is a segment of
the domestic chemical industry that produces the 350 to 400 basic and intermed-
iate organic chemicals used to produce other, finished chemicals and products.
Organic chemicals not included in the SOCMI are refinery by-products, coal tar
and other "naturally" derived organic chemicals, polymers, and end-use (finished)
organic chemicals.
The data base for the synthetic organic chemicals manufacturing industry (SOCMI)
for this report was compiled from state emissions inventory questionnaires (EIQ)
and information obtained from plant site visits.
Texas and Louisiana State EIQs supplied an estimated 80% of the tanks in the
data base, which covered approximately 4000 tanks used for the storage of organic
1 2
chemicals. ' The use of only two state EIQs to supply the bulk of the storage
tank data base was reasonable since production in these states is estimated to
be 65 wt % of the total SOCMI production.3
The data collected included various tank characteristics, type, use, and con-
tents. This information was assembled on forms and stored in a computer data
base developed by Mitre Corp., Metrek Division, McClean, VA, who used the MRI
system 2000 to manage the data base. The program has the capability of computing
emissions from each tank and of characterizing the data base in terms of various
4
input parameters, such as tank size, vapor pressure, tank type, and emissions.
Extrapolation of the data base to the SOCMI was done on a weight basis. It was
estimated that the 4000 tanks represented approximately 20% of the total SOCMI
production. Individual tanks in the data base were assigned to a specific process
at a given site. The production from each process was summed to yield a total
production for the data base. A comparison of the reported SOCMI production
with the industrial production derived from the data base indicated that a scaleup
factor of 5.0 was required for completion of the SOCMI storage characterization
projection for 1977.
-------
II-2
B. STORAGE TANKS
There are three basic designs for tanks storing volatile organic compounds
(VOC) in the SOCMI: fixed roof, floating roof [open or covered (internal)],
and pressure (low or high) vessel. These tanks are usually loaded by submerged
or bottom filling and are unloaded into marine vessels, tank cars, tank trucks,
or pipe lines. It is estimated that 20,300 VOC storage tanks, each with a
capacity of 3.8 m or greater, are located throughout the United States and
Puerto Rico. The total VOC storage volume is estimated to be 11,650 X 10 m
(see Table II-l and Fig. II-l). Approximately 14,800 fixed-roof tanks make up
61% (7100 X 10 m ) of the total storage volume, 700 floating-roof tanks make
3 3
up 20% (2340 X 10 m ), and 4800 pressure tanks make up the remaining 19% (2210
X 103 m3).
The tanks are used for storing a variety of materials: raw materials, inter-
mediates, final product, and/or usable by-products, as well as waste tars,
residues, and nonusable by-products. It is estimated (see Fig. II-2) that of
the total number of tanks the feed storage tanks comprise 23%, or 27% of the
total storage volume; product storage tanks comprise 45%, or 53% of the total
storage volume,- intermediate tanks comprise 29%, or 17% of the storage volume;
and waste storage tanks comprise 3% of both the total number of tanks and the
total storage volume.
The vapor pressure of the material stored is a major factor in the choice of
tank type used. Table II-2 gives the number of tanks and the total volume by
vapor pressure of the chemicals stored in each of the three basic tank types.
Fixed-roof tanks are normally used for storing materials with vapor pressures
up to 34.5 kPa, floating-roof tanks when vapor pressures are in the range of
6.9 to 34.5 kPa, and pressure tanks when vapor pressures are greater than 51.7
kPa. Other factors, e.g., material stability, safety hazards, health hazards,
and multiple use, can also affect the choice of tank type for a particular
organic chemical.
Small tanks are commonly used for high-turnover service, such as intermediate
storage. Very large tanks are used for low-turnover service, such as terminal
product storage. A correlation has been found between tank turnovers and tank
size. In general, as the tank size increases, the turnovers decrease. For
very small tanks (less than 35 m ) turnovers can approach 700 per year. Very
-------
Table II-l. Number of Tanks and Storage Volume by Tank Size*
Fixed Roof
Tank Size
(m3)
4 to 45
45 to 90
90 to 190
190 to 380
380 to 950
950 to 1900
1900 to 3800
3800
Total
No. of
Tanks
5,105
3,010
1,980
1,375
1,560
975
520
290
14,815
Storage Volume
(in )
124,000
205,000
282,000
396,000
968,000
1,416,000
1,428,000
2,278,000
7,097,000
Floating Roof
No. of Storage Volume
Tanks (m )
0
10
40
85
140
95
125
175
670
0
600
6,000
26,000
86,000
140,000
331,000
1,752,000
2,341,600
Pressure Vessel
No. of
Tanks
1,370
785
1,190
450 '
330
410
125
125
4,785
Total
Storage Volume No. of
(m ) Tanks
34,000
54,000
153,000
119,000
209,000
473,000
283,000
887,000
2,212,000
6,475
3,805
3,210
1,910
2,030
1,480
770
590
20,270
Storage Volume
(m )
158,000
259,600
441,000
541,000
1,263,000
2,029,000
2,042,000
4,917,000
11,650,600
*See ref. 1 (includes only tanks with capacity greater than 3.8 m ).
U)
-------
5200
4800
e
*>
o
0
4000
3bOO
3200
2800
2400
2000
1600
1200
800
400
I
6500
6000
5500
5000
4500
3500
3000
2500
2000
1500
1000
500
H
H
I
23 68 140 284 662 1419 2839 8327
Mid-Range Tank Size (m )
23 68 140 284 662 1419 2839 8327
Mid-Range Tank Size (m )
Pressure Vessel
Floating Roof
Fixed Roof
Fig. II-l. Number of Tanks and Storage Volume vs Tank Size (ref. 1)
-------
13,000
12,000
11,000
10,000
9,000
'E 8,000
i
o
i — i
7,000 T
> 6,000 '"£
g, |
in ^
5,000 s
i/j \
i
4,000
3,000
2,000
1,000 I
0 1
— 1
1
J
3
is
1 1
^^^ r^ i
I
i
; n i
2 LJ LJ b
21 ,OOO
£S 19,500
jl 18,000
I
^^ MiJ! 16,500
^ 15,000
^
Y//////////A
I
M 1
Y/////////A
1
z
1.3,500
LH
12,000 -g
IB
10,500 0
0)
9,000 |
7,500
H
H
6,000 1
4,500
3,000
1,500
Fixed Floating Pressure All Fixed Floating Pressure All
Roof Roof Vessel Tanks Roof Roof Vessel Tanks
Feed
Intermediate
Product
Waste
Fig. I1-2. Number of Tanks and Storage Volume vs Tank Type and Use
-------
Table II-2. Number of Tanks and Storage Volume by VOC Vapor Pressure*
Fixed Roof
Vapor Pressure
(kPa)
0.0001
1.4
3.5
6.9
10.3
20.7
34.5
51.7
69.0
103.4
172.3
344.7
to 1.4
to 3.5
to 6.9
to 10.3
to 20.7
to 34.5
to 51.7
to 69.0
to 103.4
to 172.3
to 344.7
to 517.0
517.0+
Total
No. of Storage -Volume
Tanks (m )
7,400
2,070
1,955
980
1,475
600
165
90
80
0
0
0
0
14,815
2,718,000
1,035,000
951,000
653,000
1,188,000
344,000
102,000
47,000
59,000
0
0
0
0
7,097,000
Floating Roof
No. of
Tanks
55
55
45
35
305
140
20
0
15
0
0
0
0
670
Pressure Vessel
Storage Volume No. of
(m ) Tanks
134,000
71,000
144,000
319,600
1,009,000
602,000
41,000
0
21,000
0
0
0
0
2,341,000
330
180
140
140
405
225
190
155
230
390
945
600
855
4,785
Total
Storage -Volume No. of
(m ) Tanks
25,000
20,000
59,000
45,000
43,000
49,000
31,000
108,000
63,000
390,000
501,000
682,000
196,000
2,212,000
7,785
2,305
2,140
1,155
2,185
965
375
245
325
390
945
600
855
20,270
Storage Volume
(in )
2,877,000
1,126,000
1,154,000
1,017,600
2,240,000
995,000 H
174,000 ^
155,000
143,000
390,000
501,000
682,000
196,000
11,650,600
*See ref. 1 (includes only tanks with capacity greater than 3.8 m3) .
-------
II-7
large tanks (greater than 7000 m ) normally have 12 turnovers or less per year.
All the data collected and the smooth curve fitting the correlation appear in
Fig. II-3.1
Fixed-Roof Tanks
Fixed-roof tanks consist of a steel cylindrical shell with a stationary roof.
The design may vary from cone-shaped roofs to flat roofs. The tank is designed
to operate at only slight internal pressure or vacuum and thus is highly sus-
ceptible to emissions from breathing, filling, and emptying. The fixed-roof
tank is the least expensive to construct and of all the tank designs now in use
is generally considered to be the minimum acceptable standard for storage of
volatile organic liquids. Some fixed-roof tanks are equipped with a conserva-
tion vent, which is a type of a pressure- and vacuum-relief valve designed to
contain minor vapor volume changes. The conservation vent reduces the amount
of breathing loss by not venting unless the design pressure differential is
exceeded. However, since the fixed-roof tank can be operated only within very
small pressure differentials (±0.2 kPa), the emission reduction may be as low
6 7
as 5%, depending on the vapor pressure of the stored material. ' Generally,
7
conservation vents are used for low-vapor-pressure VOC storage (<10.5 kPa).
Most fixed-roof tanks are small; 55% are smaller than 91 m and contain only 5%
of the total volume of liquid stored. Large tanks, 946 m or greater, which
comprise only 12% of the existing fixed-roof tanks, contain 72% of the total
volume of VOC stored (see Tables II-l and II-2). It is estimated that 36% of
the fixed-roof tanks (5345 tanks) contain liquids with vapor pressures in
excess of 3.5 kPa. These tanks
fixed-roof tank storage volume.
excess of 3.5 kPa. These tanks comprise 47% (3344 X 103 m3) of the total VOC
Floating-Roof Tanks
The open-top floating-roof tank consists of a welded cylindrical steel wall
equipped with a metal deck or roof that is free to float on the surface of the
stored liquid, rising and falling with the liquid level in the tank. This
feature reduces evaporation losses by minimizing the vapor space. The liquid
surface is completely covered by the floating roof except for a small annular
space between the roof and the shell. A sliding seal attached to the floating
-------
II-8
10,000
01
N
-H
-H ~: -t--~-4—
"-i -
30
i
i. .4-:-! - i--i
20
10
10
20
30 40 50 100 200 300 400 500
Number of Tank Turnovers Per Year
1000
Fig. 11-3. Average Number of Turnovers of VOC Storage Tanks vs Tank Size (ref. 1)
-------
II-9
roof fits against the tank wall and covers the remaining area. The primary
source of VOC emissions is through gaps in this seal.
Most emissions through the seal are attributable to wind effect. Wind blowing
across the tank creates pressure differentials around the edges of the floating
roof so that air flows into the seal on the leeward side, and air saturated
with VOC flows out on the windward side. Improper or loose fit of the seal
Q
significantly increases the emission rate.
The internal-floating-roof tank is a fixed-roof tank that has an internal roof
or deck floating on the liquid surface. If the internal roof is made of steel,
the tank is referred to as a "covered-floating-roof" tank. If the roof is made
of a material other than steel, the tank is referred to as an "internal-floating-
cover" tank.
Like the open-top floating roof, the internal floating roof has an attached
sliding seal that fits against the tank wall. Antirotational guides are provided
to maintain proper roof alignment. Special vents are installed on the fixed
roof or on the walls at the top of the shell to minimize the flammable vapor
concentrations in the vapor space. The fixed roof protects the floating roof
or deck and the seal from deterioration due to meterological effects, eliminates
the possibility of the roof sinking because of excess rain or snow loads, and
decreases the wind-induced pressure differential around the roof.
Floating-roof tanks comprise only 3.0% of the total number of storage tanks in
the industry. They generally have very large capacities, with 80% of the total
having capacities in excess of 379 m and 26% in excess of 3785 m . The major-
ity (72%) are used for storing VOC with vapor pressures of 6.9 to 34.5 kPa.
raj
1
This comprises 82% (1930 X 103 m3) of the total floating-roof tank storage
volume.'
Pressure Tanks
Pressure tanks are designed to withstand relatively large internal pressures.
They are generally used for storing highly volatile and/or toxic materials and
are constructed in various sizes and shapes, depending on the operating pres-
sure range. Noded spheroid and hemispheroid shapes are generally used for
-------
11-10
low-pressure tanks (117 to 207 kPa); horizontal cylinder and spheroid designs
are generally used for high-pressure tanks (up to 1827 kPa). Because pressure
tanks are generally operated in a closed system at the pressure of the stored
material, losses are not generally incurred. Pressure tanks with inert gas
blanketing are a minor exception and are discussed in Sect. IV.
C. LOADING AND HANDLING
Loading and transportation operations of the SOCMI are very similar to those of
the petroleum liquids (refinery) industry except that they are on a relatively
smaller bulk scale. Synthetic organic chemicals are transported from the initial
producing sites to consumers. The consumers in most cases are other industrial
organic chemical producers, who use the basic and intermediate organic chemi-
cals as raw materials to manufacture additional intermediates or finished products.
An estimated 33,925,000 Mg (35,390 X 10 m ) of synthetic organic chemicals was
shipped in 1976. Water transportation was the type of shipment most often
used, accounting for 50% (17,695 X 10 m ) of the total volume,- tank-car rail
3 3
shipments accounted for 30% (10,617 X 10 m ) of the total, and truck shipments
for the remaining 20% (7078 X 10 m ). The quantities of chemicals at specific
ranges of vapor pressure that were shipped were estimated on the basis of propor-
tional volume and vapor pressure ranges of the chemicals stored. Shipment volumes
based on the percentages given in Table II-2 are shown in Table II-3.
An estimated 81% of all VOC shipped have a vapor pressure of less than 34.5 kPa.
Approximately 14.8% (5246 X 10 m ) have a vapor pressure of greater than
2
103.4 kPa and are assumed to be stored and loaded in closed systems with negli-
gible VOC loss.
Transit losses were calculated by computing a breakdown of the time required
9
for each shipment based on the 1972 Census of Transportation. The time required
for most shipments (45 wt %) (see Table II-4) is approximately one day.
-------
11-11
Table II-3. Vapor Pressure and Volume of Synthetic Organic
Chemicals Shipped*
Vapor Pressure
(kPa)
0.001 to 1.4
1.4 to 3.5
3.5 to 6.9
6.9 to 10.3
10.3 to 20.7
20.7 to 34.5
34.5 to 51.7
51.7 to 69.0
69.0 to 103.4
>103.4
Total
*Refs. 1 and 5.
Table II-4.
Fraction^
of a Week
1/7
2/7
3/7
7/7
Total
Percentage of
Volume Stored
24.7
9.7
10.0
8.8
19.4
8.6
1.5
1.3
1.2
14.8
100.0
Time Required for Shipments
Volume
Shipped
(X 10* n3)
8,743.
3,429
3,543
3,111
6,866
3,043
530
454
424
5,246
35,389
a
Percentage0 of Volume Shipped
Material Shipped (X 10 m3)
45.0
24.0
14.0
17.0
100.0
15,925
8,493
4,955
6,016
35,389
Refs. 1, 5, and 8.
3Ref. 8.
:Ref. 1.
-------
11-12
D. REFERENCES*
1. SOCMI storage-tank data base compiled by IT Enviroscience, Inc., Knoxville, TN,
using a computer program developed by Mitre Corporation, McLean, VA, April 1978
(on file at EPA, ESED, Research Triangle Park, NC; compiled from State EIQ
files, Site Visit Trip Report, and USITC report on 1976 SOC Production and
Sales).
2. Texas/Louisiana State Emission Inventory Questionnaires (1975) (on file at EPA,
ESED, Research Triangle Park, NC).
3. 1976 Directory of Chemical Producers, United States of America, Stanford Research
Institute, Menlo Park, CA (1977).
4. E. G. Webster, Description of the Chemical Storage Tank Emissions (Tankemis)
Data Base, Working Paper No. 12993, Mitre Corp., Metrek Division, Mclean, VA
(May 1978).
5. Synthetic Organic Chemicals, 1976 United States Production and Sales, United
States International Trade Commission, GPO, Washington (1977).
6. C. C. Masser, "Storage of Petroleum Liquids," p 4.3-1 in Compilation of Air
Pollutant Emission Factors, AP-42, Part A, 3d ed. (August 1977).
7. Control of Volatile Organic Compounds from Storage of Petroleum Liquids (draft
copy) (prepared by EPA, ESED, Research Triangle Park, NC) (November 1977).
8. C. C. Masser, "Storage of Petroleum Liquids," pp. 4.3-5 in Compilation of Air
Pollutant Emission Factors, AP-42, Part A, 3d ed. (August 1977).
9. Cited in Kline Guide to the Chemical Industry, 3d ed., p 58, edited by M. K.
Meegen, Charles H. Kline and Co., Inc., Fairfield, NJ, 1977.
*Usually, when a reference is located at the end of a paragraph, it refers to
the entire paragraph. If another reference relates to certain portions of
that paragraph, that reference number is indicated on the material involved.
When the reference appears on a heading, it refers to all the text covered by
that heading.
-------
III-l
III. EMISSIONS
SOURCES
Storage Tanks
There are five emission sources that need to be considered in organic chemical
storage: fixed-roof breathing losses, fixed-roof working losses, floating-roof
standing-storage losses, floating-roof withdrawal losses, and pressure-tank
losses.
Fixed-roof breathing losses occur when VOC is expelled from a tank because
vapors expand and contract as a result of a diurnal temperature change or
barometric pressure changes and additional VOC vaporization.
Fixed-roof working losses result from VOC being expelled from a tank during
filling and emptying operations. During the filling operation vapor is displaced
by the incoming liquid, and during the emptying operation air is drawn into the
tank and becomes saturated with organic vapor and is discharged during the next
filling.
Floating-roof standing storage losses occur when VOC is expelled from the tank.
The primary source of this loss is a defective fit of the seal and shoe to the
tank wall, which exposes some liquid surface to the atmosphere.
Floating-roof withdrawal losses occur when VOC is evaporated from the liquid
film that wets the tank wall while the roof descends during emptying operations.
This loss will be small compared to the other types of losses if the number of
tank turnovers is low.
No VOC losses will be experienced from pressure-tank operation under normal
conditions. Losses may occur if an inert-gas pad is used for safety or oper-
ating reasons or if the pressure inside the tank exceeds the relief (safety)
operating pressure.
The following main-storage operating factors affect VOC loss: vapor pressure
of the liquid stored, temperature changes in the tank, tank vapor space (tank
-------
III-2
outage), tank diameter, frequency of turnovers, mechanical condition of tank
and seals, tank type, tank color, and weather conditions.
2
Loading and Handling
There are three emission sources from the transportation of volatile synthetic
organic chemicals: loading losses, transit losses, and ballasting losses.
Both loading and transit losses occur when VOC is transported by marine ves-
sels, tank cars, or tank trucks. With marine vessels there can also be bal-
lasting losses. VOC losses from pipe lines are negligible under normal oper-
ating conditions.
Loading losses are the primary source of emissions from marine vessel, tank-
car, and tank-truck operations and occur when vapors residing in empty cargo
tanks are displaced to the atmosphere by the liquid being loaded. These vapors
are a mixture of the vapors formed in the empty tank by evaporation of residual
product from a previous load and those generated in the tank as a new product
is being loaded.
The total evaporative loss from loading operations is a function of the physi-
cal and chemical properties of the previous and new cargos, the method used for
loading or unloading the cargos, and the service history of the cargo carrier
("clean tank; normal, dedicated service; or dedicated, vapor balance service").
Transit losses will occur if vapor is expelled because of temperature and
barometric pressure changes while the cargo tank is. en route. These transit
losses are eliminated if closed pressure carriers are used.
An additional practice, specific to marine vessels, that increases shipping
losses is ballasting. After a cargo is unloaded, the empty tanker will nor-
mally carry several cargo tanks filled with water to improve stability during
the return voyage. VOC vapors remaining in the empty tanks are expelled when
the ballast water is added.
-------
III-3
B. CALCULATIONS
1. Storage Tanks
Emissions from organic chemicals stored in fixed-roof and floating-roof tanks
were calculated with use of the following empirical equations.
a. Fixed-Roof Tanks Breathing losses and working losses occur from storage of
chemicals in fixed-roof tanks. The AP-42 emissions equations for breathing and
working losses were used to estimate emissions from fixed-roof tanks. However,
breathing losses calculated with these equations were discounted by a factor of
4
4 based on test results reported by EPA, the Western Oil and Gas Association
(WOGA), and the German Society for Petroleum Science and Carbon Chemistry (DGMK).6
These results indicate that AP-42 tends to overestimate the actual breathing
losses from fixed-roof tanks by a factor of roughly 4. The working losses have
not been adjusted, because initial testing indicates that AP-42 is fairly accurate
for this estimation.
Breathing losses The following equation can be applied to all fixed-roof tanks
and includes the assumption that no conservation vent is installed on the tank:
0 68
L - 9 15 X 10"6 M p .. " D1'73 H°-51 AT0'5 v PK
LB - 9.15 X 10 H 14 ? _ p D H AT FpCKc ,
where
L_ = fixed roof breathing loss (Mg/yr) ,
D
M = molecular weight of vapor in storage tank (Ib/lb mole),
P = true vapor pressure at bulk liquid conditions (psia),
D = tank diameter (ft),
H = average vapor space height (ft; used 1/2 tank height),
AT = average diurnal temperature change (°F; used 20°F),
F = paint factor (dimensionless, assuming a white tank in good
condition; factor =1),
C = adjustment factor (dimensionless) for large-diameter tanks (D > 30 ft),
c = 1; for small-diameter tanks (less than 30 ft) =
0.0771 (D) - 0.0013 (D2) - 0.1334 (ref 7),
K = crude-oil factor (dimensionless; used 1.0).
-------
III-4
The average diurnal (day/night) temperature change is variable and can be
determined from National Weather Service data for a plant at a specific geographic
location.
The paint factor (F ) for fixed-roof tanks, taken from AP-42, can be obtained for
the specific example given in Table A-l (Appendix A) of AP-42.
3
Working losses The following equation can be applied to all fixed-roof
tanks with or without conservation vents:
L = 1.09 X 10"8 MPK K VN ,
w n c
where
L = fixed-roof working loss (Mg/yr),
N = turnovers per year,
K = turnover factor (dimensionless)
= (180 + N)/6N for N > 36 and = 1 for N < 36,
V = tank volume (gal),
M, P, and K = same as for fixed-roof breathing losses.
Tank turnovers per year, if not stated, can be obtained by dividing the annual
throughput through the tank by the tank's capacity.
b. Floating-Roof Tanks Standing-storage losses and withdrawal losses were
calculated for floating-roof tanks as follows.
9
Standing-storage losses The following equation can be applied to both
open-top and internal-floating-roof tanks with single primary seals:
T - n nn«9i M P °-7 n1-5 „ °'7 K K K K 365
Ls - 0.00921 M 14>? .. p D vw \\^G 22Q5 ,
where
L = floating-roof standing-storage loss (Mg/yr),
s
v = average wind velocity (miles/hr; used 9.0),
K = tank type factor (dimensionless; used 0.045
based on welded tank),
-------
III-5
K = seal factor (dimensionless; used 1.0),
s
K = paint factor (dimensionless; used 0.9 based on white
paint),
M, P, D, and K = same as for fixed-roof breathing losses.
An average wind speed (v ) should be used for determining losses from open-top
floating-roof tanks. A wind speed of 4 mph should be used for calculating losses
from internal-floating-roof tanks.
The factors K., K , and K depend on the type of tank construction, tank seal,
and paint used, respectively. Table A-2 gives the values for these factors,
taken from AP-42.
Withdrawal losses The following equation ' can be applied to both internal-
and open-top floating-roof tanks that are equipped with either a primary seal
or a primary and a secondary seal:
Lwd = 0.000198 DTHd ^ ,
where
L , = floating roof withdrawal loss (Mg/yr),
D = tank diameter (ft),
T = turnovers (per yr),
H = tank height (ft),
d = density of stored liquid at bulk liquid conditions (Ib/gal;
used 8.0).
Floating-roof-tank emissions can also be estimated with the equations shown in
Appendix B. These equations were not used to estimate emissions for this report
since they were not available when the original report was issued and sufficient
time was not available to incorporate them into this revision. The equations
/• . • '•• • •••:• ' j •'-. ' ... TO
were derived from a pilot-test tank study conducted for the EPA.
c. Pressure Vessels Pressure tanks were considered to be closed systems with no
estimated VOC emissions.
-------
III-6
Problems may exist in applying these correlations to organic chemical storage
because they are based on empirical relationships developed primarily from data
13
on gasoline storage. Most stored organic chemicals have vapor pressures out-
side the range of the data base used by the American Petroleum Institute. How-
ever, in the absence of correlations developed specifically for volatile syn-
thetic organic chemicals, the existing equations were considered to be adequate
for this study.
2. Loading and Handling
Emissions from organic chemical transportation were calculated with the fol-
lowing empirical equations.
14
a. Loading Losses - The following equation can be applied to tank-car, tank- truck,
and marine-vessel loading and assumes that no supplemental emission control
device is in use:
__
. T 1000 2205 ,
where
LL = loading loss (Mg/yr) ,
S = saturation factor (dimensionless; used 1.0 for tank truck and tank
car and 0.2 for marine vessels),
P = true vapor pressure of the liquid loaded (psia),
M = molecular weight of vapors (Ib/lb mole) ,
T = bulk temperature of liquid loading [°R; used 540°R (80°F)],
V = volume loaded (gal/yr) .
The saturation (S) factor represents the expelled vapor's fractional approach
to saturation and accounts for the variations observed in emission rates from
the different unloading and loading methods. An S factor of 1.0 for tank cars
and tank trucks and an S factor of 0.2 for marine vessels give an average
estimate of emissions for the industry, which uses a variety of loading methods,
with no add-on emission control devices assumed to be in use. Table A-3 gives
14
the S values to be used for specific loading methods, taken from AP-42.
-------
III-7
b. Transit Losses The following equation can be applied to tank-car, tank-truck,
and marine-vessel transport and assumes that the carrier is not a pressure vessel:
Lt = °-1 PW lOOO 2205 T '
where
L = transit loss (Mg/yr),
P = true vapor pressure of the liquid loaded (psia),
W = density of condensed vapor (Ib/gal; used 8.0),
V = volume transported per year (gal/yr),
T = fraction of a week required for transport.
Ballasting losses could not be estimated because data were not available on
chemical transport by ships that are ballasted. This is not expected to be a
major emission for organic chemical transportation.
C. INDUSTRY EMISSIONS
SOCMI VOC emissions from both storage and loading are presented in two ways:
(1) Uncontrolled VOC emissions are calculated with the equations described in
Sect. B. Uncontrolled VOC emissions are defined as those emissions that would
occur as a result of no add-on controls being used on storage tanks or during
loading operations. (2) Current or present VOC emissions from storage and
loading are calculated by estimating the number and operating efficiency of
add-on controls currently in use and applying them to the uncontrolled emis-
sions. Because floating-roof tanks are VOC emission control devices, current
and uncontrolled VOC emission quantities from floating-roof tanks are the same.
Pressure tanks are considered to be closed systems with no estimated VOC emis-
sions.
1. Uncontrolled Emissions (1977)
a. Storage Tanks — Uncontrolled VOC emissions from existing fixed-roof tanks are
calculated to be 45,122 Mg/yr and for floating-roof tanks to be 1851 Mg/yr, for
a projected total of 46,973 Mg/yr.15 Tables III-l and III-2 show the distribution
of the calculated emissions by tank capacity and vapor pressure, respectively.
-------
Table III-l. 1977 Storage-Tank VOC Emissions by Tank Size and Type (Assuming No Controls)
b
Tank Size
(m3)
4 to 40
40 to 75
75 to 115
115 to 150
150 to 190
190 to 380
380 to 950
950 to 1900
1900 to 3800
>3800
Total
Fixed-Roof
No. of
Tanks
4,495
2,825
1,425
550
800
1,375
1,560
975
520
290
14,815
Tanks
Emissions
(Mg/yr)
2,483
2,860
1,639
1,154
1,083
4,024
5,710.
7,991
9,528
8,050
45/122
Floating-Roof
d
Tanks
No. of Emissions
Tanks (Mg/yr)
0
10
0
15
25
85
140
95
125
175
670
0
7
0
7
15
209
157
131
417
908
1,851
No. of
Tanks
4,495
2,750
1,360
625
915
1,460
1,700
1,070
645
465
15,485
Total6
„ • . c
Emissions
(Mg/yr)
2,483
2,867
1,639
1,761
1,098
4,233
5,867
8,122
9,945
8,958
46,973
alncludes only tanks with capacity greater than 4.0 m .
See ref 15.
°Calculated from equations detailed in this section (III.B).
jFloating-roof tanks are actually control devices and therefore have no uncontrolled VOC emissions.
Emissions from pressure vessels assumed to be negligible.
i
03
-------
Table III-2. 1977 Storage-Tank VOC Emissions by Material Vapor Pressure and Tank Type (Assuming no Controls)
Fixed-Roof Tanks
b
Vapor Pressure
(kPa)
0.001
1.4
3.5
6.9
10.3
20.7
34.5
51.7
to
to
to
to
to
to
to
to
69.0 to
Total
1.4
3.5
6.9
10.3
20.7
34.5
5.1.7
69.0
103.4
No. of
Tanks
7,400
2,070
1,955
980
1,475
600
165
90
80
14,815
c
Emissions
(Mg/yr)
2
2
3
4
11
7
3
3
6
45
,023
,454
,985
,199
,048
,414
,913
,146
,940
,122
Floating-Roof Tanks
No. of
Tanks
55
55
45
35
305
140
20
0
15
670
Emissions
(Mg/yr)
36
52
69
194
702
578
73
0
147
1,851
Total6
No. of
Tanks
7,455
2,125
2,000
1,015
1,780
740
185
90
95
15,485
c
Emissions
(Ma/yr)
2
2
4
4
11
7
3
3
7
46
,059
,506
,054
,393
,750
,992
,986
,146
,087
,973
H
H
H
VO
aincludes only tanks with capacity greater than 4.0 m .
bSee ref 15.
CCalculated from equations detailed in this section (III.B).
Floating-roof tanks are actually control devices and therefore have no uncontrolled VOC emissions.
eEmissions from pressure vessels assumed to be negligible.
-------
111-10
Large fixed-roof storage tanks more than 1900 m are the source of an
estimated 39% (17,580 Mg/yr) of the total uncontrolled fixed-roof-storage VOC
emissions, although they comprise only 5.5% of the total number of tanks. Al-
though smaller fixed-roof storage tanks less than 75 m constitute 50% of
the total industry tankage, they contribute only 12% (5343 Mg/yr) of the emissions
(see Table III-l).
Approximately 60% (26,650 Mg/yr) of the total VOC emissions is from the storage
of materials having vapor pressures between 3.5 and 34.5 kPa. Materials stored
in fixed-roof tanks and having vapor pressures higher than 34.5 kPa contribute
30% (14,000 Mg/yr) of the total estimated uncontrolled VOC emissions, and materials
with vapor pressures of less than 3.5 kPa, although comprising 53% of the total
volume stored, contribute only 10% (4470 Mg/yr) of the total estimated VOC emissions
from fixed-roof storage (see Table III-2).
Pressure tanks are not a significant source of VOC emissions under normal operating
conditions.
b. Loading and Handling '
Uncontrolled VOC emissions from loading are calculated to be 5848 Mg/yr and
from transit to be 1727 Mg/yr, for a total of 7575 Mg/yr (see Table III-3).
Rail and truck shipments account for 83% (4874 Mg/yr) of the uncontrolled
loading VOC emissions. Shipments of materials whose vapor pressures range from
6.9 to 34.5 kPa contribute 66% of the total uncontrolled VOC emissions for the
industry.
Approximately 15% (5250 X 103 m3) of the VOC shipments listed in Table II-3
(above 103.4 kPa vapor pressure) are considered to be stored arid loaded in
closed systems with negligible loss of VOC.
2. Present Industry Emissions (1977)
a. Storage Tanks Current emissions from fixed-roof tanks for the SOCMI are
estimated to be 29,562 Mg/yr and for floating-roof tanks to be 1851 Mg/yr, for
a projected total of 31,413 Mg/yr. This is approximately 34% less than the
-------
Table III-3. 1977 Loading and Transportation VOC Emissions by Material Vapor Pressure'
(Assuming No Control)
Vapor
0.001
1.4
3.5
6.9
10.3
20.7
34.5
51.7
Pressure
(kPa)
to 1.4
to 3.5
tor 6.9 "
to 10.3
to 20.7
to 34.5
to 51. 7
to 69.0 ;
69.0 to 103.4 :
Total
Annual Volume
Shipped
(103 m3/yr)
8,743
3,429
3,543
; 3,111
6,866
3,043
530
454
424
30,143
Loading
Loss
(Mg/yr)b
174
240
440
581
2,052
1,213
330
396
422
5,848
Transportation
Loss
(Mg/yr)b
31
43
96
141
559
440
120
144
153
1,727
Total VOC
Emissions
(Mg/yr)b
205
283
536
722
2,611
1,653
450
540
575
7,575
H
M
•H
H
See refs 14 and 16.
Calculated from equations detailed in this section (III.B)
-------
111-12
calculated uncontrolled emission levels. Tables III-4 and III-5 show the
distribution of the estimated emissions by tank capacity and vapor pressure,
respectively.
An estimated 2148 fixed-roof tanks have add-on controls. This represents about
15% of the total number of fixed-roof tanks in the industry.
Large tanks, above 950 m , account for about 41% of the total number of fixed-
roof tanks with add-on controls.
Approximately 40% of the fixed-roof tanks containing material with a vapor
pressure of 10.3 kPa or greater have add-on control devices.
Approximately 80% of the emissions from floating-roof tanks are from materials
having a vapor pressure of between 6.9 and 34.5 kPa. These materials are
usually stored in large tanks, with 72% of the floating-roof tank emissions
coming from tanks larger than 1890 m .
As stated in Sect. III-C-1, pressure tanks are not a significant source of VOC
emissions under normal operating conditions.
b. Loading and Handling Current VOC emissions from loading are estimated to be
4830 Mg/yr and from transit to be 86 Mg/yr, for a total of 4917 Mg/yr (see
Table III-6). This is a 35% reduction from the uncontrolled-emission levels.
The major cause for reduction is the use of pressurized cargo carriers, esti-
mated to be used at least 95% of the time, that have no VOC emissions during
transport.
As stated in Sect. C-l, approximately 15% of the VOC shipments are considered
to be stored and loaded in closed systems with negligible loss of VOC since the
materials have vapor pressures equal to or greater than atmospheric pressure.
c. Miscellaneous It is recognized that there are other sources of VOC emissions
related to storage and handling of SOCMI materials. They include tank and
cargo carrier clean-outs, spills from both loading and unloading operations,
-------
Table III-4. Current 1977 Storage-Tank VOC Emissions by Tank Size and Type (with Controls)
Fixed-Roof Tanks
b
Tank Size
(m3)
4 to 40
40 to 75
75 to 115
115 to 150
150 to 190
190 to 380
380 to 950
950 to 1900
1900 to 3800
>3800
Total
No. of
Tanks
4,495
2,825
1,425
550
800
1,375
1,560
975
520
290
14,815
No.
Controlled
339
310
154
53
39
161
215
382
300
195
2,148
c
Emissions
(Mg/yr)
1,731
1,988
1,189
1,304
757
2,742
4,000
5,106
5,542
5,203
29,562
Floating-Roof Tanks
No. of
Tanks
0
10
0
15
25
85
140
95
125
175
670
c
Emissions
(Mg/yr)
0
7
0
7
15
209
157
131
417
908
1,851
No. of
Tanks
4,495
2,750
1,360
625
915
1,460
1,700
1,070
645
465
15,485
Total6
No.
Controlled
339
320
154
68
64
246
355
477
425
370
2,818
c
Emissions
(Mg/yr)
1,731
1,995
1,189
1,311
772
2,951
4,157
5,237
5,959
6,111
31,413
H
H
H
1
W
Includes only tanks with capacity greater than 4.0 m .
See ref 15.
CDetermined by estimating the add-on controls in use and applying to the uncontrolled emissions at 80% control
efficiency.
Floating-roof tanks are actually control devices and therefore have no uncontrolled VOC emissions.
eEmissions from pressure vessels assumed to be negligible.
-------
Table III-5. Current 1977 Storage-Tank VOC Emissions by Material Vapor Pressure and Tank Type (with Controls)
Fixed-Roof Tanks
b
Vapor Pressure
(kPa)
0.001 to 1.4
1.4 to 3.5
3.5 to 6.9
6.9 to 10.3
10.3 to 20.7
20.7 to 34.5
34.5 to 51.7
51.7 to 69.0
69.0 to 103.4
Total
No. of
Tanks
7,400
2,070
1,955
980
1,475
600
165
90
80
14,815
No.
Controlled
. 370
207
391
245
443
240
99
77
76
2,148
Emissions
(Mg/yr)
1,999
2,315
3,404
3,416
8,450
5,099
2,092
1,064
1,723
29,562
Floating-Roof Tanks
No. of
Tanks
55
55
45
35
305
140
20
0
15
670
Emissions0
(Mg/yr)
36
52
69
194
702
578
73
0
147
1,851
No. of
Tanks
7,455
2,125
2,000
1,015
1,780
740
185
90
95
15,485
Total6
No.
Controlled
425
262
436
280
748
380
119
77
91
2,818
Emissions
(Mg/yr)
2,035
2,367
3,473
3,610
9,152
5,677
2,165
1,064
1,870
31,413
c
H
H
M
1
*-
alncludes only tanks with capacity greater than 4.0 m .
bSee ref 15.
CDetermined by estimating the add-on controls in use and applying to the uncontrolled emissions at 80% control
efficiency.
Floating-roof tanks are actually control devices and therefore have no uncontrolled VOC emissions.
SEmissions from pressure vessels assumed to be negligible.
-------
Table III-6. Current 1977 Loading and Transportation VOC Emissions by Material Vapor Pressure (with Controls)
Vapor
Pressure
(kPa)
0.001 to
1.4
3.5
6.9
10.3
20.7
34.5
51.7
69.0
to
to
to
to
to
to
1.4
3.5
6.9
10.3
20.7
34.5
51.7
to 69.0
to 103.4
Total
Annual Volume
Shipped
(X 103 m3/yr)
8,743
3,429
3,543
3,111
6,866
3,043
530
454
424
30,143
Loading
Loss
(Mg/yr)
174
230
405
525
1,806
1,019
264
238
169
4,830
Transportation
Loss
(Mg/yr )c
. 1
2
5
7
28
22
6
7
8
86
Total VOC
Emissions
(Mg/yr )b
175
232
410
532
1,834
1,041
270
245
177
4,917
H
H
H
1
M
U1
See refs 14 and 16.
Determined by estimating the add-on controls used and applying them to the uncontrolled emissions at 80% control
efficiency.
•« ^ .
"Transportation loss estimated to be 5% or less of uncontrolled loss due to use of pressurized cargo carriers.
-------
111-16
losses from breaking of hose and coupling connections, purging of tanks and
cargo carriers to eliminate explosive mixtures, etc. These emissions are not
controlled and have not been estimated because they are believed to be small
compared to the total VOC emissions lost from storage tanks and loading opera-
tions .
-------
111-17
D. REFERENCES*
1. C. C. Masser, "Storage of Petroleum Liquids," pp 4.3-5 and 4.3-6 in
Compilation of Air Pollutant Emission Factors, AP-42, Part A, 3d ed.
(August 1977).
2. C. C. Masser, "Transportation and Marketing of Petroleum Liquids," pp 4.4-1 to
4.4-6 in Compilation of Air Pollutant Emission Factors, AP-42, Part A, 3d ed.
(August 1977).
3. C. C. Masser, "Storage of Petroleum Liquids," pp 4.3-6, 4.3-10, and 4.3-11 in
Compilation of Air Pollutant Emission Factors, AP-42, Part A, 3d ed. (April 1977)
4. U.S. Environmental Protection Agency, Emission Test Report-Breathing Loss
Emissions from Fixed-Roof Petrochemical Storage Tanks, EMB 78-OCM-5,
Research Triangle Park, NC (February 1979).
5. Western Oil and Gas Association, Hydrocarbon Emissions from Fixed-Roof Petroleum
Tanks, prepared by Engineering-Science, Inc., Los Angeles, CA (July 1977).
6. German Society for Petroleum Science and Carbon Chemistry (DGMK) and the Federal
Ministry of the Interior (BMI), Measurement and Determination of Hydrocarbon
Emissions in the Course of Storage and Transfer in Above-Ground Fixed Cover
Tanks With and Without Floating Covers, BMI-DGMK Joint Projects 4590-10 and
4590-11; translated for EPA by Literature Research Company, Annandale, VA.
7. A. Goldfarb, Small Diameter Factor for Floating Roof Breathing Loss Equation,
Mitre Corp., Metrek Division, McLean, VA (data on file by D. Erikson, IT Enviro-
science, Inc., Knoxville, TN) (Apr. 4, 1978).
8. C. C. Masser, "Storage of Petroleum Liquids," p 4.3-10 in Compilation of Air
Pollutant Emission Factors, AP-42, Part A, 3d ed. (August 1977).
9. Ibid., pp 4.3-12 and 4.3-13.
10. Chicago Bridge and Iron Co., SOHIO/CBI Floating Roof Emission Testing Program,
Supplemental Report (Feb. 15, 1977).
11. Equation for floating-roof withdrawal loss derived by R. Burr, EPA, ESED,
Research Triangle Park, NC (May 1978).
12. Chicago Bridge and Iron Company, Plainfield, IL, Hydrocarbon Emission Loss
Measurements on a 20 foot Diameter Pilot Test Tank with an Ultraflote and a
CBI Weathermaster Internal Floating Roof, R-0113/R-0191 (June 1978).
13. American Petroleum Institute, Evaporation Loss Committee, Evaporation Loss
from Fixed-Roof Tanks, Bulletin 2518, Washington (1962).
14. C. C. Masser, "Transportion and Marketing of Petroleum Liquids," pp 4.4-5 and
4.4-7 in Compilation of Air Pollutant Emission Factors. AP-42, Part A, 3d ed.
(August 1977).
-------
111-18
15. SOCMI storage-tank data base compiled by IT Enviroscience, Inc., Knoxville, TN,
using computer program developed by Mitre Corp., McLean, VA, April 1978 (on
file at ESED, EPA, Research Triangle Park, NC) (compiled from State EIQ files,
Site Visit Trip Reports, and USITC report on 1976 SOC Production and Sales).
16. Texas/Louisiana State Emission Inventory Questionnaires (1975) (on file at EPA,
ESED, Research Triangle Park, NC).
*Usually, when a reference is located at the end of a paragraph, it refers to
the entire paragraph. If another reference relates to certain portions of that
paragraph, that reference number is indicated on the material involved. When
the reference appears on a heading, it refers to all the text covered by that
heading.
-------
IV-1
IV. CONTROL TECHNOLOGY
There are several different methods, with varying ranges of efficiency, for the
control of emissions from the storage and handling of organic chemicals.
A. STORAGE TANKS
1. Emission Control Techniques
Fixed-roof tanks represent minimal control technology. They are inadequate for
storing the more volatile organic compounds, as is demonstrated by the high
emission levels presented in Sect. III. Emissions are significant even with
vapor pressures as low as 3.4 kPa.
a- Internal Floating Roofs in Fixed-Roof Tanks Fixed-roof tank emissions can be
reduced by installing internal floating roofs and seals in the tanks to minimize
evaporation of the product being stored. Three floating-roof and seal combina-
tions have been tested for use in fixed-roof tanks, including a noncontact
internal floating roof with shingled, vapor-mounted, primary and secondary
seals; a contact internal floating roof with a liquid-mounted primary seal; and
a contact internal floating roof with a liquid-mounted primary seal and a
continuous secondary seal. Based on the test results a noncontact internal
floating roof with shingled, vapor-mounted, primary and secondary seals is not
as effective in reducing emissions as a contact internal floating roof with a
liquid-mounted primary seal. Consequently, a larger emissions reduction can
be achieved by fitting a fixed-roof tank with a contact internal floating roof
and a liquid-mounted primary seal rather than with a noncontact internal floating
roof and shingled, vapor-mounted, primary and secondary seals. Installation of
a continuous secondary seal on a contact internal floating roof yields even
more emissions reduction.
Several other roof and seal combinations, which have not been tested, are also
available for controlling the emissions from fixed-roof tanks. Some of these
include (1) a noncontact internal floating roof with a vapor-mounted primary
seal; (2) a contact internal floating roof with a vapor-mounted primary seal;
and (3) a contact internal floating roof with a vapor-mounted primary seal and
a continuous secondary seal. Based on engineering judgment, a noncontact roof
-------
IV-2
with a vapor-mounted primary seal would be less effective in reducing emissions
than the noncontact roof tested, which was equipped with both primary and
secondary seals. In addition, information presented in American Petroleum
Institute (API) Bulletin 2517 regarding the effectiveness of liquid-mounted
and vapor-mounted primary seals on external floating roofs indicates that the
contact internal floating and liquid-mounted primary seal tested would be more
effective at reducing emissions than a contact roof with a vapor-mounted primary
seal. As was also indicated in the test of a contact internal floating roof, a
secondary seal over a vapor-mounted primary seal may be expected to result in
even larger emissions reduction.
b. Rim-Mounted Secondary Seals on External Floating Roofs A rim-mounted secon-
dary seal on an external floating roof is a continuous seal that extends from
the floating roof to the tank wall, covering the entire primary seal. Instal-
led over a mechanical shoe seal, this secondary seal has been demonstrated to
effectively control VOC emissions that escape from the small vapor space between
the shoe and the wall or through any openings or tears in the seal envelope
(see Fig. IV-la). Rim-mounted secondary seals should also be effective in
controlling emissions from the liquid- and vapor-mounted primary seals shown in
Figs IV-lb, IV-lc, and IV-ld.3
Another type of secondary seal, which has not been tested, is a shoe-mounted
secondary seal. A shoe-mounted seal extends from the top of the shoe to the
tank wall (see Fig. IV-2). These seals do not provide protection against VOC
leakage through the envelope. Holes, gaps, tears, or other defects in the
envelope can allow direct communication between the saturated vapor under the
envelope and the atmosphere; the wind can enter this space through envelope
defects, flow around the circumference, and exit with saturated or nearly
saturated VOC vapors.
c. Fixed Roofs on External Floating Roof Tanks Installing a fixed-roof on an
existing external-floating-roof tank would reduce emissions by reducing the
effect of wind sweeping vapors out of the vapor space and into the atmosphere.
An alternative to the construction of a new external-floating-roof tank is the
construction of an internal-floating-roof tank with a primary seal or both
primary and secondary seals.
-------
IV-3
RIM-MOUNTED
SECONDARY SEAL
TANK
WALL
RIM-MOUNTED SECa:JDAflY SEAL
a. Shoe 353! .vi;h rinvmounted secondary seal
b. Liquid-filled sea! with rim-mounted
secondary seat.
RIM-MOUNTcD
SECONDARY SEAL
RiM-MOUNT:3
SECONDARY SEAL
c Resilient foam seal (v22OJ*-mouru;?d)
with rim-mounted secondary seal
d. Resilient foarf. se^: {liquid-mounted)
with-fim-mounted secondary seal
Fig. iv-1. Rim-mounted Secondary Seals on External Floating Roofs (from ref 3)
-------
IV-4
SECONDARY SEAl
(VVIPERTYPE)
Fig. IV-2. Metallic Shoe Seal with
Shoe-Mounted Secondary Seal (from ref 3)
-------
IV-5
d. Rim-Mounted Secondary Seals on Noncontact Internal Floating Roofs—Because
there are some noncontact internal-floating-roof tanks that have only vapor-
mounted primary seals, another control technique is to install rim-mounted
secondary seals over the primary seals. This seal, which is typically a wiper
seal, minimizes the effects from air currents inside the tank sweeping vapors
out of the annular vapor space. This type of seal can be either continuous or
shingled and extends from the floating roof to the tank wall, covering the
primary seal. Although the benefit of using a secondary seal cannot be quan-
tified because a noncontact roof with only a vapor-mounted primary seal has not
been tested, engineering judgment indicates that this modification would reduce
the emissions from noncontact internal-floating-roof tanks.
e. Contact Internal Floating Roofs in Noncontact Internal-Floating-Roof Tanks
Recent pilot test tank studies sponsored by EPA have demonstrated that non-
contact internal-floating-roofs with shingled, vapor-mounted, primary and
secondary seals may not be as effective in reducing emissions as contact
internal floating roofs with liquid-mounted primary seals. Based on these
studies one emission control technique for internal-floating-roof tanks is the
use of contact internal floating roofs with liquid-mounted primary seals
instead of noncontact internal floating roofs with shingled, vapor-mounted,
primary and secondary seals. The use of a continuous secondary seal on the
contact internal-floating-roof has been demonstrated to result in an even
larger emission reduction.
Two roof and seal combinations, which have not been demonstrated, are a contact
internal-floating-roof with a vapor-mounted primary seal and a contact internal-
floating-roof with vapor-mounted primary and secondary seals. Engineering
judgment indicates that the use of either of these roof and seal combinations
would result in lower emissions than those associated with the use of a noncontact
roof with a vapor-mounted primary seal or vapor-mounted primary and secondary
seals.
f- Liquid-Mounted Primary Seals on Contact Internal Floating Roofs Based on
2
information reported in API Bulletin 2517 and engineering judgment, vapor-
mounted primary seals are not as effective in reducing emissions as liquid-mounted
-------
IV-6
primary seals. As a result one technique to reduce the emissions from tanks
having contact internal floating roofs is the use of liquid-mounted rather than
vapor-mounted primary seals.
g. Rim-Mounted Secondary Seals on Contact Internal Floating Roofs Contact internal
floating roofs, like other types of floating roofs, can have not only a primary
seal to cover the annular vapor space but also a rim-mounted secondary seal
(Fig. IV-3). This secondary seal, which is typically a wiper seal or a resilient
foam-filled seal, minimizes the effects from the air currents inside the tank
sweeping vapors out of the annular vapor space. This type of seal is continuous
and extends from the floating roof to the tank wall, covering the entire primary
seal.
h. Carbon Adsorption—Carbon adsorption utilizes the principle of carbon's affinity
for nonpolar hydrocarbons to remove VOC from the vapor phase. Activated carbon
is the adsorbent and the VOC vapor that will be removed from the air stream is
referred to as the adsorbate. Adsorption of the VOC vapor occurs at the surface
of the adsorbent and is a physical process since no chemical change takes
place. The VOC carbon adsorption unit consists of a minimum of two carbon beds
plus a regeneration system. Two or more beds are necessary to assure that one
bed will be available for use while the other one is being regenerated.
4
Regeneration can be performed with either steam or vacuum (Fig. IV-4). In
steam regeneration, steam is circulated through the bed, raising the VOC vapor
pressure. The vaporized VOC is thus removed along with the steam. The steam-VOC
mixture is condensed, ususally by an indirect cooling-water stream, and routed
to a separator. The VOC is then decanted and returned to storage, and the
contaminated water is sent to the plant wastewater system for disposal. Cooling
water, electricity, and steam are the required utilities for a steam regeneration
system. The other method of regenerating the carbon, vacuum regeneration,
consists of a high vacuum being pulled on the carbon bed. The VOC vapor desorbed
by this process is condensed and returned to storage.
The size of the carbon adsorption units depends on the type and concentrations
of the VOC components in the vent gas, on the gas flow rate, and on the time
required for regeneration.
-------
IV-7
! SECONDARY SEAL
PRIMARY SEAL
IMMERSED IN
VOC
c^T«»'- r TYPE
INTERNAL FLOATING ROOF
Fig. IV-3. Rim Mounting of a Secondary Seal on an
Internal Floating Roof (from ref 1)
-------
IV-8
CL.EAM
EX-HAU6T
t i
-00—
PROCESS
ACT \ VATSO
AlP.
COOLIKJ6
AMSISUT
&OL.VEMT
WASTE
Fig. IV-4. Activated-Carbon Adsorption System (from ref 4)
-------
IV-9
i- Thermal Oxidation In a typical thermal oxidation system the air-vapor mixture
is injected via a burner manifold into the combustion area of the incinerator.
Pilot burners provide the ignition source and supplementally fueled burners add
heat when the flame temperature is required to be maintained between 1030 K and
1090 K (1400°F and 1500°F).
The amount of combustion air needed is regulated by temperature-controlled
dampers. The concentration of VOC in the tail gas of an oxidizer can be limited
to 10 ppm. Figure IV-5 shows a typical thermal oxidation unit.
Flashback prevention and burner stability are achieved by saturating the VOC
vapors to a concentration above the upper explosive limit. In addition two
water-seal flame arrestors are used to assure that flashbacks do not propagate
from the burner to the rest of the piping system.
A significant advantage of thermal oxidizers is that they can dispose of a wide
range of hydrocarbons. This is especially important at a tank farm where
numerous hydrocarbon liquids are stored.
Normally, thermal oxidation as a control for storage-tank VOC emissions cannot
be justified economically unless it is combined with a more significant process-
vent emission control system.
j- Refrigerated Vent Condensers Refrigerated vent condensers for controlling
fixed-roof storage-tank VOC emissions are one of the most common types of
emission reducion equipment found in the industry.
Condensers are sized to handle the maximum vapor rate expected at any given
time, which normally occurs during the tank filling operation. In actual
service one condenser can be used to collect the vapors from a manifold system
of tanks containing the same material.
> The refrigerated condenser must be properly designed to handle freezing of
moisture or VOC. This is accomplished by a system that routinely defrosts the
condenser and then separates the recovered water-VOC mixture. Package units
Lie
8
are available in a range of sizes for various applications. An example of a
refrigerated vent condenser is shown in Fig. IV-6.
-------
OUTLET VAPOR
PILOT
BURNER
, EUik—,*
VAPOR
BURNER
AIR DAMPER
STACK
MAIN BURNER
H
<
VAPOR SOURCE
WATER SEAL-
Fig. IV-5. Thermal Oxidation Unit (from ref 5)
-------
IV-11
INCOMING
VAPOR
1
CONDENSER
OVENT
S.
^~
COOLANT
RETURN
x<—rib
COOLANT
SUPPLY
VENT
/*—^
RECOVERY
TANK
V
/A
REFRIGERATION
UNIT
RECOVERED VOC
T TO STORAGE
—i'i +
PUMP
Fig. IV-6. Refrigerated Vent Condenser System
-------
IV-12
k- Pressure Vessels Pressure vessels as described in Sect. II can be considered
as a control alternative for small fixed-roof tanks (under 15 m3).
1- Conservation Vents A conservation vent is a pressure and vacuum vent that
provides a large gas flow area with a small differential pressure setting of
±0,2 kPa. The seal is provided by gasketed plates held in place over an opening
by a system of levers and adjustable weights.
Conservation vents will achieve low-cost VOC emission control for fixed-roof
tanks that do not warrant the floating-roof or other types of control options
covered above. Exceptions would be their application on fixed-roof tanks
containing chemicals that could partly or completely plug the vents in a short
period of time (i.e., shorter than a biannual inspection period) and result in
structural damage to the tank.
2. Control Efficiencies of Emissions Control Techniques
This section establishes the typical control efficiencies expected with the use
of external floating roofs with both primary and secondary seals, internal
floating roofs with primary seals, internal floating roofs with both primary
and secondary seals, and vapor control systems. The efficiencies are estimated
for only those emissions control techniques that have been demonstrated to be
effective in reducing emissions. Equations shown in Sect. Ill, which are based
on AP-42 emissions equations and test results recently reported by the Western
Oil and Gas Association (WOGA), EPA, and the German Society for Petroleum
Science and Carbon Chemistry (DGMK), are used to estimate fixed-roof-tank
emissions. Equations shown in Appendix B, which are based on testing conducted
for EPA, are used in estimating the emissions from external-floating-roof tanks
with primary seals; from external-floating-roof tanks with primary and secondary
seals; from noncontact internal-floating-roof tanks with vapor-mounted primary
and secondary seals,- from contact internal-floating-roof tanks with liquid-
mounted primary seals; and from contact internal-floating-roof tanks with both
liquid-mounted primary seals and continuous secondary seals.
It is emphasized that the emissions equations are based on limited amounts of
empirical data, and therefore should be used only to estimate, rather than to
-------
IV-13
precisely predict, the emissions. In addition the efficiency for each emission
control technique should be used only for making comparisons of the relative
effectiveness of the control techniques.
a. Internal Floating Roofs in Fixed-Roof Tanks If a noncontact internal floating
roof with shingled, vapor-mounted, primary and secondary seals is installed in
a fixed-roof tank 18 m in diameter, 12 m in height, and undergoing 13 turnovers
per year, the annual VOC emissions are reduced from 21 Mg to 8.8 Mg, a 58%
reduction. Installation of a contact internal floating roof with a liquid-
mounted primary seal in the same fixed-roof tank reduces VOC emissions from
21 Mg/yr to 3.5 Mg/yr an 83% reduction. Installing a contact internal floating
roof with a liquid-mounted primary seal and a continuous secondary seal in the
fixed-roof tank will reduce the annual VOC emissions by 90% to 2.2 Mg.
b. Rim-Mounted Secondary Seals on External Floating Roofs If the baseline is an
18-m-diam, 12-m-high external-floating-roof tank with 13 turnovers per year, the
annual emissions are estimated to be about 18 Mg. If this tank has a secondary
seal over the primary seal, annual emissions are about 13 Mg, a 28% reduction.
c. Fixed-Roofs on Extemal-Floating-Roof Tanks If a roof is installed on an
existing external-floating-roof tank 18 m in diameter, 12 m high, with 13
turnovers per year, the tank has emissions of the same order of magnitude as a
contact internal-floating-roof tank with a liquid-mounted primary seal. As a
result the annual benzene emissions are about 3.5 Mg, which is an 81% reduction
of the 18 Mg/yr emitted from the external-floating-roof tank. If the same
tank is retrofitted with a secondary seal and a fixed roof, the annual emissions
are reduced by 88%, or to 2.2 Mg.
If a new external-floating-roof tank with both primary and secondary seals is
the baseline, the construction of a contact internal-floating-roof tank with a
liquid-mounted primary seal would reduce annual emissions to 3.5 Mg. This
amount is a 73% reduction from the baseline emissions of 13 Mg/yr. An 83%
reduction equivalent to annual emissions of 2.2 Mg would be achieved with
construction of a contact internal-floating-roof tank with both primary and
secondary seals.
-------
IV-14
Another alternative to the construction of a new external-floating-roof tank is
a noncontact internal floating-roof tank. With an external-floating-roof tank
with a primary seal used as the baseline, this control option results in a 51%
reduction of the annual emissions, or from 18 to 8.8 Mg. If a new external-
floating-roof tank with both primary and secondary seals is the baseline, the
annual emissions are reduced from 13 Mg to 8.8 Mg, a 32% reduction.
d. Contact Internal Floating Roofs in Noncontact Internal Floating-Roof Tanks
Installing a contact internal floating roof with a liquid-mounted primary seal
in a noncontact internal-floating-roof tank 18 m in diameter, 12 m high, and
undergoing 13 turnovers per year reduces the annual VOC emissions to 3.5 Mg.
This amount is a 60% reduction from the 8.8 Mg/yr emitted from the noncontact
internal-floating-roof tank.
If a contact internal floating roof with a liquid-mounted primary seal and a
continuous secondary seal is installed in a noncontact internal-floating-roof
tank, the annual emissions are reduced by 75%, or to 2.2 Mg.
e. Rim-Mounted Secondary Seals on Contact Internal Floating Roofs Installation
of a secondary seal on an 18-m-diam, 12-m-high contact internal-floating-roof
tank with 13 turnovers per year reduces the annual VOC emissions from 3.5 Mg to
2.2 Mg, which is a 37% reduction in emissions.
f. Garbon Adsorption The carbon-adsorption vapor control system is estimated to
reduce the storage-tank emissions by approximately 96%. This efficiency is
based on a measured carbon-adsorption unit efficiency of 98% during gasoline
loading operations and an assumed collection efficiency of 98% of the VOC
emissions.
If a carbon-adsorption system with this efficiency is used on a fixed-roof tank
18 m in diameter, 12 m in height, and undergoing 13 turnovers per year, the VOC
emissions are reduced from 21 Mg/yr to 0.89 Mg/yr, a 96% reduction.
Because the emissions from the carbon-adsorption unit are directly related to
the volume of saturated vapor entering the unit, the use of an internal floating
roof in the same fixed-roof tank does not reduce the overall emissions. Conse-
-------
IV-15
quently the emissions from a carbon-adsorption unit that is fitted to a noncontact
or a contact internal-floating-roof tank remain unchanged at 0.89 Mg/yr.
However, because the annual emissions from these internal-floating-roof tanks
are only 8.8 Mg and 3.5 Mg, the emission reduction efficiencies from the use of
a carbon-adsorption system are only 90 and 75% respectively. These efficiencies
are lower than the 96% efficiency expected, because the vapor emitted from each
tank is saturated with VOC for safety reasons before its introduction to the
carbon-adsorption unit. Consequently the emission reductions achieved by the
carbon adsorber are based on the VOC concentration leaving the saturator and
not on the concentration from the storage tank.
In order to use a carbon-adsorption system on an external-floating-roof tank, a
fixed roof with pressure-vacuum vents must be installed over the floating roof.
Thermal Oxidation The thermal-oxidation vapor control system is estimated to
reduce the VOC emissions by approximately 97%. This efficiency is based on a
measured thermal-oxidation-unit efficiency of 99% during gasoline-loading
9
operations and an assumed 98% collection efficiency of the emissions.
Connecting a thermal-oxidation system of 97% efficiency to a fixed-roof tank
18 m in diameter, 12 m high, and undergoing 13 turnovers per year reduces
benzene emissions from 22 Mg/yr to about 0.67 Mg/yr, a 97% reduction.
If a thermal-oxidation system is used on a noncontract internal-floating-roof
tank, the emissions are reduced from 8.8 Mg/yr to 0.67 Mg/yr, a 92% reduction.
As with carbon adsorption, the percentage of emissions reduced can be less than
the overall control efficiency indicated, because the vapors may have to be
saturated.
The use of this system on a contact internal-floating-roof tank reduces emissions
from 3.5 Mg/yr to 0.67 Mg/yr, an 81% reduction.
Before a thermal oxidation system can be used on an external-floating-roof
tank, the tank must be modified by the installation of a fixed roof with pressure-
vacuum vents over the floating roof.
-------
IV-16
h. Refrigerated Vent Condenser The efficiency of vent condensers depends on the
vapor concentration of the VOC in the vent gas and on the condensing temperature,
Normally the VOC partial pressure in the effluent gas streams can be reduced to
3.4 to 0.7 kPa. Theoretical recovery efficiencies can be in excess of 90% for
VOC whose vapor pressures are over 34.5 kPa, and 60 to 90% for materials with
vapor pressures in the range of 3.4 to 34.5 kPa.
i. Pressure Vessels Under normal conditions no VOC losses will be experienced
from pressure vessels. If an inert-gas pad is used for safety or operating
reasons, losses will be similar to those experienced from variable vapor-space
tanks (i.e., reductions of 98% in VOC emissions from a fixed-roof tank).
j. Conservation Vents The use of a conservation vent reduces the amount of
breathing losses from fixed-roof tanks by venting only when the design pressure
differential is exceeded. Estimated emission control efficiencies of tank
breathing losses range from 5 to 25%+, depending on the vapor pressure of the
material stored.
The effectiveness of any of the control devices described above will be reduced
significantly if a proper maintenance and inspection program is not followed.
A biannual inspection program with followup repairs is suggested for all control
devices, including seals on floating roofs, to ensure maximum control of VOC
emissions from storage tanks, regardless of the type.
3. Retrofitting Problems
This section covers possible problems that fixed-roof-tank owners and operators
may have in retrofitting their tanks with internal floating roofs. In addition
problems associated with the retrofitting of rim-mounted secondary seals on
external floating roofs, conversion of external-floating-roof tanks to internal-
floating-roof tanks and the retrofitting of vapor control systems to tanks are
discussed.
a. Fixed Roof Tanks with Internal Floating Roofs Several modifications may be
necessary on a fixed-roof tank before it can be equipped with an internal
floating roof. Tank shell deformations and obstructions may require correc-
tion, and special structural modifications such as bracing, reinforcing, and
-------
IV-17
plumbing vertical columns may be necessary. Antirotational guides should be
installed to keep cover openings in alignment with roof openings. Special
vents must be installed on the fixed roof or on the walls at the top of the
shell to minimize the possibility of VOC vapors approaching the flammable range
in the vapor space.
b. Rim-Mounted Secondary Seals on External Floating Roofs Retrofitting problems
may be encountered when a secondary seal is installed above a primary seal,
which can accommodate a large amount of gap. Some secondary seals may not be
able to span as large a gap, and consequently excessive gaps may result between
the secondary seal and the tank shell.
c. Fixed Roofs on External Floating Roof Tanks In order to install a fixed roof
on an existing external-floating-roof tank, several tank modifications may be
required. For example, special structural modifications such as bracing and
reinforcing may be necessary to permit the external-floating-roof tank to
accommodate the added weight of a fixed roof. Vertical columns may be required
to support the fixed roof, and as a result modifications to the floating roof
will be necessary to accommodate these columns. In addition antirotational
guides should be installed to keep cover openings in alignment with roof openings,
d. Carbon Adsorption If a carbon-adsorption system is employed for reducing
emissions, steam can be used for regeneration of the carbon beds. However, a
source of steam may not be readily available at some facilities. In addition
water for cooling may also be in short supply. Furthermore, even if cooling
water is available, disposal of VOC-contaminated condensate could be a problem.
If water for cooling is not easily obtained or if steam is not readily available,
a vacuum regeneration system with a closed-loop freon refrigeration unit can be
used to regenerate the carbon. This method would also eliminate the problem of
disposal of VOC-contaminated water.
e. Thermal Oxidation The biggest problem with the use of thermal oxidation is
the requirement for supplemental fuel to maintain the flame temperature. Some
facilities may not have adequate fuel supplies readily available.
-------
IV-18
B. LOADING AND HANDLING
Loading emissions can be controlled by any of the methods described below.
1. Submerged Loading
The princinpal methods of loading cargo carriers are splash loading and sub-merged
loading. The two types of submerged loading are the submerged-fillpipe and the
bottom-loading methods. In the submerged-fillpipe method the fillpipe descends
almost to the bottom of the cargo tank. In the bottom-loading method the
material enters the cargo tank from the bottom. Submerged loading signifi-
cantly reduces liquid turbulence and vapor-liquid contact and results in much
lower VOC losses than are experienced with splash loading. Emission reductions
12
can range from 0 to 65%, depending on the method used by the cargo carrier.
2. Vapor Recovery
Vapor recovery equipment recovers the VOC vapors displaced during loading and
ballasting operations by use of refrigeration, adsorption, and/or absorption,
as described above for fixed-roof tanks. Control efficiencies range from
90 to 95%, depending on the nature of the VOC emissions and the type of recovery
equipment used.
3. Vapor Oxidation
Vapor oxidation with thermal-oxidation devices can give greater than 97-99+%
VOC emissions control. As an example, in the petroleum industry gasoline
vapors from a terminal are displaced to a vapor holder as they are generated.
When the gasoline vapors reach satuaration in the holder, they are released to
the oxidizer, after being mixed with a properly metered air stream, and are
then combusted. The thermal oxidizer is not a true afterburner; rather, it
operates in the manner of an enclosed flare.
Twelve to fifteen thermal oxidizers have reportedly been installed by gasoline
terminal operators. Later models of this type of control equipment do not
require vapor holders; vapors from the tank trucks created during loading
operations are vented directly to the thermal oxidizer.
-------
IV-19
4. Vapor Balance
Another method of loading that can be used to reduce VOC emissions is called
dedicated balance service. Cargo carriers in dedicated balance service pick up
vapors displaced from the receiving tank during unloading operations and trans-
12
port them in the empty cargo tanks back to the loading terminal. These
vapors can then be displaced, during subsequent refilling of the cargo carrier,
to an emission control device installed at the loading terminal.
-------
IV-20
C. REFERENCES*
1. U.S. Environmental Protection Agency, Measurement of Benzene Emissions from a
Floating Roof Test Tank, EPA-450/3-79-020, Research Triangle Park, NC (June 1979).
2. American Petroleum Institute, Evaporation Loss from External Floating Roof
Tanks, API Bulletin 2517 (February 1980).
3. U.S. Environmental Protection Agency, Control of Volatile Organic Emissions
from Petroleum Liquid Storage in External Floating RoofTanks, EPA-450/2-78-047,
Research Triangle Park, NC (December 1978).
4. H. S. Basdekis, IT Enviroscience, Inc., Carbon Adsorption Control Device
Evaluation (in preparation for EPA, ESED, Research Triangle Park, NC).
5. U.S. Environmental Protection Agency, Evaluation of Control Technology for
Benzene Transfer Operations, EPA-450/3-78-018, Research Triangle Park, NC
(April 1978).
6. Booz, Allen & Hamilton, Inc., Florham Park, NJ, Final Report. Volume I of II,
Cost of Hydrocarbon Emissions Control to the U.S. Chemical Industry (SIC 28)
(December 1977).
7. Information from technical bulletin, Hydrocarbon Vapor Recovery Unit, Form
8-VRC-16, by Edwards Engineering Corp., Pompton Plains, NJ (May 1, 1976).
8. D. G. Erikson, IT Enviroscience, Inc., Control Device Evaluation.Condensation
(in preparation for EPA, ESED, Research Triangle Park, NC) (July 1980).
9. Letter dated May 3, 1979, from McLaughlin, Nacy D., U.S. Environmental Protection
Agency, to David Ailor, TRW, May 3, 1979, commenting on the benzene storage
model plants package.
10. C. C. Mosser, "Storage of Petroleum Liquids," pp 4.3-13 and 4.3-14 in Air
Pollutant Emission Factors, AP-42, 3d ed., Part A (August 1977).
11- Control of Volatile Organic Compounds from the Storage of Petroleum Liquids
(draft copy) (data on file at EPA, ESED, Research Triangle Park, NC) (November
1977).
12. C. C. Mosser, "Transportation and Marketing of Petroleum Liquids," pp 4.4-4 to
4-4-7 in Air Pollutant Emissions Factors, AP-42, 2d ed., Part A (August 1977).
13• Control of Hydrocarbons from Tank Truck Gasoline Loading Terminals, Guideline
Series, EPA-450/2-77-026 (OAQPS No. 1.2-082), Research Triangle Park, NC
(October 1977).
*Usually, when a reference is located at the end of a paragraph, it refers
to the entire paragraph. If another reference relates to certain portions of
that paragraph, that reference number is indicated on the material involved.
When the reference appears on a heading, it refers to all the text covered by
that heading.
-------
V-l
V. IMPACT ANALYSIS
A. ENVIRONMENTAL AND ENERGY IMPACT
1 - Storage Tanks
To assess the impacts of the various control options, models that are represen-
tative of the industry storage tanks were devised; see Table V-l for the param-
eters used in the evaluation.
a- Fixed-Roof-Tank Emissions The quantity of VOC emitted from a fixed-roof tank
is primarily a function of the capacity of the tank itself and of the vapor
pressure of the chemical stored in the tank. Table V-2 shows the VOC emissions
in Mg/yr that would be expected for the nine selected model-tank sizes at seven
selected vapor pressures.
The smallest tank selected was 40 m and the largest was 8330 m . The vapor
pressures selected ranged from 1.0 kPa to 51.7 kPa. VOC emissions increase
with increasing tank size and increasing vapor pressure.
Emissions were calculated based on the fixed-roof breathing- and working-loss
equations discussed in Sect. Ill of this report.
Internal floating roofs used on fixed-roof tanks (new and retrofitted) sources
Table V-3 shows the VOC emission reduction that can be achieved by fitting a
contact internal floating roof with a liquid primary seal and a continuous secondary
seal to a fixed-roof tank.
A control efficiency of 85% was used to calculate emission reductions, which
represents the average control efficiency that could be expected for the range
of tank sizes (40—8330 m ) and vapor pressures (1.0—51.7 kPa) evaluated.
In operation, floating-roof storage tanks do not consume any raw materials or
utilities and have no adverse environmental or energy impacts.
-------
V-2
Table V-l. Model Parameters Used for Assessing Impacts of
Industry Storage Tanks*
Model
Tank Capacity
(nr5)
40
75
115
150
285
660
1420
2840
8330
Dimensions
Diameter
4.5
4.5
4.5
5.1
7.0
8.3
12.2
17.2
29.5
(m)
Height
2.4
4.9
7.3
7.3
7.3
12.2
12.2
12.2
12.2
Turnovers
Per Year
200
200
200
200
100
50
36
24
12
Average liquid density of materials controlled = 792 kg/m .
Table V-2. Uncontrolled VOC Emissions vs Absolute Vapor Pressures of
Material Stored for Model Fixed-Roof Tanks*
Tank
Capcity
(m3)
40
75
115
150
285
660
1420
2840
8330
VOC Emissions
1.0
kPa
0.16
0.30
0.43
0.58
0.86
1.62
3.24
4.61
7.92
1.4
kPa
0.21
0.39
0.57
0.77
1.12
2.12
4.25
6.02
10.25
3.4
kPa
0.41
0.78
1.16
1.55
2.23
4.24
8.50
11.88
19.65
(Mg/yr) for Vapor Pressures of
6.9
kPa
0.72
1.39
2.04
2.74
3.91
7.44
14.92
20.71
33.66
10.3
kPa
1.07
2.06
3.04
4.07
5.80
11.03
22.12
30.60
49.33
34.5
kPa
2.33
4.51
6.68
8.94
12.65
24.09
48.34
66.51
105.85
51.7
kPa
3.52
6.81
10.06
13.48
19.11
36.38
72.99
100.65
160 . 99
Calculated with equations discussed in Sect. III.
-------
Table V-3. VOC Emission Reduction from Replacing Fixed-Roof Tanks with Internal-Floating-
Roof Tanks vs Absolute Vapor Pressures of Materials
VOC Emission Reduction (Mg/yr)
Model Tank
Size
(m3)
40
75
115
150
285
660
1420
2840
8330
1.0 kPa
(Mg/yr)
0.14
0.26
0.37
0.49
0.73
1.38
2.75
3.92
6.73
1.4 kPa
(Mg/yr)
0.18
0.33
0.48
0.65
0.95
1.80
3.61
5.12
8.71
3.4 kPa
(Mg/yr)
0.35
0.66
0.99
1.32
1.90
3.60
7.23
101.10
16.70
6.9 kPa
(Mg/yr)
0.61
1.18
1.73
2.33
3.32
6.32
12.68
17.60
28.61
for Vapor Pressures
10.3 kPa
(Mg/yr)
0.91
1.75
2.58
3.46
4.93
9.38
18.80
26.01
41.93
34.5 kPa
(Mg/yr)
1.98
3.83
5.68
7.60
10.75
20.40
41.09
56.53
89.97
51.7 kPa
(Mg/yr)
2.99
5.79
8.55
11.46
16.24
30.92
62.04 i
85.55
136.84
a control efficiency of 85%.
-------
V-4
Refrigerated vent condenser (new and retrofitted sources) Table V-4 shows
the model parameters used for assessing the impacts from addition of a refrig-
erated vent condenser to a fixed-roof tank. Three model tank sizes (150, 660,
and 2840 m ), representing small, me<
tively were selected for evaluation.
and 2840 m ), representing small, medium, and large SOCMI storage tanks respec-
Three different cases representing three commmon conditions of condenser opera-
tion were selected for each of the model tanks and are also shown in Table V-4.
For case 1 it was assumed that the chemical stored has a low vapor pressure,
3.5 kPa, and that the vapor pressure of the exit gas stream leaving the con-
denser has been reduced to 1.4 kPa. Case 2 is an evaluation of the storage of
an average-vapor-pressure (10.3-kPa) chemical with a final (exit) vapor pres-
sure of 1.4 kPa. For case 3 it was assumed that the chemical stored has a high
vapor pressure, 51.7 kPa, and that the exit stream vapor pressure has been reduced
to 3.5 kPa.
The physical properties of three different chemicals with vapor pressures nearly
equal to the initial vapor pressures described for cases 1, 2, and 3 were averaged-
These averages were used to calculate the size of each condenser and its accom-
panying refrigeration system. Sizing details are shown in Appendix B.
Table V-5 shows the VOC emissions expected from the three model fixed-roof tanks
before and after a refrigerated vent condenser was added. Also included in the
table is the amount of VOC reduced. Quantities are shown for all three cases
of operation selected.
Refrigerated vent condensers do not consume any raw materials but do require
electricity. The electrical requirements for each model tank selected are shown
in Table V-6 as joules per gram of VOC recovered. Values range from 331 to
1854 J/g.
b. Floating-Roof-Tank Emissions The quantity of VOC emitted from a floating-
roof tank is primarily a function of the capacity (size) of the tank itself and
of the vapor pressure of the chemical stored in the tank. Table V-7 shows the
VOC emissions in Mg/yr that would be expected for the nine selected model tank
sizes at five selected vapor pressures.
-------
V-5
Table V-4. Model Parameters and Operating Conditions for Assessing
Impacts from Refrigerated Vent Condensers
Fixed-Roof
Tank Capacity
, 3>
(m )
150
660
2840
150
660
2840
150
660
2840
Final
Turnovers Vapor Pressure (kPa) Temp.
Per Year Initial Final (°C)
Case 1
200 ^
50 \ 3.5 1.4 10
24 J
Case 2
200^
50 \ 10.3 1.4 -9.6
24 J
Case 3
200 "N
50 \ 51.7 3.5 -20.3
24 J
Emission
Reduction
(%)
60
86.7
93.3
-------
Table V-5. VOC Emissions from Fixed-Roof Tanks Before and
After Refrigerated Condenser Add-on
VOC Emissions (Mg/yr)
Model Tank
Capacity (m )
150
660
2840
Before Add-On
Case I
1.55
4.24
11.88
Case II
4.07
11.03
30.60
Case
13
36
100
III
.48
. 38
.65
Case I
0.62
1.70
4.75
After Add-On
Case II
0.54
1.47
4.07
Case
0.
2.
6.
Ill
90
44
74
Case I
0.93
2.54
7.13
Reduction
Case II
3.53
9.56
26.53
Case III
12.58
33.94
93.91
I
CTi
-------
V-7
Table V-6. Energy Impact from Refrigerated
Vent Condenser
Electrical Requirements (J/g)
Capacity (m ) Case I Case II Case III
150 1854 688 511
660 885 456 467
2840 468 331 464
-------
V-8
Table V-7. VOC Emissions from Model Floating-Roof Tank
vs Absolute Vapor Pressures of Material Stored
VOC Emissions (Mg/yr) for
Model Tank
Capacity (m )
40
75
115
150
285
660
1420
2840
8330
51.7
kPa
0
0
0
0
1
1
2
4
9
.58
.60
.61
.75
.16
.47
.59
.31
.62
69.0
kPa
0
0
0
1
1
2
4
7
15
.94
.96
.98
.20
.88
.40
.23
.08
.83
76.6
kPa
1
1
1
1
2
3
5
9
20
.22
.23
.25
.54
.43
.10
.48
.17
.51
Vapor
Pressures of
86.2
kPa
1
1
1
2
3
4
8
14
31
.86
.87
.89
.33
.70
.74
.38
.05
.45
100.7
kPa
17
17
17
22
35
45
80
135
304
.88
.90
.92
.11
.59
.66
.96
.96
.88
-------
V-9
Emissions were calculated based on the standing-storage- and withdrawal-loss
equations discussed in Sect. Ill of this report for an internal-floating-roof
tank.
Two pressure-vessel conditions were evaluated for pressure vessels used in place
of internal-floating-roof tanks or open-top floating roof tanks with primary
and secondary seals (new sources): no VOC loss (100% emission control) and
98%-controlled VOC emissions. The latter condition would apply to pressure
vessels with an inert-gas pad used for safety or operating reasons and that
would therefore require periodic relief venting. Table V-8 shows the VOC
emission reduction that can be achieved by installation of a new pressure
vessel instead of a new floating-roof tank. Both the 98%- and 100%-control
conditions are shown.
In operation, pressure vessels used as emission controls do not consume any raw
materials or utilities and have no adverse environmental or energy impacts.
2. Loading and Handling
A model loading terminal sized for handling 950 m of organics per day was selected
for evaluation of the cost effectiveness of two control methods. The model was
selected from the EPA Guideline Series document Control of Hydrocarbons from Tank
Truck Gasoline Loading Terminals. The use of this model was considered to be
acceptable because petroleum loading operations compare reasonably well with
those of the SOCMI.
Two controls taken from the cited document are evaluated: a refrigerated con-
denser system that condenses VOC vapors at atmospheric pressure and -73°C
temperature, and an oxidation system based on a thermal oxidizer.
Table V-9 shows the theoretical control efficiencies of the condenser for the
four vapor pressures selected. The control efficiency of the thermal oxidizer
was estimated to be 99%.
Table V-10 gives the VOC emissions expected from the model terminal before
controls and after either the refrigeration or teh oxidizer system was
installed.
-------
Table V-8. VOC Emission Reduction from Installation of New Pressure Vessels Instead of New
Floating-Roof Tanks as a Function of Stored-Material Vapor Pressures
Model Tank
Size
(m3)
40
75
115
150
285
660
1420
2840
8330
VOC Emission Reductions (Mg/yr)
51.7
(98%)
0.57
0.58
0.60
0.74
1.13
1.44
2.53
4.23
9.42
kPa
(100%)
0.58
0.60
0.61
0.75
1.16
1.47
2.59
4.31
9.62
69.0
(98%)
0.92
0.94
0.96
1.18
1.84
2.36
4.15
6.94
15.51
kPa
(100%)
0.94
0.96
0.98
1.20
1.88
2.40
4.24
7.08
15.83
76.6
(98%)
1.19
1.21
1.23
1.51
2.38
3.04
5.37
8.99
20.10
kPa
(100%)
1.22
1.23
1.25
1.54
2.43
3.10
5.48
9.17
20.51
86.2
(98%)
1.82
1.84
1.85
2.28
3.63
4.65
8.21
13.77
30.82
kPa
(100%)
1.86
1.88
1.89
2.33
3.70
4.74
8.38
14.05
31.45
100.7
(98%)
17.53
17.54
17.56
21.66
34.88
44.75
79.34
133.24
298.79
kPa
(100%)
17.88
17.90
17.92
22.11
35.59
45.66
80.96
135.96
304.88
<
i
o
-------
V-ll
Table V-9. Control Parameters of Model
Loading Terminal for Refrigeration System
Theoretical
Control
Efficiency (%)
60
86.7
93.3
94.9
Vapor Pressure
Initial
3.5
10.3
34.5
69.0
(kPa)
Final
1.4
1.4
3.5
3.5
Table V-10. VOC Emissions from Model Loading Terminal
Before and After Addition of Controls
VOC Emission (Mq/yr)
Vapor Pressure
(kPa)
a
3.5
b
10.3
34. 5C
69. Od
Before Controls Added
15.8
76.3
123.8
267.8
After
Ref riqerati
6.3
10.5
8.3
13.7
Controls Added
.on Oxidizer System
0.2
0.8
1.2
2.7
Uncontrolled emission rate, 55 mg per liter loaded (derived from Table III-3).
Uncontrolled emission rate, 265 mg per liter loaded (derived from Table III-3).
"Uncontrolled emission rate, 430 mg per liter loaded (derived from Table III-3).
Uncontrolled emission rate, 930 mg per liter loaded (derived from Table III-3).
-------
V-12
Emissions were calculated with the equations discussed in Sect. Ill and averaged
from the vapor pressures given in Table III-3.
The energy effect of vapor recovery systems at terminals is considered to be
minimal. Although energy is required to drive compressors, pumps, and other
equipment, in most systems a valuable product is recovered that would otherwise
be lost to the atmosphere. In thermal-oxidizer systems additional energy will
be required in the form of gaseous fuel to convert the VOC vapor to carbon dioxide
and water. An estimated 85 liters of propane per Mg of product per year is
required by the oxidizer.
B. CONTROL COST IMPACT
This section presents estimated costs and cost-effectiveness ratios for the
additional control of VOC emissions from the storage and loading of organic
chemicals in the synthetic organic chemicals industry. Present emission sources
and emission quantities are discussed in Sect. III.
The capital cost estimates represent the total investment involved in purchasing
and installing all equipment and material necessary for a complete control sys-
tem that will perform as defined for a new installation at a typical location.
These estimates do not include such highly variable costs as loss of production
during installation or startup, research and development, land acquisition, and
preparation and cleaning of the tanks for retrofitting. Cost estimate calculations
are included in Appendix C.
Cost factors for calculation of annual operating costs are shown in Table V-ll-
The capital recovery factor of 0.18 is based on a 10-year depreciable life and
a 12% annual interest rate. Emission recovery credit is based on the average
market value of all synthetic organic chemicals that have potential recovery.
Annual costs are for a 1-year period beginning in December 1979.
1. Storage Tanks
a. Fixed-Roof-Tank Emission Controls The capital cost and the cost effectiveness
of controlling VOC emission losses with two control techniques were evaluated:
-------
V-13
Table V-ll. Cost Factors for Computing Annual Costs
Item
Factor
Electrical power
Operating time
Operating labor
Maitenance material and labor
Capital charges
Capital recovery
Miscellaneous {taxes, insurance,
and administration)
Liquid-waste disposal
Recovery credit
8.33 $/GJ
8760 hr/yr
Minor; not considered
$0.06 X capital cost
$0.18 X capital cost
$0.05 x capital cost
Minor; not considered
$330.8/Mg
Table V-12. Installed Capital and Operating Costs for
Internal-Floating-Roo fsa
Model
Tank Size
(m3)
40
75
115
150
285
660
1420
2840
8330
Installed Capital Cost (X
Floating
5.6
5.6
5.6
6.0
7.1
8.0
12.0
20.4
50.3
Secondary
Roof Seal =
0.6
0.6
0.6
0.6
0.9
1.0
1.5
2.1
3.6
103)
Total
6.2
6.2
6.2
6.6
8.0
9.0
13.5
22.5
53.9
Annual Operating Cost
(X 103)
1.8
1.8
1.8
1.9
2.3
2.6
3.9
6.5
15.6
A contact internal floating roof with a secondary seal installed in a fixed-
roof tank.
^Before credit.
-------
V-14
Internal floating roof used on fixed-roof tanks—Table V-12 shows the installed
capital cost and the annual operating cost for the installation of a contact
internal floating roof with a liquid primary seal and a continuous secondary
seal to an existing fixed-roof tank. The annual operating cost is the cost of
the internal floating roof before recovery credit is taken for the emission
reductions. The installed capital costs for the roof and seal were taken from
Table C-l of Appendix C.
The installed capital costs for tanks shown in Table C-l were prepared primar-
ily from data presented in the EPA report Volatile Organic Compound Emissions
2
from Organic Liquid Storage Tanks, which was prepared by TRW.
Table V-13 gives the cost effectiveness for installing a floating roof to an
existing fixed-roof tank. The cost effectiveness is affected by the vapor
pressure of the material stored in the model tank. The cost effectiveness
versus tank size is plotted in Fig. V-l.
Refrigerated vent condensers (new and retrofit sources) To evaluate the con-
trol cost effectiveness of adding a refrigerated vent condenser to a new or
existing fixed-roof tank, the models and conditions discussed in the environ-
mental and energy impact portion of this section were used; see Tables V-4—V-6.
Table V-14 gives the cost effectiveness for adding refrigerated vent condensers
to the fixed-roof model tanks for all three cases. Cost estimation details are
given in Appendix C. Figure V-2 is a plot of the cost effectiveness for the
three model conditions.
The cost effectiveness for multiple storage tanks containing the same material
and using a common add-on refrigeration unit should be more favorable than that
for equivalent storage tanks using individual controls. These highly variable
situations were not modeled.
Other control techniques Due to time limitations cost-effectiveness calculations
were not completed for carbon adsorption and thermal oxidation. Installed capital
cost curves for both techniques are, however, included in the cost estimate
details (Appendix C).
-------
Table V-13. Cost Effectiveness of Installing an
Internal Floating Roof in a Fixed Roof-Tank
Model Tank
Size
40
75
115
150
285
660
1420
2840
8330
1.0 kPa
12,530
6,590
4,535
3,545
2,820
1,555
1,085
1,325
1,985
1.4 kPa
9665
5125
3420
2590
2090
1115
750
940
1460
Cost Effectiveness
3.4 kPa 6
4815
2395
1485
1190
880
390
40
310
605
($/Mg)
.9 kPa
2625
1195
710
485
360
80
(25)
40
215
at Vapor Pressures
10.3 kPa
1650
700
370
220
135
(55)
(125)
(80)
40
of
34.5 kPa
580
140
(15)
(80)
(115)
(205)
(235)
(215)
(155)
51.7 kPa
270
(20) a
(120)
(165)
(190)
(245)
(265)
(255)
(215)
I
M
l/I
Values in parentheses represent credits.
-------
Cn
13
0)
-P
G
O
u
3
o
-H
W
W)
U)
O
O
13,000
12,000
10,000
8,000
6,000
n
O
_ 4,000
•rH
T)
OJ
u
2,000
tn
3
10
o
1.0 kPa
1.0 kPa
34.5 kPa
51.7 kPa
M
(TV
J_L
100
1000
10,000
Tank Size (m )
Fig. V-l. Cost Effectiveness of Controlling Emissions by Retrofitting
Existing Fixed-Roof Tank with Internal Floating Roofs
-------
V-17
Table V-14. Cost Effectiveness for Add-On Refrigerated Vent
Condenser to Fixed-Roof Tank (New and Retrofitted Sources)
Model Tank
Size
(n>3)
150
660
2840
Case 1
$10,430
8,030
5,245
Cost Effectiveness* (per Mg)
Case 2
$3,031
2,470
1,890
Case 3
$1,693
2,585
2,085
*Case conditions described in Table V-4.
-------
V-18
10,000
9,000
8,000
2
^: 7,ooo
T)
3 6,000
o
-p
5,000
o
CO
c
•% 4,000
w
•H
g
w
0 3,000
£ 2,000
4-1
01
O
u 1,000
0 _
100
:ase
-Case 2
1000
Tank Size (ra3)
5000
Fig. V-2. Cost Effectiveness of Controlling Emissions by
Refrigerated Vent Condenser Added On to Fixed-Roof Tank
-------
V-19
b. Floating-Roof-Tank Emission Controls The capital cost and the cost effective-
ness of controlling VOC emission losses from an internal floating-roof tank
were evaluated:
To evaluate the control cost effectiveness of installing new, low-pressure service
(172 kPa maximum) vessels instead of new floating-roof tanks, the models and
emission reductions discussed in Sect V-A were used. The annual operating cost
was calculated from the incremental capital cost difference between a new, internal-
floating-roof tank and a new pressure vessel.
Table V-15 shows the installed capital cost and the annual operating cost, before
credit was taken for emission recovery, for each of the model tank sizes.
Installed capital costs are based on the cost data for fixed-roof tanks factored
by engineering estimates to represent current low-pressure service vessel costs.
Cost estimation details are given in Appendix C.
Table V-16 gives the cost effectiveness for installing new pressure vessels at
the 98%- and 100%-control conditions. Figures V-3 and V-4 are plots of the
cost effectiveness for the new-pressure-vessel installation at 98%- and 100%-
control efficiency conditions respectively.
2. Loading and Handling
To evaluate the control cost effectiveness of installing either the refrigera-
tion or oxidizer system for loading terminals, the model and emission reductions
described in Sect. V-A were used.
All installed capital cost and operating cost data shown in Table V-17 were
taken from the EPA Guideline series document Control of Hydrocarbons from Tank
Truck Gai
dollars.
Truck Gasoline Unloading Terminals and adjusted to reflect December 1979
Table V-18 gives the cost effectiveness for installing either the refrigeration
or the oxidizer system at the model terminal for loading chemicals with four
selected vapor pressures.
-------
V-20
Table V-15. Installed Capital and Operating Costs for
Pressure Vessels
Model Tank
Capacity (m )
40
75
115
150
285
660
1420
2840
8330
Installed Capital
Cost (X 103)a
$ 7.8
5.8
4.2
6.8
14.0
34.5
64.6
102.9
242.7
Annual Operating
Cost (X 103)b
$ 2.3
1.7
1.2
2.0
4.1
10.0
18.7
29.8
70.4
Incremental cost difference between new pressure vessel and new
floating-roof tank.
29.0% of capital cost before recovery credit.
-------
Table V-16. Cost Effectiveness for Installing a New Pressure Vessel Instead of a New Floating-Roof Tank
(Incremental Cost Difference) as a Function of Stored-Material Vapor Pressures
Model Tank
Size
(m3)
40
75
115
150
285
660
1420
2840
8330
Cost Effectiveness (per
98% Control Efficiency at
51.7
kPa
$3704
2573
1667
2446
3263
6579
7067
6722
7142
69.0
kPa
$2167
1489
941
1361
1898
3907
4169
3963
4209
76.6
kPa
$1593
1075
652
995
1387
2959
3148
2982
3169
Vapor Pressures of
86.2
kPa
$ 933
600
324
526
800
1830
1948
1838
1953
100.7
kPa
($200)*
(234)
(262)
(240)
(212)
(107)
(95)
(107)
(95)
Mg)
100% Control Efficiency at
51.7
kPa
$3627
2521
1634
2397
3198
6450
6886
6588
6989
69.0
kPa
$2123
1460
922
1333
1860
3829
4085
3883
4119
76.6
kPa
$1561
1053
639
975
1359
2900
3085
3422
3100
Vapor Pressures of
86.2
kPa
$ 914
587
317
515
756
1772
1897
1794
1908
100.7
kPa
($202)
(235)
(265)
(240)
(214)
(112)
(100)
(112)
(100)
I
to
*Values in parentheses represent credits.
-------
7500
7000
6000
^ 5000
o
M
.p
o
u
en
C
0
•^
in
to
-M
cn
0
U
4000
3000
2000
1000
T
50
10
51.7 kPa
86.2 kPa
100.7 kPa -
I I L
to
to
I I I I I I
100
1000
10,000
Model Tank Size (m )
Fig. V-3. Cost Effectiveness of Controlling Emissions (100%) by
Installing New Pressure Vessel vs New Floating-Roof Tank
-------
r-l
O
-P
C
O
u
w
C
o
w
UJ
•rH
CJ
M
O
in
O
a
8000
7000
6000 _
5000 —
4000 —
3000 -
2000 -
1000 -
to
00
10,000
Model Tank Size (m )
Fig. V-4. Cost Effectiveness of Controlling Emissions (98%) by
Installing New Pressure Vessel vs New Floating-Roof Tank
-------
V-24
Table V-17. Installed Capital and Operating Cost for Model-
Loading-Terminal3 Control Systems (950 m /day)
Control System Installed Capital Cost (X 1Q3) Annual Operating Cost (X 10 )
Refrigeration $186.6 $54.1
Oxidizer 148.4 43.0
.Top-submerged or bottom-filled.
Based on 29.0% of capital before recovery credit.
Table V-18. Cost Effectiveness for Loading-Terminal
(950 m /day) Control Systems vs Material Vapor Pressures
Cost Effectiveness (per Mg)
Vapor Pressure Refrigeration Oxidizer
(kPa) System System
3.5 $5368 $2756
10.3 491 570
34.5 138 351
69.0 (118)* 162
'^Values in parentheses represent credits.
-------
V-25
C. REFERENCES*
1- Control of Hydrocarbons from Tank Truck Gasoline Loading Terminals, Guideline
Series, EPA-450/2-77-026 (OAQPS No. 1.2-082), Research Triangle Park, NC
(October 1977).
2. Volatile Organic Compound Emissions from Volatile Organic Liquid Storage
Tanks, draft, chapters 3—6 and preliminary capital costing, Apr. 28, 1980
(on file at EPA, ESED, Research Triangle Park, NC).
*A reference located at the end of a paragraph usually refers to the entire para-
graph. If another reference relates to certain portions of the paragraph, the
reference number is indicated on the material involved. When the reference
appears on a heading, it refers to all the text covered by that heading.
-------
VI-1
VI. ASSESSMENT
A. SUMMARY
1. Storage
The synthetic organic chemicals industry currently uses a total of 20,270 fixed-
roof tanks, floating-roof tanks, and pressure vessels for VOC storage. The
primary type of storage vessel utilized by the industry is the fixed-roof tank.
It comprises 73% of the total number of storage vessels and contributes 61% of
the total storage volume. Its primary drawback is its high susceptibility to
breathing and working losses. The emission sources and control levels for industry
storage are shown in Tables III-4 and III-5. Current emissions for all indus-
trial storage are estimated to be 31,413 Mg/yr. Fixed-roof tanks account for
an estimated 94% of the total storage VOC losses.
Pressure vessels are the most efficient storage tanks and have negligible emis-
sions during normal operation. The use of internal-floating-roof tanks is a
significant control measure, with internal floating roofs being an average of
85% more effective in controlling VOC emissions than fixed-roof tanks. Refriger-
ated vent condensers are one of the most common types of control devices now in
use on fixed-roof tanks with control.
Fixed-roof-tank emissions can be controlled by means of internal-floating-roof
tanks and refrigerated vent condensers in a new plant or by retrofitting float-
ing roofs or refrigerated vent condensers on existing fixed-roof tanks. The
emission reduction would be 76% to 97% for floating roofs and 60% to 95%+ for
refrigerated vent condensers. The cost effectiveness for installing internal
floating roofs can yield credits ($200+/Mg) at high vapor pressures. Refriger-
ated vent condensers on individual fixed-roof tanks are not as cost effective
as floating-roof tanks.
Floating-roof-tank emissions can be eliminated by installation of pressure ves-
sels in new plants. Cost effectiveness for pressure vessels is most favorable
at high vapor pressures and small tank sizes.
-------
VI-2
2. Loading
An estimated 3,393,100 Mg of synthetic organic material is shipped by the industry.
Cargo carriers include tank cars, tank trucks, and marine vessels, with water
transportation accounting for 50% of the total volume of chemicals shipped. An
estimated 81% of all chemicals shipped have vapor pressures of less than 34.5 kPa.
The emission sources and control levels for the industry loading are shown in
Fig. III-6. Current emissions for all chemical loading and transport are esti-
mated to be 4917 Mg/yr.
The industry currently uses both splash-filling and submerged- or bottom-filling
methods for loading all types of carriers with a wide vapor pressure range of
chemicals. The use of refrigerated condensers or of a vapor oxidizer is a signifi-
cant control measure when applied to chemicals with vapor pressures in excess
of 3.5 kPa. VOC emission reductions of 70 to 90% are possible. The use of
pressurized carriers eliminates transit losses.
Refrigerated condensers systems can have favorable cost effectiveness if used
on high-vapor-pressure materials, yielding credits as high as $120 per Mg of
VOC controlled.
B. SUPPLEMENTAL INFORMATION
The primary basis for characterization of the storage and handling segment of
the SOCMI was the State Emissions Inventory files obtained from Texas and Louisiana.
It would be appropriate to update the present data base when supplemental tank
data from additional site visits resulting from new-product study assignments
and other State Emissions Inventory files become available.
-------
APPENDIX A
EMISSION CALCULATION SUPPLEMENTS
-------
A-3
Table A-l. Paint Factors for Fixed-Roof Tanks
Paint Factors (F )
Tank
Roof
White
Aluminum (specular)
White
Aluminum (specular)
White
Aluminum (diffuse)
White
Light gray
Medium gray
Color
Shell
White
White
Aluminum (specular)
Aluminum (specular)
Aluminum (diffuse)
Aluminum (diffuse)
Gray
Light gray
Medium gray
Paint
Good
1.00
1.04
1.16
1.20
1.30
1.39
1.30
1.33
1.40
Condition
Poor
1.15
1.18
1.24
1.29
1.38
1.46
1.38
1.44*
1.58*
*Estimated from the ratios of the seven preceding paint factors.
-------
Table A-2. Tank, Type, Seal, and Paint Factors for Floating-Roof Tanks
Tank Type
Welded tank with pan or pontoon
roof, single or double seal
Riveted tank with pontoon roof,
double seal
Riveted tank with pontoon roof,
single seal
Riveted tank with pan roof,
double seal
Riveted tank with pan roof,
single seal
Painted Color of
K * Seal Type K * Shell and Roof K*
*- S y
0.0045 Tight fitting (typical of modern 1.00
metallic and non-metallic seals)
0.11 Loose fitting (typical of seals 1.33
built prior to 1942)
0.13
0.13 Light gray or 1.0
aluminum
0.14 White 0.9
*See "Calculations" in Sect. Ill
-------
A-5
Table A-3. S Factors for Calculating Petroleum Loading Losses
Cargo Carrier
Mode of Operation
S Factor
Tank trucks and tank cars
Marine vessels
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 1.45
dedicated service
Submerged loading: 1.00
dedicated, vapor balance
service
Splash loading: dedicated 1.00
vapor balance service
Submerged loading: ships 0.2
Submerged loading: barges 0.5
*Saturation factor,- see "Calculations" in Sect. III.
-------
APPENDIX B
ALTERNATIVE FLOATING-ROOF-TANK EMISSION ESTIMATION EQUATIONS
-------
B-3
FLOATING-ROOF-TANK EMISSIONS EQUATIONS
VOC emissions from external-floating-roof, noncontact internal-floating-
roof, and contact internal-floating-roof storage tanks can be estimated with
equations based on a pilot test tank study conducted for EPA.
From the equations presented below, it is possible to estimate the total
evaporation loss (L ), which is the sum of the withdrawal loss (L ), the
seal loss (L,,), and the fitting loss (L ) :
S r
LT = LWD + LS + V
L = K V M D
LS KS V "V D f , P .0.5 2 2205
1 U " 14.7 ' J
T - MK vn M (14.7)
LF - mF V My . P 0.5,2
where
LT = total loss (Mg/yr);
L = withdrawal loss (Mg/yr);
WD
L = seal loss (Mg/yr);
L = fitting loss (Mg/yr);
M^ = molecular weight of product vapor (Ib/lb-mole); 78.1 Ib/lb-mole
for VOC;
P = true vapor pressure of product (psia); 2 psia assumed;
D = tank diameter (ft);
W = density of product (Ib/gal); 7.37 Ib/gal for VOC;
V = average wind speed for the tank site (mph); 10 mph assumed average
wind speed;
Q = product average throughput (bbl/yr) ,- tank capacity (bbl/turnover) X
turnovers/yr;
"''Chicago Bridge and Iron Co., Plainfield, IL, Hydrocarbon Emission Loss on
a 20 Foot Diameter Pilot Test Tank with an Ultraflash and a CBI Weathermaster
Internal Floating Roof, R-0113/R-0191 (June 1978).
-------
B-4
K = seal factor; see Table B-l;
KF = fitting factor; see Table B-2;
n = seal wind speed exponent; see Table B-l,-
m = fitting wind speed exponent; see Table B-2;
C = product withdrawal shell clingage factor [bbl/(ft X 10 )],- use
0.0015 bbl/(ft X 10 ) for VOC in a welded steel tank with light
rust;
N = fitting multiplier; see Table B-3.
-------
B-5
Table B-l. Emission Factors K and n
Roof and Seal Combinations
Contact internal floating roof
Liquid-mounted primary seal only
Liquid-mounted primary seal and
continuous secondary seal
Non-contact internal floating roof with
vapor-mounted primary and secondary seals
External floating roof
Primary seal only
Primary and secondary seals
v a
KS
12.7
3.6
10.3
48.6
57.7
b
n
0.4
0.7
1.0
0.7
0.2
K = seal factor.
O
3n = seal wind speed exponent.
Table B-2. Summary of Emission Factors KF and m For Floating Roofs
Case
Number
1
2
3
Roof Description
Contact internal floating roof
Non-contact internal floating
roof
External floating roof
*/
132
309
0
b
m
0
0.3
0
iv, = fitting factor.
m = fitting wind speed exponent.
-------
B-6
Table B-3. Fitting Multipliers
Tank
20 •
75 •
100 «
125 •
150 «
175 "
D
Diameter
(ft)
< D -
« D <
< D «
« D -
« D <
: D <
: D <
e 20
c 75
s 100
: 120
: 150
: 175
: 200
N
Fitting
Multiplier
0.5
1
2
3
4
5
6
Chicago Bridge and Iron Company. Hydrocarbon Emission Loss
Measurements on a 20 Foot Diameter Pilot Test Tank with an Ultra-
flote and a CBI Weathermaster Internal Floating Roof. Report
No. R-0113/R-0191. Plainfield, Illinois. Juen 1978.
-------
APPENDIX C
COST ESTIMATE DETAILS AND CALCULATIONS
FLOATING ROOF - REFRIGERATED VENT CONDENSER - PRESSURE VESSEL
-------
C-3
COST ESTIMATE DETAILS
This appendix contains the details of the estimated costs presented in this
report.
Installed capital costs for fixed-roof tanks, floating roofs and secondary
seals were derived from Tables C-l through C-4. These tables were obtained
from an EPA, ESED draft report, Volatile Organic Compound Emissions from
Volatile Organic Liquid Storage Tanks.
The refrigerated vent condenser installed capital costs shown on the evaluation
sheets were derived from the cost curves shown in Figs. C-l and C-2 and
includes the capital required for installing a condenser and a refrigeration
system. The curves were taken from the Condensation, Control Device Evaluation
report prepared by IT Enviroscience for EPA, ESED in Research Triangle Park,
NC.
Pressure vessel installed capital costs were obtained by applying engineering
estimates to floating-roof tank costs. The methodology is shown in Table C-5.
Also included in this Appendix are installed capital cost curves for a carbon
adsorption unit (Fig. C-3) and thermal oxidation unit (Fig. C-4). They were
obtained from the Carbon Adsorption Control Device Evaluation report, and from
The Thermal Oxidation Control Device Evaluation report, both prepared by
IT Enviroscience for the EPA, ESED in Research Triangle Park, NC.
-------
C-4
Table Cl. COST3 OF BUILDING A FIXED ROOF TANKb
Diameter (ft)
20
40
45
50
60
70
80
90
Height (ft)
24
40
40
40
40
40
40
40
Installed cost0
26,000
63,000
73,000
89,000
113,000
139,000
168,000
202,000
Costs obtained from Thomas Tremblay, of Chicago Bridge and Iron,
February 27, 1980.
Equation derived from the available data.
Cost ($) = -1783.02 + (1.099 x 10J x D) + (14.028 x D^) - (.01301 x D3)
where D = diameter in feet.
Costs in first quarter 1980 dollars.
-------
05
Table G2. COST9 OF INSTALLING A CONTACT SINGLE SEAL
INTERNAL FLOATING ROOFb
Diamter (ft)c
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
Installed cost ($)d
5,630
6,485
7,510
8,780
10,280
11,970
14,180
16,190
19,630
21 ,480
25,120
27,770
31,240
34,360
37,910
41,530
aCost obtained from Bill Blumquist, of Petrex, February 20, 1980.
Equation derived from available data. 9
Cost ($) = 5761 - (58.15 x D) + (5.335 x D^).
cCost of adding a secondary seal, per foot of circumference is $11.85,
dCosts in first quarter 1980 dollars.
-------
C-6
Table C3. COST3 OF INSTALLING A NON-CONTACT SINGLE
SEAL INTERNAL FLOATING ROOFb
Diameter (ft)c
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
Installed cost ($)d
4,500
6,000
7,300
9,300
10,000
12,000
13,300
14,700
16,500
17,400
19,700
21,500
23,500
25,200
27,800
30,000
Cost obtained from Henry Reiss, of Altech Industries, Inc., February 26, 1980.
Equation derived from available data. ~
Cost (S) = 1614.3 + (198.5 x D) + (1.26 x Dd).
Installed cost of adding a secondary wiper seal per foot of circumference
is $22.50.
Costs in first quarter 1980 dollars.
-------
C-7
Table C4. COST3 OF BUILDING AN EXTERNAL FLOATING
ROOF TANKb
Diameter (ft)c
30
35
42.6
52
60
67
73.4
100
Height (ft)
24
30
40
40
40
40
40
40
Installed Costd
40,000
54,000
81 ,000
117,000
126,000
140,000
153,000
212,000
aCosts obtained from George Stilt, of Pittsburg Des Moines,
April 10, 1980.
Equation derived from available data.
D < 60 COST = -22.2587 x D2 + 4869.95 x D - 86065.68
D > 60 COST = 5.078 x D2 + 1337.52 x D + 27467.96
where D = diameter in feet
°Cost of adding a secondary seal per foot of circumference
is $13.50.
Costs in first quarter 1980 dollars.
-------
C-8
10,000
T3
c
o
O
_C
4ft
Q.
C
O
_o
2
CO
_c
en
OJ
O
O)
Q
1,000 -
100
10
10
100 1,000
Condenser System Area (Ft2)
10,000
Fiq. C-l. Installed Capital Cost vs Condenser Area for Various
Materials of Construction for a Complete Condenser Section
-------
C-9
1,000
10 100
Refrigeration Capacity -Tons (12,000 BTU/Hr)
1,000
Fig. 6-2. Installed Capital Costs vs Refrigeration
Capacity at Various Coolant Temperatures for a
Complete Refrigeration Section
-------
Table C-5. Pressure Vessel vs Floating-Roof Tank Installed Capital and Operating Cost Differential
a
Tank Fixed Roof
Size Installed Capital
1UJ aal (103 S)
10
20
30
40
75
175
375
750
2200
From Table C-l.
17.8
17.8
17.8
20.9
30.8
37.9
63.8
104.0
224.9
Floating-Roof Plus
Secondary Seal
Installed Capital
(103 5)
6
6
6
6
8
9
13
22
53
.2
.2
.2
.6
.0
.0
.5
.5
.9
Q
Floating-Roof
Tank
Installed Capital
UO3 S)
24.
24.
24.
27.
38.
46.
77.
126.
278.
0
0
0
5
8
9
3
5
8
25 psid
Pressure Vessel
Scaleup Factor
1
1
1
1
1
1
1
1
1
.32
.24
.18
.25
.36
.74
.84
.81
.87
25 psi6
Pressure Vessel
Installed Capital
(103 S)
31.8
29.8
28.2
34.3
52.8
81.4
141.9
229.4
521/5
25 psi
Pressure Vessel (29.0% of Capital)
Incremental Annual Operating
Capital Difference Cost
(103 S) (103 S/yr)
7.8
5.8
4.2
6.8
14.0
34.5
64.6
102.9
242.7
2.
1.
1.
2.
4.
10.
IB.
29.
70.
3
7
2
0
1
0
7
8
4
n
i
Total of fixed-roof tank plus floating-roof and secondary seal.
Based on IT Enviroscience engineering estimates.
EObtained by multiplying scaleup factor times floating-roof tank installed capital.
Difference between pressure vessel and floating-roof tank installed capital.
-------
A- 2 beds, vertical:
O 900 Ib of carbon; bed, 4-ft Jj .im by 3 ft deep
A 450 Ib carbon; bed, 4-ft diam by 1} ft deep
B- 2 beds, vertical:
O 4500 Ib of carbon; bed, 8-ft diam by 3 ft deep
A 2250 Ib of carbon; bed, 8-ft diam by 1J ft deep
C- 3 beds, horizontal
O 9OOO Ib of carbon; bed, 8-ft diam, 15 ft long, 3 ft deep
A 4500 Ib of carbon; bed, 8-ft diam, 15 ft long, 1J ft deep
D- 3 beds, horizontal
O 22,500 Ib of carbon; bed, 11-ft diam, 26 ft long, 3 ft deep
A 11,250 Ib of carbon; bed, 11-ft diam, 26 ft long, 1) ft deep
E- 4 beds, horizontal
O 30,000 Ib of carbon; bed, 12-ft diam, 30 ft long, 3 ft deep
A 15,000 Ib of carbon; bed, 12-ft diam, 30 ft long, 1! ft deep
— n
100
FLOW (1000 SCfm)
Fig. C-3. Installed Capital Cost of Carbon Adsorption Systems
-------
7,000 i
o
o
q_
to
<
tz
CL
-------
TANK, Si I
|OB SAL.
Ao
160
. - I -
5* E , 0-5 IVP, O.'XPVP (P
\2£,7 I .Temp, IO F.Tem p (.' -
O.llTon f? el r i
foo P f 2 Condenser
I GPM 8 HP, Pomp
Cap
-3M*
O-loTon Re
F -I-1 Conde
nser
l SPM , 8 HP Pump
Cap = IOSM
-2.. i Ton tPeJngerjTor
I iso Ft* Condenser
ISFM , 8 HP Pump
Cap-- ROM*
2.325 TPV >-<3
-TF>Y HC Recovered
EC.SLbJHr »
I9.36S TPY HC Emission^
25.3 Hr /Yr Operating
11.381 TPY HC Recovered
32.5 Lb/Wr '
, 1.5 IVP,O.EFVP (P3IA)
O. 39 Ton R=-fri
°i°> F+2 Condenser
I &PM S HP Pump
'i.
Pu
l.«3 Ton
5O3 ^V* Condenser
g
m
Cap = l^<£ ,V\"*
7.1 Ton ReJngera+or
ie>&"7 F-? Condenser
IGPfvV a HP PamP'4
Cap = •2^o M
HC Emissions
HC "Recovered
18.3 VbjHr "
1€>,OZ3TPY HC Emission's
Hr/^r Ofxsratmcj
Recovered
96, S Lb/nr
TPY HC Emissions
E3O Mr/Yr Opera + ing
4I.IS7TPY HC Recovered
.S Lb/Hr '
C ASE -3 ~
E:,T.S IVP.0.5 FVP (PSIA)
=>I */. Sa-rura+ion
I.& Ton
101 Ft1 Condenser
> &PM , § HP Pump
8.5 Ton Relngerator
1063 Ft1 Condenser
31. T Ton Re4rigera+or
Ftz Condenser
, 4" HP Pump
Cap 5 errs M
IT.SaiTP'Y HC Emissions
SOT Hr/Yr O pera-tmc.
ife.TSSTPY HC Recovered
&6.I Lb/Hr "
5U7Z6 TPY HC Emissions
H.SO Hr/Y<- Opera+mc^
4Q.48O TPY HC Recovered
».-a Lb/Hr"
TPY HC Emissions
aQl H>/Y i- Opera+mr,
142.o&z "T'PY HC Recovered
LbjHr »
.M's( U-'
\ AIM I Ml S INt
2 . Pi J r-. p H r L . v. i • ) • >y i-
£ p.
tacK. in n 1 j I -i ^ . -11
l/s + ed eq^)> [-• i... n»
pipi ng , in 2 L.I Ij » ioi-
en+a + ^n.vj)..r;^
STOKAGE
n
,-i i- -l &
-------
C-14
* o T i \
J-i-mti-s ^~ ibujer Use
>O I O.R F. 5- I I- 73
SE Z All
( r\
«a-V. Hotter = O0(o~l x OollTon X " "__._..... 4 I XwM
'..ard Liqh-f & ) nstromen-f s O.I KW x 876O ^5O". ON . . 4 3 S >v
I - (4o) ------- c- 4 7 ^ "
Ga|,
Pomp
. oouer- = O^^ x .ITon X ^5hr/u __.._. I 84
t Ins 4-. >SS
O v\
- -- - •*/
Hbruc, FUJT - 5.*2x IO"S x Sa,5/Kx 35^ K/u _ _ ___ -~x~~^_
"Pe^, " - 0.a 1.- • 5.^x )O"S x 18,*} X 50"? _ _ , „
"""pei.Puu^. (.231% . SS x 507 _ _. ._. , _ . ^37
Yard. L fc lnt>+ ____ ^j 3S
3 - (4q)
115 M Gal,
~^
-------
C-15
A Kl D V- 1'
&rpf.-re-r:- gN nr
'./ Mil,
w H T Er.T.
WASTE: £ RATE, o£ H'DLG.
PeC. l3-7«t INSTALLED
TARITA 1 rO&T
UTILITIES :
Jt
Pon.)/M^
i j
3oo ! &OO '• Z.4-OQ
o.q^ ' 2.54- l i. 13
;
9.10O ! "Zc2,4-OO ^7,4-OCi
\ ;
»O4"5o ' £L£>3C : 5^45
i
i
i ,
-------
...z: rzsA'
C-16
\A G^iT'p.
or
A G-«si Tk
W/ASTE tt. RPvTE. OP MDLG.
INSTALLED
!!CAE1TAL_£QSI_ .. ..
41, Coo
730 Ga
f-i,oo<2 -
UTILITIES :
^
R.AW KAATLS:
N o T
MANPOWER
\j
!(.
j_MAJ_£
CAPITAL R.ECQVER.V
^00
MISC. CAP. COSTS
_v_j
WASTE DISPO^R
----J^9-
S.ECQYJLfLY_
O, ZOO
\ OC R r- _• o\' F.K !'- O ( M£-, / VR
3, *5 3
- 5 k
NE1T A^slNUAL,^ZEO COST
10,700
XT
"26.
COST E _F F E CT IV E N E Sg^./Afc;
-------
C-17
i_~_tll
....CAS^-NO^- -
WASTE 9- RATE. OF MDLG.
^P^C. 12
UTILITIES 1
TCVLLED
1"
i
i i
£50 ooo : 34- Looo ' ~72':£^e
\
\ \
^j— -^— V **! '
U .
1 R.AW MAT L5 '
MANI PO W E F^
r> ^ A I M T F" M /X N C F $ ' ''a '
• \
i CAPITP\L RpCOVER-Y ! 6 ^ /^
!r i
i - — "
il /VMSC CAP« COSTS C"-> '
!i VAs/^\c,TFr Ol ^ P O .^i P\ 1
iv)pq- M^-^ i 4 co
vl ; xJ _ :
, |
Kin-- ! N^^^ ^ X^^-.
-
,
i
i
I
i
i
•
Nlie^ ! Ncq. ; Nc-tj
vi . xl J
LECOyJL
NET ANNUAHZED COST
COST EFFECT IV ENE SS(tAfe
O
-------
2-i
REPORT 2
FUGITIVE EMISSIONS
D. G. Erikson
V. Kalcevic
IT Enviroscience
9041 Executive Park Drive
Knoxville, Tennessee 37923
Prepared for
Emission Standards and Engineering Division
Office of Air Quality Planning and Standards
ENVIRONMENTAL PROTECTION AGENCY
Research Triangle Park, North Carolina
September 1980
D28I
-------
2-iii
CONTENTS OF REPORT 2
I. ABBREVIATIONS AND CONVERSION FACTORS
II- CHARACTERIZATION AND DESCRIPTION
A. Introduction
B. Sources
C. Methodology
D. References
III- EMISSIONS
A. Introduction
B. Fugitive-Emission Estimates
C. References
IV. CONTROL TECHNOLOGY
A. Control Devices
B. Leak Detection Methods
C. Maintenance
D. References
V. COST ANALYSIS
A. Introduction
B. Control Costs
C. References
VI. ASSESSMENT
A. Summary
Page
1-1
II-l
II-l
II-2
II-9
11-16
III-l
III-l
III-3
III-6
IV-1
JV-1
IV-8
IV-11
IV-13
V-l
V-l
V-l
V-14
VI-1
VI-1
APPENDICES OF REPORT 2
A. TOTAL NUMBER OF SOCMI PRODUCT SITE LOCATIONS
B. CHARACTERIZATION OF MODEL PLANTS AND NUMBER OF PUMPS AND
VALVES VERSUS PLANT CAPACITY
C. COST ESTIMATE DETAILS
D. LIST OF EPA INFORMATION SOURCES
A-l
B-l
C-l
D-l
-------
2-v
TABLES OF REPORT 2
Number
II-l SOCMI Valve Characterization 11-10
II-2 SOCMI Pump Seal Characterization 11-10
II-3 Equipment Data for Three Model Plants 11-12
II-4 Current SOCMI Equipment Component Estimate 11-15
III-l Uncontrolled Fugitive Emission Factors for the Synthetic III-2
Organic Chemical Manufacturing Industry (SOCMI)
III-2 Fugitive Emission Estimates for Three Model Plants III-4
III-3 Current SOCMI Uncontrolled Fugitive Emissions Estimate III-5
V-l Equipment Data for Three Model Plants V-2
V-2 Cost Factors Used in Computing Annual Costs for Equipment V-4
Modifications (Control Devices)
V-3 Cost Estimates for Installation of Control Devices in Model V-4
Plants
V-4 Basis for Determining Monitoring/Maintenance Manpower V-7
Requirements
V-5 Annual Monitoring Manpower Requirements for Individual- V-9
Component Survey for Three Model Plants
V-5 Maintenance Manpower vs Requirements for Initial Leak Repair V-10
for Three Model Plants
V-7 Factors Used in Computing Annual Costs for Monitoring Programs V-12
V-8 Control Cost Estimate for Leak Detection for Three Model Plants V-13
V-9 Annual Monitoring and Maintenance Costs for Leak Detection for V-13
Three Model Plants
v-io Control Cost Estimates for a Fixed-Point Monitoring System with V-13
an Individual-Component Survey
A-l Total Number of SOCMI Product Site Locations A-3
B-l Model Plants Characterized to Date B-3
-------
2-vii
FIGURES OF REPORT 2
Number
II-l Diagram of a Simple Packed Seal II-3
II-2 Diagram of a Basic Single Mechanical Seal II-3
II-3 Diagram of a Double Mechanical Seal II-3
II-4 Diagram of a Gate Valve II-5
II-5 Diagram of a Spring-Loaded Relief Valve II-7
II-6 Liquid-Film Compressor Shaft Seal II-8
IV-1 Diagram of a Rupture-Disk Installation Upstream of a Relief Valve IV-3
IV-2 Diagram of Simplified Closed-Vent System with Dual Flares IV-5
IV-3 Diagram of Two Closed-Loop Sampling Systems IV-7
B-l Plot of Total Number of Pumps per Plant vs Plant Capacity B-4
B-2 Plot of Total Number of Valves per Plant vs Plant Capacity B-5
-------
[-1
ABBREVIATIONS AND CONVERSION FACTORS
EPA policy is to express all measurements used in agency documents in metric
units. Listed below are the International System of Units (SI) abbreviations
and conversion factors for this report.
To Convert From
Pascal (Pa)
Joule (J)
Degree Celsius (°C)
Meter (m)
Cubic meter (m3)
Cubic meter (m3)
Cubic meter (m3)
Cubic meter/second
(m3/s)
Watt (W)
Meter (in)
Pascal (Pa)
Kilogram (kg)
Joule (J)
To
Atmosphere (760 mm Hg)
British thermal unit (Btu)
Degree Fahrenheit (°F)
Feet (ft)
Cubic feet (ft3)
Barrel (oil) (bbl)
Gallon (U.S. liquid) (gal)
Gallon (U.S. liquid)/min
(gpm)
Horsepower (electric) (hp)
Inch (in.)
Pound-force/inch2 (psi)
Pound-mass (Ib)
Watt-hour (Wh)
Multiply By
9.870 X 10"6
9.480 X 10~4
(°C X 9/5) + 32
3.28
3.531 X 101
6.290
2.643 X 102
1.585 X 104
1.340 X 10"3
3.937 X 101
1.450 X 10"4
2.205
2.778 X 10~4
Standard Conditions
68°F = 20°C
1 atmosphere = 101,325 Pascals
PREFIXES
Prefix
T
G
M
k
m
Symbol
tera
giga
mega
kilo
milli
micro
Multiplication
Factor
Example
10
10
10
10
10"
10"
12
9
6
3
3
6
1
1
1
I
1
1
Tg =
Gg =
Hg =
km =
mV =
pg =
1
1
1
1
1
1
X
A
X
X
X
X
10
10
10
10
10
10
12
9
G
r\
0
•3
~6
grams
grams
grams
meters
volt
grara
-------
II-l
II. CHARACTERIZATION AND DESCRIPTION
A. INTRODUCTION
The Synthetic Organic Chemicals Manufacturing Industry (SOCMI) is a segment of
the domestic chemical industry that produces 378 basic and intermediate organic
chemicals used to produce other intermediates and finished chemicals. Organic
chemicals not included in the SOCMI are refinery by-products, coal-tar products,
and other naturally derived organic chemicals or polymers.
For the purpose of this report fugitive emissions are defined as those volatile
organic compound (VOC) emissions that result from leaking plant equipment. The
potential sources of VOC emissions in the SOCMI that are considered in this
report include pump seals, in-line (process and control) valves, relief valves,
open-ended (sample, drain, vent) valves, compressor seals, flanges, and cooling
towers. VOC emissions that result from the transfer, storage, treatment, and
disposal of liquid, solid, and aqueous process wastes are defined as secondary
emissions and are covered in a separate report1.
The data base that was used for this report to characterize the industrial
types of equipment was compiled from 53 plant site visits2 and 2 EPA reports:
Equipment Component Analysis for Identification of Potential Fugitive Emission
Sources,3 prepared by Pullman Kellogg, and Data Package for Formaldehyde Plant
Fugitive Emissions Study,4 prepared by Walk, Haydel and Associates, Inc.
The data supplied from the site visits and the two EPA reports include a quanti-
fication of the number of equipment components associated with a specific chemica]
or co-products manufactured from a specific process. The various equipment
components, all of which are considered to be potential sources of fugitive
emissions, were grouped into five major equipment categories: pumps, valves,
compressor seals, flanges, and cooling towers. Pumps were further subdivided
as to the type of seal used, with four subcategories considered: those with
single mechanical seals, with double mechanical seals, with packed seals, or
with no seals (sealless). Valves were subcategorized by type as saftey/relief,
open-ended, and in-line, which were further subdivided by the kind of service
as vapor (gas) or liquid.
-------
II-2
B. SOURCES
There are many kinds of equipment in the SOCMI that can contribute VOC emissions
from leakages. The types of equipment assessed for fugitive emissions in this
report are as follows.
1. Pumps
Pumps are used extensively in the SOCMI for the movement of organic fluids.
The predominant type used is the centrifugal pump, although other types, such
as the positive-displacement pump, both reciprocating and rotary action, and
the specialized canned and diaphragm pumps, are used for some applications.
Except for such pumps as the canned and diaphragm types the pumps have a shaft
that requires a seal to isolate the pump interior fluid from the atmosphere.
The possibility of a leak through this seal makes it a potential source of fugi-
tive VOC emissions.
The general types of shaft seals in use are packed, single mechanical, and double
mechanical, which are listed in order of increased effectiveness in minimizing
leaks. The sealless pump is, of course, the most effective. Proper instal-
lation and maintenance are required for all seal types if they are to function
properly and retain their ability to seal.
a. Packed Seal Figure II-l is a diagram of a simple packed seal. Packed seals
can be used on both reciprocating and rotary action types of pumps. The seal
consists of a stuffing box in the pump casing filled with specialized packing
material that is compressed with a packing gland to form a seal around the shaft.
To prevent buildup of frictional heat, lubrication is required. A sufficient
amount of either the liquid being pumped or the liquid that is injected must be
allowed to flow between the packing and the shaft to provide the necessary lubri-
cation.
t>. Single Mechanical Seal Figure II-2 is a diagram of a basic single mechanical
seal. Mechanical seals can be used on rotary-type pumps only. The rotating-
seal-ring face and the stationary-element face are lapped to a very high degree
of flatness to ensure that they maintain contact throughout their mutual surface
area. As with packing, the faces must be lubricated; however, because of their
construction, much less lubrication is needed. There are many variations to
-------
II-3
FLUID
E.MO
AR.C.V
ig. II-l. Diagram of a Simple Packed Seal
Fig
UA.WJO
RIVJOl
PUUIO
E.IUO
Fig. II-2. Diagram of a Basic Single Mechanical Seal
FL.UIO
E.UO
Fig. I 1-3. Diagram
of a Double Mechanical Seal
-------
11-4
the basic design but all have the lapped seal face between a stationary element
and a rotating seal ring.
Double Mechanical Seal Figure II-3 is a diagram of a double mechanical seal
in which the two seals are installed in opposite configuration. A liquid, such
as water or seal oil, is circulated through the stuffing-box space between the
two seals. This arrangement is an improvement over a single seal since the
sealing fluid surrounds the double seal and provides lubrication to both sets
of seal faces. It also allows a low differential pressure to be set across the
inner seal face.
Another double mechanical seal arrangement is one in which the two seals are
installed in the same, or tandem, configuration. The inner seal functions identi-
cally to a conventional single inside seal. The stuffing box space between the
two seals is flooded with a liquid from a closed reservoir. If the inner seal
fails, it will be sensed by a pressure rise at the reservoir; also, the outer
seal will take over as a backup seal.
Valves
One of the most common pieces of equipment in organic chemical plants is the
valve. The types of valves commonly used are control, globe, gate, plug, ball,
relief, and check valves. All except the relief valve and check valve are
activated by a valve stem, whose motion may be rotational or linear or both,
depending on the specific design. This stem requires a seal to isolate the
valve interior fluid from the atmosphere. The possibility of a leak through
this seal makes it a potential source of fugitive VOC emissions.
The most common type of valve stem seal in use is the packed seal. It consists
of a stuffing box in the valve housing filled with specialized packing material
that is compressed with a packing gland to form a seal around the stem.
Figure II-45 is a diagram of one type of valve with a valve stem.
Pressure-relieving devices are required by engineering codes for applications
in which the pressure on a vessel or a system may exceed the maximum allowed.
Spring-loaded safety/relief valves are typically used for this service in the
-------
II-5
PACK-IM& G,I_AKJD
PACKlNJGi
VALVE:
Fig. II-4. Diagram of a Gate Valve (from Ref. 4)
-------
11-6
SOCMI. Figure II-5 is a diagram of a relief valve. The seal is a disk on a
seat held in place by a spring during normal system operation. The possibility
of a leak through this seal makes it a potential source of fugitive VOC emissions.
There are two potential causes for leaks: simmering, a condition due to the
system pressure being close to the valve set pressure, and improper reseating
following a relieving operation.
Check valves are used to prevent fluid backflow in a system. They do not need
a seal and therefore are not a source of leakage.
Some valves are installed in a system so that they function with the downstream
side open to the atmosphere. Examples are sample valves, drain valves, and
vent valves. The possibility of a leak through the seat of these valves makes
them a potential source of fugitive VOC emissions.
3. Flanges
Flanges are bolted gasket-sealed junctions used wherever pipe or equipment com-
ponents such as vessels, pumps, valves, and heat exchangers may require isolation
or removal. The possibility of a leak through the gasket seal makes them a
potential source of fugitive VOC emissions.
Two primary causes of leakage are seal deformation due to thermal stress on the
adjoining piping or equipment and repeated opening without replacement of the
gasket.
4. Compressors
SOCMI compressors, like pumps, can be both centrifugal and positive-displacement
types. Compressors have a shaft that requires a seal to isolate the compressor
interior gas from the atmosphere. The possibility of a leak through this seal
makes it a potential source of fugitive VOC emissions. In addition to having
seal types like those for pumps, centrifugal compressors can be equipped with a
liquid-film seal as shown in Fig. II-6.6 The seal is a film of oil that flows
between the rotating shaft and the stationary gland. The oil that leaves the
compressor from the system side is under the system internal gas pressure and
is contaminated with the gas. When this contaminated oil is returned to the
oil reservoir, process gas can be released and emitted to the atmosphere.
-------
II-7
SEAT
DISK
NJOZ.Z.L.E
SIDE.
Fig. II-5. Diagram of a Spring-Loaded Relief
Valve
-------
II-8
OIL. INJ FP.OM
GAS
SHAFT 5l_£.E:ve
IKJTERklAU
COM TAMI KJATE.D
OIL_ OUT
TO R.E.SERVOIR
OIL. OUT
Fig. II-6. Liquid-Film Compressor Shaft Seal (from Ref. 5)
-------
II-9
5- Cooling Towers
Cooling towers cool the recirculating water that is used to remove heat from
such process equipment as reactors, condensers, and heat exchangers. If a leak
in the process equipment occurs and if the equipment is operating at a pressure
higher than that of the recirculating water, process material can get into the
water stream. This material can be released to the atmosphere at the cooling
tower, making it a potential source of fugitive VOC emissions. A potential
source of fugitive emissions can also occur if VOC-contaminated process water
is used as the cooling-water source.
6. Other Sources
It is recognized that other kinds of equipment in the SOCHI can contribute VOC
emissions from leakages that are not covered in this report; two of them are
threaded connections and agitator seals. The use of threaded connections in
VOC service is believed to be small, accounting for an estimated 2 to 3% of all
piping-related connections.3 Agitator seals could not be assessed for their
fugitive emission potential because of insufficient data-base information.
c• METHODOLOGY
!• Data Base
The data base was compiled in the following two steps.
a- Step 1 In the first step all the equipment component data supplied by the
reference sources for the valve and pump subcategories were averaged to yield a
typical characterization for the industry. This approach was considered to be
reasonable since the data represented both large- and small-capacity processes
using both continuous- and batch-production operations. Results of the initial
characterization of valves and pumps from the data base are given in Tables II-l
and II-2. This information is considered to be representative of the entire
industry.
From Table II-l it can be seen that safety and relief valves constitute 3.4% of
the total valve population. An estimated 83% of the total safety-relief valves
is in vapor service and 17% is in liquid service. Open-ended valves, which
include sample, vent, and drain valves, represent 27.6% of the total valve count
-------
11-10
a
Table II-l. SOCMI Valve Characterization
Typo of
Valve
Safety-relief
Open-ended (sample
Percent of
Total Valves
3.4
27.6
Percent in
Liquid Service
17.0
91.0
Percent in
Vapor (Gas) Service
83.0
9.0
vent, drain)
In-line (process, 69.0 65.0 35.0
control)b
All valves 100.0 71.0 29.0
includes only valves in VOC service.
Check valves are excluded.
Table TI-2. SOCMI Pump Seal Characterization3
Pump Seals Percent in Use
Mechanical
Single
Double
Packed
None (sealless)
Total
alncludes only pump seals in VOC service.
-------
11-11
and are used primarily in liquid service (91%), with a small application (9%)
in vapor service. In-line valves, including process and control valves, make
up the bulk of the total valve population (69%) in the industry, with 65% of in
liquid service and 35% in vapor service. In general, valves in the industry are
used primarily in liquid service (71%), the remaining 29% being used in vapor
service.
As is shown in Table II-2, the most common type of pump seal in use in the indus-
try is the mechanical seal. An estimated 88.6% of all pump seals used to handle
VOC are mechanical; 71.9% of them are single mechanical seals and 16.7% are
double mechanical seals. Pumps with packed seals represent 10.6% of the total
pumps. Less than 1.0% of the total industry pumps is sealless.
A basic assumption used in estimating the total SOCMI equipment components is
that the total number of components associated with each chemical process is
dictated more by the number of plants manufacturing the product than by the
capacity of the production facility. A plant that is twice as large in capa-
city as another, producing the same chemical, does not necessarily have twice
as many pieces of equipment. In most instances the equipment is simply larger
in size. Plots that graphically illustrate the lack or correlation between the
number of equipment components and a plant's rated capacity are shown in Appendix B.
The plots were made for the total number of both pumps and valves versus the
plant's rated capacity for selected sites in this data base.
Analysis of the data collected indicated that the plants in SOCHI varied consider-
ably in size; i.e., the number of equipment components associated with one specific
process differed by orders of magnitude from that associated with another. A
second step in the data base compilation was therefore required to enable a
scaleup.
Step 2 Three model plants were selected to represent existing small, medium,
and large SOCMI plants. They are described in Table II-3. The most complete
information obtained from the site visits and the two EPA reports was the total
number of pumps at each location; therefore 15, 60, and 185 pumps were used to
quantify small, medium, and large model plants respectively. Once the total
number of pumps for each plant had been selected, the total number of valves
-------
11-12
Table II-3. Equipment Data for Three Sizes of Model Plants
Equipment Component
Pump seals
Light-liquid service
Single mechanical
Double mechanical
Sealless
Heavy-liquid service
Single mechanical
Packed
In-line valves
Vapor service
Light-liquid service
Heavy-liquid service
Safety-relief valves
Vapor service
Light-liquid service
Heavy- liquid service
Open-ended valves
Vapor service
Light-liquid service
Heavy- liquid service
Compressor seals
Sampling connections
Flanges
Cooling towers
Small
Model Plant (A)
5
3
0
5
2
90
84
84
11
1
1
9
47
48
1
26
600
1
Number of Components in
Medium
Model Plant (B)
19
10
1
24
6
365
335
335
42
4
4
37
189
189
2
104
2400
1
Large
Model Plant (Cj.
60
31
1
73
20
"1117
1037
1037
130
13
14
115
581
581
8
320
7400
1
52% of existing plant units are similar to model-plant A; 33% are similar to model-plant
B; 15% are similar to model-plant C.
Includes equipment components in VOC service only.
*t ' ' • '
'Sample, drain, purge valves.
Based on 25% of open-ended valves.
-------
11-13
for each plant was calculated by using a valve-to-pump ratio of 25:1. This
ratio is an average of the entire data base. A breakdown of the valves and
pumps into the various types was completed based on the characterizations given
in Tables II-l and II-2.
It was assumed that pump seals are split 50/50 between light- and heavy-liquid
service and that all packed pumps are used in heavy-liquid service, and all
pumps with double mechanical seals are used in light-liquid service.7 In-line
valves, safety-relief valves and open-ended valves were assumed to be split
50/50 between light- and heavy-liquid service.7 The heavy-liquid/light- liquid
split was necessary because equipment components in heavy-liquid service have a
lower leak frequency and smaller leak rates than equipment components in light-
liquid service. A review of the data indicated a breakpoint between naphtha
and kerosene. Heavy liquid, therefore, is defined as a fluid with vapor pressure
equal to or less than 0.3 kPa at 20°C.
The total number of compressor seals for each of the three model plants was
selected from the data base and is considered to be, in general, representative
of SOCMI.
The total number of sampling connections in the model plants was estimated by
assuming that 25% of the open-ended valves are in sampling service.6 This per-
centage is considered to be representative of the data collected.
Flange estimates were made by developing a flange-to-valve ratio from the two
EPA reports and the site visit information. A flange-to-valve ratio of 1.6:1,
which reflects the use of welded and screwed valves and the existence of a signi-
ficant number of open-ended valves, was considered to be representative of the
data base.
Analysis of the data base indicated the use of normally one cooling tower per
product site. Larger capacity operations usually had larger flow rates instead
of additional cooling towers.
-------
11-14
2. Industry Scaleup
A total of 1345 unique product units were identified with the production of the
378 organic chemicals that comprise SOCMI in 1978. A previous study9 lists
2000 production units for SOCMI but when adjustments are made for the fact that
some chemicals are co-produced in the same facility or are campaigned produced
in a common facility, the number of actual or unique units reduces to 1345.
An analysis of the data base, with the total number of pumps used as the basis
for selection, indicated that 52% of the SOCMI plant units would be classified
as small, 33% would be classified as medium, and 15% classified as large.
The total industry estimate of equipment components was obtained by first multi-
plying the total number of SOCMI units (1345) by the estimated percentage of
small, medium, and large plants. With this approach 700 plant units would be
considered small, 445 plant units would be considered medium, and 200 plant
units would be considered large. Then the average number of each of the specific
equipment components for the three model plants was multiplied by the corresponding
number of estimated units. The totals for each type of model plant were summed
to yield a total estimate for SOCMI.
This extrapolation is considered to be valid because the data base consists of
62 plant units representing approximately 5% of the total SOCMI plant units and
incorporates both large and small capacities and both batch- and continuous-
production operations. The current total SOCMI equipment component estmate
resulting from the scaleup is shown in Table II-4.
-------
11-15
Table II-4. Current SOCMI Equipment Component Estimate
£
Equipment Component
Pumps
Light- liquid service
Single mechanical seals
Double mechanical seals
Sealless
Heavy- liquid service
Single mechanical seals
Packed seals
In-line valves
Vapor service
Light- liquid service
Heavy- liquid service
Safety-relief valves
Vapor service
Light- liquid service
Heavy-liquid service
Open-ended valves
Vapor service
Light- liquid service
Heavy- liquid service
Compressor seals
Sampling connections
Flanges
Cooling towers
Agitators
Small
Plants
3,500
2,100
0
3,500
1,400
63,000
58,800
58,800
7,700
700
700
6,300
32,900
33,600
700
18,200
420,000
700
d
Number of
Medium
Plants
8,460
4,450
445
10,680
2,670
162,430
149,080
149,080
18,690
1,780
1,780
16,470
84,100
84,100
890
46,280
1,068,000
445
d
Components in
Large
Plants
12,000
6,200
200
14,600
4,000 '
223,400
207,400
207,400
26,000
2,600
2,800
23,000
116,200
116,200
1,600
64,000
1,480,000
200
, d
Total
SOCMI
23,960
12,750
645
28,780
8,070
448,830
415 , 280
415,280
52,390
5,080
5,280
45,770
233,200
233,900
3,190
128,480
2,968,000
1,345
d
Includes equipment components only in VOC service.
Sample, drain, purge valves.
CBased on 25% of open-ended valves.
NO data available.
-------
11-16
D. REFERENCES*
1. J. Cudahy and R. Standifer, IT Enviroscience, Inc., Secondary Emissions (June 1980)
(EPA/ESED Report, Research Triangle Park, NC).
2. Site visits conducted and used to quantify various equipment components for the
Fugitive Emissions report; see Appendix D.
3. Pullmann Kellogg, Houston, TX, Equipment Component Analysis for Identification
of Potential Fugitive Emissipn Sources (data on file at EPA, ESED, Research
Triangle Park, NC) (June 1978).
4. Walk, Haydel and Associates, Inc., New Orleans, LA, Data Package for Formaldehyde
Plant Fugitive Emissions Study (data on file at EPA, ESED, Research Triangle
Park, NC) (June 27, 1978).
J>. Emissions from Leaking Valves, Flanges, Pump and Compressor Seals, and Other
Equipment at Oil Refineries, LE-78-001, State of California Air Resources Board
(April 1978).
6. R. F. Boland et al., Monsanto Research Corp., Screening Study for Miscellaneous
Sources of Hydrocarbon Emissions in Petroleum Refineries, EPA-450/3-76-041
(December 1976).
7. Personal communication October 19, 1979, between K. C. Hustvedt, EPA, ESED,
Research Triangle Park, NC, and D. Erikson, IT Enviroscience, Inc., Knoxville, TN
8. R. G. Wetherold et al., Emission Factors and Frequency of Leak Occurrence for
Fittings in Refinery Process Units, p 22, EPA Report No. 600/2-79-044 (February
1979).
9. Organic Chemical Producers' Data Base (OCPDB) Program, USEPA Cincinnati, OH,
Updated November 28, 1979.
^Usually, when a reference is located at the end of a paragraph, it refers to
the entire paragraph. If another reference relates to certain portions of
that paragraph, that reference number is indicated on the material involved.
When the reference appears on a heading, it refers to all the text covered by
that heading.
-------
III-l
III. EMISSIONS
A. INTRODUCTION
Data on the measurement of fugitive VOC emissions from SOCMI (synthetic organic
chemical manufacturing industry) sources are limited. However, recent testing
efforts have generated a considerable amount of information from petroleum
refinery operations.1 SOCMI fugitive VOC emissions are assumed to be very
similar to petroleum refinery VOC emissions for all types of assessed sources.
Data from petroleum refineries indicate that emission rates of sources decrease
as the vapor pressure (volatility) of the process fluid decreases. Three
classes of volatility have been established based on the petroleum refinery
data: vapor service, light-liquid service, and heavy-liquid service. As noted
in Sect. II, the split between light and heavy liquids for refinery data is
between the streams of naphtha and kerosene. Table III-l1 gives the uncontrolled
fugitive VOC emission factors that resulted from testing conducted at petroleum
refineries and that have been applied to SOCMI equipment components to generate
VOC fugitive emission estimates.1
All pumps in light-liquid service, except for sealless pumps, have an uncon-
trolled emission factor of 0.12 kg/hr. It was assumed that pumps with double
mechanical seals leak VOC through the inner seal at the same rate as single
mechanical and packed seals and that the VOC is emitted from the seal-oil
degassing vent. All pumps in heavy-liquid service, except for sealless pumps,
also have the same uncontrolled emission factor of 0.02 kg/hr based on the same
assumptions stated above.
Uncontrolled emission factors for all the remaining sources were obtained
directly from the refinery study.1'2
Cooling towers were not assessed for emissions because data from petroleum
refineries indicate that cooling towers are a very small source of VOC fugitive
emissions. Differences in SOCMI operating procedures, such as recirculation of
process water, might result in cooling-tower VOC fugitive emissions but no data
are available to verify this.
-------
III-2
Table III-l. Uncontrolled Fugitive Emission Factors for the SOCMI
Fugitive-Emission Source
Uncontrolled Emission
Factor (kg/hr)a
Pumps
Light liquids
With packed seals
With single mechanical seals
With double mechanical seals
With no seals
Heavy liquids
With packed seals
With single mechanical seals
With double mechanical seals
With no seals
Valves (in-line)
Gas
Light liquid
Heavy liquid
Safety-relief valves
Gas
Light liquid
Heavy liquid
Open-ended valves
Gas
Light liquid
Heavy liquid
Flanges
Sampling connections
Compressors
Cooling towers
Agitators
0.12
0.12
0.12b
0.0
0.020
0.020
0.020b
0.0
0.021
0.010
0.0003
0.16
0.006
0.009
0.025
0.014
0.003
0.0003
0.015
0.44
c
c
These uncontrolled emission levels are based upon the refinery data presented
. in ref 1.
Assumes that the inner seal leaks at the same rate as single seal and that the
VOC is emitted from the seal-oil degassing vent.
C
No data available.
-------
III-3
Agitators also were not assessed for VOC fugitive emissions because the number
of agitator seals in SOCMI in not known. Furthermore the emission rate from
SOCMI agitator seals has not yet been measured. Since there are no data on
similar sources in other industries, no estimates of the emission rate were
available. Because of these uncertainties, cooling towers and agitator seals
are not included in the model plant or industry emission estimates.
B. FUGITIVE-EMISSIONS ESTIMATES
SOCMI VOC fugitive emissions are presented in two ways: uncontrolled emissions
have been calculated for each of the three model plants described in Table II-3,
and uncontrolled emissions have been calculated for the entire industry based
on the industry scaleup and resulting equipment component estimate given in
Table II-4.
Uncontrolled VOC fugitive emission estimates for the small, medium, and large
model plants described in Sect. II are given in Table III-2. Emission estimates
were calculated by multiplying the equipment components in Table II-3 by the
emission factors in Table III-l.
Uncontrolled VOC fugitive emissions that were estimated for SOCMI are shown in
Table III-3. The total emissions are currently estimated to be 321,000 Mg
annually, which accounts for approximately 40% of the total VOC emissions from
SOCMI in 1980.
The largest source of VOC fugitive emissions are valves in vapor service that
contributed a total of 166,000 Mg of VOC emissions, or 52% of the total estimated
VOC fugitive emissions resulting from SOCMI.
-------
III-4
Table III-2. Fugitive Emission Estimates for Three Model Plants
Equipment Component
Pumps
Light- liquid service
Single mechanical seals
Double mechanical seals
Sealless
Heavy- liquid service
Single mechanical seals
Packed seals
In-line valves
Vapor service
Light-liquid service
Heavy-liquid service
Safety- re lief valves
Vapor service
Light-liquid service
Heavy-liquid service
Open-ended valves
Vapor service
Light-liquid service
Heavy- liquid service
Compressor seals
Sampling connections
Flanges
Total
Uncontrolled VOC Emissions
Small .Medium
Model Plant Model Plant
5,256
3,154
0
876
350
16,556
7,358
221
15,418
53
79
1,971
5,764
1,261
3,854
3,416
1,577
67,164
19,973
10,512
0
4,205
1,051
67,145
29,346
880
58,867
210
315
8,103
23,179
4,967
7,709
13,666
6,307
256,435
(kg/yr) in
Large
Model Plant
63,072
32,587
0
12,790
3,504
205,483
90,841
2,725
182,208
683
1,104
25,185
71,254
15,269
30 , 835
42,048
19,447
799,035
Equipment in VOC service only.
-------
III-5
Table III-3. Current SOCMI Uncontrolled Fugitive-Emissions Estimate
Uncontrolled Emissions
Fugitive-Emission Source (Mg/yr)
Pumps
Light-liquid service
Single mechanical seals 25,187
Double mechanical seals 13/403
Sealless 0
Heavy-liquid service
Single mechanical seals 5,042
Packed seals 1,414
In-line valves
Vapor service 82,567
Light-liquid service 36,379
Heavy-liquid service 1,091
Safety-relief valves
Vapor service 73,430
Light-liquid service 267
Heavy-liquid service 416
Open-ended valves
Vapor service 10,024
Light-liquid service 28,600
Heavy-liquid service 6,147
Compressor seals 12,296
Sampling connections 16,882
Flanges 7,800
Cooling towers c
Agitators c
Total 320,945
Includes sources in VOC service only.
Based on emission factors listed in Table XXI-1.
CNot estimated; no data available.
-------
III-6
C. REFERENCES
1. R. G. Wetherold e_t al., Emission Factors and Frequency of Leak Occurrence
for Fittings in Refinery Process Units, p 22, EPA Report No. 600/2-79-044
(February 1979).
2. The Assessment of Environmental Emissions from Oil Refining, Appendix B, Vol 3,
EPA Report No. 600/2-80-075C (April 1980).
*Usually, when a reference is located at the end of a paragraph, it refers to
the entire paragraph. If another reference relates to certain portions of
that paragraph, that reference number is indicated on the material involved.
When the reference appears on a heading, it refers to all the text covered by
that heading.
-------
IV-1
IV. CONTROL TECHNOLOGY
This section describes the existing technology for the control of fugitive emis-
sions in the synthetic organic chemicals manufacturing industry (SOCMI). The
control of equipment leaks (fugitive emissions) can be accomplished not only by
the installation of control devices but also by the use of monitoring (leak
detection) and maintenance procedures.
A. CONTROL DEVICES
In some cases the equipment can be modified to limit VOC emissions from SOCMI
plants. The modifications included in this report that are considered to be
control devices consist of double mechanical seals on pumps, rupture disks and
a closed-vent system for safety/relief valves, blinds and a closed-loop sampling
system for open-ended valves, and a closed-vent system for compressor liquid-film
seals.
1 - Double Mechanical Seals
Double mechanical seals can be superior to single mechanical seals and packed
seals in preventing leakage from rotary-type pumps and compressors. By design
double mechanical seals have a chamber between the two seal faces that is either
flushed with a sealing fluid that allows control of the conditions under which
the seal operates or that flooded with a fluid whose pressure can be monitored
for seal failure.
Any leaks through the inner seal may be dissolved or suspended in the sealing
fluid, and subsequent degassing of the sealing fluid may result in the emission
of VOC. Therefore sealing-fluid degassing vents must be controlled in order to
provide maximum control effectiveness of double mechanical seals. After extended
periods of use double mechanical seals may also develop leaks at the outer-seal
shaft junction.
Emissions of VOC from degassing vents can be controlled by a closed-vent system,
which consists of piping and, if'necessary, flow-inducing devices to transport
the degassing emissions 'to a control device such as a flare, a process heater,
or a vapor-recovery system; Control effectiveness as high as 99% for a double
mechanical seal and closed-vent system is possible depending on the effectiveness
-------
IV-2
of the auxiliary control device and on the frequency of mechanical seal failure.
Failure of both the inner and outer seals can result in relatively large VOC
emissions at the seal area of the pump. As noted, the pressure monitoring of
the sealing fluid may be used to detect failure of the seals. In addition, visual
observation of the seal area can be effective in detecting failure of the outer
seal.
Mechanical seals, single or double, are limited in their applications. They
can be used only on shafts with a rotary motion. Also, the maximum service
temperature is usually limited to less than 260°C.1 Despite their limitations
it is estimated that about 90% of the SOCMI pumps handling VOC are equipped
with mechanical seals, with 17% being double mechanical seals (Sect. II). There-
fore double mechanical seals could be used in most new-pump applications and
probably could be retrofitted to replace many of the present single mechanical
seals and some of the packed seals. When the VOC being handled by the pump has
a low vapor pressure, the emission potential is reduced, which could limit the
necessity for double mechanical seals in these applications.
2. Rupture Disks
A rupture disk can be installed upstream of a relief valve in order to prevent
fugitive emissions through the relief-valve seat.2 Figure IV-1 is a diagram of
a rupture-disk and relief-valve installation. Under normal conditions the rupture
disk seals the system tightly, but if its set pressure is exceeded, it will
break, at which time the relief valve will relieve the pressure. This proce-
dure may require the use of a larger size relief valve because of operating
codes. The disk/valve combination may also require appropriate piping changes
to prevent disk fragments from lodging in the valve and precluding it from being
reseated should the disk rupture. A block valve upstream of the rupture disk
is also required in order to permit in-service replacement of the disk after
rupture. If the disk were not replaced, the first overpressure would result in
the relief valve being the same as an uncontrolled relief valve, and it may
actually be worse since disk fragments may prevent proper reseating of the relief
valve. In some chemical plants installation of a block valve upstream of a
pressure-relief device may be common practice. In others it may be forbidden
by operating or safety procedures. Tandem pressure-relief devices with a three-
way valve can be used to prevent operation without relief protection.
-------
IV-3
BUJOO FLAWGjE.
REUE.F VAUVE. ATTACHE** WERE.
COMUECTIOKJ FOR C;
PR-E-SSURE. GAUGiE.
VAL.V£
I
RUPTURE. D\5V<.
FROM SYSTE.M
Fig. IV-1. Diagram of a Rxipture Disk Installation Upstream of a Relief Valve
-------
IV-4
In rupture-disk/relief-valve combinations there must be some provision for testing
the integrity of the disk. It is important that no pressure be allowed to build
up in the pocket between the disk and the relief valve; otherwise the disk will
not function properly. The pocket must be connected to a pressure indicator,
recorder, or alarm. If the process fluid is not hazardous or toxic, a simple
bubbler apparatus could be used to test the integrity of the disk by connection
of the bubbler to the pocket. The control efficiency of the disk/valve combina-
tion is assumed to be 100% for fugitive emissions resulting from improper
seating and from simmering of the relief valve. If the disk integrity is not
maintained or if the disk is not replaced after rupture, the control efficiency
would be lowered. The disk/valve combination has no effect on emissions result-
ing from over-pressure relieving.
Manufacturers of relief valves state that resilient-seat or 0-ring relief valves
provide better reseating qualities compared to standard relief valves. No test
data are available to verify these statements. These improvements would have
no effect on rupture emissions or fugitive emissions due to seal failure or
simmering.
3. Closed-Vent System
A closed-vent system is used to collect and dispose of gaseous emissions in
many industrial operations, particularly those processes in which large volumes
of combustible volatile materials are handled. These gases can result from
such conditions as plant upsets, which require that the material be vented to
prevent overpressuring from plant startups and shutdowns, from disposal of waste-
gas streams, and from equipment leaks. Such emissions are typically intermittent,
and their flow rates during major upsets can be large. The usual method of
disposing of them is by flaring. Figure IV-2 is a diagram of a dual-flare system.
The small flare operates more efficiently with routine smaller exhausts. The
large flare is normally on standby to handle large emergency exhausts.
By connecting relief-valve discharges and compressor seal-oil degassing vents
to a closed-vent system, their emissions can be effectively controlled. The
effectiveness of VOC destruction will depend on the flare design and turn-down
capability. A dual-flare system should be more effective than a single-flare
system for the relatively low flows from relief-valve leaks and compressor seal-
oil degassing.
-------
IV-5
PILOT GAS
J FLARE
HE.ADE.RS->
n
RELIEF VALVE.
OIL_
EL.E.VATE.D FLARE.
FL.ARE.
/ \
0
Fig. IV-2. Diagram of Simplified Closed Vent System With Dual Flares
-------
IV-6
4. Blinds, Caps, and Plugs
Blinds, caps, plugs, or a second valve are devices for closing off the ends of
valves and pipes. When installed downstream of an open-ended valve, they are
effective in preventing leaks through the seat of the valve from reaching the
atmosphere. Open-ended valves in SOCHI, about 28% of the total valves handling
VOC, are used mostly in intermittent service for sampling, venting, or draining.
If a blind, cap, plug, or second valve is used downstream of a valve when it is
not in use, VOC emissions can be reduced. If a second valve is used in conjunction
with the first valve, the upstream valve should always be closed before the
second valve is closed. A further modification to minimize VOC emissions would
be to install a bleed valve between the two valves (double block and bleed) to
remove any VOC that might have leaked through the first valve and then accumulated
in the area between the two valves. Each time that the blind, cap, plug, or
second valve is opened, any VOC that leaked through the first valve will be
released. These emissions have not been quantified. The control efficiency of
these devices is assumed to be 100%. The actual control efficiency will depend
on how frequently the cap or plug is removed. The installation of a blind,
cap, or plug does not prevent the leakage that may occur through the valve-stem
seal.
5. Closed-Loop Sampling
A frequent operation in many SOCMI plants is to withdraw a sample of material
from the process for analysis. To ensure that the sample is representative,
purging of the sample lines and/or sample container is often required. If this
purging is done to the atmosphere or to open drains and if there are incidental
handling losses, VOC emissions can result. A closed-loop sampling system is
designed so that the purged VOC is returned to the system or sent to a closed
disposal system and so that the handling losses are minimized. Figure IV-33
shows two examples of closed-loop sampling systems, in which the purged VOC is
flushed from a point of higher pressure to one of lower pressure in the system
and where sample-line dead space is minimized. Reduction of emissions from the
use of closed-loop sampling is dependent on the efficiency of the control device
to which the purge is directed, i.e., 99% for a flare or 100% if the purge is
returned to the process. Flare efficiencies have not veen satisfactorily
documented except for specific designs and operating conditions using specific
fuels. An efficiency of 99% was used for tentative comparison purposes.
-------
IV-7
PROCESS LIME
LJ
i
SAMPLE.
COMTAIME.R
PROCESS LIME
COMTXIMER
Fig. IV-3. Diagram of Two Closed-Loop Sampling Systems (from Ref. 5)
-------
IV-8
B. LEAK DETECTION METHODS (MONITORING)
The EPA has defined a leak as one having a VOC concentration of over 10,000 ppm
at the potential leak source.4 The emission rate predicted by linear regression
analysis for 10,000 ppm at 0 cm is 0.42 kg/day for pump seals and 4.5 kg/day
for gas valves.5
Leak detection can be accomplished by three types of monitoring: individual-
component survey, unit-area survey, and multiple fixed-point monitoring systems.
Monitoring methods, their advantages and disadvantages, and their leak detection
efficiencies are discussed. The leak detection efficiency is defined as the
number of leaks detected by the monitoring method on a given date divided by
the total number of leaks known to exist on that same date.
1. Individual-Component Survey6
Each type of equipment listed in Sect. II can be monitored for leaks by use of
a portable VOC detection instrument to sample the ambient air in proximity to
the potential leak point.
In a complete individual-component survey each leak source is screened by a
portable VOC detection device that measures the concentration at the surfaces
where leakage could occur. The instrument probe should traverse the length of
the leak interface at the leak surface. Some potential sources, such as process
drains, cooling towers, pressure-relief devices, and open-ended valves or pipes,
have an exhaust area open to the atmosphere rather than a seal interface. For
these sources the probe should be placed at the centroid of the open area, as
well as around the perimeter of the open area. The major advantage of the com-
plete individual-component survey is that all significant equipment leaks are
located. By checking each component individually at its surface there should
be no false indicating leaks that are emenating from another leak source.
2. Unit Area Survey6
A unit area survey entails measuring the ambient VOC concentration within a
given distance, for example, 1 m, of all equipment located on the ground and
other accessible levels within a processing area. These measurements are per-
formed with a portable VOC detection instrument utilizing a strip-chart recorder.
The instrument operator walks a predetermined path to assure total available
-------
IV-9
coverage of a unit on both the upwind and downwind sides of the equipment, noting
on the chart record the location in the unit where any elevated VOC concentrations
are detected. If an elevated VOC concentration is recorded, the components in
that area would be screened individually to locate the specific leaking equip-
ment. It is estimated that 50% of all leaks in the unit are detected by the
walk-through survey, provided that all major leaking equipment has been repaired,
which reduces the VOC background concentration sufficiently to allow for reliable
detection.7
The major advantages of the unit-area survey are that leaks from ground and
other accessible leak sources can be located quickly and that the manpower
requirements are much lower than those for the individual-component survey.
Some of the shortcomings of this method are that VOC emissions from adjacent
units can cause false leak indications; high or intermittent winds (local
meteorological conditions) can increase dispersion of VOC, causing leaks to go
undetected; and additional effort is necessary to locate the specific leaking
equipment. However, two or- more consecutive (back-to-back) surveys in a unit
would minimize these problems.
Fixed-Point Monitoring Systems6
The basic concept of the fixed-point monitoring system is that sampling-point
devices can be installed at specific sites within the process area to monitor
for leaks automatically. The ambient VOC concentration can be remotely and
centrally indicated to the operator, who can respond appropriately when elevated
levels are recorded. The monitoring sites would not include the entire geographic
area of the facility, but only those areas where equipment handling VOC is located.
The performance of individual equipment can also be monitored to detect equip-
ment failures that result in leaks.
The approaches to leak detection with fixed-point monitors differ in the number
and placement of"the sample points and in the manner in which the sample is
taken and analyzed. One approach is to establish the sample points near speci-
fic pieces of equipment, such as process pumps, compressors, and cooling towers.
Another approach is to establish the sample points in a grid pattern throughout
the process area. When an elevated concentration is noted, the operator makes
an individual-component survey on equipment in that area to locate the leaking
-------
IV-10
component. In addition to these variations in the location of the sampling
points, different types of systems can be used. For example, the sampling can
be done continuously and the analysis done on-site or the samples can be col-
lected at the site and then be analyzed at a central location (an automatic
sequential system).
One feature of the fixed-point monitor approach is that the location of the
monitor and the type of sampling and analysis can be tailored to meet the spe-
cific requirements of individual plant sites and VOC. Fixed-point monitors can
sample for specific compounds by flame ionization—gas chromatography (GC/FID)
•
or infrared (IR) analysis. Of the several leak detection methods, fixed-point
monitoring systems have the highest capital cost but the lowest manpower require-
ment for monitoring. However, this approach may still require the use of a
portable VOC detection device to locate the leak, particularly if process-area
monitoring is used. Leak detection efficiency for fixed-point monitoring systems
is estimated to be 33%6 for facilities with a small number of recurring leaks,
provided that major leaking equipment has been repaired, which reduces the VOC
background concentration sufficiently to allow for reliable detection.
4. Cooling-Tower Total Organic Carbon (TOC) Analyzer
If there is a leak into a cooling-water system, detection by a portable or fixed-
point monitor may not be possible at the cooling tower. The air in close proximity
to the tower is normally very turbulent as the result of wind drafts created by
operation of the tower. This can increase the dispersion of VOC and cause the
leak to go undetected. To overcome this problem, a TOC analyzer can be used to
monitor for total organic carbon (ppm) in the cooling-tower stream. At one
plant using such a system a difference in total carbon measurements of £5 ppm
between inlet and outlet cooling-water streams at the tower is considered to be
an indication of a leak.8 Samples are then taken from various individual items
of equipment that make up the cooling-water system until the leak is found.
The leak detection efficiency for TOC monitoring of cooling water with concen-
trations §5 ppm is expected to be high.
5. Visual Inspection
Visual inspections can be performed to detect evidence of liquid leakage from
plant equipment. Any liquid leak found on equipment in VOC service should be
-------
IV-11
repaired because it represents equipment not operating properly and it represents
a potential release of VOC to the atmosphere.
C. MAINTENANCE
When leaks are located by the monitoring methods described in this section, the
leaking component must be repaired or replaced. Many components can be serviced
on-line, which is generally regarded as routine maintenance to keep operating
equipment functioning properly. Equipment failure, as indicated by a leak not
eliminated by servicing, requires isolation of the faulty equipment for either
repair or replacement.
1- Pumps
Most critical-service process pumps in SOCMI are backed up with a spare in place
so that they can be isolated for repair. Of those pumps that are not backed up
with spares, most can be isolated without major process disruption. Packed-seal
leaks frequently can be corrected by just tightening the packing, whereas mechanical-
seal leaks and packed-seal leaks that need correction require that the pump be
removed from service for seal repair. When the seal leak is small, there can
be situations in which the temporary emissions resulting from removal of the
pump from service can be larger than the emissions that would occur if the pump
remained in service until shut down for other process reasons. The maintenance
lead time required to schedule repair of a leaking pump can be an important
factor in the VOC emission-reduction efficiency obtainable.
2. Valves
Most valve leaks can probably be corrected on-line by tightening the packing
gland for valves with packed seals or by lubrication for plug valves, for example.
Based on field observation in one refinery study9 it was assumed that 75% of
leaking valves could be repaired on-line. Leaking valves that need to be repacked
or to be removed for repair must be capable of being isolated or the unit must
be completely or partly shut down. Control valves, 6% of the total valves in
SOCMI, can usually be isolated.1 Block valves, which are used to isolate or
by-pass equipment, normally cannot be isolated. One refiner estimates that 10%
of the block valves can be isolated.1 Based on the assumption that there is a
random distribution of leaks versus valve type and usage, in about 3% of the
leaks the valve can be isolated and in 22% the unit will have to be shut down
for repair of the valve.
-------
IV-12
When the leaking valves can be corrected on-line, repair servicing is usually
done soon after the leak is detected. For leaks that can be corrected by isola-
tion of the valve the maintenance lead time required to schedule repair can be
an important factor in the VOC emission-reduction efficiency obtainable. When
the leaks can be corrected only by a total or partial unit shutdown, the temporary
emissions could very likely be larger than the continuous emissions that would
result from not shutting down the unit until it was time for a shutdown for
other reasons.
3. Flanges
In one refinery field study it was noted that flange leaks could be sealed effec-
tively on-line by simply tightening the flange bolts.9 For a flange leak that
requires off-line gasket seal replacement a total or partial shutdown of the
unit would probably be required because most flanges cannot be isolated. For
many of these cases there are temporary flange repair methods that can be used.
Unless a leak is major and cannot be temporarily corrected, the emissions result-
ing from the shutting down of the unit would probably be larger than the continuous
emissions that would result from not shutting down the' unit until time for a
shutdown for other reasons. Flange leak incidences are very low and most can
be corrected by on-line maintenance.
4. Compressors
Compressors usually are in critical service and most often a spare is not provided.
In most cases shutdown for repair of a leaking seal and the subsequent startup
will involve flaring the process stream until operations are stablized.2 This
can result in shutdown emissions being larger than the continuous emissions
that would occur until it was time for the unit to be shut down for other reasons.
-------
IV-13
D- REFERENCES*
1 J- Johnson, Exxon Co., letter dated July 28, 1977, to Robert T. Walsh. EPA.
2- D. S. Kayser, "Rupture Disc Selection," Chemical Engineering Progress 68(5),
61—66 (1972). —
3- Briggs, T. and V. P J. Patel, Evaluation of Emissions from Benzene-Related
Petroleum Processing Operations EPA Report No. 450/3-79-022 (October 1978).
4- EPA, Chemical and Petroleum Branch, OAQPS Guideline Series, Control of Volatile
Organic Compound Leaks from Petroleum Refinery Equipment. EPA-450/2-78-036,
OAQPS No. 1.2-111 (June 1978).
5- R. G. Wetherold et al.. Emission Factors and Frequency of Leak Occurrence
for Fittings in Refinery Process Pnits. p 22. EPA Report No. 600/2-79-044
(February 1979).
6- K. c. Hustvedt and R. C. Weber, Detection of Volatile Organic Compound Emissions
from Equipment Leaks, paper presented at 71st Annual Air Pollution Control
Association Meeting, Houston, TX, June 25—30, 1978.
7- R. C. Weber,EPA, personal communications with D. G. Erikson, IT Enviroscience, Inc.,
Oct. 26, 1978 (analysis of preliminary results of EPA test program).
8- J. F. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Union Carbide Corp..
South Charleston. WV. Dec. 7, 1977 (data on file at EPA, ESED, Research Triangle
Park, NC).
9' Emissions from Leaking Valves, Flanges, Pump and Compressor Seals, and Other
Equipment at Oil Refineries. LE-78-001. State of California Air Resource
Board (Apr. 24, 1978).
^Usually, when a reference is located at the end of a paragraph, it refers to
the entire paragraph. If another reference relates to certain portions of
that paragraph, that reference number is indicated on the material involved.
When the reference appears on a heading, it refers to all the text covered by
that heading.
-------
V-l
V. COST ANALYSIS
This section presents estimated costs for the control of VOC fugitive emissions
from equipment leaks in the synthetic organic chemicals industry (SOCMI).
A- INTRODUCTION
Estimates of capital and annual operating costs are given for controlling emissions
from equipment leaks at SOCMI plants. The major sources of VOC emissions that
are considered in this section include pump and compressor seals, in-line valves,
open-ended valves, safety-relief valves, flanges, and cooling towers. Control
costs are presented for the five equipment modifications discussed in Sect. IV.
They include the installation of double mechanical seals, installation of a
rupture disk before a safety-relief valve, capping open-ended valves, sampling
by the closed-loop method, and incorporating a compressor-degassing vent and
safety-relief vents into an existing closed-vent system. Control costs are
also developed for a one-time-only leak detection program for the three model
plants decribed in Sect. II.
Plants in the SOCMI vary considerably in the size, configuration, and age of
facilities, the products produced, and the degree of control. Because of the
variations among plants this cost analysis is based on the three model plants
described in Sect. II and on the equipment data given in Table V-l. As was
stated earlier, the parameters considered to be representative of existing small,
medium, and large SOCMI plants are from the data base compiled for this report.
It is estimated that 15% of the SOCMI plants would be considered to be large,
33% to be medium, and 52% to be small. Although model plant costs may differ.
sometimes appreciably, from the actual costs that may be incurred, they are the
most useful means from determining and comparing emission control costs.
B- CONTROL COSTS
Capital cost estimates represent the investment required to purchase and install
monitoring instruments for leak detection and the labor and material costs to
install each of the five equipment modifications.
-------
V-2
Table V-l. Equipment Data for Three Model Plants
Equipment Component
Pump seals
Light- liquid service
Single mechanical
Double mechanical
Sealless
Heavy- liquid service
Single mechanical
Packed
In-line valves
Vapor service
Light- liquid service
Heavy- liquid service
Safety- re lief valves
Vapor service
Light- liquid service
Heavy- liquid service
Open-ended valves
Vapor service
Light- liquid service
Heavy- liquid service
Compressor seals
Sampling connections
Flanges
Cooling towers
Small
Model Plant
5
3
0
5
2
90
84
84
11
1
1
9
47
48
1
26
600
1
Number of Components in
Medium
Model Plant
19
10
1
24
6
365
335
335
42
4
4
37
189
189
2
104
2400
1
Large
Model Plant
60
31
I
73
20
1117
1037
1037
130
13
14
115
581
581
8
320
7400
1
52% of exist ing. pi ant units are similar to Model Plant A. 33% of existing plant units
are similar to Model Plant B. 15% of existing plant units are similar to Model Plant C.
Q
Includes equipment components only in VOC service.
—«•
"Sample, drain, purge valves.
Based on 25% of open-ended valves.
-------
V-3
1 - Control Devices (Equipment Modifications)
Details of the capital cost estimates for the five equipment modifications are
given in Appendix C. The cost factors used in computing annual operating costs
for all five equipment modifications are shown in Table V-2. In computing the
depreciation and interest costs, a capital recovery factor of 0.163 was used
based on a 10-yr life at an interest rate of 10% per annum. Costs for property
taxes and insurance are computed at 4% of the capital costs. All operating
costs are for 1-year periods, beginning with the first quarter of 1979. Table V-3
shows the installed capital cost per unit for each of the five equipment modifica-
tions. Costs for loss of production during installation or startup, preparation
of the equipment for retrofitting, or other highly variable costs are not included
in the estimates.
a- Double Mechanical Seals The installed cost for retrofitting a single mechanical
seal or a packed seal is estimated to be $850.l—3 This cost consists of $560
for the seal and $290 labor for field installation. The installed cost for a
new double mechanical seal is estimated to be $575. A credit of $225l—3 for a
new single mechanical seal was subtracted from the double-seal cost of $560 to
yield a net double-seal cost of $335. Labor to install the new double mechanical
seal in a shop area was estimated at $240. Actual costs for double mechanical
seals will range from $400 to $2000, depending on the pump shaft diameter, the
material of construction of the seal, the pump operating pressure and temperature,
and the physical and chemical properties of the material to be pumped. Auxiliary
equipment, such as a cooler for control of the temperature of the liquid used
to flush the seal, may be required and can range in cost from $470 to $800;l
also, an individualized pressurized-pot system nay be required to maintain pressure
•
on the liquid used to flush each double seal and can cost from $400 to $700.1<3
b- Rupture Disks
The cost for a rupture disk located upstream of a relief valve is estimated to
be $1730 for a new installation and $3110 for a retrofitted installation. One
7.6-cra carbon-steel rupture-disk holder costs $324,4'5 One 7.6-cm stainless-
steel rupture disk costs $195.4'5 One 0.6-cm dial-face pressure gauge costs
$15. One 0.6-cm carbon-steel bleed valve gate costs $25. The installation
cost for installation of these items is $240. In order to allow in-service
disk replacement, a block valve must be installed upstream of the rupture disk
-------
V-4
Table v-2. Cost Factors Used in Computing Annual Costs for
Equipment Modifications (Control Devices)
Maintenance
Capital charges
Capital recovery
Miscellaneous (taxes, insurance, and
administration)
0.05 X capital cost
0.163 X capital cost
0.04 X capital cost
aThese cost factors are different than those that are used in
other IT Enviroscience reports but were used to be consistant
with EPA's proposed standard documents.
Table V-3. Cost Estimates for Installation of
Control Devices in Model Plants
Control Device
Unit Cost
Double mechanical seals
New
Retrofitted
Flush oil system for double mechanical seals,
per pump
Rupture disks for relief valves
New
Retrofitted
Caps for open-ended valve)
Closed-loop sampling connections
Closed vents for degassing reservoirs of
compressors and double-seal pumps
Compressor
Pumps
$ 575
$ 850
§1500
$1730
$3110
$ 45
$ 460
$6530
$3265
-------
V-5
at an installed cost of $760. In.order to prevent disk fragments from damaging
the relief valve, an off-set mounting, which consists of one 10.2-cm tee and
one 10.2-cm elbow at an installed cost of $140, is required. Cost for the retrofit
installation is based on the assumption that a new relief valve is required to
replace the existing de-rated relief valve and that-there is no credit for the
used (replaced) relief valve. Installed cost for a 7.6-cm pressure-relief valve
is estimated to be $1380.
Actual costs for just the rupture disks can range from $200 to $1200 depending
on the operating pressure required, the number of disks purchased, the material
of construction required, and the size of the disk itself. The rupture disk
can become a cost-saving installation if it allows the relief valve to be tested
in place instead of having to be removed to a shop area. This is normally done
once or twice annually.
c- Caps on Open-Ended'Valves The installed cost of a blind flange on an open-
ended valve is estimated to be $30. This cost includes the blind flange and
the bolt and gasket set and is representative of either a new or retrofitted
installation. Actual costs for blinding or plugging open-ended valves could
range from $20 to $200 depending on whether a plug, cap, or blind flange is
used, on the size of the valve, and on the material of construction required.
The installed cost for a 2.5-cm screwed valve is estimated to be $45. Purchase
cost is estimated to be $30 with $15 required for 1-hr installation.
d- Closed-Loop Sampling The cost of a closed-loop sampling system is estimated
to be $460 for either a new or a retrofitted installation. The costs are based
on a 6-m length of 2.5-cm-diam, schedule 40, carbon-steel pipe and three 2.5-cm-
diam carbon-steel ball valves. It is estimated from the data base that approxi-
mately 25% of the open-ended valves are used for sampling. Actual costs will
be highly variable, depending on pipe size and material of construction required.
For example, the use of type 316 stainless steel pipe would increase installation
costs for the above system to $828.
e. Closed-Vent System The installation of a vent line from the compressor oil-seal
reservoir and from the plant safety relief valves to an existing (retrofitted)
-------
V-6
controlled closed-vent system is estimated to be $6530.6 This cost is based on
the installation of a 122-m length of 5.1-cm-diam, schedule 40, carbon-steel
pipe at a cost of $5200 and on three 5.1-cm-diam cast-steel plug valves and one
5.1-cm-diam metal gauze flame arrester at a cost of $1330. It is estimated
that only one vent line with a header system is necessary for collecting the
vapors. A cost of $3265 for double mechanical seal pumps is based on the assump-
tion that two pumps (such as a pump and its spare) are connected to a single
degassing vent. Actual costs will depend on the line size required, the loca-
tions of relief valves and compressor reservoirs, and the materials of construc-
tion. If a controlled closed-vent system is not in place, the installation
costs will be significantly higher.
Leak Detection and Maintenance
The manpower required to monitor equipment leaks varies with the methods or
combination of methods used. The manpower required for monitoring has been
estimated for each of the three model plants discussed in Sect. II. These esti-
mates are based on an industry estimates7, and the monitoring guidelines discussed
in Sect. IV. Table V-4 shows the maximum number of leaks measuring >10,000 ppm
at the leak surface available for detection per year for each of the three model
plants.1'8
Table V-5 shows the annual manpower (man-hours) required at each model plant
for monitoring an individual-component survey one time with a portable organic
analyzer.6 Two analyzers are used for example cost, one for use and one as a
spare. Flanges are not monitored since the leak occurrence is considered to be
negligible. For the purposes of these estimates only, it is assumed that this
survey will be conducted by two people: one operating the portable VOC detection
instrument and the other recording the results.
It will be noted that no matter what leak detection method is used, associated
maintenance manpower will be required for repairing the leaking component.
These requirements (shown in Table V-6) are estimated from an EPA report6 and
from the assumptions noted in the table. The actual maintenance manpower required
is directly proportional to the percentage of leaks detected; i.e., if only
one-half the leaks are detected, then the maintenance man-hours will be 50% of
those shown in Table V-6.
-------
V-7
Table V-4. Basis for Determining Monitoring/Maintenance
Manpower Requirements
Percentage
Number of of Initial
_____ Sources3 Leaks
Pumps (light liquid)
Single mechanical
Double mechanical
In-line valves
Vapor
Light liquid
Safety-relief valves (vapor)
0Pen-ended valvesd
vapor
Light liquid
Compressor seals
Pumps
Single mechanical
Double mechanical
In-line valves
Vapor
Light liquid
Safety-relief valves (vapor)
Open-ended valvesd
Vapor
Light liquid
Compressor seals
Pumps
Single mechanical
Double mechanical
Jn-line valves
Vapor
Light liquid
Small Model Plant
5
3
90
84
11
9
47
I
Medium Model Plant
19
10
365
335
42
37
189
2
Large Model Plant
60
31
1117
1037
23
23
10
12
8
10
12
33
23
23
10
12
8
10
12
33
23
23
10
12
Number of
Leaks
Detected0
2
1
9
11
1
1
6
1
5
3
37
41
3
4
23
1
14
8
112
125
-------
V-8
Table V-4. (Continued)
Percentage
Number of of Initial
Sourcesa Leaks"
Safety-relief valve (vapor) 130 8
Open-ended valves
Vapor 115 10
Light liquid 581 12
Compressor seals 8 33
From data base.
b
See ref 9.
Number of
Leaks
Detected
0
12
70
3
All leaks (>10,000 ppm) assumed to be detected.
Refers to the valve and not the open end.
-------
Table V-5. Annual Monitoring Manpower Requirements for Individual-Component Survey for
Three Model Plants
No. of Components
per Model Planta
Components
Pump seals
Compressor seals
In-line valves (gas)
In-line valves (liquid)
Open-ended valves
Safety-relief valves
pipeline flanges
Total
Small
15
1
90
168
104
13
600
Medium
60
2
365
670
415
50
2400
Large
185
8
1117
2074
1277
157
7400
Estimated Time^
Type of for Monitoring
Monitoring (min)
Instrument
Instrument
Instrument
Instrument
Instrument
Instrument
None
5
10
1
1
1
8
Times
Monitored
(No./yr)^
1
1
1
1
1
1
Annual Manpower Required
(hr) per Model Plantc
I
> Small
2.5
0.3
3.0
5.6
3.5
3.5
18.4
Medium
10
0.7
12.2
22
14
13.3
72.2
Large
30.8
2.7
37.2
69.1
42.6
41.9
224.3
report data base .
See re f 6.
Except where otherwise noted, based on 2 persons required.
Refers to valve and not open end.
VO
-------
Table V-6. Maintenance Manpower vs Requirements for Initial Leak Repair for Three Model Plants
v Components Per Model Plant
Source Type
Pumps (light liquid)
Single mechanical
Double mechanical
Valves (in-line)
Vapor
Light liquid
Safety-relief valves
Vapor
•
Open-ended valves
Vapor
Light liquid
Compressor seals
Total
Small
Plant
5
3
90
84
11
9
47
1
Medium
Plant
19
10
365
335
42
37
189
2
Large
Plant
60
31
1117
1037
130
115
581
8
Number
Small
Plant
2
1
9
11
oa
1
6
1
of Initial Leaks
Medium
Plant
5
3
37
41
oa
4
23
1
Large
Plant
14
8
112
125
oa
12
70
3
Repair -
Time
(hr)
80
80
1.13
1.13
0
1.13
1.13
40
Labor-Hours Required
Small
Plant
160
80
10
12
0
1
7
40
310
Medium
Plant
400
240
42
46
0
5
26
40
799
Large
Plant
1120
640
127
141
0
14
79
120
2241
maintenance required because rupture disks are assumed in use.
-------
V-ll
The cost factors used to compute annual operating-cost for leak detection are
shown in Table V-7.6'9 These factors"are based on an EPA report and on engineer-
ing estimates. In confuting the depreciation and interest costs, a capital
recovery factor of 0.163 was used based on a 10-year life at an interest rate
of 10% per annum for the fixed-point monitoring system. A capital recovery ,
w'l .
factor of 0.23 was used, based on a 6-yr life at an interest rate of 10%-per .
annum, for the portable VOC analyzers. Costs for property taxes and insurance
are computed at 4% of the capital costs. All operating costs are for 1-year
periods, beginning with the first quarter of 1979. Tables V-8 and V-9 show the
installed capital cost, the annual instrument capital-related costs, the annual
instrument materials, the maintenance and calibration costs, and the total annual
operating costs for each of the three model plants.
The installed capita! cost for the instruments shown in Table V-8 is $8500,
based on the use of two portable organic analyzers (one used as a spare) at a
cost of $4250 each.6 The annual cost for instrument materials, maintenance,
and calibration is $2700,6 including replacement of one battery pack and two
filter packs per year. The annual monitoring and maintenance costs are shown
in Table V-9. The annual monitoring labor cost for"each model plant was obtained
by multiplying the total monitoring manpower requirements by $15/hr.6 The annual
maintenance labor cost for each model plant was obtained by multiplying the
total maintenance manpower requirements by $15/hr.6
The installed instrument capital cost shown in Table V-10 is $136,000, which
includes $8,500 for two portable VOC analyzers and $127,5Q09 for a 20-point
gas-chromatograph fixed-point monitor. The annual instrument capital-related
cost of $28,500 consists of $2,600 for the portable analyzers and $25,900 for
the fixed-point unit. Annual instrument materials, maintenance, and calibration
costs of $9100 reflect a cost of $2700 for the portable analyzer and $6400 for
the fixed-point monitor.
-------
V-12
Table V-7. Factors Used in Computing Annual Costs
for Monitoring Systems
Monitoring-instrument capital
Portable
Fixed point (installed)
Labor
Monitoring
Maintenance
Annual instrument capital-related cost
Capital recovery
Portable monitor
Fixed-point monitor
Miscellaneous (taxes and insurance)
Annual instrument material,
maintenance, and calibrations
Portable monitor
Fixed-point monitor
Annual administrative and support
$4250
$127,500*
$15.00/hrc
$15.00/hr£
0.23 X capital
0.168 X capital
0.04 X capital
$2700
0.05 X capital cost
0.4 X annual monitoring
and maintenance cost
From ref 6.
From ref 9.
-------
V-13
Table V-8. Control Cost Estimate for Leak Detection for
Three Model Plants
Cost for Small, Medium, and
Large Model Plant
Instrument installed capital
Annual instrument, capital-related
Annual instrument materials maintenance
and calibration
Total
$8,500
2,600
2,700
$5,300
Table V-9. Annual Monitoring and Maintenance Costs for
Leak Detection for Three Model Plants
Annual monitoring labor
Annual maintenance labor
Annual administrative and support
Total
Small
Model Plant
$ 280
4,650
1,970
$6,900
Cost for
Medium
Model Plant
$ 1,090
11,980
5,230
$18,310
Large
Model Plant
$ 3,370
33,620
14,800
$51,790
Table V-10. Control Cost Estimates for a Fixed-Point
Monitoring System with an individual-Component Survey
Cost for small, Medium, and
Large Model Plant
Instrument capital
Annual instrument, capital-related
Annual instrument material
maintenance and calibration
Total
$136,000
$28,500
9,100
$ 37,000
-------
V-14
C. REFERENCES*
1. E. MacRae, Durametallic Corp., Kalamazoo, MI, personal communcation Jan. 17,
1979, with D. G. Erikson, IT Enviroscience, Inc. (documented for files of D. G.
Erikson).
2. D. Way, Crane Corp., Morton Grove, IL, personal communication Jan. 16, 1979,
with D. G. Erikson, IT Enviroscience, Inc. (documented for files of D. G. Erikson)
3. M. Pepin, Johnson City, TN, representative of Chesterton Co., personal communica-
tion Jan. 23, 1979, with D. G. Erikson, IT Enviroscience, Inc. (documented for
files of D. G. Erikson).
4. J. Smith, Atlanta, GA, representative of Continental Disc Corp., personal communi-
cations Jan. 19, 1979, with D. G. Erikson, IT Enviroscience, Inc. (documented
for files of D. G. Erikson).
5. E. Wischhusen, Knoxville, TN, representative of BS&B Safety Systems, Inc., per-
sonal communication Jan. 22, 1979, with D. G. Erikson, IT Enviroscience, Inc.
(documented for files of D. G. Erikson).
6. EPA, Chemical and Petroleum Branch, OAQPS Guideline Series. Control of Volatile
Organic Compound Leaks from Petroleum Refinery Equipment, EPA-450/2-78-036,
OAQPS No. 1.2-111 (June 1978).
7. J. Johnson, Exxon Co., letter dated July 28, 1977, to Robert T. Walsh, EPA.
8. R. G. Wetherold et al., Emission Factors and Frequency of Leak Occurrence for
Fittings in Refinery Process Units, p 22, EPA Report No. 600/2-79-044 (February
1979).
9. K. C. Hustvedt and R. C. Weber, Detection of Volatile Organic Compound Emissions
from Equipment Leaks, paper presented at 71st Annual Air Pollution Control Associa-
tion Meeting, Houston, TX, June 25—30, 1978.
*Usually, when a reference is located at the end of a paragraph, it refers to
the entire paragraph. If another reference relates to certain portions of that
paragraph, that reference number is indicated on the material involved. When
the reference appears on a heading, it refers to all the text covered by that
heading.
-------
VI-1
VI. ASSESSMENT
A- SUMMARY
Fugitive emissions are VOC emissions that result from leaks from plant equipment.
Equipment characterized in this report includes pump seals, in-line valves,
open- ended valves, safety-relief valves, compressor seals, flanges, and cooling
towers. The total estimate of equipment components in VOC service in the SOCMI
includes 72,200 pumps, 1,855,000 valves, 2,968,000 flanges, 3,190 compressors,
and 1,345 cooling towers. It is estimated that 50% of the above equipment com-
ponents are used in light-liquid service and 50% are used in heavy-liquid service
The uncontrolled fugitive emission factors listed in Table III-l and used through-
out this report are based on operating conditions found to exist in the petroleum
industry in 1978. If it is assumed that the SOCMI operates with the same degree
of equipment maintenance and monitoring, the estimated fugitive emissions from
the total SOCMI may be as high as 320,945 Mg/yr. The actual degree of main-
tenance and monitoring control used in the SOCMI has not been established.
Control of fugitive VOC emissions is primarily achieved by use of leak detec-
tion and maintenance methods. They range from a complete testing of every
potential VOC leak source to the use of fixed-point monitors installed to detect
leaks in specific process unit areas. Leak detection efficiencies range from
100 to 33%. A significant leak is defined as one that is >10,000 ppm read out
on a portable VOC analyzer at the source of the leak (0-cm distance). In addi-
tion to leak monitoring and maintenance, five specific equipment modifications
can be used as a supplement to eliminate VOC emissions specific to those sources.
Installed capital costs range from $45 to $3,110 for each equipment modifica-
tion and from $8,500 to $136,000 for leak detection methods.
-------
APPENDIX A
TOTAL NUMBER OF SOCMI PRODUCT SITE LOCATIONS
-------
A-3
Table A-l. Total Number of SOCMI Product unit Locations3
Location
Texas
New Jersey
Louisiana
Illinois
Ohio
California
Pennsylvania
New York
West Virginia
Tennessee
Alabama
North Carolina
Kentucky
Indiana
Michigan
Puerto Rico
All others
Total
Number of Product Units b
220
140
95
90
77
75
71
64
37
32
32
32
30
28
25
23
271
1345
Organic Chemical Producers Data Base (OCPDB) Program,
USEPA Cincinnati, OH, Updated Nov. 28, 1979.
A product unit is defined as a geographic location
that utilizes specific pieces of equipment to produce
one or more specific synthetic organic chemicals.
-------
APPENDIX B
CHARACTERIZATION OF MODEL PLANTS
AND NUMBER OF PUMPS AND VALVES VERSUS
PLANT CAPACITY
-------
B-3
Table B-l. Model Plants Characterized to Date0
Total Number of Components in Plant
__ Plant Type Pumps
Acrylic acid and esters 36
Acrylonitrile 50
Adipic acid 56
Chlorobenzenes 102
Chloromethanes 30
Cyclohexane-benzene 15
Cyclohexane-petroleum 35
Cyclohexanol/cyclo- 75
hexanone via cyclohexane
Cyclohexanol/cyclo- 35
hexanone via phenol
Ethylbenzene/styrene 50
Ethylene 165
Ethylene dichloride 42
Ethylene oxide via air 10
Ethylene oxide via oxygen 10
Formaldehyde 13
Maleic anhydride 15
Nitrobenzene 42
a
From Emissions .Control Options
Industry Product Reports (on
Process
Valves
820
1200
349
800
1000
300
700
1975
875
1000
4150
1100
400
400
214
500
500
Relief
Valves
44
22
12
15
15
35
56
26
65
40
6
20
for the Synthetic Organic
file at EPA,
Capacity
Compressors (Gg/yr)
76.1
180
150
96
2 45 — 180
1 30—265
1 100
100
100
300
8 408.2 — 680.3
400
2 227
2 136.1
45
22.7
30—150
Chemicals Manufacturing
ESED, Research Triangle Park, NC) .
In voc service only.
-------
B-4
Log Y = .82 + .27 Log X (Slope)
r = 0.07 (Correlation Coefficient
1OOO
BOO
IOO
ME.OIUM MODEL-
SMALL. MODE-L
5O IOO
Plant Capacity M Ibs (X)
5OO IOOO
Fiq.B-1 . Total Number of Pumps for Each Plant in the Data Base Versus
the Plant's Rated Capacity (54 Points)
-------
B-5
.OOOO
Log Y = 2.13 + 0.31 Log X (Slope)
r2 = 0.10 (Correlation Coefficient)
iO
100
Plant Capacity M Ibs (X)
\ooo
50<
Fig. B-2.
c i-^u DI ant- in the Data Base Versus
Total Number of Valves for Each Plant in tae u
the Plant's Rated Capacity (42 Points)
-------
APPENDIX C
COST ESTIMATE DETAILS
-------
PRELIMINARY CAPITAL
G-3
-------
PRELIMINARY CAPITAL
04
PHASE
£ F NUMBER
St
BATE
JOB NUuSlB
CAiE NUMBER
SECTION NUMBER
SECTOR NUMBER & NAME
NAME OF FACILITY|'.DESCRIPTION| OUANf.|
M Auc.
A-1 0 «v
'.j
-------
PRELIMINARY CAPITAL
05
:i«v o cam-v. -'^ v\ c\ c.-i r
NAME OF FACILITY! DE SCR I
-------
PRELIMINARY CAPITAL
C-6
rssrtf
PHASE
E F NUMBER
JOB
CASE NUMAEM
SECTION NUMBER
SECTOR NUMBER I NAME
NAME OF FACILITYf_DESCRIPTION/OUANr.
"Pt Pirt
20'
-------
PRELIMINARY CAPITAL
C-7
NAME OF FACILITYJD ESCRIPTI ON/OUANI-.
i
"R P\ ^ c
2-"cf> v
2.
Z" Cb Q.O.
4
^00
430
-------
APPENDIX D
LIST OF EPA INFORMATION SOURCES
-------
D-3
LIST OF EPA INFORMATION SOURCES*
1- J. W. Blackburn, IT Enviroscience, Inc., Trip Reports for Visit to Allied Chemical
Co., Hopewell, VA, on the Adipic Acid Process, the Cyclohexanol/Cyclohexanone
Process, and the Caprolactum Process, Feb. 21, 1978.
2. D. H. Bolb, CY-RO Industries, letter dated May 4, 1978, to R. E. White, IT Enviro-
science, Inc., with information on American Cyanamid's Kenner, LA, methyl methacry-
late process.
3. J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Amoco Chemicals,
Joliet, IL, on the Maleic Anhydride Process, Jan. 24, 1978.
4- S. W. Dylewski, IT Enviroscience, Inc., Trip Report for Visit to Amoco, Decatur, AL.
on the Dimethyl Terephthalate Process, Oct. 31, 1977.
5. R. L. Standifer, IT Enviroscience, Inc., Trip Report for Visit to ARCO, Channelview.
TX, on the Butadiene Process and the Ethylene Process, Aug. 16, 1977.
&. J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to BASF, Geismar,
LA, on the Ethvlene Oxide Process, July 12, 1977.
?. J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Borden,
Fayettville, NC, on the Formaldehyde Process, Aug. 24, 1977.
8. S. W. Dylewski, IT Enviroscience, Inc., Trip Reports for Visit to Carolina
Eastman, Columbia, SC, on the Dimethyl Terephthalate Process and on the Tere-
phthalic Acid Process, Dec. 6, 1977.
9. J. C. Cudahy, IT Enviroscience, Inc., Trip Report for Visit to Celanese, Clear
Lake, TX, on the Acetaldehyde Process, Sept. 22. 1977.
10. j. A. Key, IT Enviroscience, Inc., Trip Report for Visit to Celanese, Clear
Lake, TX, on the Acetic Acid Process, Oct. 12, 1977.
11. R. W. Helsel, IT Enviroscience, Inc., Trip Reports for Visit to Celanese,
Pampa, TX, on the Acetic Anhydride Process and on the Ethyl Acetate Process.
Mar. 1, 1978.
12. J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Celanese.
Clear Lake, TX, on the Ethylene Oxide Process, June 21, 1977.
13. J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Celanese,
Bishop, TX, on the Formaldehyde Process, July 26, 1977.
14. J. A. Key, IT Enviroscience, Inc., Trip Report for Visit to Celanese, Bishop,
TX, on the Methanol Process, Oct.11, 1977.
15. s. W. Dylewski, IT Enviroscience, Inc., Trip Report for Visit to Celanese,
Bay City, TX, on the Vinyl Acetate Process, Sept. 28, 1977.
*A11 Sources on file at EPA, ESED, Research Triangle Park, NC.
-------
D-4
16. J. A. Key, IT Enviroscience, Inc., Trip Report for Visit to Cosmar, Carville, LA,
on the Ethylbenzene/Styrene Processes, June 16, 1977.
17. J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Denka (Petro-Tex),
Houston, TX, on the Maleic Anhydride Process, Nov. 11, 1977.
18. J. A. Key, IT Enviroscience, Inc., Trip Report for Visit to Dow Chemical, Freeport^
TX, on the Ethylbenzene/Styrene Process, July 28, 1977.
19. J. W. Blackburn, IT Enviroscience, Inc., Trip Report for Visit to Dupont, Memphis,
TN, on the Acetone Cyanohydrin Process and on the Methyl Methacrylate Process,
Jan. 10, 1978.
20. J. A. Key, IT Enviroscience, Inc., Trip Report for Visit to Du Pont, Beaumont, TX^
on the Acrylonitrile Process, Sept. 7, 1977.
21. C. W. Stuewe, IT Enviroscience, Inc., Trip Report for Visit to Pu Pont, Beaumont,
TX, on the Aniline/Nitrobenzene Process, Sept. 7, 1977.
22. J. W. Blackburn, IT Enviroscience, Inc., Trip Report for Visit to Exxon, Baytown, T
on the Cyclohexane Process, Sept. 15, 1977.
23. R. L. Standifer, IT Enviroscience, Inc., Trip Report for Visit to Gulf Oil Co.,
Cedar Bayou, TX, on the Ethylene Process, Sept. 13, 1977.
24. S. W. Dylewski, IT Enviroscience, Inc., Trip Report for Visit to Hereofina,
Wilmington, NC, on the Dimethyl Terephthalate Process, Nov. 17, 1977.
25. C. A. Peterson, IT Enviroscience, Inc., Trip Report for Visit to Monsanto. Sauget,
on the Chlorobenzene Process, Jan. 27, 1978.
26. J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Monsanto. St. Louis,
on the Maleic Anhydride Process, Oct. 20, 1977.
27. J. A. Key, IT Enviroscience, Inc., Trip Report for Visit to Monsanto, Texas City, T
on the Acetic Acid Process and on the Methanol Process, Dec. 13, 1977.
28. C. A. Peterson, IT Enviroscience, Inc., Trip Report for Visit to Monsanto, Alvin, T
on the Alkyl Benzene Process, Nov. 8, 1977.
29. C. W. Stuewe, IT Enviroscience, Inc., Trip Report for Visit to Monsanto, Alvin, TX/_
on the Phenol/Acetone Processes, July 28, 1977.
30. J. W. Blackburn, IT Enviroscience, Inc., Trip Report for Visit to Monsanto, Pensa-
cola, FL, on the Cyclohexanol/Cyclohexanone Process, Feb. 8, 1978.
31. W. D. Bruce, IT Enviroscience, Inc., Trip Report for Visit to Nipro, Augusta, GA,
on the Caprolactam Process, Apr. 18, 1978.
32. J. W. Blackburn, IT Enviroscience, Inc., Trip Report for Visit to Phillipsf Puerto
Rico, on the CycloheKane Process, Sept. 20, 1977.
33. S. W. Dylewski, IT Enviroscience, Inc., Trip Report for Visit to PPG, New Martins-
ville, WV, on the Chlorobenzene Process, Sept. 7, 1977.
-------
D-5
34. J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Reichhold, Morris,
IL, on the Maleic Anhydride Process, July 28, 1977.
35. j. w. Blackburn, IT Enviroscience, Inc., Trip Report for Visit to Rohm and Haas,
Deer Park. TX, on the Acrylic Acid Process and on the Methyl Methacrylate Process,
Nov. 1. 1977.
36. C. W. Stuewe, IT Enviroscience, Inc., Trip Report for Visit to Rubicon, Geismar, LA,
on the Aniline/Nitrobenzene Processes, July 19, 1977.
37- C. A. Peterson, IT Enviroscience, Inc., Trip Report for Visit to Shell, Norco, LA,
on the Acrolein/Glycerine Processes, Jan. 25, 1978.
38- R. E. Van Ingen, Shell Oil Co., Houston, TX, letter dated Dec. 6, 1974, to Don
Goodwin, EPA, ESED, with information on Shell's vinyl chloride monomer process.
39- J. W. Blackburn, IT Enviroscience, Trip Reports for Visit to Union Carbide, Taft,
LA, on the Acrylic Acid Process, the Ethyl Acrylate Process, and the Heavy Acrylic
Esters Process, Dec. 8, 1977.
40 • C. A. Peterson, IT Enviroscience, Inc., Trip Report for Visit to Union Carbide.
Institute, WV, on the Linear Alkylbenzene Process, Dec. 7, 1977.
41- J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Union Carbide,
South Charleston, WV, on the Ethylene Oxide, Glycol Processes, Dec. 7, 1977.
42 • J. L. Lawson, IT Enviroscience, Inc., Trip Report for Visit to Union Carbide^
Sea Drift. TX, on the Glycol Ethers Process, Feb. 14, 1978.
43. J. A. Key, IT Enviroscience, Inc., Trip Report for Visit to Vistron, Lima, OH.
on the Acrylonitrile Process, Oct. 4, 1977.
44• Pullman Kellogg, Houston, TX, Equipment Component Analysis for Identification
of Potential Fugitive Emission Sources (on file at EPA, ESED. Research Triangle
Park, NC)(June 1978).
45- Walk, Haydel and Associates, Inc., New Orleans, LA, Data Package for Formaldehyde
Plant Fugitive Emissions Study (on file at EPA, ESED, Research Triangle Park,
NC) (June 27, 1978).
-------
3-i
REPORT 3
SECONDARY EMISSIONS REPORT
J. J. Cudahy
R. L. Standifer
IT Enviroscience
9041 Executive Park Drive
Knoxville, Tennessee 37923
Prepared for
Emission Standards and Engineering Division
Office of Air Quality Planning and Standards
ENVIRONMENTAL PROTECTION AGENCY
Research Triangle Park, North Carolina
June 1980
D73A
-------
3-iii
CONTENTS OF REPORT 3
I- ABBREVIATIONS AND CONVERSION FACTORS
II- THE SOCMI WASTES
A. Introduction
B. Process Waste Sources
C. References
HI. SECONDARY EMISSION SOURCES
A. Introduction
B. Physical Separation Methods
C. Chemical Treatment
D. Thermal Destruction
E. Biological Treatment
F. Terminal Storage
G. Discharge to Natural Waters
H. References
IV. EMISSIONS
A. Introduction
B. Physical Separation Methods
C. Chemical Treatment
D. Thermal Destruction
E. Biological Treatment
F. Terminal Storage
G. Discharge to Natural Waters
H. References
V. APPLICABLE CONTROL METHODS
A. Introduction
B. Waste Source Control
C. Resource Recovery
D. Alternative Treatment Terminal
E. Add-On Controls
F. Potential Secondary VOC Emission Reduction
G. References
Page
1-1
II-l
II-l
II-l
11-21
III-l
III-l
III-l
III-6
III-8
III-ll
111-14
111-19
111-20
IV-1
IV-1
IV-10
IV-11
IV-12
IV-13
IV-19
IV-23
IV-24
V-l
V-l
V-l
V-7
V-9
V-14
V-21
V-22
-------
3-iv
CONTENTS (Continued)
VI. IMPACT ANALYSIS
A- Introduction
B. Control Cost Impact
C. Environmental and Energy Impacts
D. References
HI. ASSESSMENT
A. Summary
B. Data Assessment
C. Reference
Page
VI-1
VI-1
VI-6
VI-11
VI-13
VII-1
VII-1
VII-1
VII-3
APPENDICES OF REPORT 3
APPENDIX A.
APPENDIX B.
APPENDIX C.
APPENDIX D.
APPENDIX E.
LITERATURE REVIEW A-l
ESTIMATION PROCEDURES FOR SECONDARY VOC EMISSIONS FROM B-l
WASTEWATER
ESTIMATE OF UNCONTROLLED SECONDARY VOC EMISSIONS FROM C-l
WASTE WATERS FOR THE SYNTHETIC ORGANIC CHEMICALS MANUFAC-
TURING INDUSTRY (SOCMI)
COMPARISON OF SECONDARY VOC EMISSIONS FOR FIVE MODEL D-l
CHEMICALS
COST CALCULATIONS FOR VOC SECONDARY EMISSION CONTROL E-l
OPTIONS
-------
3-v
TABLES OF REPORT 3
Number Page
II-l Summary of Common Process Waste Sources and Corresponding II-2
Waste Characteristics
II-2 Waste Sources in Selected SOCMI Processes II-3
III-l Summary of Waste Characteristics and Corresponding Common III-2
Treatment Operations
IV-1 SOCMI Secondary Emission Sources and Estimated Emissions IV-2
IV-2 Parameters for Model-Plant Wastewater Treatment Systems IV-15
IV-3 Operational Parameters for Model Chemical Production Plant IV-17
IV-4 Estimated Order-of-Magnitude Uncontrolled-Secondary-Emission IV-18
Wastewater Factors for 30 Organic Chemical Products
IV-5 Parameters for Model-Plant Landfill IV-20
IV-6 Comparison of Estimated Secondary VOC Emissions from a IV-21
Wastewater Treatment Plant with Those from a Chemical
Landfill at a Model Chemical Production Plant
V-l Applicable Control Methods V-2
V-2 Secondary VOC Emission Reductions by Various Control V-3
Techniques or Methods
V-3 Current Applications of Incineration to the Disposal of Liquid and V-10
Solid Wastes
VI-i cost Estimates for Control of Secondary VOC Emissions VI-7
from Model Plant
VI-2 Environmental and Energy Impacts of Controlled Secondary VI-12
VOC Emissions from Model Plant
B-l Henry's Law Constants from Thibodeaux B-8
B-2 Desorption as a Function of Henry's-Law Constant in an B-8
Activated-Sludge Wastewater Treatment System at 25°C
C-l Wastewater Parameters C-5
C-2 Summary of VOC Desorption Calculations C-6
D-l Physical Properties of Model Chemicals D-3
D-2 Calculated Properties of Model Chemicals D-4
E-l Parameters for Model-Plant Surface Impoundment E-3
E-2 Cost Factors Used for Computing Annual Costs E-4
E-3 Cost Estimates for Control of Secondary VOC Emissions from E-5
Model Plant
-------
3-vii
FIGURES OF REPORT 3
Number
l Conventional Activated Sludge Treatment Plant 111-13
IV-1 Uncontrolled Secondary VOC Emissions from Model-Plant IV-14
Wastewater
VI-1 Control Option 1: Removal of Model-Plant Wastewater VOC VI-2
with Carbon Adsorption
VI-2 Control Option 2: Cover for Model-Plant Wastewater Clarifier VI-3
vI-3 Control Option 3: Covered Clarifier Plus Carbon Adsorption VI-4
of the Secondary VOC Emissions from a CAS System Treating the
Model-Plant Wastewater
Vi-4 Control Option 4: Covered Clarifier Plus Fume Incineration VI-5
of the Secondary VOC Emissions from a CAS System Treating the
Model-Plant Wastewater
-------
1-1
ABBREVIATIONS AND CONVERSION FACTORS
EPA policy is to express all measurements used in agency documents in metric
units. Listed below are the International System of Units (SI) abbreviations
and conversion factors for this report.
_ To Convert From
Pascal (Pa)
Joule (J)
Degree Celsius (°C)
Meter (m)
Cubic meter (m3)
Cubic meter (m3)
Cubic meter (m3)
Cubic meter/second
(m3/s)
Watt (w)
Meter (m)
Pascal (Pa)
Kilogram (kg)
Joule (J)
To
Atmosphere (760 mm Hg)
British thermal unit (Btu)
Degree Fahrenheit (°F)
Feet (ft)
Cubic feet (ft3)
Barrel (oil) (bbl)
Gallon (U.S. liquid) (gal)
Gallon (U.S. liquid)/min
(gpm)
Horsepower (electric) (hp)
Inch (in.)
Pound-force/inch2 (psi)
Pound-mass (Ib)
Watt-hour (Wh)
Multiply By
9.870 X 10"6
9.480 X 10"4
(°C X 9/5) + 32
3.28
3.531 X 101
6.290
2.643 X 102
1.585 X 104
1.340 X 10"3
3.937 X 101
1.450 X 10"4
2.205
2.778 X 10"4
Standard Conditions
68°F = 20°C
1 atmosphere = 101,325 Pascals
PREFIXES
Prefix
T
G
M
k
m
M
Symbol
tera
giga
mega
kilo
milli
micro
Multiplication
Factor
IO12
IO9
IO6
IO3
IO"3
io"6
Example
1 Tg = 1 X IO12 grams
1 Gg = 1 X IO9 grams
1 Mg = 1 X IO6 grams
1 km = 1 X IO3 meters
1 mV = 1 X IO"3 volt
1 \ig = 1 X IO"6 gram
-------
II-l
II. THE SOCHI WASTES
INTRODUCTION
The synthetic organic chemical manufacturing industry (SOCHI), which produces
an estimated 350 to 400 chemicals by approximately 600 processes, generates
large quantities and a wide variety of wastes. Secondary emissions, as used in
this report, are defined as emissions of volatile organic compounds (VOC) that
result from the handling, treatment, and disposal of aqueous, liquid, and solid
wastes generated by the industry. Organic emissions from cooling towers are
considered as fugitive emissions and are discussed in a separate report.1 Emis-
sions that result from the treatment or control of process emissions are not
considered as secondary emissions unless they emanate from aqueous, liquid, or
solid wastes generated in the treatment or control of process emissions.
PROCESS WASTE SOURCES
Introduction
An overview of typical waste sources and the probable characteristics of wastes
from each type of source is given in Table II-l. It should be emphasized that
only the more prevalent combinations of sources and types of wastes are included.
The major waste categories shown include aqueous wastes, liquid organic wastes,
and solid wastes (including sludges and slurries). The major waste categories
are further subdivided according to probable characteristics (i.e., content of
organi.es, inorganics, water, solids, metal salts, and acids or bases (pH)J that
may influence or restrict treatment or disposal techniques and, in turn, affect
the resulting secondary emissions.
Table II-2 summarizes the primary operations that generate wastes and the cor-
responding types of wastes for 34 of the most significant chemicals produced by
the SOCHI and for 41 of the primary processes used. Although the products listed
account for less than 10% of the total number of chemicals produced, they account
for more than 70% of total industry production and total industry VOC emissions.
A number of the processes listed, however, do not generate significant secondary
emissions. The data in Table I1-2 were extracted from Emissions Control Options
for the SOCHI: Product Reports, which are incorporated in other volumes of this
report, and cover the products listed in the first column.
-------
II-2
Table II-l. Summary of Common Process Waste Sources and
Corresponding Waste Characteristics
Aqueous
Organic
Cone.
Process Waste Sources
Distillation
Overhead
Conventional drying
Azeotropic drying
r
a
0
o
0
I
o
10
X
X
&
"a
o
o
o
o
r-4
I
O
o
r-i
X
X
Cu
o
o
o
o
A
X
X
1 Inorganics
ssolvet
Q
in
C
•H
a
c
o
o
1 Solids
01
•a
c
u
in
p
(a
in
C
id
C
O
u
o
A
„
s,
o
•H
01
id
n
V
x
Qi
U
S
Liquid Organic
^
c
O
a>
a
•H
C
13
Cn
Vl
O
1 Inorganics
ssolvec
O
01
a
•H
a
c
8
1 Solids
•«J
-------
Table II-2. Waste Sources in Selected SOCMI Processes
Product
Acetaldehyde
Acetone /phenol
Acetic anhydride
Estimated
1978
Production
Primary Process (Gg/yr)a Type of Waste
Ethylene oxidation 450 Aqueous
Liquid
Cumene 900 Aqueous
Acetic acid pyrolysis 680 Aqueous
Solid
Waste Source
Distillation bottoms
Distillation side cut
Water scrubbing, phase
separation, distilla-
tion bottoms
Gas scrubbing, vacuum
steam- jet ejector
Distillation bottoms,
sludge evaporator
H
H
1
10
Acrylic acid and esters Propylene oxidation
410
Acrylonitrile
Adipic acid
Alkylbenzene
Propylene ammoxidation
Nitric acid oxidation
Olefins
815
855
250
Aqueous
Spent sulfuric acid
Liquid
Solid
Aqueous
Liquid
Aqueous
Aqueous
Solids
Quench-absorber, distil-
lation
Distillation
Distillation
Distillation, spent
catalyst
Distillation, steam
stripping
Distillation
Deep-well holding pond
HF scrubber, azeotropic
distillation, vacuum
jet
Calcium fluoride from HF
scrubber
-------
Table II-2. (Continued)
Product
Primary Process
Estimated
1978
Production
(Gg/yr)*
Type of Waste
Waste Source
Chlorination
Butadiene
rt-Butane dehydrogena-
tion
260
Oxidative dehydro-
genation
265
By-product of ethylene
manufacture
975
Caprolactam
Cyclohexanone
400
Aqueous
Liquid
Solid
Aqueous
Solid
Aqueous
Solid
Aqueous
Solid
Aqueous
Liquid
Distillation, phase
separation, neutrali-
zation, steam jets
Phase separation
(skimming)
Catalyst sludge spent
filters
Distillation, phase
separation
Polymeric residues from
solvent and product
purification
Distillation, phase
separation , steam
condensed in quench
tower
Polymeric residues from
solvent and product
purification
Distillation, phase
separation
Polymeric residues from
solvent and product
purification
Neutralization, distil-
lation, filtration
Neutralization, distil-
lation bottoms
M
H
I
Ca.-usti.c
-------
Table II-2. (Continued)
Product
Chloromethanes
Cumene
Cyclohexanol/cyclo-
hexanone
Ethanolamines
Ethylene
Primary Process
Methane chlorination
Phosphoric acid
catalyst
Aluminum chloride
catalyst
Cyclohexane
Phenol hydrogenation
Ethylene oxide and
ammonia
Ethane /propane feed
Naphtha feed
Estimated
1978
Production
(Gg/yr)a
930
968
289
1,012
178
165
6,100
3,000
Type of Waste
Aqueous
Liquid
Aqueous
Solid
Aqueous
Aqueous
Liquid
Liquid
Aqueous
Liquid
Aqueous
Solid
Aqueous
Waste Source
Spent sulfuric acid,
waste caustic from
scrubbers, misc.
wastewater
Distillation residues
Phase separation
Spent catalyst
Azeotrope distillation,
phase separation,
scrubbers
Stripper overhead, misc,
sources
Distillation residues
Catalyst reactivation
Steam jet condensate
Distillation residues
Quench effluent phase
separation
Pyrolysis coke
Quench effluent phase
H
Solid
separation
Pyrolysis coke
-------
Table I1-2. (Continued)
Estimated
1978
Production
Product Primary Process (Gg/yr)a Type of Waste
Waste Source
Ethylbenzene'/styrene
Ethylene dichloride/
vinyl chloride
Ethylene glycol
Ethylene oxide
Formaldehyde
Gas oil feed
Benzene alkylation/
dehydrogenation
Direct chlorination
Ethylene oxidation
(oxygen)
Metallic silver
catalyst
Metal oxide catalyst
3,000
3,220
4,900
Ethylene oxide hydration 1,960
Ethylene oxidation (air) 1,460
729
2,250
750
Aqueous
Solid
Aqueous
i
Liquid
Solid
Aqueous
Liquid and/or solid
Aqueous
Liquid
Liquid
Solid
Liquid
Solid
Aqueous
Aqueous
Quench effluent, phase
separation
Pyrolysis coke
Aluminum chloride cata-
lyst solution, spent
caustic
Distillation bottoms
Distillation bottoms
Acid scrubbers waste-
water stripper
Distillation residues
Cooling tower purge,
vent condenser con-
dens ate
Distillation residue
Distillation bottoms
Spent catalyst
Distillation bottoms
Spent catalyst
Ion exchange regenera-
tion, cooling tower
purge
Ion exchange regenera-
tion, cooling tower
purge
H
H
i
01
-------
Table II-2. (Continued)
Product
Glycol ethers
Primary Process
Reaction of alcohols
Estimated
1978
Production
(Gg/yr)a
281
Type of Waste
Aqueous
Waste Source
Vacuum system condensate
Maleic anhydride
Methanol
Methyl methacrylate
Nitrobenzene
Perchloroethylene/
trichloroethylene
with ethylene and/or
propylene oxide
Benzene oxidation
Low pressure
Acetone cyanohydrin
Benzene nitration
Chlorination and oxy-
chlorination of
ethylene dichloride
135
3,350
398
434
300
Liquid
Aqueous
Liquid
Solid
Aqueous
Liquid
Aqueous
Liquid
Solid
Aqueous
Aqueous
Liquid and/or solid
Distillation bottoms
Distillation, phase
separation
Distillation residues
Spent catalyst
Distillation bottoms
Distillation side cut
Salt solution from neu-
tralization, distilla-
tion
Spent sulfuric acid
Polymeric residues formed
in MMA reaction
Sulfuric acid from
stripper, nitrobenzene
scrubber, caustic from
neutralizer
Stripper bottoms
Distillation residues
-------
Table II-2. (Continued)
Product
Propylene oxide
Terephthalic acid
(crude)
1,1 ,1-Trichloroethane
Primary Process
Chlorohydronat ion
i-Butane peroxidation
Ethyl benzene peroxi-
dation
Air oxidation
Vinyl chloride
Estimated
1978
Production
(Gg/yrja
500
280
130
1,985
310
Type of Waste
Aqueous
Aqueous
Liquid
Aqueous
Liquid
Solid
Aqueous
Liquid
Aqueous
Waste Source
Reaction /distillation
Water wash, scrubbing.
phase separation
Distillation residues
Water wash, scrubbing,
phase separation
Distillation residues
Spent catalyst
Distillation
Distillation residues
Scrubbing, phase separa-
Ethane
18
Liquid/solid
Aqueous
Liquid/solid
tion
Distillation residues
Scrubbing,phase separa-
tion
Distillation residues
H
H
I
00
See ref 2.
TP-Butanol produced as co-product.
-------
II-9
" Frequently, two or more of the operations listed are involved in the generation
of a single waste (e.g., the waste may be formed in one operation and separated
in another operation). For this reason the waste sources shown tend to overlap
to some extent. Following is a discussion of waste sources and the corresponding
types of wastes which are common for each source category.
2. Distillation
a. Introduction Distillation is widely used in the SOCMI for the separation of
two or more components present in a liquid mixture. Separation is attained by
partial vaporization, with the more volatile components tending to concentrate
in the vaporized fraction. With conventional distillation the completeness of
the separation is determined primarily by the relative volatility^f the com-
ponents, by the distillation column height, and by the reflux ratio used. Dis-
tillation is frequently used for the separation of wastes from process streams.
The waste components may occur as either the more volatile (overhead) or the
less volatile (bottoms) fraction.
*>- Aqueous Wastes—Conventional distillation is widely used to separate water
(introduced or formed in previous process steps) from single-phase organic-
water solutions. Because separation by distillation is rarely totally com-
plete, the water fraction will almost always contain some of the organic com-
pounds present in the original solution, with the concentration determined pri-
marily by the relative volatilities of water and the organic components. As
shown by Table II-l when water is the more volatile component and is removed
overhead, the resulting aqueous waste will generally be relatively free of
inorganic compounds or other solids contained in the original solution. However,
if water is the less volatile component and is removed as the bottoms fraction,
the aqueous waste will contain any inorganic compounds or other solids present.
Steam distillation is frequently used to purify organic compounds with rela-
tively high boiling points that tend to decompose at the temperatures necessary
for conventional distillation. The heat required for vaporization is supplied
by the direct injection of steam, and the overhead stream is an organic-water
vapor mixture. After condensation the overhead stream is separated into organic
and aqueous phases (relatively low water solubility of the organics is necessary).
-------
11-10
The aqueous layer, containing any soluble organics, is usually discharged as a
waste source. From the standpoint of aqueous waste generation, steam distillation
differs from conventional distillation in the following significant respects:
(1) the concentration of organics in the aqueous waste is determined primarily
by water solubility; with conventional distillation the organic concentration
is primarily a function of the relative volatilities of water and the contained
organics; (2) the source of water, which results in the formation of the aqueous
waste, is primarily the steam distillation operation itself (i.e., the injected
steam); with conventional distillation, water is introduced or formed in previous
steps and is merely separated in the distillation operation.
Azeotropic distillation is frequently used when an organic-water separation by
conventional distillation is not feasible because of unfavorable relative volati-
lities (often the result of the formation of an organic-water azeotrope). In
this type of distillation a second organic compound, or drying agent, which
forms a lower boiling azeotrope with water, is added, permitting overhead removal
of a water-drying agent mixture. If the drying agent has low water solubility,
it may be separated from the water by phase separation and recycled to the dis-
tillation operation and the water may be discharged as an aqueous waste source.
If the drying agent has significant water solubility, an additional step will
be required to separate water and the drying agent. If the aqueous waste is
phase-separated, the organic concentration will depend primarily on the water
solubility of the drying agent. When azeotropic drying is employed, the primary
organic component in the aqueous waste is often the drying agent rather than
the original organic compounds being dried.
c. Liquid Organic Wastes Conventional distillation is frequently used to remove
those liquid organics introduced as raw material contaminants or formed as unusable
by-products in previous process steps. The liquid organic waste may occur as
either the more volatile (overhead) fraction or the less volatile (bottoms)
fraction. When the liquid organic waste is the more volatile component and is
removed overhead, the resulting waste stream will generally be relatively free
of inorganic compounds or other solids contained in the feed solution. However,
if the waste is the less volatile component and is removed as the bottom fraction,
it will retain any inorganic compounds or other solids.
-------
11-11
Steam distillation is used primarily to separate organics with relatively high
boiling points from less volatile compounds. The resulting liquid wastes, often
in the form of high-boiling residues, are usually removed as the bottoms stream
and may contain any inorganic compounds or other solids that were present in
the feed solution.
azeotropic distillation is used for water removal, significant quantities
of liquid organic wastes are not usually generated.
^- Solid Wastes - In conventional distillation the solid wastes generated are almost
always removed in the bottom fraction. They are usually removed as sludges or
slurries or as liquids with high melting points that solidify when cooled to
ambient temperature and will contain any inorganic compounds present in the
feed solution.
In steam distillation the solid wastes generally have the same characteristics
as those from conventional distillation.
Azeotropic drying does not usually generate solid wastes.
3- Water Contacting Operations
a* Introduction - Operations in which process streams are directly contacted with
water are common throughout the SOCMI. This broad category includes a number
of specific unit operations [e.g., absorption (scrubbing) , quenching, extraction
(water washing), and neutralization]. Although the purposes and the conditions
of the specific operations vary significantly, the characteristics of the resulting
wastes are generally similar. .The general waste characteristics are given in
Table II-l.
k- Aqueous Wastes- — The predominant type of wastes resulting, from water-contacting
operations are aqueous wastes. Although the contact water is often recycled, a
portion is usually discharged as * ah aqueous waste to remove impurities or by-
products, which tend to accumulate in the recycled water. Additional water,
initially present in the process stream, may be removed in the water-contact
operation and will contribute to 'the quantity of aqueous waste (e.g., quenching
-------
11-12
of reactor effleunt which contains water vapor). The concentration of organics
in aqueous wastes from water-contact operations is usually primarily dependent
on the water solubility of the organics.
A primary function of water-contact operations is often the removal of inorganic
compounds from process streams (e.g., acid gas absorption, extraction of salts,
neutralization), which result in relatively high concentrations of inorganic
compounds in the aqueous wastes.
c. Liquid Wastes Although aqueous wastes are the predominant type of wastes from
water-contact operations, organic liquid wastes with relatively low water solu-
bility may be discharged with the aqueous waste and be subsequently separated
(e.g., organic liquid wastes formed in gas-cracking operations may be discharged
with effluent quench water).
d. Solid wastes Similar to the manner in which liquid organic wastes may be gen-
erated by water-contact operations, semisolid residues (i.e., high-melting liquids,
slurries, sludges) are frequently discharged with aqueous wastes from water-contact
operations and subsequently separated. The resulting solid or semisolid waste
usually contains a significant amount of water and may contain inorganic compounds
present in the stream subjected to water contact.
4. Reaction
a. Introduction Aqueous, liquid, and solid wastes are frequently formed in process
reaction steps; however, the wastes formed are usually not discharged directly
from the reaction vessel. The reactor effluent generally contains a mixture of
wastes and usable process materials (i.e., raw materials, intermediates, products,
usable by-products), and the wastes formed during reaction are usually actually
discharged after a subsequent separation operation (e.g., distillation, phase
separation, filtration).
b. Aqueous Wastes Aqueous wastes may result from the formation of water as a
reaction product, from the introduction of water to solutions or to mixtures
of raw materials, or from the direct introduction of water into the reactor
as a liquid or as steam. The concentration of organic and inorganic compounds
-------
11-13
in the resulting aqueous wastes will depend on subsequent separation operations,
as well as on the composition of the feed to the reactor and the properties of
the other reaction products.
c- Liquid Organic Wastes Liquid organic wastes may be formed as normal reaction
by-products or they may be introduced as raw material impurities or diluents.
The intermittent generation of liquid organic wastes may also occur as a result
of the temporary production of off-grade products that cannot be recycled or
otherwise used in the process. The production of off-grade products most fre-
quently occurs during startup or shutdown, during upsets caused by uncontrolled
variations in process conditions, or from the unintentional use of off-grade
raw materials or catalysts.
As with the aqueous wastes, the characteristics of liquid organic wastes (e.g.,
concentration of water or inorganic compounds) may be influenced by subsequent
separation steps, as well as by the conditions governing the reaction and com-
position of the feed.
**• Solid Wastes Solid wastes may be generated from normal reaction by-products,
from raw-material impurities or diluents, or from the discharge of spent catalyst.
As with liquid organic wastes, solid wastes may also be generated by the temporary
production of off-grade products, which occurs during normal startups and shutdowns
or during process upsets., The characteristics of solid wastes may also be influenced
by subsequent separation steps.
5- Gas Compression and Vacuum Generation
a- Introduction——Gas compression and vacuum generation, common operations in pro-
cesses within the SOCHI, are, in general, accomplished either with rotary mechani-
cal devices (e.g., centrifugal and reciprocating compressors, vacuum pumps) or
with venturi devices (e.g., steam- or water-jet ejectors). Significant quantities
of wastes are generated in these operations. The general characteristics of
wastes and generation mechanisms associated with the two types of devices (i.e.,
rotary or venturi) differ significantly and are discussed separately.
-------
11-14
b. Rotary Devices Aqueous wastes are frequently generated when organic gases
containing water vapor are compressed and subsequently cooled to remove the
heat of compression, condensing water vapor along with an equilibrium quantity
of the compressed organic compounds. If the contained organic compounds have
low water solubility, the condensed liquid is usually phase-separated to remove
the insoluble organics before the aqueous waste is discharged. In this case
the concentration of organics in the aqueous waste is primarily dependent on
the degree of water solubility of the organics. If the contained organics have
significant water solubility and cannot be readily separated, the concentration
of organics in the aqueous waste will depend primarily on the relative volati-
lities of the water and the organic compounds. Aqueous wastes from this source
may contain inorganic compounds present in the compressed vapor (e.g., acid vapors)
Water or aqueous solutions are frequently used as sealing fluids in rotary vacuum
pumps, resulting in the generation of significant quantities of aqueous wastes.
Vacuum pumps of this type cause intimate contact between the vapor removed and
the sealing fluid and sometimes serve the dual functions of vacuum generation
and vapor scrubbing (e.g., acid vapors can be removed, using caustic solution
as the sealing fluid). The resulting aqueous wastes can contain significant
concentrations of inorganic compounds, as well as equilibrium concentrations of
any organics present in the vapor removed.
Liquid organic wastes are frequently condensed and separated during the compression
of organic vapors in the same manner as the aqueous wastes discussed in the
previous section. The injection of compressor lubricating oil, required for
sealing or for piston and packing lubrication, may also contribute to the genera-
tion of liquid waste. Liquid wastes that occur with the use of rotary vacuum
pumps are generated primarily from the discharge or loss of sealing fluid. If
the sealing fluid is an aqueous solution, the contained insoluble liquid organic
wastes are frequently phase-separated before they are discharged. If organics
are used as a sealant (e.g., oil), the discharge of spent sealant or the entrain-
ment of sealant and subsequent separation from the discharged vapor can result
in the generation of liquid waste.
Solid wastes in significant quantities are not usually generated in compression
and vacuum generation operations. Minor quantities of polymers or other solid
-------
11-15
residues may be formed during compression and may have to be removed from the
equipment periodically.
c. Venturi Devices Aqueous wastes are generated when venturi devices, such as steam-
and water-jet ejectors, are used for vacuum generation. These devices develop vacuum
by directly contacting the vapor removed from the evacuated vessel with a high-
velocity jet of steam or water in a venturi throat. The effluent water (steam is
usually subsequently condensed) will contain the soluble organic or inorganic com-
ponents that were originally present in the vapor contacted, resulting in the
generation of aqueous wastes. The concentration of organics in the discharged
aqueous waste is generally relatively low. Since jet ejectors cause intimate
contact between the effluent vapor and the motivating fluid (steam or water),
they frequently serve the dual functions of vacuum generation and vapor scrubbing
When the gases that are removed contain significant concentrations of acid vapors
caustic solution instead of water may be circulated through the venturi, which
will result in relatively high concentrations of inorganic salts in the aqueous
wastes.
Liquid organic wastes may occur when venturi devices are used for vacuum genera-
tion. Organics contained in the removed gas may be condensed and removed with
the effluent aqueous waste. If the condensed organics have low water solubility,
they may subsequently be phase-separated from the aqueous waste stream and dis-
charged as a separate waste stream composed primarily of liquid organic compounds
Solid wastes containing significant quantities of organic compounds are not
usually generated by venturi-type devices.
&. Solid-Liquid Separation
a. Introduction The separation of liquids from solids, accomplished primarily by
flotation, sedimentation, and filtration, can result in the generation of pri-
marily aqueous and solid wastes. Liquid-solid separations are frequently used
to separate wastes that are formed as normal-process by-products, introduced
with raw materials as contaminants or diluents or introduced as catalysts or
other process aids that must subsequently be removed.
-------
11-16
b. Aqueous Wastes Aqueous wastes are frequently generated when solid products or
intermediates, precipitated in aqueous solutions, are subsequently filtered.
The discarded filtrate may contain significant quantities of organic compounds.
The filtrate generally contains an equilibrium concentration of the product,
whose concentration is determined by the solubility of the product in the filtrate
solution. Other organic components in the filtrate may include unreacted organic
raw materials and soluble organic by-products.
The concentration of inorganic compounds in discarded filtrate solutions is
often significant. Inorganic compounds are frequently added in excess as raw
materials, and salts are frequently formed as normal-reaction by-products.
Inorganic salts may also be added to reduce product solubility, thereby reducing
product losses in the filtrate.
Products or intermediates that have been precipitated from aqueous or organic
solutions and subsequently separated are frequently washed with water to remove
residual organic or inorganic raw materials or by-products. The effluent wash
water, usually discharged as an aqueous waste, will contain most of the com-
ponents contained in the initial filtrate although usually in much lower con-
centrations. The concentration of the separated product in the discharged wash
water, however, will be dependent on water solubility and may be greater than
in the initial filtrate if the other components, present in much higher con-
centration in the initial filtrate, reduce product solubility.
c. Liquid Organic Wastes Significant quantities of liquid organic wastes are not
usually generated from liquid-solid separation operations. When precipitation
occurs in an organic solution or when the filtered precipitate is washed with
an organic solvent instead of water, the solvent is usually recovered by distil-
lation and then recycled.
d. Solid Wastes Solid wastes generated from liquid-solid separation operations
are of two general types: solids of primarily inorganic composition containing
relatively small quantities of liquid or solid organic compounds and solids of
primarily organic composition usually in the form of slurries or sludges, which
frequently contain significant quantities of water and inorganic salts.
-------
11-17
Solid wastes that are primarily of inorganic composition may be generated from
normal reaction by-products, from raw material impurities, from the discharge
of spent catalyst (i.e., the solid spent catalyst is discharged from the reactor
with liquid reaction products and is subsequently removed by liquid-solid sepa-
ration) , and from the discharge of inorganic compounds used for filter precoating
(e.g., diatomaceous earth).
Solid wastes composed primarily of organic compounds frequently result from the
formation of heavy residues or polymers as undesirable by-products in process
equipment and storage tanks. These wastes are frequently transported as liquid-
solid mixtures (e.g., slurries, floes, suspensions) and are subsequently sepa-
rated.
A major source of solid wastes that fall within this category is activated
sludge, which is formed in biological waste treatment systems. Activated sludge
is usually separated from the effluent wastewater by sedimentation (settling),
followed by various dewatering techniques.
?• Adsorption
a- Introduction Adsorption is frequently used in the SOCMI for the separation
of gaseous or liquid components present in relatively low concentrations from
the respective gaseous or liquid streams. At high concentrations other sepa-
ration techniques (e.g., distillation) are usually more suitable. The most
common adsorbent and the one generally used for the removal of organic compounds
from air (oxygen and/or nitrogen) and from water is.activated carbon. Other
adsorbents, such as activated alumina, molecular sieves, and silica gel, are
less common and are most frequently used for the removal of water and other
polar compounds from organic gas or liquid streams. Aqueous, liquid, or
solid wastes may be generated by adsorption processes.
b. Aqueous Wastes When aqueous streams are treated by adsorption to remove con-
tained organic compounds, the separation is rarely totally complete. The concen-
tration of organics in the effluent water (usually discharged as an aqueous waste)
will depend on a number of factors, including the inlet concentration, the
affinity of the adsorption media for the specific organic compounds, and the mass
flow rate through the bed. The effluent concentration will usually increase
-------
11-18
slowly as loading of the bed is continued, until the concentration of organics
in the bed reaches the "breakthrough" point. If loading is continued beyond
breakthrough, the effluent concentration of organics will increase rapidly and
will eventually reach the inlet concentration if not terminated. The concentration
of inorganics present in the treated aqueous stream is usually not significantly
changed as it passes through the adsorbent bed.
Aqueous wastes may also be generated during the regeneration (i.e., removal of
adsorbed organics) of adsorbents treating either gaseous or liquid streams when
in-place, nondestructive regeneration techniques are used. In-place regeneration
is more commonly used for treatment of gases than for treatment of liquids. The
most common method for in-place, nondestructive regeneration consists of passing
steam through the bed, with the adsorbed organics being eluted with the effluent
steam. After condensation of the effluent the organics may be phase-separated
and the aqueous phase discharged directly as an aqueous waste if the organics
have low water solubility. If the organics have significant solubility and
their value justifies recovery, an additional recovery step will be required
(e.g., distillation). Other regeneration techniques (e.g., pH change, solvent
regeneration), which are less common, may result in the generation of aqueous
wastes.
When adsorbents (e.g., alumina, molecular sieves) are used for the removal of
water from organic streams, the adsorbent bed is usually periodically regenerated
to remove the adsorbed water. This is usually accomplished by passing heated
gas (e.g., methane, nitrogen) through the bed. On cooling of the effluent gas
most of the contained water is condensed and is usually discharged as an aqueous
waste.
c. Liquid Wastes Liquid organic wastes resulting from the treatment of gas or
liquid streams by adsorption generally occur when the adsorbed organics are of
little or no value or when the cost of purification is excessive. The organics
recovered during regeneration may then be discharged as a liquid waste.
d. Solid Wastes Solid wastes, which may contain significant quantities of organic
compounds, generally occur when spent adsorbents (e.g., activated carbon, alumina)
are discarded. In most applications adsorbents tend to lose activity, or adsorption
capacity, with extended use as the result of a gradual loss of effective surface
area.
-------
8.
a.
11-19
When activated carbon loses activity, it is usually removed from the adsorber
and thermally treated in a furnace, where the carbon is partly oxidized. When
carbon usage is small, regeneration may not be practical and spent carbon may
be discarded as a solid waste.
Storage
Introduction The formation and/or separation of water and residues in storage
tanks for organic liquids frequently results in the generation of wastes, primarily
in the form of either aqueous wastes or semisolid sludges.
Aqeuous Wastes Water frequently accumulates as an insoluble layer in organic
liquid storage tanks and is periodically drained off and discharged as an aqueous
waste. The water phase may be introduced with the stored organic liquids as an
emulsion that separates on standing, or it may result from the leakage of rain-
water into open-top or floating-roof tanks. The concentration of organics in
the discharged aqueous waste is usually determined by the water solubility of
the organics. The discharged aqueous waste may contain inorganic compounds when
they are present in the incoming water.
Solid Wastes Solid residues (e.g., polymers) are occasionally formed in organic
liquid storage tanks, usually due to the slow degradation or polymerization of
contained liquid organics. If the residues are more dense than the liquid,
they will tend to collect in the bottom of the tank and may be periodically
drained off as a sludge or slurry. If the particles tend to remain in suspension,
the discharged organic liquid may have to be filtered or centrifuged. The col-
lected solids are then usually discharged as solid wastes.
9- General Process Area
Introduction—Aqueous, liquid, and solid wastes frequently result from a number
of miscellaneous conditions or occurrences that are common in the SOCMI. Included
are relatively small but frequent or continuous leaks from equipment, piping,
and storage tanks; accidental spills resulting from equipment failure or operating
errors; the normal generation of waste or discharge of materials occurring during
maintenance activities (i.e., shutdowns, equipment repair, equipment cleanout);
-------
11-20
washdown of process areas; and rainfall runoff. As the actual generation of
wastes will frequently result from combinations of these conditions or events,
there is some overlap in the waste sources discussed.
b. Aqueous Wastes Aqueous wastes frequently result from the flushing of equipment
for cleanout purposes or in the preparation of equipment for maintenance, process
area washdown, and storm runoff. The organics in aqueous wastes from area washdown
and storm runoff generally result from equipment leaks and/or spills.
c. Liquid Wastes Liquid organic wastes can result from equipment leaks, spills,
and drainage of equipment before or during maintenance operations. Frequent
sources of organic liquid leaks are pump mechanical seals and packing glands.
d. Solid Wastes Solid wastes can result from spills of solid materials and from
solid residues removed from equipment and piping during cleanout and maintenance
operations.
-------
11-21
C. REFERENCES*
1. D. G. Erikson and V. Kalcevic, IT Enviroscience, Inc., Emissions Control Options for
the Synthetic Organic Chemicals Manufacturing Industry. Fugitive Emissions Report
(on file at EPA, ESED, Research Triangle Park, NC) (February 1979).
2. Chemical Economics Handbook Manual of Current Indicators, Supplemental Data,
Stanford Research Institute, Menlo Park, CA.
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
III-l
III. SECONDARY EMISSION SOURCES
A- INTRODUCTION
Secondary emissions result from the handling, treatment, and disposal of the
wastes described in Sect. II. Table III-l provides an overview of the waste
treatment or disposal methods that are generally applicable to the waste cate-
gories shown in Table II-l. The resulting secondary emissions are determined
by both the characteristics of the wastes and the treatment or disposal methods
used. The following discussion covers the most significant treatment and dis-
posal methods and the general waste categories for which these methods are appli-
cable. As was the case for the waste sources, the generic classifications of
treatment and disposal methods tend to overlap to some extent (e.g., physical
separation steps are integral parts of a biological treatment system). Mention
is made of operations and processes described as waste sources in Sect. II that
are also included as waste treatment or disposal methods or secondary emission
sources in this section. The distinction between the waste sources and the
secondary emissions sources is determined by the function of the operation. If
its primary function is the treatment or disposal of aqueous, liquid, or solid
waste, then it is considered to be a secondary emission source. The distinction
is not always clear. An operation may be both a waste source and a secondary
emission source [e.g., a biological treatment system treats aqueous wastes (waste
treatment system) and discharges aqueous waste and solid waste or sludge (waste
source)].
B- PHYSICAL SEPARATION METHODS
*• Introduction
Physical separation methods do not actually destroy or dispose of aqueous, liquid,
and solid wastes but remove or reduce the concentration of components, thereby
rendering the wastes more acceptable or amenable to subsequent disposal methods.
For this reason essentially all physical separation operations for treating
wastes are also waste sources and the actual disposal of the,wastes may result
in additional secondary emissions.
-------
III-2
Table III-l. Summary of Waste Characteristics and
Corresponding Common Treatment Operations
Aqueous
Organic
Cone.
Treatment/Disposal
Operations
a
0
o
o
*
I
id
I
0
0
o
fe
o
1
o
o
r-i
ft
ft
o
0
o
o
rH
A
in
U
S1
§
H
•o
H
o
(0
in
tains
c
8
nded Solids
ft
3
to
CO
•H
id
c
8
o
!E
ft
6
•H
to
id
oa
Solids
Liquid Organic and
V
SB
ft
i
•H
T3
•H.
0
CO
U
•H
id
z
o
8
3
.8
Ul
0
-H
tn
O
H
H
O
M
in
a
en
•H
G
8
nded Solids
ft
1
a
•H
JS
8
flj
(d
S
(0
.5
id
jj
d
8
>i
H
W
O
•H
C
o
, Slurries,
Sludges
8
•H
s
o
H
CQ
C
•H
id
JJ
c
8
Metals
fr
id
-------
III-3
2- Steam Stripping
a> Aqueous Wastes Steam stripping is most frequently used to reduce the concen-
tration of organic components present in aqueous wastes. Steam stripping is
similar to steam distillation (Sect. II-B-2-b) in that volatilization is attained
by the direct sparging of steam into the liquid phase. The distinction between
steam stripping and steam distillation lies in the components that are separated.
Steam stripping is used to remove organics from aqueous solutions. Steam dis-
tillation is used primarily to separate relatively high boiling organics from
higher boiling impurities or residues.
As with steam distillation, the overhead vapor from a steam stripper is a mixture
of steam and the organic components removed from the original solution. If the
contained organic compounds have low water solubility, the condensed overhead
stream separates into aqueous and organic phases. The organic phase may be
further processed to reduce the water content (e.g., conventional distillation)
and be returned to the process or discharged as an organic liquid waste. The
aqueous phase, which often has approximately the same organic concentration as
the stripper feed has, may be recycled to the stripper. The bottoms stream
from the stripper is discharged as an aqueous waste stream.
Secondary emissions from steam stripping result primarily from the venting of
gases contained in the overhead stream, which are not condensed (primarily air).
The quantity of contained VOC that is vented is generally determined by the
quantity of noncondensable gases vented, the vapor pressure(s) of the organic
component(s), and the effective condenser temperature.
b * Liquid Organic Wastes——Steam stripping is not generally used for the treatment
of liquid organic wastes.
c- Solid Wastes—Steam stripping may be used for the treatment of solid wastes that
either are primarily inorganic but contain significant quantities of volatile
organic compounds (e.g., spent catalyst, spent adsorbents, filter aids, precipi-
tates) or are primarily composed of organic solids with low volatility but contain
significant quantities of organic compounds of relatively high volatility (e.g.f
activated sludge). The main purpose for steam stripping is to reduce emissions
-------
III-4
of VOC from the solids during subsequent handling and disposal operations. As
with the steam stripping of aqueous wastes, emissions of VOC occur mostly with
the venting of noncondensable gases.
3. Conventional Distillation
a. Aqueous Wastes Conventional distillation is seldom used in the treatment of
aqueous wastes unless the concentration and quantity of organic compounds that
can be recovered are sufficient to justify the energy requirements. Steam stripping
is generally more suitable for solutions with very low concentrations, possibly
followed by the conventional distillation of the stripper overhead if the contained
organics and water have significant mutual solubility. As with steam stripping,
emissions mainly result from the venting of noncondensable gases and are usually
relatively low.
b. Organic Liquid Wastes The conventional distillation of liquid organic wastes
may occur as a pretreatment step to render the wastes more amenable to final
disposal methods (e.g., removal of water or inorganics prior to combustion or
removal of highly volatile, noxious, or hazardous components prior to disposal).
Emissions result mainly from the venting of noncondensable gases.
4. Liquid-Liquid Phase Separation/Emulsion Breaking
a. Aqueous Wastes Phase separation and/or emulsion breaking operations are fre-
quently used to separate aqueous wastes from organic liquids that have relatively
low water solubility. The most common methods of phase separation and/or emulsion
breaking include density differential separation (e.g., API oil/water separators,
plate separators, centrifuges), filtration/coalescence separation (e.g., granular
media, fibrous media), and chemical separation (e.g., acid-alum-lime process,
polyelectrolyte addition). The most significant emissions occur from the vapori-
zation of organic compounds when separation vessels or basins such as API separators
and plate separators are open-topped, a relatively common practice throughout
the SOCMI. Emissions from enclosed or covered vessels are generally much less
than those from open-top vessels although they may be significant if vented
without additional controls.
-------
III-5
b. . Liquid Organic Wastes Phase separation of aqueous wastes and liquid organic
wastes frequently precedes final treatment or disposal operations when separate
disposal methods are used for the aqueous wastes and the liquid organic wastes
(e.g., aqueous wastes are biologically treated and liquid organic wastes are
incinerated). The comments concerning emissions from the previous section also
apply to liquid organic wastes.
c- Solid wastes Liquid-liquid phase separation is not applicable to solid wastes.
5. Solid-Liquid Separation
a. Introduction Solid-liquid separation operations, including sedimentation (clari-
fication, thickening), flotation, filtration, and centrifugation, are frequently
used in the SOCMI for the treatment of aqueous, liquid, and solid wastes. These
specific unit operations frequently occur as integral steps within more inclusive
treatment processes [e.g., following biological treatment, aqueous waste is
separated from activated sludge by clarification and thickening (sedimentation)].
The thickened sludge may be further dried (dewatered), and the residual suspended
solids in the aqueous waste effluent may be further reduced by filtration.
b. Aqueous Wastes The most significant emissions occur from the vaporization of
volatile organic components when separation vessels, such as settling basins,
clarifiers, and thickeners, are open-topped, a relatively common practice in
the SOCMI. Emissions from enclosed or covered vessels are generally much less
than those from open-topped vessels although they may be significant if vented
without additional controls.
c. Liquid Organic Wastes Liquid-solid separation operations used in the treatment
of wastes are less common for liquid organic wastes than for aqueous wastes.
Liquid organic wastes containing solids may require separation as a pretreatment
step prior to final disposal (e.g., solids are removed by filtration before
incineration of liquid organic wastes to prevent clogging of incinerator atomizing
nozzles); however, emissions from this type of source are relatively minor.
Frequently operations whose function is either liquid/solid separations (e.g.,
settling basins) or liquid/liquid phase separations (e.g., API separators) may
provide a three-way separation (i.e., aqueous liquid, insoluble organic liquids,
-------
III-6
and solids), with the liquid/liquid phase separation taking place in the upper
part of the vessel and the settled solids being removed from the bottom. The
emissions of volatile organic compounds from open-topped vessels may be signifi-
cantly greater when insoluble liquid organics separate as an upper layer.
d. Solid Wastes Solid wastes are discussed above.
C'. CHEMICAL TREATMENT
1. Introduction
The chemical treatment and disposal processes described in this section include
both those used for the terminal treatment of SOCMI wastes and those used as
pretreatment steps to render the wastes more amenable to terminal treatment.
Major categories of chemical treatment methods discussed in the following
sections include neutralization, precipitation, and chemical oxidation. As was
true for liquid/liquid and solid/liquid separation operations, chemical treatment
operations frequently take place in open-topped vessels or basins or in covered
vessels that are vented without control. Emissions result chiefly from the
vaporization of organic components contained in the wastes.
2. Neutralization
a. Aqueous Wastes In many manufacturing operations within the SOCMI large quanti-
ties of acidic or alkaline aqueous wastes are discharged. Most regulatory codes
specify that the pH of any effluent discharge should not be lower than 6 or
higher than 9 units. Neutralization to a similar pH range is also generally
required in biological treatment systems when acid or alkaline characteristics
of the wastewaters would otherwise interfere with or retard biological activity.
Neutralization is accomplished by the addition of alkali to acids or of acid to
alkalis as required to effect the desired pH adjustment. Neutralization process
equipment generally consists of open-topped or covered neutralization basins,
mixing devices, and neutralizing agent feed, storage, and control facilities.
Emissions generally result from the vaporization of contained organic compounds
from the surface of the liquid in the neutralization basin.
b. Liquid Wastes Neutralization as a waste treatment operation is not generally
applicable to liquid organic wastes.
-------
III-7
c- Solid Wastes Neutralization as a waste treatment operation is not usually appli-
cable to solid wastes; however, solid wastes, often composed of relatively insoluble
inorganic salts, may be formed as precipitates during neutralization and may require
a subsequent solid/liquid separation step.
3- Precipitation and Coagulation
a' Aqueous Wastes—-Chemical precipitation is a process in which chemicals are
added to react with specific constituents to form insoluble products that can
be removed in a subsequent solid-liquid separation step, such as by sedimentation
or filtration. In wastewater treatment, precipitation processes are perhaps
most commonly associated with the removal of such heavy metals as zinc, nickel,
cadmium, and lead. Metals are generally precipitated as the hydroxide or carbonate
species. If the precipitate is formed as a colloidal dispersion that cannot be
removed by a conventional sedimentation or filtration process, a coagulation
step may be added. Coagulation may be defined as the addition of chemicals to
cause aggregation of the dispersed colloidal particles and to permit separation
by conventional sedimentation or filtration. Precipitation/coagulation operations
and the ensuing sedimentation or filtration operations frequently take place in
open-topped basins or tanks. Emissions result chiefly from the vaporization of
contained organics.
Organic Liquids Precipitation/coagulation as a waste treatment operation is
not generally applicable to liquid organic wastes unless the wastes also contain
a significant concentration of water.
C' Solid Wastes Precipitation is not applicable to solid wastes. Coagulation
may be used to aid in the removal of finely dispersed particles of solids con-
tained in aqueous wastes.
4- Chemical Oxidation
a- Aqueous Wastes—Chemical oxidation is used to chemically modify waste streams
either by completely oxidizing the Organics to C02 and HgO or by partially oxi-
dizing the organics to detoxify them. These waste streams are usually aqueous
wastes that cannot be handled directly by biological-oxidation (either because
of organic strength or bioinhibitory characteristics) or incineration (because
-------
III-8
the organic concentration is too low or there is excessive corrosion caused by
the presence of inorganic salts or halogens). Oxidation is also used to purify
aqueous streams (such as waste sodium chloride brines) to permit recycling.
Most of the chemical oxidation processes are based on one of two oxidizing
agents chlorine or oxygen. Because chemical oxidation processes currently
have relatively limited use and enclosed vessels are generally required (to
conserve oxidizing agents or because elevated temperatures and pressure are
required), secondary emissions of VOC from these processes are generally rela-
tively low compared with emissions from the more conventional aqueous waste
treatment methods.
b. Liquid Organic Wastes Chemical oxidation is not generally applicable.
c. Solid Wastes Chemical oxidation is not generally applicable.
D. THERMAL DESTRUCTION
1. Introduction
Thermal destruction or incineration is widely used within the SOCMI for the
disposal of liquid and solid organic wastes. Because of high supplemental energy
requirements to vaporize and heat the contained water, thermal oxidation is not
generally used for the disposal of large quantities of aqueous wastes unless
the more conventional treatment methods (usually biological) are not suitable.
Emissions of VOC from thermal destruction processes result primarily from the
incomplete combustion of hydrocarbons, with residual organic compounds being
discharged in the flue gas. The destruction efficiency, or VOC removal efficiency,
is dependent on a number of factors, including the combustion zone temperature,
the residence time at the combustion temperature, the degree of turbulence in
the combustion zone, the percentage of excess air or oxygen (above stoichiometric
requirements), the composition and physical characteristics of the wastes, and
the specific design features of the incinerator.
In general, incinerator design and operating requirements for the effective
destruction of liquid wastes are less demanding than those than for the destruc-
tion of solids. The chemical composition of the contained organic compounds in
-------
III-9
both liquid and solid wastes is very significant. The difficulty of attaining high
VOC removal efficiency generally increases with increasing molecular weight and
with the increasing carbon:hydrogen ratio of the organic compounds.
The presence or concentration of the following waste components may determine
the suitability of thermal destruction as a disposal method or influence the
incinerator design and flue-gas treatment requirements:
Water, if present in an excessive amount, may result in the heat of combustion
of the contained organics being insufficient to attain the required temperature
and auxiliary fuel will be required.
Sulfur will result in the presence of S02 in the flue gas and may require flue
gas treatment.
Halogens (primarily Cl~) will result in the presence of HC1 (HBr, Br2, C12) in the
flue gas, which may cause severe corrosion problems and may require flue-gas
treatment.
Salts may cause excessive salt particulates in the flue gas, which are diffi-
cult to remove by flue-gas treatment, and may form low-melting slag, possibly
causing operating problems with conventional incinerators.
Phosphorous may form particulates that are difficult to remove by flue-gas treat-
ment.
Combined nitrogen may form excessive NO in the flue gas, which is difficult to
--------- "^ X
remove by flue-gas treatment.
The concentrations of highly toxic compounds, e.g., mercury, arsenic, and cyanides,
in vented flue gas must be extremely low; they are difficult to reduce to accept-
able levels by flue-gas treatment. Heavy metals may be leached from ash to form
highly toxic aqueous wastes.
Many types of incinerators are manufactured, and each type is basically designed
for specific waste profiles. Most systems will handle liquids to some degree
-------
111-10
but significant variations exist in sludge and solids handling capabilities.
The most prevalent types of incinerators are single-closed-chamber, rotary-kiln,
multiple-hearth, and fluid-bed units. In addition, liquid wastes with favorable
flue-gas characteristics (i.e., low toxicity, noncorrosive, low particulate
concentration) are frequently burned as supplementary fuel in direct-fired boilers
and process furnaces.
Other thermal destruction processes that currently have rather limited industrial
application include catalytic oxidation and pyrolysis processes.
2. Aqueous Wastes
Because of their high energy requirements, large quantities of aqueous wastes
containing relatively low concentrations of organic compounds are not generally
treated by thermal destruction methods; however, if the wastes are not amenable
to biological treatment (usually due to toxicity or poor biodegradability),
thermal destruction may be the most suitable alternative. When possible, aqueous
wastes are usually burned with organic liquid wastes to minimize or eliminate
the need for auxiliary fuel (i.e., oil, gas). Other than the supplementary
fuel requirements, incineration requirements and VOC emissions resulting from
the combustion of aqueous wastes are generally similar to those for liquid wastes,
discussed in the next section. Aqueous wastes frequently contain significant
quantities of soluble inorganic compounds, which may result in the problems
and/or corresponding incinerator requirements discussed in Sect. III-D-1.
3. Liquid Wastes Most of the types of incinerators discussed in Sect. III-D-1
can handle liquid wastes,- however, if concurrent thermal destruction of solid
wastes is not required, a relatively inexpensive closed-chamber incinerator or
a boiler or process furnace combustion chamber can generally be used. The thermal
destruction of liquid wastes in boilers or process furnaces is generally limited
to wastes that do not form flue gas that is severely corrosive or requires
extensive treatment before discharge.
The characteristics of the emissions resulting from the thermal destructon of
liquid wastes are dependent on the waste characteristics and incinerator param-
eters discussed in Sect. III-D-1.
-------
III.-ll
4- Solid Wastes
Of the types of incinerators discussed in Sect. III-D-1, the rotary-kiln, multiple-
hearth, and fluid-bed incinerators are all designed for the thermal destruction
of solid wastes, although all of them can concurrently handle liquid wastes.
Features incorporated in solid-waste incinerators, which are generally not necessary
for the combustion of liquids, include a feed mechanism, an ash removal system,
and a means of tumbling or agitating the wastes to achieve complete burnout.
Rotary-kiln incinerators also usually require a secondary combustion chamber
(complete combustion of vapors released when the solid wastes are heated cannot
generally be attained in the primary chamber).
The characteristics of emissions resulting from the thermal destruction of solid
wastes are dependent on the waste characteristics and incineration parameters
discussed in Sect. III-D-1.
r>
BIOLOGICAL TREATMENT
Introduction
Biological waste treatment is a generic term applied to a variety of processes
that utilize active microorganisms,to convert primarily organic constituents in
aqueous wastes to more stable forms. Biological waste treatment processes essen-
tially simulate the biological reactions that would occur in the environment.
However, since biological processes generally employ high concentrations of
active microorganisms under controlled conditions, the decomposition rates of
degradable materials are vastly accelerated.
*n general, biological processes are applicable to the treatment of soluble,
degradable organics in the concentration range of 0.01 to 1%. Removal efficiencies
can vary from 50 to 99+% (BOD or COD), depending on the process configuration,
the loading factor, and the nature and distribution of organic material present
in the wastewater.
The major process equipment in,biological treatment is basically a reactor to
provide contact between the wastewater and the microorganisms and to introduce
air or oxygen (aerobic processes only), a solids-liquid separation device (such
as a settling tank or clarifier), sludge recycle pumps, and monitoring and control
-------
111-12
devices. Equipment for pH control and/or nutrient addition may also be required.
Highly concentrated or variable waste loads may necessitate the use of an equali-
zation basin before treatment. Finally, solids handling devices for final sludge
treatment and disposal may be necessary. A simplified flow diagram for a conven-
tional activated sludge treatment plant is shown in Fig. III-l.
Biological waste treatment operations (i.e., aeration, biomass-wastewater contact,
solid-liquid separation) usually take place in open-topped basins or vessels.
(Some of the newer and less widely used processes such as fluid-bed and enriched
oxygen processes utilize enclosed vessels.) Most of the VOC emissions result
from the vaporization of organic compounds in the aqueous wastes as they pass
through the various process steps. The quantity of VOC emissions is for the
most part a function of the organic concentration in the aqueous waste, of the
solubilities of the organic components, and of their vapor pressures. (The
factors affecting emissions are discused in more detail in Sect. IV.)
2. Aqueous Wastes
Biological waste treatment is used almost exclusively for the treatment of aqueous
wastes (0.01 to 1.0% concentration). The amenability of specific aqueous wastes
to biological treatment is dependent on the composition and concentration of
both the organic and the inorganic components present.
Many chemical compounds that are biodegradable in dilute solutions may inhibit
biological activity when they appear in a concentrated form. Other components,
such as metallic or cyanide ions, are toxic to biological organisms at certain
concentration levels. The actual concentration at which biological activity is
inhibited is difficult to determine for each separate compound.
Biological processes generally operate effectively only within a relatively
narrow pH range (approximately 6 to 9). The biological treatment of alkaline
or acidic wastes usually requires preneutralization to within this pH range by
the addition of either acid or alkali (see Sect. III-C-2). An open-topped basin
is also usually used for neutralization, and therefore additional emissions of
VOC may occur.
-------
111-13
Secondary Emissions
Wastewater
Clarifier
Conventional
Activated -
Sludge (CAS)
Wastewater
Effluent
Fig. III-l. Conventional Activated Sludge Treatment Plant
-------
111-14
3. Liquid Organic Wastes
Biological treatment processes are not generally used for the treatment of concen-
trated liquid organic wastes.
4. Solid Wastes
Biological treatment is not generally used for the treatment of solid wastes;
however biological processes are almost always sources of solid wastes, which
are in the form of sludges. The disposal of activated sludge discharged from
biological treatment processes may result in additional secondary emissions of
VOC. (Terminal disposal of activated sludge is usually either by storage in
landfill or by incineration.)
F. TERMINAL STORAGE
1. Introduction
Terminal storage, which includes landfilling, impoundment, and deep-well injec-
tion, is widely used in the SOCMI for the ultimate disposal of aqueous, liquid,
and solid wastes. The significant number of incidents involving the contamina-
tion of natural waters (groundwater and surface water) by toxic chemicals, which
were caused by poor waste storage practices, has brought terminal storage as a
means of waste disposal under criticism. Although the prevailing environmental
objectives are aimed at minimizing the use of terminal storage for the disposal
of hazardous or toxic wastes, there are currently no feasible alternatives for
certain wastes. However, if approved storage practices are observed, wastes
can be stored with minimal adverse effects on the environment.
Emissions of VOC from the landfilling and impoundment of wastes containing organic
compounds may be significant, depending on the concentration and properties of
the compounds in the wastes and on the storage procedures used. Well-designed
deep-well disposal systems for liquid wastes generally result in minimal secondary
emissions.
-------
111-15
2- ^ Landfill
a- General A landfill, as opposed to an open dump, is the engineered burying of
wastes in which consideration is given to the geological, hydrogeological, and
climatic conditions at the site, to the type, quantity, and physical properties
of the materials being dumped, and to the ultimate redevelopment of the land
after the landfill site has been closed. The most important single consideration
in selection of the landfill site is protection of the quality of associated
natural waters.
A naturally secure landfill is one situated on bedrock, shale, or other impervious
materials that prevent the seepage of leachate into groundwater, or the site
may be one where there is no groundwater nearby. When naturally secure sites
do not exist, liners made of natural or synthetic materials are used.
The important secondary-emission parameters are the physical and chemical proper-
ties of the wastes, the water content of the landfill soil, the physical and
chemical properties of the soils, the rate of air flow across the surface of
the landfill, and the biodegradability of the organic waste components. Secondary
emissions from landfilling are discussed more fully in Sect. IV-F.
Landfarming,1 which may be defined as the use of the upper soil layer (upper
6 to 8 in.) for the biological treatment/disposal of wastes, is receiving con-
siderable attention although it is not currently widely used. Landfarming is
primarily suitable for the disposal of nonhazardous liquid and solid wastes with
relatively low vapor pressures. Although no specific data on secondary emissions
from landfarming are available, the similarity of landfarming to landfilling
would indicate comparable secondary emissions and the following discussion of
secondary emissions from landfilling also generally applies to landfarming.
b. Aqueous Wastes Landfilling is generally not suitable and is seldom used for
the disposal of large quantities of aqueous wastes. When not amenable to bio- ,
logical treatment (due to poor biodegradability of organics or to significant
concentrations of toxic inorganic components), when impoundment is not an accept-
able alternative, or when deep-well disposal facilities are not available, land-
filling may be used for the disposal of relatively small quantities of aqueous
-------
111-16
wastes. The comments regarding landfilling practices and the corresponding
secondary emissions for liquid organic wastes generally also apply to the land-
filling of aqueous wastes.
c. Liquid Wastes Bulk liquids normally require special approval to be landfilled.
One method of minimizing the potential for migration of liquids is to keep the
liquid content below the absorptive capacity of the solid materials (solid wastes
and back-filled soil) in the landfill. When this is done, the secondary emissions
resulting from the landfilling of liquid wastes are essentially the same as for
solid wastes or sludges.
Stabilization or encapsulation of free liquids is required at many landfill sites
before liquid wastes are accepted for storage. Although various chemical fixation
and encapsulation methods have existed for some time, current concerns with hazard-
ous materials in landfills and new regulations have resulted in extensive research
being undertaken in these areas. Although the main purpose of stabilization is
usually to prevent the leaching of water-soluble components, it is also often
effective in controlling secondary emissions. Stabilization is further discussed
in Sect. V-E-5.
Drums and other containers are also used when liquids are landfilled, but problems
can occur in the landfill when the containers deteriorate, causing leakage of
stored liquids; therefore drums of liquids are limited to the absorptive capacity
of the filled solids in most landfills, as are bulk liquids.
d. Solid Wastes The greatest quantity of wastes disposed of by landfill are solid
wastes and sludges. Generally, terminal disposal options for solid organic
wastes are limited to either landfilling or incineration, and for solid wastes
composed principally of inorganic compounds landfilling is usually the only
feasible option, with the possible exception of ocean dumping.
Secondary emissions resulting from the landfilling of solid wastes and sludges
depend primarily on the concentration and vapor pressure of the organic material
and .on the rate of diffusion through the air surrounding the chemically covered
surface. Secondary emissions resulting from landfilling solid wastes are dis-
cussed in more detail in Sect. IV-F.
-------
111-17
3- ^ Surface Impoundment
*• Introduction Surface impoundment in open ponds or basins is frequently used
for the storage of aqueous and liquid wastes and sludges. Impoundment is usually
considered a terminal disposal method only when contained water is removed by
evaporation from the pond and the residues are landfilled in place.
As with landfills, the most important consideration is protection of the adjacent
natural water; however, in contrast to landfills, migration of toxic components
is not inhibited by adsorption by solid wastes or soil, and the impermeability
of the containing pond or basin is of even greater importance.
b- Aqueous Wastes Open ponds or basins are frequently used for the storage of
aqueous wastes that are not amenable to biological treatment or acceptable for
discharge to receiving waters. These wastes (e.g., brines) frequently contain
high concentrations of dissolved or suspended solids and are often permitted to
evaporate to dryness. This is not considered as an acceptable practice if the
wastes contain significant concentrations of VOC; unless there is significant
biological activity, essentially all the contained VOC will eventually result
in secondary emissions.
c- Liquid Organic Wastes Open ponds or basins can be used only for temporary
storage of liquid organic wastes. Basically, the only manner in which liquid
organic wastes can be terminally disposed of by impoundment is from the evapora-
tion of the contained organic compounds, a totally unacceptable practice.
d. Solid Wastes Impoundment basins are frequently used for the storage and drying
of biological and chemical sludges. Activated sludge, as discharged from a
biological waste treatment plant, will typically contain 95 to 98% water. Sludge
is usually dewatered to produce a cake with a solids concentration of 12 to 40%
before final disposal by landfill or incineration. When sludge is dried by
evaporation of the contained water in an open basin, secondary emissions may
occur as the result of the concurrent evaporation of VOC contained in the sludge.
-------
111-18
3. Deep-Well Injection
a- General The underground injection of liquid wastes into permeable rock forma-
tions isolated from potable water and mineral-bearing strata is in some cases
the only safe method for disposing of certain indestructible, nonconvertible, or
otherwise hard-to-treat hazardous substances.2 The use of deep-well injection
is generally selected only after all reasonable alternative disposal methods have
been evaluated and found to be less desirable in terms of environmental protec-
tion and dependability.
Secondary emissions resulting from actual deep-well injection operations are
generally minimal; however, preinjection storage or pretreatment operations
may result in significant secondary emissions. The most prevalent pretreat-
ment requirements are those aimed at preventing the injection wells from being
plugged by solids that are either suspended in the liquid wastes or are formed
by precipitation when dissolved waste components react with the well formation
fluid. The secondary emissions result primarily from the evaporation of contained
organics that occurs during the required preinjection treatment operations (e.g.,
neutralization, precipitation, solid/liquid separation).
b. Aqueous Waste Deep-well injection of aqueous wastes is'primarily used when
the wastes contain high concentrations of inorganic compounds, including acids
such as sulfuric, hydrochloric, and phosphoric; bases, such as sodium hydroxide,-
and salts, such as sodium chloride, sodium sulfate, aresenic sulfide, ammonium
bisulfate, sodium bromide, and calcium carbonate.
c. Liquid Wastes Organic chemicals that are disposed of by deep-well injection
include acids, such as maleic, formic, adipic, cresylic, salicylic, and acetic;
alcohols, such as methanol, tertiary butanol, phenol, and isopropanol; solvents,
such as acetone, toluene, xylene, formaldehyde, ethylbenzene, benzaldehdye, and
methyl ethyl ketones; and other compounds, such as sodium naphthenate, sodium
cresylate, calcium and sodium acetate, and large molecular structures, such as
styrene polymers and various polymeric resins.
d. Solid Wastes Not applicable; solids generally cause well-plugging problems.
-------
111-19
G- DISCHARGE TO NATURAL WATERS
1. Introduction
The primary concern stemming from the discharge of wastes to natural waters
(including the dilution of aqueous effluents and the ocean dumping of containerized
or bulk wastes) is with the effect of the wastes on the water quality. Generally,
if pretreatment methods are acceptable from the standpoint of water quality,
secondary emissions will be minimal.
2. Dilution of Aqueous Effluents
a. Aqueous Wastes The prevailing method for the terminal disposal of aqueous
wastes is by discharge and dilution in natural receiving waters. Generally, if
pretreatment is acceptable from the standpoint of the effect of the wastes on
the water quality, secondary emissions are minimal.
b- Liquid Waste The direct discharge of liquid organic wastes into natural receiving
waters is never an acceptable practice.
c- Solid Wastes—The discharge of solid wastes (with the exception of deep-ocean
dumping) is not an acceptable practice.
3- Ocean Dumping
a. Aqueous Wastes Deep-ocean dumping of aqueous wastes is not generally applicable.
b. Liquid Wastes Ocean dumping of bulk liquid organic wastes is not an acceptable
practice. Deep-ocean dumping of containerized, stabilized, or encapsulated
liquid wastes, although possibly undesirable from the standpoint of long-term
water quality, probably results in minimal secondary emissions.
c- Solid Wastes-^Secondary emissions are minimal.
-------
111-20
H. REFERENCES*
1. R. L. Huddleston, "Solid-Waste Disposal: Landfarming," Chemical Engineering
86(5), 119—124 (Feb. 26, 1979).
2. M. E. Smith, "Solid-Waste Disposal: Deepwell Injection," Chemical Engineering
86(8), 107—112 (Apr. 9, 1979).
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
IV-1
IV. EMISSIONS
INTRODUCTION
Secondary emissions are estimated to account for 2 to 5% of the total VOC emis-
sions from the SOCHI in 1978. The primary secondary emission sources and esti-
mates of the corresponding total secondary emissions for 34 of the most signifi-
cant organic chemicals produced by the SOCHI are given in Table IV-1. The data
in Table IV-1 were extracted from the Emissions Control Options for the SOCHI
Product Reports (on file at EPA, ESED. Research Triangle Park. NC). The product
chemicals listed (same as in Table II-2) account for less than 10% of the total
number of chemicals produced but account for more than 70% of the total quantity
produced.
The estimates of secondary emissions listed in Table IV-1 are based on very
limited data. Secondary emissions are, in general, difficult to measure. The
estimates given were based primarily on the physical and chemical properties of
the primary organic waste components and on estimates of the quantities and the
composition of wastes. A review of the pertinent literature on secondary emis-
sions and estimating methods is given in Appendix A.
One of the major difficulties in sampling for secondary VOC emissions results
from the dispersion of secondary VOC emissions in the ambient air. A promising
technique for assessing the effect of dispersion on secondary VOC emissions has
been tried in two studies, one on fugitive emissions1 and the other on secondary
emissions.2 The technique involves sampling upwind and downwind of the emission
source at different horizontal and vertical points in the secondary-VOC-emission
plume. Dispersion mathematics can then be used to calculate the emission rate
of a hypothetical source that would be necessary to produce the plume.
Following is a brief discussion of emissions from the sources described in Sect. III.
The bulk of secondary emissions (50 to 90%) are estimated to result from the handling,
storage, pretreatment, and terminal treatment (primarily biological treatment) of
aqueous wastes. Other sources of major significance include landfilling, surface
impoundment, and incineration of liquid and solid wastes. Only the sources of major
significance are discussed in detail in this section; this is also true in Sects. V
("Applicable Control Methods") and VI ("Impact Analysis").
-------
Table IV-1. SOCMI Secondary. Emission Sources and Estimated Emissions
Product
Acetaldehyde
Acetone /phenol
Acetic anhydride
Acrylic acid and
esters
Total
Acrylonitrile
Estimated
1978
Production Type of
Primary Process (Gg/yr) Waste
Ethylene oxidation 450 Aqueous
Liquid
Cumene 900 Aqueous
Acetic acid 680 Aqueous
pyro lysis
Solid
Propylene oxidation 410 Aqueous
Liquid
Solid
Propylene oxidation 815 Aqueous
Liquid
Estimated Secondary
Emissions
Secondary Emission Source
Biotreatment or deep- we 11
Incineration or deep-well
Biotreatment
Biotreatment
i
Landfill or incineration
Biotreatment
Incineration vj
}
Incineration J
J
Holding pond, deep-well
Incineration
Ratio
(kg/Mg)a
b
b
1.6
b
b
0.05
0.001
0.051C
5.34
0.36
Rate
(Mg/yr)
b
b
1300
b
b
40C
JH
Total
5.71
4650
Adipic acid
Nitric acid
oxidation
855
Aqueous
akg of emission per Mg of product produced.
Information not available.
'Does not include emissions resulting from H2SO4 recovery.
Holding pond, deep-well
Neg.
Neg.
-------
Table IV-1. (Continued)
Product
Alkylbenzene
Total
Butadiene
Total
Caprolactam
Estimated
1978
Production Type of
Primary Process (Gg/yr) Waste
Olefins 250 Aqueous
Solid
Chlorination Aqueous
Liquid
Solid
n-Butane dehydro- 260 Aqueous
genation
Solid
Oxidative dehydro- 265 Aqueous
genation
Solid
By-product of ethy- 975 Aqueous
lene manufacture
Solid
Cyclohexanone 400 Aqueous
Liquid
Secondary Emission Source
Deep-well injection \
Landfill " j
Biotreatment x.
}
Incineration \
Landfill J
Biotreatment ^l"
\
Incineration J
Biotreatment^
I
)
incineration J
Biotreatment 'i
1
>
Incineration J
r
Biotreatment
Incineration
Estimated Secondary
Emissions
Ratio Rate
(kg/Mg)a (Mg/yr)
0.033 3.0
0.253 45.0
48
0.150 40
0.240 64
0.150 146
250
H
-------
Table IV-1. (Continued)
Product
Chlorobenzenes
Total
Chlorome thanes
Cumene
Total
Cyclohexanol/
eye iohexanone
Estimated
1978
Production
Primary Process (Gg/yr)
Benzene chlorina- 220
tion
Methane chlorina- 930
tion
Phosphoric acid 968
catalyst
Aluminum chloride 289
catalyst
Cyclohexane 1,012
Phenol hydrogenation 178
Estimated Secondary
Emissions
Type of
Waste
Aqueous
Solid
Aqueous
Liquid
Aqueous
Solid
Aqueous
Aqueous
Liquid
Aqueous
Liquid
Secondary Emission Source
Biotreatment
Landfill
Disposal I
Incineration j
Biological treatment
Landfill
Biological treatment
Biological treatment^
Incineration 1
Biological treatment
Not indicated
Ratio
(kg/Mg)a
0.28
Neg.
0.13
0.008
b
0.23
0.0911
Neg.
Neg.
Rate
(Mg/yr)
62
Neg.
62
120
o H
8 <
*.
b
67
—
75
92
Neg.
Total
92
-------
Table IV-1. (Continued)
Product
Ethanol amines
Ethylene
Total
Ethy Ibenzene /
styrene
Estimated
1978
Production Type of
Primary Process (Gg/yr) Waste
Ethylene oxide and 165 Aqueous
ammonia
Liquid
Ethane /propane 6,100 Aqueous
feed
Solid
Naphtha feed 3,000 Aqueous
Solid
Gas oil feed 3,000 Aqueous
Solid
Benzene alkylation 3,200 Aqueous
dehydrogenation
Liquid^1)
Solid J
Estimated Secondary
Emissions
Secondary Emission Source
Biotreatment
Landfill, incineration, sales
Phase separation and bio-
treatment
Landfill
Phase separation and bio-
treatment
Landfill
Phase separation and bio-
treatment
Landfill
Incineration
Ratio
(kg/Mg)a
b
b
0.02
Neg.
0.04
Neg.
0.07
Neg.
0.079
0.009
Rate
(Mg/yr)
b
b
122
b
122
b
214 <
-------
Table IV-1. (Continued)
Product
Ethylene dichlo-
ride/vinyl
chloride
Total
Ethylene glycol
Total
Ethylene oxide
Estimated
1978
Production Type of
Primary Process (Gg/yr) Waste Secondary Emission Source
Direct chlorina- 4,900 Aqueous Biotreatment
tion
Liquid Incineration
and /or
solid
Ethylene oxide 1,960 Aqueous Biotreatment
hydration
Liquid Incineration of landfill
Ethylene oxidation 1,460 Liquid Incineration
(air)
Solid Landfill
Ethylene oxidation 729 Liquid Incineration
Estimated Secondary
Emissions
Ratio
(kg/Mg) a
0.027
0.095
0.122
0.772
Neg.
0.0116
Neg.
0.013
Rate
(Mg/yr)
130
470
600
1513
H
1
b ^
1513
17
10
(oxygen)
Solid
Landfill
Neg.
27
-------
Table IV-1. (Continued)
Product
Formaldehyde
Total
Glycol ethers
Total
Maleic anhydride
Total
Methanol
TVH-al
Estimated
1978
Production
Primary Process (Gg/yr)
Metallic silver 2,250
catalyst
Metal oxide 750
catalyst
Reaction of 281
alcohols with
ethylene and /or
propylene oxide
Benzene oxidation 135
Low pressure 335
Estimated Secondary
Emissions
Type of
Waste
Aqueous
Aqueous
Aqueous
Liquid
Aqueous
Liquid
Solid
Aqueous
Liquid
Ratio
Secondary Emission Source (kg/Mg)a
Biotreatment 0.01
Biotreatment 0.05
Biotreatment 0.0019
Not specified 0.025
Handling and biotreatment^l
T . . f 0.05
Incineration )
Catalyst reclamation Neg.
Biotreatment 0.00044
Incineration 0.00006
0.0005
Rate
(Mg/yr)
23
38
61
0.5
H
1
7.0 ""*
7.5
6.8
Neg.
6.8
1.7
-------
Table IV-1. (Continued)
Product
Primary Process
Estimated
1978
Production
(Gg/yr)
Estimated Secondary
Emissions
Type of
Waste
Secondary Emission Source
Ratio
(kg/Mg)a
Rate
(Mg/yr)
Methyl methacrylate Acetone cyanohydrin 398
Nitrobenzene
Benzene nitration
Perchloroethylene/ Oxychlorination
trichloroethylene
Propylene oxide
Chlorohydronation
i-Butane peroxida-
~ tiond
Ethylbenzene per-
oxidation
Total
434
300
135
500
280
130
t-Butanol produced as co-product.
Aqueous
Liquid
Solid
Aqueous
Aqueous
Liquid
and/or
solid
Aqueous
Aqueous
Liquid
Aqueous
Liquid
Solid
Biological treatment
Sulfuric acid recovery
Incineration or landfill
Sulfuric acid recovery^
Wastewater treatment J
i
Biotreatment
Incineration
Biotreatment
Biotreatment
Incineration
Biotreatment
Incineration
Landfill
b
b
b
0.33
b
2.6
2.6
b
b
b
143
b
2100
2100
03
-------
Table IV-1. (Continued)
Product
Primary Process
Estimated
1978
Production
(Gg/yr)
Estimated Secondary
Emissions
Type of
Waste
Secondary Emission Source
Ratio
(kg/Mg)a
Rate
(Mg/yr)
Terephthalic acid
(crude)
Total
Air oxidation
1,989
Aqueous
Liquid
Biotreatment
Incineration
0.004
0.006
8
20
1,1,1-Trichloro-
ethane
Vinyl chloride
310
Ethane
18
Aqueous
Liquid/
solid
Aqueous
Liquid/
solid
Biotreatment
Incineration
Biotreatment
Incineration
0.001
<0.001
0.001^
<0.001J
0.62
0.04
H
I
Total
0.66
-------
IV-10
B. PHYSICAL SEPARATION METHODS
1. Steam Stripping
a. Aqueous Wastes Steam stripping is frequently used to control secondary emissions
from subsequent treatment operations. Secondary emissions primarily result from
venting noncondensed gas from the stripper condenser (primarily air) and are
generally relatively low.
b. Liquid Organic Wastes Steam stripping is not generally used for the treatment
of liquid organic wastes.
c. Solid Wastes As with aqueous wastes, stream stripping, or purging, of solids
is frequently used to reduce secondary emissions from subsequent treatment or
disposal operations. Emissions resulting from the venting of noncondensable
gases (air) are relatively low.
2. Conventional Distillation
a. Aqueous Wastes Conventional distillation is not generally used for the treat-
ment of aqueous wastes unless the concentration of organics is significant. As
with steam stripping, emissions chiefly result from the venting of noncondensable
gases. Secondary emissions are relatively minor.
b. Liquid Wastes Although conventional distillation of organic liquids is a common
process operation, it is infrequently used for the treatment of wastes. Secondary
emissions are minor and result from the venting of noncondensable gases.
c. Solid Wastes Not applicable.
3. Liquid-Liquid Phase Separation/Emulsion Breaking
a. Aqueous Wastes The most significant emissions occur when open-topped and atmos-
pherically vented separation vessels (e.g., API oil/water separators, plate
separators) are used, when the organic phase is the upper layer, and when the
vapor pressures of the organic compounds are significant.
-------
IV-11
b- Liquid Wastes The comments concerning emissions from aqueous wastes also apply
to liquid wastes.
c- Solid Wastes Not applicable.
4- Solid-Liquid Separation
l- Aqueous Wastes The solid-liquid separation operations used in the treatment
of aqueous wastes are usually integral steps in biological wastewater treatment
processes (i.e., sedimentation, clarification, thickening) and are significant
sources of secondary emissions. A primary clarifier is included in a model
wastewater treatment plant using the continuous activated sludge system (CAS)
described in Sect. IV-E-l-b. Sample calculations for estimating emissions of
VOC from soluble aqueous wastes in open-topped vessels are given in Appendix B.
This type of calculation is applicable to the estimation of secondary emissions
from all aqueous waste treatment and storage operations occurring in open-topped
vessels in which the primary emission mechanism is the vaporization of soluble
organics (i.e., solid-liquid separation, aeration, neutralization, impoundment).
k• Liquid Wastes Secondary emissions resulting from the separation of solids
from liquid organic wastes are relatively minor.
c- Solid Wastes Discussed above.
C- CHEMICAL TREATMENT
1• Neutralization
a- Aqueous Wastes The neutralization/equalization of aqueous wastes, frequently
a pretreatment step to biological wastewater treatment and usually occurring in
open-topped basins or vessels, may result in significant secondary emissions.
Secondary emissions, which result chiefly from the vaporization of contained
organics, may be estimated by the methods shown in the sample calculations
(Appendix B).
-------
IV-12
b. Liquid Wastes Not generally applicable.
c. Solid Wastes Not generally applicable.
2. Precipitation and Coagulation
a. Aqueous Wastes Significant secondary emissions may result from precipitation/
coagulation operations conducted in open-topped vessels. The emissions, which
result chiefly from the vaporization of contained organics, may be estimated by
the methods shown in the sample calculations (Appendix B).
b. Liquid Wastes Not generally applicable.
c. Solid Wastes Secondary emissions from solid wastes generated by precipitation/
coagulation operations occur in solid-waste terminal treatment or disposal opera-
tions (e.g., incineration, landfilling) (see Sects. IV-D and F).
3. Chemical Oxidation
a. Aqueous Wastes Chemical oxidation processes are used relatively infrequently
and generally take place in enclosed or pressurized vessels. Secondary emis-
sions are relatively insignificant.
b. Liquid Wastes Not generally applicable.
c. Solid Wastes Not generally applicable.
D. THERMAL DESTRUCTION
1. Aqueous Wastes
As thermal destruction or incineration is not generally used for the disposal
of aqueous wastes, secondary emissions are relatively insignificant.
2. Liquid Wastes
Combustion in incinerators or boiler combustion chambers is widely used in the
SOCMI for the terminal disposal of large quantities of liquid organic wastes.
Combustion efficiencies in excess of 99% can as a rule be readily attained if
-------
IV-13
the recommended residence time and combustion zone temperature requirements are
met. Secondary emissions resulting from incomplete oxidation of organics are
generally relatively low compared to emissions from alternative terminal treat-
ment methods. Although the concentration of VOC in the vented flue gas is usually
very low, the quantity of contained organics vented may be significant because
of the extremely large quantities of flue gas vented. Estimates of secondary
emissions resulting from the combustion of liquid organic wastes were generally
based on an assumed VOC removal efficiency of 99%.
3. Solid Wastes
Combustion requirements necessary for the satisfactory thermal destruction of
solid wastes are generally more stringent than those for liquid wastes. Incin-
erators designed for specific waste profiles are generally required; however,
at suitable conditions, destruction efficiencies for organics in solid wastes
are also usually 99% or better.
E- BIOLOGICAL TREATMENT
1. Aqueous Wastes
a. Introduction Emissions resulting from the handling and treatment of aqueous
wastes or wastewater probably represent 50 to 90% of the uncontrolled secondary
emissions from the SOCMI, b?sed on the estimates described in the following
sections. The bulk of the wastewater generated by the industry is treated bio-
logically, with the resulting secondary emissions occurring during the separate
operations involved in biological treatment and any necessary pretreatment.
Some of the separate unit operations that are normally included in biological
treatment and pretreatment (e.g., liquid/solid separation, neutralization) are
described in Sect. II.
b. Model Plants A theoretical order-of-magnitude estimate of total SOCMI uncon-
trolled secondary VOC emissions from wastewater was made based on 30 high-volume
chemicals with available wastewater data3—7 and on the wastewater secondary VOC
emission estimation models of Mackay8 and Thibodeaux9 applied to the wastewater
data and the model wastewater treatment plant shown in Fig. IV-1 and detailed
in Table IV-2. Actual wastewater treatment plants are designed to meet indi-
vidual requirements and may differ significantly from the model.
-------
IV-14
t
Secondary Emissions
Model-Plant
Wastewater
Clarifier
Conventional
Activated -
Sludge (CAS)
Wastewater
Effluent
Fig. IV-1. Uncontrolled Secondary VOC Emissions from Model-Plant Wastewater
(Total, 1060 Mg of VOC Emissions Per Year)
-------
IV-15
Table IV-2. Parameters for Model-Plant Wastewater Treatment System
Primary clarifier
Depth (m) 3
2
Surface area (m ) 700
Retention time (hr) 3
Activated-sludge system (CAS)
Depth (m) 6
2
Surface area (m ) 6300
Air flow (m3/min) 612
-------
IV-16
The model wastewater treatment system used in the analysis consists of a clari-
fier and activated sludge system sized according to industrial practices for
the wastewater flow from a model chemical production plant (MCPP). The opera-
tional parameters for the MCPP, given in Table IV-3, are based on production-
weighted averages for the 30 chemicals with available wastewater data.3—7
c. Emissions—An estimate of total uncontrolled secondary wastewater emissions
resulting from the production of 30 high-volume chemicals is summarized in
Table IV-4. The average uncontrolled secondary VOC wastewater emission factor
for the 30 chemicals is estimated as 650 g of VOC per Mg of product. Applying
this estimated secondary emission factor to the 1978 estimated industry produc-
tion of 100,000 Gg results in an estimate of 65 Gg/yr for uncontrolled secondary
emissions from SOCMI wastewater. Because the 30 chemicals represent 55 wt % of
1978 SOCMI production, this is considered to be reasonable order-of-magnitude
estimate for the industry wastewater emissions.
It should be emphasized that, although the estimate of total emissions given in
Table IV-4 is probably of reasonable accuracy, the estimates of secondary emis-
sion rates shown for the individual products are each based on a single organic
component in the wastewater (the primary product) and may be significantly in
error. The estimates of secondary emissions from the product reports (summarized
in Table IV-1) cannot be compared with those from Table IV-4 as the tables are
based on the following differences in criteria: (1) Table IV-1 is for total
secondary emissions; Table IV-4 is for emissions from wastewater only; (2) Table IV
is for uncontrolled emissions; Table IV-4 estimates the actual current emissions.
A discussion of the estimating procedures used is presented in Appendix B. The
criteria on which the estimates were based, sample calculations, and summaries
of the calculations are given in Appendix C.
2. Liquid Organic Wastes Biological treatment is not generally used for liquid
wastes containing more than 1% organic compounds.
3. Solid Wastes—Not generally applicable.
-------
IV-17
Table IV-3. Operational Parameters for Model Chemical Production Plant
Production (Gg/yr) 1,650
Wastewater flow (m3/day) 17,000
Wastewater COD content (ing/liter) 3,000
Wastewater chemical content (mg/liter) . 1,360
Amount of chemical in wastewater leaving plant (Gg/yr) 8.44
Henry"s-law constant, M ., for chemical in wastewater 25
Molecular weight of chemical 58
-------
IV-18
Table IV-4. Estimated Order-of-Magnitude Uncontrolled-Secondary-Emission Wastewater
Factors for 30 Organic Chemical Products
Approximate 1978 Secondary VOC
Production Secondary VOC Emission Rate
Chemical (Gg/yr) Emission Factor (SEf ) * (Gg/yr)
Ethylene
Ethylene dichloride
Styrene
Methanol
Ethylbenzene
Vinyl chloride
Ethylene oxide
Ethylene glycol
1 , 3-Butadiene
Dime thy 1 ter eph tha 1 a t e
Cumene
Acetic acid-
Cyclohexane
Phenol
Chlorinated methanes
Isopropy,! alcohol
Acetone
Propylene oxide
Acetic anhydride
Acrylonitrile
Ethanol
Vinyl acetate
Terephthalic acid (crude)
Adipic acid
t-Butanol
Acetaldehyde
Cyclohexanol/cyclohexanone
Phthalic anhydride
Adiponitrile
Caprolactam
Total
11,800
4,900
3,200
2,900
3,700
3,100
2,200
1,900
1,500
1,300
1,500
1,200
1,100
1,200
900
800
900
900
700
800
600
800
2,000
800
600
500 ;
1,000
1,000
400 -
400
54,600
67
2
16
876
1143
0
959
3
0
0
0
0
2246
31
1575
2623
260
7283
201
10
0
0
2868
910
6
0
0
0
2.6
12.2
0.2
0
0
2.7
2.5
0
1.4
0
0
0
0
0
2.0
0
1.4
2.4
0.2
5.8
0.1
0
0
0
1.7
0.5
0
0
0
0
35.7
*g of secondary VOC emissions per Mg of product produced.
-------
IV-19
F. " TERMINAL STORAGE
1. Landfill
a- Aqueous Waste Not generally applicable.
b' Liquid Wastes Significant secondary emissions are believed to result from the
landfilling of liquid organic wastes, primarily from those with significant
vapor pressures that are not containerized or stabilized before landfilling.
Actual emission data are extremely scarce.
c- Solid Wastes The comments concerning liquid wastes (Sect. IV.F.l.b) also apply
to solid wastes.
**• Comparison of Estimated Secondary VOC Emissions from a Wastewater Treatment
System and a Chemical Landfill Because of the scarcity of data for secondary
emissions resulting from the landfilling of liquid and solid wastes, a theoreti-
cal sensitivity analysis was made to permit an order-of-magnitude comparison
between the potential emissions from wastewater treatment and those from landfilling
operations. The sensitivity analysis was performed for hexachlorobenzene (HCB),
£-dichlorobenzene (DCB), benzene (Bz), acetone, and acetic acid. These model
chemicals were chosen because they represent a wide range of volatilities and
physical properties. The secondary VOC emission estimation was performed by
using the Mackay,8 Thibodeaux,9 and Farmer10 estimation models (see Appendix D).
The operational parameters for the model landfill are given in Table IV-5.
The organically contaminated wastewater, liquid organic wastes, and solid organic
wastes generated are treated as single-component wastes with average physical and
chemical properties. The results of the analysis are summarized in Table IV-6,
and the bases and calculations for the analysis are given in Appendix C.
The sensitivity analysis for wastewater treatment and landfill summarized in
Table IV-6 suggests that for low-water-soluble, low-vapor-pressure chemicals
most of the secondary VOC emissions of the chemical will be from wastewater
treatment. The analysis also suggests that the secondary VOC emissions from
landfill compared with wastewater emissions increase for low-water-soluble,
high-vapor-pressure chemicals such as benzene. Comparison of the secondary VOC
-------
IV-20
Table IV-5. Parameters for Model-Plant Landfill
Solid waste (Gg/yr) 4-2
Waste bulk density (g/cm ) 1-0
fj
Landfill area (m per year) 4047
Landfill soil moisture Dry
Depth of soil cover (m) 0-6
-------
IV-21
Table IV-6. Comparison of Estimated Secondary VOC
Emissions from a Wastewater Treatment Plant with Those from
a Chemical Landfill at a Model Chemical Production Plant
Secondary VOC Emissions (Mg/yr)
— Source
Landfiua
Model wastewater treatment system
Clarifier
Activated-sludge system
Easy to biodegrade
Difficult to biodegrade
Wastewater totals
HCB
3.2 X 10~4
2.4 X 10~3
3.6 X 10"2
3.8 X 10"2
DCB
12
46.1
42.4b
447b
88 — 493
Benzene
840
1488
2020
3508
Acetone
1675
186
12
198
Acetic Acid
88
8.4
0.4
8.8
of wastewater emission
rate to sanitary landfill
emission rate
119
7—41
4.2
0.1
Based on 4047 m2 per year of sanitary landfill used.
DCB
0.1
was estimated both ways to show the relative importance of biodegradability on the
secondary VOC emission rates from an activated-sludge system.
-------
IV-2 2
emissions from liquid chemicals that have higher vapor pressures and are more
water stable, such as acetone and acetic acid, suggests that more of the secondary
VOC emissions for these chemicals will be from landfill than from wastewater.
The analysis described in Appendix C for landfill is a worst-case estimate based
on dry soil. In actual landfill situations in which there are moist soil, biologi-
cal decomposition, and some downward percolation of water, the secondary VOC
emissions for highly water soluble chemicals such as acetone and acetic acid
will be lower than the values shown.
2. Surface Impoundment
a. Aqueous Wastes A preliminary sensitivity analysis, similar to that shown in
Table IV-6, was done for the secondary VOC emissions from landfill and surface
impoundments. The analysis suggested that the two most important physical property
indicators, relative to secondary VOC emission potential, are the vapor pressure
and the aqueous solubility of the waste chemicals. The analysis also suggested
that secondary VOC emissions from surface impoundments are potentially significant
relative to landfill secondary VOC emissions.
The calculational models used for the sensitivity analysis were particularly
sensitive to the average SOCMI vapor pressures assumed for the wastes being put
into landfill and for the aqueous solubilities assumed for the wastes being put
into surface impoundments. Since no data were available on the average vapor
pressure and average aqueous solubilities of SOCMI wastes going to landfill and
to surface impoundments, no meaningful estimate of secondary VOC emissions from
these sources could be made.
b. Liquid Wastes The storage of liquid organic wastes that have significant vapor
pressures in open basins is not an acceptable practice.
c. Solid Wastes No information available.
-------
IV-2 3
G. DISCHARGE TO NATURAL WATERS
1. Dilution of Aqueous Effluents
a. Aqueous Wastes Generally, if pretreatment is acceptable from the standpoint
of the effect of the wastes on water quality, secondary emissions are minimal.
b. Liquid Wastes The discharge of liquid organic wastes into natural receiving
waters is never an acceptable practice.
c. Solid Wastes The discharge of solid wastes (with the exception of deep-ocean
dumping of containerized wastes) is not an acceptable practice.
2. Ocean Dumping
a. Aqueous Wastes Not generally applicable.
b. Liquid Waste The ocean dumping of bulk liquid organic wastes in not an accept-
able practice. The ocean dumping of containerized or stabilized liquid wastes
probably results in minimal secondary emissions.
c. Solid Wastes Secondary emissions are minimal.
-------
IV-2 4
H. REFERENCES*
1. P. R. Harrison, Meteorology Research, Inc., Characterization of Fugitive
Emissions from the Exxon Benica, California, Process Facility, MR 77 FR-1522
(May 30, 1978).
2. T. W. Hughes and D. A. Horn, Source Assessment; Acrylonitrile Manufacture (Air
Emissions), EPA-600/2-77-107J, pp. 37, 38, 100, 101, 105 (September 1977).
3. Development Document for Effluent Limitations Guidelines and New Source Per-
formance Standards for the Major Organic Products Segment of the Organic
Chemicals Manufacturing Point Source Category, EPA 440/1-74-009-a (April 1974).
4. Development Document for Interim Final Effluent Limitations Guidelines and New
Source Performance Standards for the Major Organic Products Segment of the Organic
Chemicals Manufacturing Point Source Category, EPA 440/1-75-045, Group 1, Phase II
(November 1975).
5. Monsanto Research Corp. and Research Triangle Institute, Chapter 6. The Industrial
Organic Chemicals Industry, Part I.
6. W. D. Bruce, IT Enviroscience, Inc., Trip Report for Visit to Monsanto Textiles
Co., Pensacola, FL, Feb. 8, 1978 (on file at EPA, ESED, Research Triangle Park,
NC).
7. Synthetic Organic Chemicals, 1976 United States Production and Sales, United
States International Trade Commission, GPO, Washington, 1977.
8. D. Mackay and P. J. Leinonen, "Rate of Evaporation of Low-Solubility Contami-
nants from Water Bodies to Atmosphere," Environmental Science and Technology
9(13), 1178—1180 (December 1975).
9. L. J. Thibodeaux, "Air Stripping of Organics from Wastewater. A Compendium,"
pp. 358—378 in Proceedigs of the Second National Conference on Complete
Watereuse. Water's Interface with Energy, Air and Solids, Chicago, IL,
May 4—8, 1975, sponsored by AIChE and EPA Technology Transfer.
10. W. J. Farmer et a!., "Land Disposal of Hexachlorobenzene Wastes: Controlling
Vapor Movement in Soil," pp. 182—190 in Land Disposal of Hazardous Wastes.
Proceedings of the Fourth Annual Research Symposium, edited by D. W. Schultz,
EPA-600/9-78-016 (August 1978).
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
V-l
V. APPLICABLE CONTROL METHODS
A. INTRODUCTION
The primary categories of methods by which secondary emissions can be con-
trolled include (1) waste source control the elimination or reduction of
aqueous, liquid, or solid wastes or the reduction in the concentration of VOC
in the wastes through changes in the waste generating processes; (2) resource
recovery the reduction in the quantity or VOC concentration of wastes requiring
terminal treatment or disposal through the addition of recovery operations between
the waste generating processes and terminal treatment; the recovered organics may
be recycled, sold as by-products, or burned (providing fuel value); (3) alternate
disposal the reduction of secondary emissions by the use of alternate terminal
treatment or disposal methods, which generate lower secondary emissions; and
(4) add-on controls the reduction of secondary emissions released to the
atmosphere by the addition of controls to waste treatment and disposal operations.
Table V-l is a summary of specific control methods within these categories and
of the secondary emission sources where they are applicable; however, only the
more prevalent combinations of control methods and significant emission sources
are discussed in the text. As with the emission sources, the general categories
within which specific control methods fall are not always clear-cut, and they
tend to overlap to some extent. It may also be noted that operations which are
considered as secondary emission sources may also occur as emission control
methods. Following is a discussion of the specific control methods. The esti-
mated ranges of emission reductions provided by the most prevalent control methods
are given in Table V-2.
B. WASTE SOURCE CONTROL
1. Introduction
Secondary emissions can frequently be reduced through changes in the waste sources
or waste generating processes. These changes can be effected by reducing the
quantity of wastes generated, by reducing the concentration of VOC in the wastes,
or by altering the waste characteristics to make the wastes amenable to alterna-
tive disposal methods that emit less secondary emissions. Because a reduction
-------
V-2
Table V-l. Applicable Control Methods
Waste
Secondary Emission
Source Operation
Physical separation
Distillation/steam stripping
Liquid/liquid phase separation
Solid/liquid separation
Chemical treatment
Neutralization
Precipitation /coagulation
Chemical oxidation
Thermal destruction
Incineration
Biological treatment
Activated sludge reaction
Clari £ ication
Thickening/dewatering
Terminal storage
Landfill
Surface impoundment
Deep-well disposal
Discharge to natural waters
Ocean dumping
ative Waste Source Processes!
c
-j.
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
ative Raw Materials
c
M
0)
4J
r-l
X
X
X
X
x
X
X
x
x
x
x
x
x
x
X
tfl
c
o
4J
"S
a,
•a
V
o
*4
1
X
x
X
X
X
X
X
X
X
X
X
X
X
X
X
Source Control
u
1
rH
ID
C
W
S
H
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
limination
M
10
X
X
X
X
X
X
X
X
X
X
X
X
X
X
d Water/Steam Use
e;
1
X
X
X
X
X
X
X
X
X
X
X
X
X
t Water/Organic Contact
c
Preve
X
X
X
X
X
X
X
X
X
X
X
X
X
ed Process Control 1
;»
a
M
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Resource
Stripping
tional Distillation
c
§ u
V ti
•u O
I/I (J
X
X
X X
X X
x x
X
X
X
X
X X
x x
X X
c
0
Di
o
W
s
X
X
X
X
X
X
x
x
X
X
Recovery
§
o
ID
M
X
X
X
X
X
X
X
X
X
X
an Breaking
ne Separations
•A > l4
-------
V-3
Table V-2. Secondary VOC Emission Reductions by
Various Control Techniques or Methods *
Reduction Range
Control Option (%)
Organically contaminated wastewater
Source control 10—90
Resource recovery 10—90
Immiscible organic cover 50—90
Floating plastic spheres 70—90
Floating plastic cover >99
Collection/carbon adsorption 95—>gg •
Collection/thermal destruction >99
Organic liquid and solid
Landfill
Soil cover 75—99+
Plastic cover >99
Surface impoundment
Immiscible organic cover 50—90
Floating plastic spheres 70—90
Floating plastic cover >99
*Based on the limited data available plus engineering experience and
judgement.
-------
V-4
in the quantity of waste or VOC concentration almost always results in a cor-
responding reduction in VOC emissions, waste source control is effective in
reducing emissions from almost all secondary sources.
2. Alternative Waste Source Processes or Process Steps
a. Aqueous Wastes Secondary emissions from alternative processes or process steps
by which the same product is made are often significantly different. Although
the use of an alternative process is often not feasible as a secondary emission
control method in an existing plant, potential secondary emissions should be
considered in process selection for a new facility.
b. Liquid Wastes See Sect. V-B-2-a.
c. Solid Wastes See Sect. V-B-2-a.
3. Alternative Raw Materials
a. Aqueous Wastes Frequently the use of alternative raw materials or raw materials
with lower concentrations of impurities or diluents will result in reduced quantities
of wastes generated or reduced VOC concentration.
b. Liquid Wastes See Sect. V-B-3-a.
c. Solid Wastes See Sect. V-B-3-a.
4. Improved Separations
a. Aqueous Wastes The concentration of VOC in aqueous wastes is frequently dependent
on the efficiency of physical separation steps (e.g., distillation, phase sepa-
ration) within the source processes. Improvement in separation efficiency can
result in significant reduction in the VOC concentration of the wastes and a
corresponding reduction in secondary emissions.
-------
V-5
b- Liquid Wastes Generally not significant.
c- Solid Wastes Generally not significant.
5- Internal Recycle
a- Aqueous Wastes The quantity of wastes discharged can frequently be reduced
with modifications to source processes that permit the internal recycle of waste
streams with a corresponding reduction in makeup requirements. The treatment
of aqueous wastes necessary to permit recycle to the process may be less demanding
than the treatment required for acceptable discharge and may result in reduced
secondary emissions.
b- Liquid Wastes^—See Sect. V-B-5-a.
c- Solid Wastes See Sect. V-B-5-a.
6- Leak Reduction
a- Aqueous Wastes The reduction of small and intermittent leaks of organics from
process equipment, storage tanks, and piping can often significantly reduce
secondary emissions, as well as fugitive emissions.
Leaks in heat exchanger tubes often contribute to increased aqueous wastes and/or
increased VOC concentration. The introduction of water into process streams
through leaks in heat exchanger tubes eventually results in increased aqueous
wastes. The leakage of organics into cooling water results in increased VOC
concentration in wastewater or in increased fugitive emissions from cooling
towers. Frequent systematic inspection and maintenance and effective equipment
specifications are usually the most effective means of minimizing leaks. Moni-
toring organic levels in process cooling wastewater can be used to indicate
the presence of process leaks.
b- Liquid Wastes See Sect. V-B-6-a.
c' Solid Wastes—Not generally applicable.
-------
V-6
7. Reduction of Water/Steam Usage
a. Aqueous Wastes Most of the water or steam directly introduced into source
processes is eventually discharged as aqueous wastes. Usage can frequently be
reduced with the systematic examination of requirements and actual usage.
b. LiquidWastes Not applicable.
c. Solid Wastes Not applicable.
8. Prevention of Water/Organic Contact
a. Aqueous Wastes The quantity of aqueous wastes and/or the concentration of VOC
can often be reduced by eliminating operations that require the direct contact
of organic process streams with water that is eventually discharged as wastewater.
This is most frequently accomplished by the use of heat exchangers instead of
direct water contact operations.
b. Liquid Wastes Not applicable.
c. Solid Wastes Not applicable.
9. Improved Process Control
a. Aqueous Wastes The reduction in process upsets and improved control of process
parameters can result in significant reductions in process wastes and the corres-
ponding secondary emissions. Improvements are generally attained with better
process control equipment, improvement in control application, and improved
training of operating personnel.
b. Liquid Wastes See Sect. V-B-9-a.
c. Solid Wastes See Sect. V-B-9-a.
-------
V-7
C. RESOURCE RECOVERY
1. Introduction
The recovery of organic raw materials, products, or by-products from waste streams
prior to terminal treatment is often an effective method of reducing VOC concentra-
tion and the corresponding secondary emissions. Most resource recovery operations
utilize physical separation methods, as described in Sects. II and III. Some
recovery operations are relatively complex and may require reactions, as well
as separation steps. Resource recovery may not be economically feasible for
streams containing very low VOC concentrations or for streams containing a
number of organic components that are not usable as mixtures and that cannot be
readily separated.
When applicable, the primary advantage of resource recovery as a method for
controlling emissions is that the value of the recovered materials will usually
at least partially offset the cost of control. Frequently a net profit can be
realized. In addition to the recovery of products, raw-material and salable
by-product recovery operations may be used to concentrate organic wastes pre-
sent in aqueous wastes, permitting the organic wastes to be burned as fuel with
positive heating value. Recovery operations are primarily used for treating
aqueous wastes, and the resulting reductions in secondary emissions occur
largely in biological treatment and pretreatment.
2. Steam Stripping
a. Aqueous Wastes Steam stripping is used primarily to recover VOC from aqueous
wastes.
b. Liquid Wastes Not generally applicable.
c. Solid Wastes The steam stripping of solids, although used in reducing VOC
content, is seldom used for recovery.
3. Conventional Distillation
a. Aqueous Wastes—HNot generally applicable.
-------
V-8
b. Liguid Wastes Conventional distillation as a recovery method is chiefly used
for treating liquid organic wastes for separating usable products, by-products,
or raw materials.
c. Solid Wastes Not generally applicable.
4. Adsorption
a. Aqueous Wastes Adsorption, usually with activated carbon, can effectively
remove organic compounds even when present at very low concentrations from aqueous
wastes. As discharge requirements for wastewater become increasingly stringent,
activated carbon adsorption is being used more widely both for resource recovery
and for final terminal treatment.
b. Liquid_Was_tes Not generally applicable.
c. Solid Wastes Not generally applicable.
5. Solvent Extraction
a. Aqueous Wastes Extraction is not presently widely used as a recovery process;
however, with the development of more efficient extraction solvents (i.e., higher
extraction coefficients, lower water solubility) combined with the higher cost
of organic chemicals and more stringent wastewater discharge requirements, the
use of extraction for the recovery of organics even when present in very low
concentrations in aqueous wastes is receiving wider attention.
b. Liquid Wastes Not generally applicable.
c. Solid Wastes Not generally applicable.
6. Emulsion Breaking
a. Aqueous Wastes Significant concentrations of VOC in aqueous wastes are fre-
quently caused by the formation of waterrorganic emulsions, which makes phase
separation difficult. Specific emulsion breaking or separating methods such as
-------
V-9
centrifugation, flotation, filtration/coalescence separation, and chemical treat-
ment can frequently be used for the recovery of organics occurring as emulsions,
resulting in lower VOC concentrations entering terminal treatment and corres-
pondingly lower secondary emissions.
b- Liquid Wastes See Sect. V-C-6-a.
c- Solid Wastes Not applicable.
Membrane Separations
a- Aqueous Wastes Although not presently widely used, membrane separation techniques,
such as reverse osmosis, ultrafiltration, and electrodialysis, are receiving
considerable attention as potential methods for the recovery of chemicals primarily
from aqueous wastes containing high concentrations of dissolved inorganic com-
pounds (e.g., brines). The chief advantage of these techniques is their ability
to treat wastes that are not normally amenable to treatment by the more common
methods.
• Liquid Wastes—Not generally applicable.
c- Solid Wastes Not applicable.
D- ALTERNATIVE TERMINAL TREATMENT
1• Introduction
Secondary emissions can frequently be reduced by using alternative treatment or
disposal methods. The bulk of current secondary emissions result from the vapori-
zation of VOC from aqueous wastes, occurring primarily during biological treatment
and pretreatment operations conducted in open-topped vessels or basins. Because
most processes in the SOCMI discharge extremely large quantities of wastewater,
the terminal treatment and disposal methods discussed below are seldom practical
as alternatives to biological treatment for the total wastewater effluent.
However, specific aqueous wastes that contain relatively high concentrations of
VOC or that are not readily amenable to biological treatment frequently can
be separated and treated by other methods to reduce secondary emissions.
-------
V-10
Table V-3. Current Applications of Incineration to
the Disposal of Liquid and Solid Wastes
Product/Process
Company
Location
Type of Waste
Acetic acid
Acetic anhydride
Acrolein
Acrylic acid/esters
Adipic acid
Aniline
Butadiene
Caprolactarn
CC/4/perchloroethylene
(clorinolysis)
Chloroprene
Borden
Monsanto
Celanese
Union Carbide
Tennessee Eastman
Celanese
Union Carbide
Union Carbide
Rohm and Haas
Celanese Chemicals Co.
Du Pont
Du Pont
Petro-Tex
Dow Badische
Dow Chemical
Vulcan
Vulcan
Denka
Geismer, LA
Texas City, TX
Clear Lake, TX
Brownsville, TX
Kingsport, TN
Pampa, TX
Taft, LA
Taft, LA
Deer Park, TX
Clear Lake, TX
Orange, TX
Beaumont, TX
Houston, TX
Freeport, TX
Plaquemine, LA
Geismar, LA
Wichita, KA
Houston, TX
Distillation
heavy ends
Distillation
heavy ends
Distillation
bottoms
Distillation
light and
heavy ends
Distillation
light ends
Still residue
(solid)
Distillation
light ends
Polymeric
residues
Heavy ends
Organic liquids;
polymeric
residues
Waste oil stream
Aniline tars
Furfural purifi-
cation residue
Extraction-tower
bottoms;
benzene extrac-
tion raffinate
Waste chloro-
carbons; solids
wastes
Distillation
heavy ends
Distillation
heavy ends
Dehydrochlori-
nation residue
(incinerated
off-site)
-------
V-ll
Table V-3. Cont'd
•.-..Product/Process
Company
Location
Type of Waste
Chloroprene
Cumene
cyclohexanol/
cyclohexanone
Ethyl acetate
Ethylene dichloride
examethylenediamine
alkylbenzene
anhydride
Meth
anol
Methyi ethyl ketone
Methyi methacrylate
erchloroethylene
Du Pont
Monsanto
Union Carbide
Dow Badische
Tennessee Eastman
PPG Industries
Dow Chemical
Conoco
Borden
Shell
Du Pont
Monsanto
Union Carbide
Amoco
Rohm and Haas
Celanese
Air Products
Tenneco
Arco Chemical Co.
Rohm and Haas
Du Pont
Dow Chemical
La Place, LA
Alvin, TX
Taft, LA
Freeport, TX
Kingsport, TN
Lake Charles, LA
Oyster Creek, TX
Westlake, LA
Geismar , LA
Deer Park, TX
Victoria, TX
Decatur, AL
Institute, WV
Chicago, IL
Deer Park, TX
Bishop, TN
Pensacola, FL
Houston, TX
Channelview, TX
Deer Park, TX
Memphis, TN
Freeport, TX
Organic layer
from waste-
water,- solid
polymeric waste
Distillation
bottoms
Light and
heavy ends
Light and
heavy ends
Reactor sludge
Chlorinated
residues
Chlorinated
residues
Distillation
heavy ends
Chlorinated
residues
Distillation
light and
heavy ends
Clorinated
organics
Light and
heavy ends
Residual tars
Distillation
bottoms
Purification
waste liquid
Distillation
light ends
Distillation
Extraction and
refining
Organic polymers
Light ends
Spent acid (10%
VOC); polymeric
residue
Chlorinated tar
-------
V-12
Table V-3. Cont'd
Product/Process
Company
Location
Type of Wast
Perchloroethylene
Perchloroethylene/
trichloroethylene
Phenol acetone
1,1,1-Trichloroethane
Vinyl acetate
Diamond Shamrock
PPG Industries
Georgia Pacific
Vulcan
Celanese
Union Carbide
Celanese
U.S.I. Chemicals
Du Pont
Deer Park, TX
Lake Charles, LA
Plaquemine, LA
Geismar, LA
Bay City, TN
Texas City, TX
Clear Lake, TX
Deer Park, TX
La Porte, TX
Chlorinated
organics and
tars
Distillation
residues
Light and
heavy oils
Chlorinated
residues
Heavy ends from
acetic acid
purification
Polymeric
residues
Heavy ends
Acid purifica-
tion; heavy
ends
Distillation
light ends;
tars
-------
V-13
2. Incineration
a- General Secondary emissions resulting from the incineration of wastes are
usually relatively low compared to alternative terminal disposal methods, and
incineration is frequently the primary alternative disposal method to be con-
sidered as a means of reducing secondary emissions. Incineration is used for
the disposal of liquid and solid wastes containing relatively high concentra-
tions of organic compounds. Table V-3 presents some specific examples in which
incineration is used for the terminal disposal of liquid and solid wastes.
b- Aqueous Wastes Because of the high fuel usage necessary, incineration is usually
restricted to relatively small quantities of aqueous wastes (see Sect. V-D-2-a).
c- Liquid Wastes Although incineration is generally a versatile method, wastes
containing significant concentrations of certain inorganic compounds (e.g.,
heavy metals, halogens, low-melting salts, phosphorous) may not be readily amenable
to incineration.
d- Solid Wastes See Sect. V-D-2-c.
3. Deep-Well Disposal
a- Aqueous Wastes Secondary emissions resulting from deep-well disposal are gen-
erally low. Significant emissions may result from the storage of wastes (e.g.,
open basins, lagoons) prior to deep-well injection. Because the availability
and capacity of deep-well storage are generally limited, its use is usually
restricted to aqueous wastes that are not amenable to biological treatment
(e.g., brines).
b- Liquid Wastes Deep-well disposal is usually restricted to liquid wastes that
are not amenable to disposal by incineration.
c- Solid Wastes Not generally applicable except for solids in solution.
-------
V-14
4. Landfill
*
a. Aqueous Wastes Landfilling is not generally used for the disposal of large
quantities of aqueous wastes. Landfilling may be used for the disposal of rela-
tively small quantities of aqueous wastes that are not amenable to biological
treatment.
b. Liquid Wastes Since landfilling can result in significant secondary emissions,
it is not generally considered as a likely alternative for secondary emission
reduction; however, if the wastes are stabilized before being landfilled (see
Sect. V-E-5), secondary emissions are generally minimal.
c. Solid Wastes The disposal alternatives for solid wastes are generally limited
to either landfilling or incineration. If the solid wastes contain significant
concentrations of VOC, the secondary emissions are usually less from incineration
than from landfilling.
5. Advanced Biological Treatment Methods
a. Aqueous Wastes Secondary emissions from some of the more recently developed
biological treatment processes (e.g., enriched oxygen, fluidized bed) are probably
relatively low compared with those from conventional biological treatment processes.
Although no secondary emission data for these processes are available, the general
use of enclosed reaction vessels probably results in lower secondary emissions
than those that occur from the open-topped basins commonly used for conventional
biological treatment.
b. Liquid Wastes—Not applicable.
c. Solid Wastes Not applicable.
E. ADD-ON CONTROLS
1. Introduction
The addition of "tail-end" controls to existing terminal treatment/disposal
operations probably provides the greatest potential for the reduction of current
-------
V-15
secondary emissions from the SOCMI. As the bulk of controllable secondary emissions
emanate from the biological treatment and/or pretreatment of aqueous wastes and
from the landfilling of liquid and solid wastes, the emphasis in this section
and in Sect. VI ("Impact Analysis") is on the control of secondary emissions
from these sources. Although Table V-l shows a number of possible control options
for most secondary emission sources, only those sources with significant emis-
sion potential and the most probable control options are discussed here.
2. Covers
a- Aqueous Wastes Covers can be effectively used to reduce secondary emissions
occurring during the biological treatment, the surface impoundment, and the
phase separation of aqueous wastes.
Biological treatment Most secondary emissions are caused by the evaporation
from open-topped basins commonly used in biological wastewater treatment systems.
Secondary emissions from open-topped basins can be effectively controlled with
covers. A floating cover is being used for odor control in anaerobic lagoon
wastewater treatment.1'2 The covers are used to contain the anaerobically gen-
erated gas and to channel the gas to an odor control device, such as a flare.
Floating covers could also be used to reduce secondary VOC emissions from some
types of nonaerated wastewater treatment systems, such as clarifiers and neturali-
zation basins. In aerated wastewater treatment systems with submerged aerators
a floating cover can also be used to convey secondary VOC emissions to a control
device, or a rigid equipment cover fabricated from steel, aluminum, or fiber-rein-
forced plastic could also be used although at a higher cost.3'4 The use of
covers may not be feasible for the control of emissions from aerated systems
equipped with surface aerators.
Actual field experience is limited on the floating-cover concept as a secondary
VOC emissions control or conveying system. An immiscible organic liquid, plastic
spheres, and plastic covers are three types of floating covers in use that may
be feasible for secondary VOC emission control for nonaerated basins.
Three acrylonitrile manufacturing plants in Texas use an approximately 0.1-m-thick
layer of high-molecular-weight lubrication oil to reduce secondary VOC emissions
-------
V-16
from a settling basin. The settling basin is a suspended-solids removal device
located ahead of the wastewater deep well.5 The lubrication oil is periodically
replaced as it is lost through solubilization and evaporation.6 Sampling and
analysis of a controlled settling basin and of an identical uncontrolled basin
showed that the controlled basin had reduced the secondary VOC emissions approxi-
mately 90%. Within the estimated statistical limits of sampling accuracy, however,
the VOC emission reduction for the controlled settling basin ranged from 70 to
100%.5
The floating, immiscible, organic liquid is probably the most economical of the
three cover types; however, the organic liquid layer will evaporate and be a
source of secondary VOC emissions. A plant located in Texas reports an emission
factor of approximately 2.5 to 5 kg/(yr)(m2) for the lubrication oil cover.6
Plastic spheres have been used to reduce emissions of hydrocarbons from fixed-
roof crude-oil storage tanks in the petroleum industry. The plastic spheres
are microscopically small, hollow, and filled with gas. Tests made in the United
States and Canada have shown an evaporation loss reduction of 50 to 70% for
working tanks and 70 to 90% for static tanks. A 1.25- to 2.5-cm-thick blanket
of spheres is generally used.7
Larger floating plastic spheres are used to reduce the heat loss, evaporation,
and odors from a sewage digester in an European wastewater plant. The spheres
are 45 mm in diameter, are made of polypropylene, and have a molded-in circum-
ferential rim. The rim causes the spheres to interlock and reduces the problem
of rotating balls exposing a film of wastewater to the atmosphere. The spheres
come in diameters of 20, 45, and 150 mm. Tests showed that a single-layer blanket
of 45-mm-diam spheres reduced heat losses by 70%, evaporation losses by 90%,
and odor by 98%.8
Floating plastic spheres are also used for VOC emission control in the SOCMI.
For example, they are used on an API separator in a Texas ethylbenzene plant to
reduce VOC emissions.9
A floating, plastic, gas-tight cover has been used primarily for odor control,
methane collection, algae control, evaporation control, dilution control, and
-------
V-17
potable-water protection.10 One potential technical difficulty with this concept
for secondary VOC emission control is that an adhesive must be found that will
allow repairs of the floating cover and withstand the corrosive action of the
chemicals present in wastewaters from the SOCMI.11 Another concern is the gen-
eration of pockets of flammable and explosive vapor mixtures under the cover,
although this has not been a problem with floating plastic covers used to collect
anaerobically generated methane gas.11 The largest floating plastic covers
presently installed are about 37,000 m2 in area.
Surface impoundment Since no data were found in the literature review on the
control of secondary VOC emissions from surface impoundments, the floating plastic
cover concept discussed in Sect. V-E-2-a will be considered as the control tech-
nology. While used primarily for odor control, methane collection, algae control,
evaporation control, dilution control, and potable-water protection,10 the floating
plastic cover appears to be a feasible alternative for control of secondary VOC
emissions from surface impoundments.
Liquid-liquid phase separation See Sect. V-E-2-b.
b. Liquid Wastes Covers can also be effectively used to reduce secondary emissions
occurring in surface impoundment and landfilling of liquid waste and in liquid-
liquid phase separations.
Surface impoundment See Sect. V-E-2-a.
Landfill The volatilization rate of organic liquids and solids from a landfill
soil cover can be greatly reduced by increasing the compaction and water content
of the soil cover.12 In general secondary VOC emissions from properly operated
landfills with adequate soil cover will be low. For special situations with
particularly hazardous organic liquids or solids in a landfill, secondary VOC
emissions can probably be further reduced by use of an artificial cover such
as polyurethane foam over the landfill cell.13 The type of cover material will
depend on the corrosive nature of the material being landfilled. In general
plastic and asphaltic materials can be applied.
-------
V-18
Liquid-liquid phase separation Secondary emissions from open-topped phase
separators (e.g., API separators) may also be controlled with floating covers
(see Sect. V-E-2-a).
c. Solid Wastes The use of covers to reduce secondary emissions resulting from
the treatment or disposal of solid wastes is limited primarily to landfill
operations.
Landfill See Sect. V-E-2-b.
3. Carbon Adsorption
a. Aqueous Wastes Carbon adsorption can be effectively used to reduce secondary
emissions resulting from the following aqueous waste treatment operations.
Biological treatment/pretreatment Carbon adsorption has been used to control
odors from municipal wastewater treatment equipment such as clarifiers, sludge
thickeners, and trickling filters. The wastewater equipment is covered and
vented to the carbon adsorption system.14 The use of carbon adsorption equip-
ment to reduce nonodorous secondary VOC emissions from wastewater treatment
systems in the SOCMI is not common practice. Also, the potential for a flam-
mable and explosive vapor mixture being generated during the handling of sec-
ondary VOC emissions is a significant design consideration.
Surface impoundment The comments on the use of carbon adsorption to control
secondary emission from biological treatment operations (Sect. V-E-3-a) also
apply to surface impoundment.
b. Liquid Wastes Carbon adsorption can be effectively used to control secondary
emissions resulting from the following liquid waste treatment/storage operations.
Surface impoundment See Sect. V-E-3-a.
Liquid-liquid phase separation See Sect. V-E-3-a.
-------
V-19
4. Thermal Destruction
a. Aqueous Wastes Although thermal destruction is not generally used for the
direct disposal of aqueous wastes, it can be effectively used to control
secondary emissions from other aqueous treatment/disposal operations.
Biological treatment/pretreatment The two thermal destruction control options
discussed below are a fume incinerator and a nearby fire box to reduce secondary
VOC emissions from a wastewater treatment system.
Fume incinerators, both thermal and catalytic, have been used to reduce odor
from both municipal and industrial wastewater systems. A recent thermal instal-
lation, for example, collects off-gas from two covered biological aeration lagoons.
The unit is designed to handle approximately 1700 m3 of off-gas per hour and
recovers energy as steam.15 A pharmaceutical plant also thermally oxidizes the
odors from its wastewater treatment plant.16 The use of fume incineration to
reduce nonodorous secondary VOC emissons from wastewater treatment systems in
the SOCMI is not common practice. Moreover, the potential for a flammable and
explosive vapor mixture being generated during the handling of secondary VOC
emissions is a significant design consideration.
A pharmaceutical plant uses a steam boiler fire box to thermally oxidize foul
odors from its fermentation systems.16 The cost-effective use of this control
technology is dependent mainly on the following factors: the distance of the
source of the secondary VOC emissions from the fire box, the combustion air
requirement of the fire box relative to the amount of foul air, the oxygen con-
tent of the collected secondary VOC emissions, and the chemical composition of
the secondary VOC emissions. VOC emissions containing organic halogens, sulfur,
phosphorous, and nitro compounds could result in unacceptable boiler corrosion
or stack emissions of hydrogen halide, sulfur dioxide, phosphorous pentoxide,
or nitrogen oxides. Generation of flammable and explosive gas mixtures during
handling is also a potential hazard for this control technology.
Surface impoundment The comments on the use of thermal destruction for the
control of secondary emissions from biological treatment operations (Sect. V-E-4-a)
also apply to surface impoundment.
-------
V-20
Liquid-liquid phase separation See Sect. V-E-4-a.
b. Liquid Wastes Thermal destruction can also be used for the control of secondary
emissions resulting from the following liquid waste treatment and storage opera-
tions .
Surface impoundment See Sect. V-E-4-a.
Liquid-liquid phase separation See Sect. V-E-4-a.
c. Solid Wastes Probably not applicable.
5. Stabilization
a. Introduction Stabilization is the term used for any process that converts a
liquid or solid waste into a chemical or physical form that will prevent or
limit the release of hazardous material to the environment. Although stabili-
zation is most often used to prevent the release of chemicals from landfill
sites into natural waters, it is probably concurrently effective in controlling
secondary emissions. No data are currently available on application or effect-
iveness .
b. Aqueous Wastes Not generally applicable.
c. Liquid Wastes. Landfill See Sect. V-E-5-a.
d. Solid Wastes. Landfill See Sect. V-E-5-a.
6. Refrigerated Condensers
a. General Refrigerated condensers are commonly used in the SOCMI to control
emissions from process storage tanks containing VOC.17 Although specific examples
in which refrigerated condensers have been used for the control of secondary
emissions are not available, condensers would probably be suitable for appli-
cations in which the concentration of VOC in emitted vapor is relatively high.
The VOC removal efficiency of a refrigerated condenser is primarily dependent
-------
V-21
on the condenser temperature, the concentration of VOC in noncondensable gases
vented (i.e., air, nitrogen), and the vapor pressure of the contained VOC.
b. Aqueous Wastes Applications of the refrigerated condensers to the control of
secondary emissions from aqueous waste treatment are discussed below.
Biological treatment/pretreatment Application would be limited to the control
of secondary emissions from non-aerated basins (e.g., clarifiers, thickeners,
neutralization basins) and would require the basins to be covered and vented to
the refrigerated condenser.
Surface impoundment See Sect. V-E-6-b.
Liquid-liquid phase separation See Sect. V-E-6-b.
c. Liquid Wastes Since the concentration of organics in the secondary emissions
occurring with the following liquid waste treatment or storage operations is
frequently greater, the use of refrigerated condensers may be more suitable than
for corresponding aqueous waste treatment applications.
Surface impoundment See Sect. V-E-6-b.
Liquid-liquid phase separation See Sect. V-E-6-b.
d- Solid Wastes Not applicable.
p- POTENTIAL SECONDARY VOC EMISSION REDUCTION
Table V-2 summarizes the range of estimated secondary VOC emission reduction
through the use of the more significant control options discussed in this
section.
-------
V-22
G. REFERENCES*
1. J. R. Miner, Control of Odors from Anaerobic Lagoons Treating Food Processing
Wastewaters, Contract CC69935-J, pp. 3—35, Industrial Environmental Research
Laboratory, EPA, Cincinnati, Ohio.
2. J. A. Chittenden, "Anaerobic Treatment of Meat Packing Wastes Preventing Odor
Problems and Recovering Energy," paper presented at the 85th National AICHE
Meeting, Philadelphia, PA, June 1978.
3. Personal communication from Temcor Co., Torrance, CA, to J. J. Cudahy, Jan. 8,
979 (documented for files of J. J. Cudahy, IT Enviroscience, Inc.).
4. Personal communication from N. C. Olsen, Preform, Inc., Environmental Systems
Division, Minneapolis, to J. J. Cudahy, Jan. 8, 979 (documented for files of
J. J. Cudahy, IT Enviroscience, Inc.).
5. T. W. Hughes and D. A. Horn, Source Assessment; Acrylonitrile Manufacture (Air
Emissions), EPA-600/2-77-107J, pp. 37, 38, 100, 101, and 105 (September 1977).
6. Personal communication from J. Key, IT Enviroscience, Inc., to J. J. Cudahy,
Nov. 5, 1978 (documented for files of J. J. Cudahy, IT Enviroscience, Inc.).
7. Evaporation Loss Committee of the American Petroleum Institute, Evaporation
Loss in the Petroleum Industry Causes and Control, API Bulletin 2513,
p. 16 (1973).
8. "Plastic Balls Suppress Odors in a Large European Plant," Chemical Processing 39, 78
(March 1976). —
9. J. Key, IT Enviroscience, Inc., Trip Report on Visit to Dow Chemical Co., Freeport,
TX, on Ethylbenzene Styrene Production, July 28 and 29 and Sept. 7—9, 1977 (on
file at EPA, ESED, Research Triangle Park, NC).
10. Globe Linings, Inc., Long Beach, CA, Development Sales Installation, bulletin.
11. Personal communication from Bill Kays, Globe Linings, Inc., Long Beach, CA,
Nov. 15, 1978 (documented for files of J. J. Cudahy, IT Enviroscience, Inc.).
12. W. J. Farmer et al., "Land Disposal of Hexachlorobenzene Wastes: Controlling
Vapor Movement in the Soil," pp. 182—190 in Land Disposal of Hazardous Waste.
Proceeding's of the Fourth Annual Research Symposium, EPA-600/9-78-016 (August 1978).
13. T. Fields and A. W. Lindsay, Landfill Disposal of Hazardous Wastes: A Review of
Literature and Known Approaches, EPA/530/SW-165 (September 1975).
14. Activated Carbon Division, Calgon Corporation, Pittsburgh, Air Purification
with Granular Activated Carbon, Bulletin No. 23—55, pp. 18 and 19.
15. Peabody Engineering, Stamford, CT, Case History for Peabody Engineering Odor Abatement
Project at Armak Chemical Company, McCook, Illinois.
-------
V-23
16. G. A. Herr, "Odor Destruction: A Case History," Chemical Engineering Progress
70(5), 65—69 (1974).
17. D. G. Erikson, IT Enviroscience, Inc., Emissions Control Options for the Synthetic
Organic Chemicals Industry. Storage and Handling Report (on file at EPA, ESED,
Research Triangle Park, NC) (October 1978).
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
VI-1
VI. IMPACT ANALYSIS
A. INTRODUCTION
The impact analysis of secondary emissions is based on a model chemical pro-
duction plant (MCPP) with a model wastewater treatment system, landfill, and
surface impoundment; the model parameters for these units are summarized in
Tables IV-2, IV-3, and IV-5 and E-l in Appendix E. The MCPP can be considered
to be a large chemical complex producing 1650 Gg of chemicals per year with a
production based on the average production of the 30 chemicals shown in Table IV-4.
For simplifying purposes the organically contaminated wastewater was treated as
a wastewater containing a single-component waste chemical with a Henry1s-law
constant (M .) of 25 and a molecular weight of 58. The other wastewater param-
AJ.
eters (shown in Table IV-3) are production-weighted averages for the 30 chemicals
based on available wastewater data.1—4
The impact analyses for the landfill and surface impoundment are based on two
model solid chemicals selected for widely varying secondary VOC emission poten-
tial. Chemical A, similar in physicochemical properties to hexachlorobenzene,
has a low vapor pressure (0.0025 pascal) and chemical B, similar in physico-
chemical properties to dichlorobenzene, has a high vapor pressure (130 pascals).
The impact analyses for landfill and surface impoundment were done separately
for each chemical.
Uncontrolled secondary VOC emissions for the model wastewater treatment plant
are shown in Fig. IV-1. Uncontrolled emissions of 1.06 Gg/yr are based on the
industry secondary emissions factor of 640 g of VOC per Mg of product multiplied
by the model-plant production of 1650 Gg/yr. The emissions shown in Fig. IV-1
from the clarifier and conventional activated-sludge (CAS) system were estimated
by the methods described in Appendix B.
For the impact analysis the following options were evaluated for the control of
secondary VOC emissions from the model-plant wastewater: a carbon adsorption
system for recovery of the VOC from the wastewater, a cover to reduce secondary
VOC emissions from the wastewater clarifier, a cover for the clarifier plus a
carbon adsorption system, and a cover for the clarifier plus a CAS system using
a fume incinerator. The control options are shown in Figs. VI-1—VI-4.
-------
VI-2
VOC removal.
8.36 Gg of VOC
per year
Model-Plant
Wastewater
8.44 Gg of
VOC per
year
9.3 Mg
of VOC
per year
1.3 Mg
of VOC
per year
Carbon
Adsorption
84.4 Mg
of VOC
per year
^^Secondary Emissions
Clarifier
CAS
Wastewate
Effluent
Fig. VI-1. control Option 1: Removal of Model-Plant Wastewater VOC with
Carbon Adsorption (Total, 10.6 Mg of VOC Emissions Per Year)
-------
VI-3
9.3 Mg
of VOC
per year
Model-Plant
Wastewater
8.44 Gg of VOC
per year
147.5 Mg
of VOC
per year
•Secondary Emissions
Covered
Clarifier
8.43 Gg of
VOC per year
CAS
Wastewater
Effluent
Pig. VI-2. Control Option 2: Cover for Model-Plant Wastewater Clarifier
(Total, 157 Mg of VOC Emissions Per Year)
-------
VI-4
Secondary Emission
9.3 Mg of VOC
per year
Carbon
Adsorption
Model-Plant
Wastewater,
8.44 Gg of
VOC per
year
Covered
Clarifier
8.43 Gg of
vex: per year
Recovery or
Disposal
. .146 Mg of
VOC per year
Secondary
Emission
1.5 Mg of
VOC per year
147.5 Mg of VOC
per year
CAS
f- or*
Effluent
Fig. VI-3. Control Option 3: Covered Clarifier Plus Carbon Adsorption
(With and Without Chemical Recovery) of the Secondary VOC Emissions
from a CAS System Treating the Model-Plant Wastewater
(Total, 10.8 Mg of VOC Emissions Per Year)
-------
VI-5
Steam
Secondary Emission
9.3 Mg of VOC
per year
Fume
Incinerator
Model-
Plant
Wastewater
8.44 Gg of1
VOC per
year
Covered
Clarifier
8.43 Gg of
VOC per year
Energy
Recovery or
Stack
Secondary
Emission
1.5 Mg of ^
VOC per year
147.5 Mg of VOC
per year
CAS
Wastewater
Effluent
Fig. VI-4. Control Option 4: Covered Clarifier Plus Fume Incineration
(With and Without Energy Recovery) of the Secondary VOC Emissions
from a CAS System Treating the Model-Plant -Wastewater
(Total, 10.8 Mg of VOC Emissions Per Year)
-------
VI-6
The impact analysis for the liquid and solid organic wastes is based on the
model-plant parameters summarized in Table IV-5 and Table E-3 in Appendix E.
The reduction of secondary VOC emissions from an existing landfill and surface
impoundment by means of gas-tight plastic covers is evaluated.
B. CONTROL COST IMPACT
1. Introduction
The estimated costs and cost-effectiveness data for control of secondary VOC
emissions resulting from the model-plant production are presented in Table VI-1.
Details of the model chemical production plant are given in Tables IV-2, IV-3,
IV-5, and E-3; cost estimate calculations are included in Appendix E. The
following model-plant analyses are for estimation purposes only and not for
standards support.
2. Aqueous Wastes
a. Control Option 1 With this control option (see Fig. VI-1) a carbon adsorption
system is used to remove 99% or greater of the VOC from the model-plant waste-
water. In some instances the removal efficiency will be slightly decreased by
VOC losses during regeneration or subsequent decanting operations. The VOC in
the carbon adsorption system wastewater effluent then enters the model-plant
wastewater treatment system described in Table IV-2. Recovery of the VOC from
the wastewater before it enters the model wastewater treatment plant results in
a 99% reduction of secondary VOC emissions relative to the uncontrolled base
case shown in Fig. IV-1. The bases for control option 1 are given in Sect. A-l
of Appendix E.
The estimated capital cost of a carbon adsorption system designed to reduce by
99% or greater the model-plant wastewater is $3,900,000 (see Table VI-1). This
cost is based on the installation of a carbon adsorption system consisting of
skid-mounted beds, blowers, condensers, decanter, interconnecting piping, valving,
instrumentation, and all necessary piping and utility connections.
-------
Table VI-1. Cost Estimates for Control of Secondary VOC Emissions for Model Plant
Annual Operating Cost, December 1979
Cost Item
Wastewater
c
Carbon adsorption (CA)
d
Cover for clarifier
Cover and CA with chemical
e
recovery
Cover and CA without
chemical recovery
Cover and fume incineration
(FI) without energy recovery
Cover and FI with energy
..*. „,._...,
Landfill
Cover/chemical A9
Cover /chemical B
Surface impoundment
Cover/chemical A°
Cover/chemical B
Total Contract
Installed Utilities Capital Waste
Capital and Raw Related Disposal
Cost Materials Manpower Costs Costs
$3,900,000 647,000 39,400 1,131,000
75,320
1,363,200 25,700 39,400 21,840
1,363,200 25,700 39,400 395,300 $17,500
1,583,200 909,300 27,000 459,100
2,153,200 910,100 27,000 624,400
435,600 126,300
435,600 126,300
435,600 126,300
435,600 126,300
Met
Recovery Annual
(Credits)* Costs
$1,817,400
21,840
(26,300) 434,100
477,900
1,395,400
(598,700) 962,800
126,300
126,300
126,300
126,300
Emission
Reduction
(Mg/yr)
1049
903
1049
1049
1049
1049
3 X 10~4
12
0.014
274
(%)
99
85
99
99
99
99
99
99
99
99
Effectiveness
(per Hg)
$ 1,733
24
414
456
1,330
918
^
H
1
420,000,000
10,500
9,000,000
460
Net annual cost and cost effectiveness depend strongly on recovery credits.
Cost per Mg of VOC emission reduction.
CSee Fig. VI-1.
d
See Fig. VI-2.
CSee Fig. VI-3.
£See Fig. VI-4.
Q
Low-vapor-pressure solid.
High-vapor-pressure solid.
-------
VI-8
To determine the cost effectiveness of the carbon adsorption system, estimates
were made of the direct operating cost, the capital recovery cost, and the miscel-
laneous capital costs. Three cases were evaluated using recovery credits of 0,
$0.18, and $0.36 per kg of recovered VOC. For control option 1 the net annual
cost is $1,817,000 based on a zero recovery credit. The cost effectiveness for
this control option is $1733 per Mg of VOC reduction.
The net annual cost and the cost effectiveness for this control option are directly
proportional to the VOC recovery credit. For a VOC recovery credit of $0.18 per kg
of VOC the net annual cost is $312,400 and the cost effectiveness is $298 per Mg
of VOC reduction. For a $0.36 per kg of VOC recovery credit the net annual
savings would be $1,137,000 and the cost effectiveness would be a savings of
$1,137 per Mg of VOC reduction. The cost effectiveness for this control option
is also strongly dependent on the quality of the recovered VOC. Additional
unit operations such as distillation to upgrade the recovered VOC will result
in higher net annual costs than those given.
b. Control Option 2 With this control option (see Fig. VI-2) a floating gas-tight
plastic cover on the model-plant wastewater clarifier, 99% or greater of the
uncontrolled secondary VOC emissions from the clarifier are reduced. This means
that more VOC is treated in the CAS unit, which slightly increases the secondary
emissions from the CAS unit. The secondary VOC emissions from the CAS system
are uncontrolled in this opt?on. Covering the clarifier results in an 85%
reduction of secondary VOC emissions relative to the uncontrolled base case
shown in Fig. III-l.
The estimated installed capital cost of such a cover designed to reduce by 99%
or greater the clarifier secondary VOC emissions is $75,320 (see Table VI-1).
To determine the cost effectiveness of this control option, estimates were made
of the capital-related costs (see Appendix E). For this control option the
annual cost is $21,840. The cost effectiveness for the control option is $24 per
Mg of VOC emission reduction for the model chemical having an M . of 25.
XX
c. Control Option 3 This control option (shown in Fig. VI-3) consists of carbon
adsorption (with and without chemical recovery) of the secondary VOC emissions
from a CAS system treating the model-plant wastewater. The clarifier and CAS
-------
VI-9
system are covered with a gas-tight cover, and the off-gas VOC emissions
from the CAS are removed by a carbon adsorption system. The VOC removal efficiency
of the carbon adsorption system is 99% or greater. The clarifier cover and the
carbon adsorption system result in a 99% reduction of secondary VOC emissions
relative to the uncontrolled base case shown in Fig. III-l.
The estimated capital cost of a carbon adsorption5 and cover system (floating
plastic cover) designed to reduce by 99% or greater the model-plant wastewater
VOC is $1,360,000 (see Table VI-1).
To determine the cost effectiveness of the covers and carbon adsorption system,
estimates were made of the direct operating cost, the capital related cost, and
the miscellaneous capital costs. For this control option without chemical recovery,
the net annual cost, including disposal charges, is $478,000. The net annual
cost for the option with chemical recovery is directly proportional to the value
of the recovered chemical and ranges from $460,500 to $407,900 for recovery
credits ranging from zero to $0.36 per kg of recovered VOC.
The cost effectiveness for the option with chemical recovery and a recovery
credit of $0.18 per kg is $414 per Mg of VOC emission reduction. Without chemical
recovery the cost effectiveness is $456 per Mg of VOC emission reduction.
For. the option without chem:'.cal recovery the incremental cost effectiveness for
the additional 146 Mg of VOC emission reduction resulting from the addition of
the CAS cover and carbon adsorption system to the covered clarifier of control
option 2 is $3124 per Mg of VOC emission reduction. For the option with chemical
recovery at $0.18 per kg of recovered chemical, the incremental cost effective-
ness for control option 3 over control option 2 is $2824 per Mg of emission reduction.
-------
VI-10
result in a 99% reduction of secondary VOC emissions relative to the uncontrolled
base case shown in Fig. III-l.
The estimated capital cost of a fume incineration6 and cover system designed to
reduce by 99% or greater the model-plant wastewater VOC secondary emissions is
$1,583,200 (see Table VI-1). If waste-heat recovery is included to reduce the
operating cost, the estimated installed capital cost is $2,153,200. These costs
are based on a thermal oxidizer designed for a residence time of 0.5 sec at
871°C, completely installed, and include a blower, all controls and instrumenta-
tion, a complete energy recovery steam boiler, and a stack.
To determine the cost effectiveness of a thermal oxidizer, estimates were made
of the direct operating cost, the capital-related cost, and the miscellaneous
capital costs, both with and without heat recovery. The recovery credit was
calculated for the heat-recovery case based on a boiler outlet temperature of
260°C, or the recovery of approximately 60% of the energy in the flue gases.
The value of the recovered energy is credited as the cost of natural gas at
$1.90 per GJ. The net annual cost without energy recovery is $1,395,400 and
with energy recovery is $962,800 (see Table VI-1). The cost effectiveness without
energy recovery is $1330 per Mg of VOC emission reduction and with energy recovery
is $918 per Mg of VOC emission reduction (see Table VI-1).
The incremental cost effectiveness for control option 4 over control option 2
(covered clarifier) is $6440 per Mg of emission reduction with energy recovery
and is $9408 per Mg of emission reduction without energy recovery.
2. Liquid Wastes
a. Landfill Plastic Cover In this control option the model-plant landfill param-
eters listed in Table IV-5 were used as the basis for the estimation. Two cases
are evaluated, one for model chemical A, a low-vapor-pressure solid, and the
other for model chemical B, a relatively high-vapor-pressure solid. The secondary
VOC emission estimates are based on the estimation methods described in Appendix D.
The plastic cover is installed after a landfill cell is completed and is designed
to reduce the secondary VOC emissions by 99% or greater, relative to the uncontrolled-
emission case.
-------
VI-11
The estimated capital cost of a plastic cover to reduce the landfill secondary
VOC emissions is $435,600 (see Table VI-1). To determine the cost effectiveness
of this control option, estimates were made of the capital recovery cost and
the cover maintenance and miscellaneous costs (see Appendix E). For the model
plant the annual cost is $126,300 each for chemicals A and B. The cost effective-
ness for the model plant is $420,000,000 per Mg of VOC emission reduction for
chemical A and is $10,500 per Mg for VOC emission reduction for chemical B.
b. Surface Impoundment Cover In this control option the model-plant impoundment
surface parameters used as the basis for the estimation are those given in Table E-l
in Appendix E. Two cases are evaluated, one for model chemical A, a low-vapor-
pressure solid, and the other for model chemical B, a relatively high vapor-pressure
solid. The secondary VOC emission estimates are based on the estimation methods
described in Appendix C. The plastic cover is installed after the surface impound-
ment is filled and is designed to reduce the secondary VOC emissions by 99% or
greater, relative to the uncontrolled-emission case.
The estimated capital cost of a plastic cover to reduce the surface-impoundment
secondary VOC emissions is $435,600 (see Table VI-1). To determine the cost
effectiveness of this control option, estimates were made of the capital recovery
cost and the cover maintenance and miscellaneous costs (see Appendix E). For
the model plant the annual cost is $126,300 each for chemicals A and B. The
cost effectiveness for the model plant is $9,000,000 per Mg of VOC emission
reduction for chemical A and is $460 per Mg of VOC emission reduction for chemi-
cal B.
3. Solid Wastes
a. Landfill Plastic Cover See Sect. IV-B-2-a.
C. ENVIRONMENTAL AND ENERGY IMPACTS
Table VI-2 gives the environmental and energy impacts of reducing the secondary
VOC emissions by application of the indicated control options to the model plant.
Secondary VOC emissions are shown to be reduced by as much as 903 Mg/yr to 1049 Mg/yr
for the model-plant wastewater and by 3 X 10-4 Mg/yr to 274 Mg/yr for the model-
plant liquid and solid organic wastes. The net energy usage for the control
options ranges from zero to 479 TJ per year as natural gas.
-------
Table VI-2. Environmental and Energy Impacts of Controlled Secondary VOC Emissions from Model Plant
Emission Source
Model wastewater
treatment plant
Sanitary landfill
Surface impoundment
Control Option
Carbon adsorption (CA)
Cover for clarifier
Cover and CA with and without
chemical recovery
Cover and fume incineration (FI)
e
without energy recovery
Cover and FI with energy recovery6
Cover/chemical A^
Cover/chemical B
Cover/ chemical A"
Cover/chemical B
Emission
(%)
99
85 '
99
99
99
99
99
99
99
Reduction
(Mg/yr)
1049
903
1049
1049
I
1049
-4
3 X 10
12
0.01
274
Energy Usage
(TJ/yr)
196 as steam
7 as steam
479 as natural gas
479 as natural gas
usage; 316 as re-
covered steamf
See Fig. VI-1.
bSee Fig. VI-2.
CSee Fig. VI-3.
Does not include any energy necessary for disposal.
SSee Fig. VI-4.
Net energy usage for this control option is 163 TJ/yr.
^Low-vapor-pressure solid.
High-vapor-pressure solid.
H
I
to
-------
VI-13
D- REFERENCES*
11 Development Document for Effluent Limitations Guidelines and New Source Per-
formance Standards for the Major Organic Products Segment of the Organic
Chemicals Manufacturing Point Source Category, EPA 440/1-74-009-a (April 1974).
2* Development Document for Interim Final Effluent Limitations Guidelines and New
Source Performance Standards for the Significant Organic Products Segment of the
Organic Chemical Manufacturing Point Source Category, EPA 440/1-75-045, Group I,
Phase II (November 1975).
3- Monsanto Research Corp. and Research Triangle Institute, Chapter 6. The Industrial
Organic Chemicals Industry, Part I.
4- W. D. Bruce, IT Enviroscience, Inc., Trip Report for Visit to Monsanto Textiles Co.,
Pensacola, FL, Feb. 8, 1978 (on file at EPA, ESED, Research Triangle Park, NC).
H. L. Basdekis, IT Enviroscience, Inc., Control Device Evaluation, Carbon
Adsorption (June 1980) (on file at EPA, ESED, Research Triangle Park, NC).
J. W. Blackburn, IT Enviroscience, Control Device Evaluation, Thermal Oxidation
(June 1980) (on file at EPA, ESED, Research Triangle Park, NC).
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
VII-1
VII. ASSESSMENT
A- SUMMARY
Secondary VOC emissions, defined as emissions that result from the treatment
and disposal of aqueous, liquid, and solid wastes, may account for as much as
5% of the total SOCMI emissions. Most secondary emissions result from the
storage, treatment, and disposal of aqueous wastes, with the balance originating
mainly from chemical landfill and surface impoundment of primarily liquid and
solid wastes. Based on 1976 SOCMI production1 the order-of-magnitude uncon-
trolled secondary VOC emissions from aqueous wastes are estimated to be 58 Gg/yr.
Most of the secondary emissions from aqueous wastes that are currently uncon-
trolled are caused by the evaporation of VOC from open-topped basins, which are
commonly used in the handling and treatment of aqueous wastes.
The add-on control alternatives for aqueous wastes that are evaluated in Sect. VI
are all based on the use of covers to minimize VOC emissions and/or to collect
the emissions and route them through terminal treatment/disposal devices. The
range of estimated control costs for the large model plant considered varies
widely from relatively low to prohibitively expensive, depending primarily on
the degree of control provided. Although the use of the various types of
covers to control secondary emissions from aqueous wastes is currently not
widely used by industry, it is a promising approach.
B. DATA ASSESSMENT
Because secondary emissions are very difficult to measure directly, practically
no direct emission data are available.
The conclusions presented in this report are based primarily on theoretically
developed, order-of-magnitude estimates of emissions. The accuracy of these
estimates is dependent on the availability of information on the characteristics
and quantities of aqueous, liquid, and solid wastes. This information is also
very scarce, particularly information pertinent to landfilling and surface
impoundment operations.
In order to attain estimates of greater accuracy it will be necessary to develop
a great deal more information on the quantity and characteristics of liquid and
-------
VII-2
solid wastes sent to landfill and on surface impoundment based on actual indus-
trial experience.
-------
VI I-3
C. REFERENCE*
1- Synthetic Organic Chemicals, 1976 United States Production and Sales, United
States International Trade Commission, GPO, Washington" (1977)"
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
APPENDIX A
LITERATURE REVIEW
-------
A-3
Appendix A
LITERATURE REVIEW
Although the area of secondary emissions is starting to receive a lot of research
attention, the topic is not a new one. For example, in 1957 Eckenfelder et
al.1 presented a paper on some theoretical aspects of aerating industrial waste-
waters, and in 1961, Gaudy et al.2'3 started publishing a series of papers on
the stripping of volatile components from conventional wastewater treatment
systems. The following review highlights the literature of secondary emissions
relative to wastewater, liquid organic, and solid organic wastes.
A. ORGANICALLY CONTAMINATED WASTEWATER
In collaboration with others,2'4 Gaudy did much of the early published work on
secondary emissions. In other early researches5—7 the concentration was on
the removal of chemical oxygen demand (COD) from refinery wastewaters by aeration.
Gaudy et al. concentrated on laboratory work with volatile ketones and aldehydes,
for example, acetone, butanone, and propionaldehyde, and investigated air stripping
of synthetic solutions during biological oxidation8 and without biological
oxidation.2 They also investigated the air stripping of acetone in the presence
of dissolved salts and found no appreciable effect on stripping caused by the
salts.3 In a recent paper9 Gaudy investigated the formation of strippable bio-
degradation products in a glucose solution and found that a maximum of 3 to 4%
of the glucose COD input was converted to air-strippable COD.
Much of the recent secondary-emission literature (refs. 10—12, for example)
has been concerned with the estimation of secondary emissions from organically
contaminated wastewaters.
• Estimation Methods
There are basically two types of wastewater secondary-emission estimation methods
described in the literature: estimation methods using pure-component physicochemical
properties and methods using stripping constants experimentally derived in the
laboratory from actual industrial multicomponent wastewater samples.
Secondary-emission estimation methods based on the pure-component physicochemical-
property approach have been developed by Mackay and Leinonen,10 Thibodeaux,11
-------
A-4
and Smith and Bomberger.12 The procedures developed in refs. 10 and 11 for the
estimation of secondary VOC emissions from organically contaminated wastewaters
are detailed in Appendix A.
a. Mackay Method Mackay's estimation method is based on the two-film theory of
mass transfer.13 This method has been compared with experimental data for the
evaporation of selected chlorinated hydrocarbons from dilute aqueous solutions
and found to produce reasonably close results.14 The equations in ref. 10 can
be used to estimate the desorption of low-solubility contaminants from nonaerated
process basins without biological oxidation and from surface impoundments.
b. Thibodeaux Method The Thibodeaux paper11 is basically a computer-simulation
study for the desorption of oxygen-containing polar compounds from wastewater
treatment systems. Most of the compounds are highly soluble in water. In his
paper mass balance equations are developed for the desorption of the compounds
from different wastewater treatment systems, which are summarized as a function
of Henry's-law constant (M .) for the compounds.
Aa.
c. Smith Method The procedure described in ref. 12 is based on the observation
that for highly volatile compounds the experimentally obtained ratio of the
desorption rate constant to the oxygen reaeration rate constant is a constant
for a range of turbulent conditions. In that paper an activated-sludge process
is modeled, and the calculated desorption of various chemical compounds in the
modeled activated-sludge process is summarized.
Secondary-emission estimation studies based on laboratory-derived stripping
constants for multicomponent wastewaters have been published by Thibodeaux and
Millican15 and by Richardson and Ledbetter.16
d. Thibodeaux Study In this study wastewater is pumped from the bottom of a flask
back into the flask, where the liquid irrigates a section of packing. Air is
forced into the flask below the packing to produce countercurrent contact in
the packing and desorption of the volatile fractions of the organics in the
wastewater. As many as 75 wastewater samples representing 26 industry types
were tested by the stripping-constant method.
-------
A-5
e. Richardson Study In this study 5-liter samples of biologically inert multicom-
ponent wastewater samples were aerated by a centrifugal air blower. Before the
air supply entered the sample bottle, it was presaturated with water to minimize
water evaporation. Different air flows were used in the laboratory work, and a
regression equation was obtained for the stripping-rate constant as a function
of stripping air flow. The industrial effluent tested consisted mainly of waste-
waters from a compressed-gas plant, two petroleum refineries, a petrochemical
plant, and a kraft pulp and paper mill.
2. Field Studies
There are very few published studies of full-scale industrial wastewater treatment
systems that have been sampled and analyzed for secondary emissions. Most of the
published studies have been done for petrochemical and petroleum refinery waste-
waters. These studies are summarized below.
a. Los Angeles County Study on Petroleum Refineries17 In this pioneering study
of petroleum refinery emissions, field inspections were made at six major
refineries in Los Angeles County for secondary emissions from process drains
and wastewater treatment systems. During the inspections a combustible gas
indicator was used to determine the hydrocarbon vapor concentrations, and a
titanium tetrachloride smoke tube was used to indicate the approximate air flows
associated with the hydrocarbon concentration.
The wastewater systems inspected were drain sumps (collection basins), interceptor
systems, oil-water separators, and open secondary flotation tanks or ponds.
Hydrocarbon vapor concentrations were found above the surface of all these waste-
water systems except the open secondary flotation tanks or ponds. Since air
flows during the inspections were not quantitatively determined, secondary emis-
sions could not be calculated. The secondary emissions estimated in the report17
were based on a total-pollution survey by the Air Pollution Control District.18
The secondary emissions for all six refineries were estimated to be 2.7 Mg of
hydrocarbons per day, which is approximately equal to an emission factor of
2.4 kg of hydrocarbons per 1000 m3 of wastewater.
b. German Study on Refineries and Chemical Factories19 In this paper the author
states that the average quantities of exhaust gas exiting the wastewater ducts
of refineries and chemical factories are estimated to be 12 kg per 1000 m3 of
-------
A-6
plant wastewater throughput. No comments are made in the paper as to how this
emission factor was obtained.
c. EPA Study on Acrylonitrile Manufacture20 In the manufacture of acrylonitrile
by the catalytic vapor-phase amoxidation of propylene a settling basin is used
as a solids removal device for process wastewater before it is injected into a
deep well. As a VOC control device three of the manufacturing plants cover the
surface of the settling basin with about 0.1 m of lubricating oil having a molecu-
lar weight of about 500. Field sampling during this study was done on both
controlled and uncontrolled secondary emissions from the settling basin. Sampling
was done to obtain ambient air samples upwind and downwind of the basin.
d. Richardson Study on Mixed Industrial Wastes16 In this work, which was discussed
earlier, both laboratory and field observations were made. In the field work
an inverted funnel that was floated on the surface of an aeration basin was
used to minimize atmospheric dilution. The funnel conveyed the off-gas directly
to a portable gas chromatograph. The aeration basin sampled was treating waste-
water from many facilities, most of which were petrochemical industries. Four
sampling sites were selected: at the ends and at points one-third the distance
from each end of an interior bay of a six-bay aeration basin. The secondary
emissions found were reported as methane.
3. On-Going EPA Work
The EPA is currently assessing secondary emissions from industrial and municipal
wastewaters in the following on-going studies:
a. Oil Refinery Wastewater Systems In this study, conducted by the Research
Triangle Park, North Carolina, office, two dissolved air flotation (DAF) systems
and five wastewater separators in oil refineries have been evaluated. A mass-
balance method is being used for secondary-emission determination. Wastewater
influent and effluent samples are analyzed for hydrocarbon concentration, and
the difference is taken as the hydrocarbon secondary emission. At this time no
data have been made public from the DAF and API units.
b. Robert S. Kerr Environmental Research Laboratory22 In this assessment by the
Ada, Oaklahoma, office, priority pollutants in the wastewater treatment systems
of ten plants from five different industries have been studied. As part of the
-------
A-7
study a mobile air-stripping column has been used to collect volatile organics
from the mixed liquor-suspended solids in five industries. At this time no
results have been made public relative to the air stripping of organics from
these wastewater streams.
c. Municipal Environmental Research Laboratory23 The Wastewater Research Division
in Cincinnati has started to assess the secondary emissions of priority pollutants
from publicly owned treatment works. The program, which will include laboratory
and field studies of priority pollutant secondary emissions, will evaluate pre-
dictive techniques, development of techniques, and field sampling and analysis.
d. Water Quality Analysis Branch23—This study, which is being done by the Washington,
D.C., office, will concentrate on priority pollutant identification in publicly
owned treatment works and will also include field sampling and analysis of priority-
pollutant secondary emissions from 40 such works.
B. LIQUID AND SOLID ORGANIC WASTES
The sampling, analysis, and characterization of secondary emissions from the
treatment and disposal of liquid and solid organic wastes have not received as
much attention as secondary emissions from wastewaters. Most of the secondary-
emission studies found in a brief literature review were concerned with methane
generation in sanitary landfills handling municipal refuse, which is outside
the scope of this report. A few references were found for secondary emissions
from sanitary landfills and surface impoundments, and are summarized in the
following sections.
1. Estimation Methods
Secondary emissions from sanitary landfills and from surface impoundments were
studied by Farmer e_t al.24—26 and by Mackay10 respectively.
The most comprehensive work to date on emissions from landfills has been done
by Farmer et al. in a series of papers concerned with the land disposal of
hexachlorobenzene (HCB). Using HCB waste, the authors determined in the labora-
tory the important soil parameters controlling HCB vapor movement out of the
soil cover. Based on these research findings they developed predictive equations
for use in designing landfill soil covers to reduce organic vapor movement out
of a landfill.
-------
A-8
The technique developed by Mackay10 is the most reasonable method of estimating
secondary emissions from a surface impoundment. The surface impoundment should
be considered to be an organic liquid or organic solid waste covered by a layer
of water. The Mackay model is a diffusion model only and does not incorporate
aerobic or anaerobic degradation mechanisms.
2. Field Studies
Vinyl chloride monomer (VCM) emissions at a landfill used for polymerization
sludges have been assessed by bag-sampling techniques.27
No information on field studies of secondary emissions from surface impoundments
was found.
3. On-Going EPA Work
The Solid and Hazardous Waste Research Division of the Municipal and Environmental
Research Laboratory in Cincinnati, Ohio, is funding a study to sample and measure
the levels of air pollutants that might be emitted from hazardous waste management
facilities located throughout the United States.28 Among the areas tested will
be landfill, chemical/biological treatment, transfer operations, and storage
facilities.
Under a grant to the University of Arkansas, wastewater treatment basins at
pulp and paper mills will be sampled to determine the magnitude of selected
chemical air emissions. The compounds selected for study are methanol, acetone,
acetaldehyde, hydrogen sulfide, methyl mercaptan, and dimethyl disulfide. A
sampling method for upwind/downwind measurements of the basin is also being
developed.
-------
A-9
B. REFERENCES*
1. W. W. Eckenfelder et al., "Some Theoretical Aspects of Solvent Stripping and
Aeration of Industrial Wastes," pp. 14—25 in Proceedings of the llth Indiana
Waste Conference; Purdue University, published in Extension Series 91, 14—25
(1957). —
2. A. F. Gaudy, Jr., et al., "Stripping Kinetics of Volatile Components of Petrochemical
Wastes," Journal of Water Pollution Control Federation 33(4), 382—392 (April 1961).
3. A. F. Gaudy, Jr., et al., "Diffused Air Stripping of Volatile Waste Components
of Petrochemical Wastes," Journal of Water Pollution Control Federation 33(2),
127—135 (February 1961). —
4. S. R. Goswami and A. F. Gaudy, Jr., "Removal of Aldehydes and Ketones by Stripping
and by Combined Stripping and Microbial Metabolism," pp. 253—266 in Proceedings
of the Fourth Mid-Atlantic Industrial Waste Conference, Nov. 18, 19, 20, 1970,
published by the University of Delaware, Newark, Dept. of Civil Engineering,
December 1971.
5. B. V. Prather and A. F. Gaudy, Jr. "Purifying Refinery Waste Water," Oil and
Gas Journal 62, 96—99 (Aug. 10, 1964).
6. B. V. Prather, "Waste-Water Aeration May Be Key to More Efficient Removal of
Impurities," Oil and Gas Journal 57(49), 78 (1959).
7. B. V. Prather, "Will Air Flotation Remove the Chemical Oxygen Demand of Refinery
Waste Water?" Hydrocarbon Processing and Petroleum Refining 40(5), 177 (May 1961).
8. A. F. Gaudy, Jr., et al., "Biological Treatment of Volatile Waste Components,"
Journal of Water Pollution Control Federation 35(1), 75—93 (January 1963).
9. A. F. Gaudy, Jr., "Formation of Strippable Metabolic Products in Biological
Waste Treatment," pp. 1239—1244 in Biotechnology and Bioengineerinq, vol. XIX,
Wiley, New York, 1977.
10. D. Mackay and P. J. Leinonen, "Rate of Evaporation of Low-Solubility Contaminants
from Water Bodies to Atmosphere," Environmental Science and Technology 9(13),
1178—1180 (December 1975).
11. L. J. Thibodeaux, "Air Stripping of Organics from Wastewater. A Compendium,"
pp. 358—378 in Proceedings of the Second National Conference on Complete
Watereuse. Water's Interface with Energy, Air and Solids, Chicago, IL, May 4—8,
1975, sponsored by AIChE and EPA Technology Transfer.
12. J. H. Smith and D. C. Bomberger, "Prediction of Volatilization Rates of Chemicals
in Water," paper presented at the 85th National AIChE Meeting, Philadelphia,
PA. June 4—8. 1978.
13. R. E. Treybal, Mass Transfer Operations, 1st ed., pp. 80—90, McGraw-Hill,
New York, 1955.
-------
A-10
14. W. L. Dilling, "Interphase Transfer Processes," Environmental Science and
Technology 11(4). 405—409 (April 1977).
15. L. J. Thibodeaux and J. D. Millican, "Quantity and Relative Desorption Rates
of Air-Strippable Organics in Industrial Wastewater," Environmental Science and
Technology 11(9), 879—883 (September 1977).
16. C. P. Richardson and J. 0. Ledbetter, "Hydrocarbon Emissions from Refinery
Wastewater Aeration," Industrial Wastes 24(4), 26—28 (July/August 1978).
17. "Emissions to the Atmosphere from Eight Miscellaneous Sources in Oil Refineries,"
pp. 32—38 in Report No. 8, Joint District, Federal and State Project for the
Evaluation of Refinery Emissions, Los Angeles County Air Pollution Control District
(June 1958).
18. N. R. Shaffer and C. J. Seymour, Emissions of Hydrocarbons from Refineries in
Los Angeles County, County of Los Angeles, California, Air Pollution Control
Ditrict, April 1957.
19. R. Schwanecke, "Air Pollution Resulting from Leakage from Chemical Facilities,"
pp. 9—15 in Luftverunreinigung, 1970 (translation by Scitran on file at EPA,
ESED, Research Triangle Park, NC).
2Q. T. W. Hughes and D. A. Horn, Source Assessment: Acrylonitrile Manufacture (Air
Emissions), EPA-600/2-77-107J, pp. 37—38, 100—101, 105 (September 1977).
21. Environmental Protection Agency Contract No. 68-02-2147, Industrial Environmental
Research Laboratory, Office of Research and Development, Research Triangle Park, NC.
22. Personal communication from L. Myers, Robert S. Kerr Laboratory, to J. J. Cudahy,
Oct. 27, 1978 (documented for files of J. J. Cudahy, IT Enviroscience, Inc.).
23. Personal communication from J, Cohen, EPA, Cincinnati, to J. J. Cudahy, Oct. 27,
1978 (documented for files of J. J. Cudahy, IT Enviroscience, Inc.).
24. W. J. Farmer et al., Problems Associated with the Land Disposal of an Organic
Industrial Hazardous Waste Containing HCB, EPA-600/9-76-015, pp. 182—190
(July 1976).
25. W. F. Farmer et al., "Land Disposal of Organic Hazardous Wastes Containing HCB,"
Proceedings of the National Conference on Disposal of Residues on Land, St. Louis,
Missouri, EPA, Office of Research and Development, published by Information
Transfer Inc., Rockville, MD, 1977.
26. W. J. Farmer et al., "Land Disposal of Hexachlorobenzene Wastes: Controlling
Vapor Movement in Soil," pp. 182—190 in Land Disposal of Hazardous Wastes.
Proceedings of the Fourth Annual Research Symposium, edited by D. W. Schultz,
EPA-600/9-78-016 (August 1978).
-------
A-ll
27. R. Markle, R. B. Iden, and F. A. Sliemers, "Preliminary Examination of Vinyl
Chloride Emissions from Polymerization Sludges During Handling and Land Disposal,"
pp. 186—194 in Residual Management by Land Disposal. Proceedings of the
Hazardous Waste Research Symposium, edited by W. H. Fuller, EPA-600/9-76-015
(July 1976).
28. Personal communication from D. Oberacker, EPA, Cincinnati, to J. J. Cudahy,
Oct. 27, 1978 (documented for files of J. J. Cudahy, IT Enviroscience, Inc.).
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
APPENDIX B
ESTIMATION PROCEDURES FOR SECONDARY VOC EMISSIONS FROM WASTEWATER
-------
B-3
Appendix B
ESTIMATION PROCEDURES FOR SECONDARY VOC EMISSIONS FROM WASTEWATER
This appendix summarizes three procedures that were obtained from the literature
for estimating secondary VOC emissions from wastewater treatment systems and
describes how they are used in this report: the Mackay1 and Dilling2 methods
for nonaerated flow systems such as a clarifier; the Thibodeaux3 method for aerated
flow systems such as activated-sludge treatment; and the Mackay1 method for
nonaerated and nonflow systems such as a surface impoundment. Because of the
limitations discussed later in this appendix the secondary VOC emission estimates
based on the three methods should be considered to have an order-of-magnitude
value.
A. HENRY'S-LAW CONSTANT (M .)
XI
Inherent in the use of the procedures to estimate the secondary VOC emissions
from wastewater is the need for a Henry's-law constant for the organic contami-
nant. Either of two methods may be used to estimate this constant: the use of
vapor-liquid equilibrium data at a temperature close to the wastewater temperature
or the use of pure-component vapor pressure and solubility data for compounds
that have low water solubility.
1. Estimation by Vapor-Liquid Equilibrium Data
The vapor-liquid equilibrium method for estimating M . is limited because the
equilibrium data needed generally must be at ambient temperature to simulate
wastewater temperatures. However, it is the best estimation method for highly
water soluble compounds.
The following derivation describes how vapor-liquid equilibrium data can be
used to calculate Henry's-law constant:
Since the fugacity coefficient is equal to unity for many dilute organic aqueous
systems, the constant, in units of atmosphere,3 can be expressed as
(1)
-------
B-4
where
Y. = the mole fraction of the organic compound in the vapor phase,
x. = the mole fraction of the organic in the aqueous phase,
P = the total pressure of the system in atmospheres.
An example of the use of equilibrium data for estimating the above constant is
shown below for acetone in water at 25°C:
xi Yi P (mm Hg)
0.01936 0.5234 50.1
0.0289 0.6212 61.8
0.04495 0.7168 81.3
At x..^ = 0.01936
YiP (0.5234K50.1) 1
x. 0.01936 760
M
xi
= 1.782.
is then developed from a plot of x. versus Y.P/x.. The intercept of Y.P/x.
for x. equal to zero is Henry's-law constant. In this case the intercept of
Y.P/x. at x. equal to zero, using nonlinear curve fitting, is about 2. This
value should then be corrected from 23.7 mm Hg, the saturation pressure of the
water at 25°C, to 760 mm. This correction is small and in light of the order-
of-magnitude type of analysis being performed can be ignored.
2. Estimation by Solubility Data
For compounds with low water solubility (about 10% or less by weight) M . can
A«L
be estimated from the saturated solubility of the compound in water and the
vapor pressure of the pure compound, both at the temperature of the system.
For a saturated solution it is assumed that x. = x. and that the partial pres-
3L IS
sure of the compound in solution, P.^, is the same as the vapor pressure, P*a/
of the pure compound. This will be true for systems in which the water is rela-
tively insoluble in the compound and fugacity coefficient, 4>, is equal to 1.
This can be expressed as
-------
B-5
. .
M . i i ria
~ x x ~ x (2)
x x x
y.p P.
.
i i is
where
P*a = the pure-component vapor pressure of the compound at the
temperature of the system in atmospheres,
xis = the mole fraction at saturated solubility of the compound
in the water at the temperature of the system,
M . = Henry1 s-law constant in atmospheres.
AX
Since most solubility data are given in mg/liter or in ppm and most vapor pres
sure data in mm Hg, Eq. (2) can be rewritten as the following approximate
equation (valid when Si < 100,000 mg/liter) by applying conversion factors for
the units involved:
73
M . =
-xi Si(18)(760) ~ S^^ ' (3)
where
P* = the pure-component vapor pressure of the compound
at the temperature of the system, in mm Hg,
M . = the molecular weight of the compound,
W JL
S..^ = the saturated solubility (in mg/liter or ppm) of the compound at
the temperature of the system,
M . = Henry's-law constant in atmospheres.
Using Eq. (3), the Henry's-law constant for nitrobenzene at 25°C is calculated
in the following manner:
= (73)(0.27)(123)
xi 2000
= 1.21.
B. ESTIMATION OF SECONDARY VOC EMISSIONS FROM WASTEWATER SYSTEMS
As an example, secondary VOC emissions will be estimated for a nonaerated flow
system, an aerated flow system, and a nonaerated, nonflow system with nitro-
benzene as the model compound.
-------
B-6
1. Nonaerated Flow System
The model clarifier discussed in Sect. III-B will be used as an example of a
nonaerated flow system. For nonaerated systems without biological degradation,
such as clarifiers and settling basins, the estimation methods described by
Mackay1 and modified by Dilling2 should be used in the following manner:
Using the equation developed in ref. 2, recalculate Henry's-law constant:*
16.04 P*M .
H. =
i wi , (4)
1 TSi
where
H. = Henry's-law constant, dimensionless,
P* = the compound's pure-component vapor pressure in mm Hg at T,
M . = the molecular weight,
Mf piL.
T = the absolute temperature of the wastewater in K,
S^ = the compound's solubility in mg/liter at T.
Then for nitrobenzene:
H - 16.04(0.27X123) - fl q y in 4
"i ~ (298)(2000) " °'y A iu~ '
Calculate the overall liquid mass-transfer coefficient K., by the following
equation from ref. 2:
K. (221.1X0.6) , in m/hr, (5)
ft2*100) ("T
where H. is Henry's-law constant, M . is the molecular weight, and K., is the
J™ Vr lit J.J.
overall liquid mass-transfer coefficient in m/hr.
Then
9.4 X 10-3
/ im
(a.9 x ?o-* * 10") <123> '
*Dilling defines H, as dimensionless (i.e., mg of chemical per liter of air
divided by mg of cRemical per liter of water), whereas M . is in units of
atmospheres. .
-------
B-7
Calculate the percent desorption from the Mackay model (ref. I)-.
fl = exp / i\ f
o V 1 /
where
C. = the concentration at time t of the nitrobenzene,
C = the initial nitrobenzene concentration,
o
L = the liquid depth (in m),
t = the retention time (in hr) of the liquid in the wastewater system.
For a clarifier with a 3-hr retention time and a 3-m depth,
C. / 3v
~ = exp (-9.4 X 10 3 |j = 0.99.
o '
I Ci\
The percent desorption is equal to ( 1 - ~ J X 100, or 1%.
\ o /
2. Aerated Flow System
For this example the model activated-sludge system discussed in Sect. IV-E and
the activated-sludge system model described by Thibodeaux3 are used.
Thibodeaux's work is basically a paper study using computer simulations for the
desorption of oxygen-containing polar compounds from wastewater. Most of the
compounds are infinitely soluble in water. Equations are developed for the
desorption of these compounds from different wastewater treatment systems, and
the desorption values are summarized either in tables or figures. The compounds
and associated Henry's-law constants from Thibodeaux's paper are summarized in
Table B-l.
-------
B-8
Table B-l. Henry's Law Constants from Thibodeaux*
Mxi
Compound (atmospheres)
Acetaldeyde 5.88
Acetone 1.99
Isopropanol 1.19
n-Propanol 0.471
Ethanol 0.363
Methanol 0.300
n-Butanol 0.182
Formic acid 0.0247
Propionic acid 0.0130
*See ref. 3.
Table B-2, adapted from Thibodeaux's work,3 shows the relationship of desorption
(secondary VOC emission) versus Henry's-law constant and the ease of biodegrada-
bility in an activated-sludge system. In an activated-sludge system easy-to-
biodegrade chemicals are not desorbed as much as difficult-to-biodegrade chemicals
are.
Table B-2. Desorption as a Function of
Henry's-Law Constant in an Activated-Sludge
Wastewater Treatment System at 25°C
Henry 's-Law Constant
5.88
1.99
1.19
0.471
0.363
0.300
0.182
0.0247
0.0130
Desorption
Of High-Rate
Biodegradable
Material
0.41
0.13
0.055
0.023
0.022
0.025
0.0071
0.0021
0.0006
(%)
Of Low-Rate
Biodegradable
Material
11.0
3.76
1.62
0.685
0.660
0.749
0.212
0.062
0.019
-------
B-9
The desorption of nitrobenzene (Mxi = 1.21) from an activated-sludge system was
estimated by interpolation from Table B-2 as 0.06% based on nitrobenzene being
easily biodegraded. Within the order-of-magnitude limits of the estimation
techniques discussed in this Appendix, the data in Table B-2 can be presented
as the following approximate linear equations:
% desorption (high-rate biodegradability) = 0.07 M . . (8)
A!
% desorption (low-rate biodegradability) = 1.87 M . . (9)
XI
Equations (8) and (9) are used throughout this report to estimate secondary VOC
emissions from activated-sludge wastewater treatment systems for chemicals other
than those listed in Table B-l. High-rate biodegradable chemicals are those
that are easily biodegraded in a typical industrial activated-sludge system.
Low-rate biodegradable chemicals are those that are more difficult to biodegrade
in a typical industrial activated-sludge system. Examples of low-rate biodegrad-
able chemicals would be certain highly chlorinated organics.
3. Nonaerated Nonflow System
The model surface impoundment discussed in Sect. VI will be used as an example
of a nonaerated, nonflow system. The estimation method described by Mackay1 is
used, and has the following bases:
The surface impoundment system is a basin containing an organic liquid or solid
that is covered by an aqueous layer.
A steady-state situation exists. The rate of solution of the organic material
into the aqueous cover equals the rate of desorption of the organic material
into the atmosphere from the aqueous cover.
Sufficient atmospheric dispersion exists above the surface impoundment so that
the partial pressure (P*) of the organic material above the aqueous cover is
negligible.
The chemical is present in the aqueous cover at the solubility limit concentra-
tion (C ). For nitrobenzene at 25°C, C is equal to 2000 Mg/liter. The area
S ' S .
of the surface impoundment is 4047 m2.
-------
B-10
The secondary VOC emission (SE ) from the aqueous cover into the atmosphere is
Set
obtained from the following equation by Dilling2 from Mackay's model:
SEsa • Kil Cs - iT I ' <8)
Since P* is negligible, Eq. (8) reduces to
where
SE = the secondary VOC emission rate from the surface impoundment in
g/(m2)(hr),
K., = the overall liquid-phase mass transfer coefficient in m/hr,
C = the chemical concentration in the aqueous cover in mg of chemical
s
per liter of wastewater (mg/liter is equivalent to g/m3).
For nitrobenzene (NB) at 25°C SE is calculated in the following way:
Scl
SE a s (9.4 X 10-3 m/hr)(2000 g of NB/m3 of WW)
Scl :
= 18.8 g of NB per hr per m2 of wastewater.
For 4047 m2 of surface impoundment area and 8760 hr/yr the secondary VOC emis-
sion rate becomes 667 Mg of nitrobenzene secondary emissions per year.
C. GENERAL COMMENTS ON THE PROCEDURES FOR ESTIMATING SECONDARY VOC EMISSIONS
The procedures used in this Appendix for estimating the secondary VOC emissions
should be considered as only an order-of-magnitude estimate for the following
reasons:
1. They are based on a pure-component analysis and do not incorporate any
possible chemical interreaction, hydrolysis, or acid/base caused reactions,
such as the conversion of phenol to sodium phenate.
2. The Thibodeaux work3 is a theoretical analysis based mostly on average
parameters obtained from the literature.
3. The Thibodeaux simulation3 does not take into account, for instance, absorp-
tion of organics on sludge and/or other suspended solids in the wastewater.
-------
B-ll
The simulation also does not account for possible desorption of biochemical
reaction products or organic degradation via catalyzed or uncatalyzed photo-
degradation.
-------
B-12
D. REFERENCES*
1. D. Mackay and P. J. Leinonen, "Rate of Evaporation of Low-Solubility Contaminants
from Water Bodies to Atmosphere," Environmental Science and Technology 9(13),
1178—1180 (December 1975).
2. W. L. Billing, "Interphase Transfer Processes," Environmental Science and
Technology 11(4), 405—509 (April 1977).
3. L. J. Thibodeaux, "Air Stripping of Organics from Wastewater. A Compendium,"
pp. 358—378 in the Proceedings of the Second National Conference on Complete
Watereuse. Water's Interface with Energy, Air and Solids, Chicago, IL, May 4—8,
1975, sponsored by AIChE and EPA Technology Transfer.
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
APPENDIX C
ESTIMATE OF UNCONTROLLED SECONDARY VOC EMISSIONS FROM
WASTEWATERS FOR THE SYNTHETIC ORGANIC
CHEMICALS MANUFACTURING INDUSTRY (SOCMI)
-------
C-3
Appendix C
ESTIMATE OF UNCONTROLLED SECONDARY VOC EMISSIONS FROM WASTEWATERS FOR THE SOCMI
Data on secondary VOC emissions from wastewater for the SOCMI are very limited.
Therefore an order-of-magnitude theoretical assessment has been made based on
available SOCMI wastewater data and on the model chemical production plant and
the model wastewater treatment systems described in Sect. IV-E.
The wastewater data used are from the 30 highest volume SOCMI chemical products.1—4
The model clarifier emissions were estimated based on the Mackay5 and Dilling6
methods as discussed in Appendix A. The model activated-sludge system emissions
were estimated based on the Thibodeaux7 method modified as discussed in Appendix B.
CRITERIA
The following criteria were used in the estimations:
The wastewater from the production of each chemical enters a model-plant waste-
water treatment system at 20 or 25°C, which includes, in series, a 3-hr retention
time, a 3-m-deep clarifier, and an activated-sludge system.
Wastewater total organic carbon (TOC), chemical oxygen demand (COD), and flow
parameters were average values taken from refs. 1—4 for the major process given
in the reference.
All wastewater TOC and COD were attributed to only the chemical produced.
In cases where, at 20 or 25°C, the solubility of the chemical produced was lower
than that of the process TOC or COD data, the chemical solubility was used in
making the estimate.
The desorption of each chemical in the model-plant wastewater treatment system
was calculated based on estimated Henry's-law constants and the procedures described
in Appendix B.
-------
C-4
B. ESTIMATION OF WASTEWATER PRODUCT CONCENTRATION
Table C-l summarizes the estimation of the wastewater product concentration for
each of the 30 products listed.
Column A summarizes the average amount of wastewater TOC or COD per Mg of waste-
water for each product.
Column B is the theoretical ratio of the molecular weight of the product to the
TOC or COD for each mole of chemical listed. For example, the molecular weight
ratio of ethylene to organic carbon is 28:24, or 1.17.
Multiplication of column A by column B results in column C, the amount of the
product present, in grams, per Mg of wastewater. Column C in Table C-l is com-
pared with column E in Table C-2 (the product solubility at 20 or 25°C), and
the smaller of the two is used in the next estimation.
C. ESTIMATION OF THE AMOUNT OF PRODUCT DESORBED (D ) FROM THE MODEL-PLANT WASTEWATER
SYSTEM T
Using the procedures described in Appendix A, Henry's-law constants (M . and
A J.
H.) and the overall liquid mass-transfer coefficients (K.,) were estimated for
each product and used to estimate the desorption fraction (D ) for the clarifier
and the activated-sludge system (D ) (see Table C-2).
3
The total desorption fraction (DT) is the sum of D and D (1 - D ) and is defined
1 C 3, C
as the amount (in grams) of product desorbed from the wastewater per gram of
product originally present in the wastewater.
D. ESTIMATION OF THE SOCMI UNCONTROLLED SECONDARY VOC EMISSION FACTOR (SEf)
Column D in Table C-l summarizes the average amount of wastewater (WW) used to
make 1 g of each product.
Multiplication of column D by DT and the smaller of column C or E results in
the secondary VOC emission factors (SEf) for the products shown in Table IV-4.
For ethylbenzene (EB), SEf would be estimated as follows:
_ 0.31 q of WW 0.34 g of EB desorbed 152 g of EB in WW
f g of EB produced g of EB in WW Mg of WW
= 16 g of EB desorbed per Mg of EB produced.
-------
C-5
Table C-l. Wastewater Parameters
(Col A)
(Col B)
g of TOC
(in WW) g of
Chemical
Ethylene
Ethylene dichloride
Styrene
Methanol
Ethylbenzene
Vinyl chloride
Ethylene oxide
Ethylene glycol
1, 3-Butadiene
Dimethylterephthalate
Cumene
Acetic acid
Cyclohexane
Phenol
Chlorinated methanes
Isopropal alcohol
Acetone
Propylene oxide
Acetic anhydride
Acrylonitrile
Ethanol
Vinyl acetate
Terephthalic acid
Adipic acid
t-Butanol
Acetaldehyde
Cyclohexanol/cyclohexanone
Phthalic anhydride
Adiponitrile
Caprolactam
Mg of WW
758
1,106
22
345
2,091
120
11,175
929
502
4,606
180
3,843a
38,800a
198
519
38,800a
345
153,140a
19,633
30,000a
220
10
19,483b
18,560
580
26,217
4,500
109
Prod.
g of TOC
1
4
1
2
1
2
1
2
1
1
1
0
0
8
1
0
1
0
1
0
1
1
1
1
1
1
1
1
.17
.08
.08
.67
.10
.60
.83
.58
.13
.62
.11
.94a
.813
.52
.67
.62
.61
.80
.47
.48
.79
.73
.54
.57
.38
.54
.50
.57
(Col C)
g of Prod.
(in WW)
Mg of WW
887
4,516
24
920
2,300
313
20,487
2,400
564
7,446
200
3,600
31,337
1,686
865
23,862
556
122,031
28,900
14,375
394
17
30,036
29,166
800
40,418
6,750
171
(Col D)
g of WW
g of
1.
0.
14.
0.
0.
2.
0.
4.
1.
4.
3.3 X
2.
0.
1.
1.
2.
3.
61.
2.
4.
2.
0.
4.
19.
0.
1.
1.
9.
Prod.
71
80
50
42
31
80
62
87
70
24
io~4
1
002
85
33
54
00
26
13
20
80b
23
24
c
lb
52
15
44
76
11.13
Data
Source
for Cols
A and D
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
Ref.
1
1
1
1
1
1
1
1
1
1
2
1
1
1
2
2
1
2
3
2
3
1
1
2
1
4
2
2
1
a b
COD. Estimate.
-i
"Essentially zero secondary VOC emission rate.
-------
C-6
Table C-2. Summary of VOC Desorption Calculations
rhvruc.-il
EtMy lene
Ethylene dlchloride
3 tyrene
Methanol
EUiylbrnzene
Vinyl cnloride
Ethylene oxide
Ethylene glycol
1 , 3-Butadiene
Dimethylterephthalate
Ci^nene
Acetic acid
Cyclohexane
Phenol
Chlorinated nethanes
Isopropal alcohol
Acetone
Propylene oxide
Acetic anhydrij''
Acrylonitrile
Ethanol
vinyl acetate
Terephthalic acid
Adipic acid
t-3utanol
Acetalcehyde
Cyclohcxanol/'
cyclo.'inxanono
Pnthilic anhydrid'-
•\dir on it r i !'.•
Capcoi icta.".
(Col a
131
3^0
300
Inf.
152
1100
Inf.
Inf.
735
<19
50
Inf.
55
S2.00U
5394
Inf.
Inf.
405 ,000
Inf.
73.5CJ
Inf.
20,000
19
15,000
Inf.
Inf.
30,000
6200
30,000
340,000
1
30,400
61
5
123
-
2G60
1095
0.05
1900
0.011
3.2
11.4
77
0.2
986
32
179
4OO
3 . 5
IOC
43.9
83
5 X 10~fi
-4
1 X 10
31
740
2.3
2 X 10"
1 x :o-3
l x 10"
M
Para:
H.
J
23 35fi
99
104
32
106
62.5
44
62
54
194
120
60
84
94
112.4
60
58
58
102
53
46
36
166
146
74
44
99
14P .
108
113
0
0.
2.
O,
8.
8.
4.
7.
6
0.
4.
6.
2.
1.
8.
1.
7.
7 .
4.
2.
C.
2 .
5.
3.
4 .
4.
2.
1.
7.
.038
.095
,3 X 10
,267
3
2 X lo"
-7
3 X 10
6
x io"3
419
6 X lo"
4
2 X 10"
01
8 X 10"4
5 X 10~
4 X 10~3
-5
4 X 10
8 X 10"3
6 X Ij""*
0194
3 X lo'6
3 X 10"
3 X 10"
3 X 10"
8 X lo"
_ 7
6 X 10
9 X 10" '
_q
2 X 10 "
not-
0
0
0
5
0
0
0
6
0
0
0
7.
0.
3.
0.
0.
n.
0.
9.
0.
4.
0.
2.
5.
4.
0.
6.
2.
2.
8.
a
K;I
.25
. 105
.117
x icf 3
.124
.168
.088
-6
.9 X 10
. 13
.035
.118
.6 X 10"
.145
.6 X lo"4
.124
.0133
,0215
,072
,2 X 10"4
057
,9 X 10"3
093
3 X 10
_7
6 X 10
7 X 10"3
0587
i x io"3
7 X IO"6
4 X 10"6
-9
6 X 10
0.
•1.
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
D_
22
10
11
005
12
15
084
17
034
11
14
12
013
021
07
001
055
005
09
005
057
006
Mx;
474, 300
51
127
0.3
356
11,030
11
-4
5.8 X 10
10,190
8
560
0.063
8,585
0.03
1373
1.19
1.99
10
0.1
6.5
0. 36
26
3 X 10"3
-5
7 X 10
0.45
5.88
0.65
3.5 X 10~4
2.6 X lo"-
-6
9.8 X 10
Da
1.0
0.66
0.09
0
0.25
1.00
0.003
0
1.00
C .005
0.39
0
1.00
0
1.00
0.001
0.001
0.007
0
0 .004
0
0.018
0
0
0
0.004
0
0
0
0
DT
1.0
0.69
0.19
0.005
0.3-
1.0
0.09
0
1.00
0.04
0.46
0
1.0'
0
1.0
0.014
0.022
0.077
0.001
? .06
0 .005
0 .11
0
0
0.005
0 .06
0 . 006
0
0
0
Tee Sect.
in this appendix for def ir.it »*?n.
-------
C-7
Multiplication of SEf by the estimated 1978 production8 results in an estimate
of secondary VOC emissions for each product (Table IV-4). The sum of these esti-
mated emissions (35.7 Gg) divided by the total production (54,600 Gg) results in
an estimated average SOCMI uncontrolled secondary VOC wastewater emission factor
of 650 g of VOC per Mg of product produced.
Multiplication of the SOCMI emission factor (650 g/Mg) by the total SOCMI 1978 pro-
duction (100,000 Gg) results in an estimated 65 Gg/yr of uncontrolled secondary
emissions from wastewater.
E. SYMBOLS AND DEFINITIONS
The symbols that are used in this Appendix are defined below:
Symbol Definition
S. Solubility of the product in water, generally at ambient temperature
(mg/liter or g of product per Mg of WW)
P* Pure-component vapor pressure of product at 20 or 25°C (mm Hg)
M . Molecular weight of product
W J.
H^ Henry's-law constant: mg of chemical per liter of air T mg of chemical
per liter of water (see Appendix A)
Kil Overall liquid mass-transfer coefficient as defined in Appendix A
DC Mass fraction of VOC desorbed in the clarifier of the model-plant
wastewater treatment system
Mxi Henry's-law constant in atmospheres
Da Mass fraction of VOC desorbed in the activated-sludge system of the
model-plant wastewater treatment system
D_, Total mass fraction of VOC desorbed in the total model-plant wastewater
treatment system; defined as D plus D (1 - D )
C 3 C
-------
C-8
F. REFERENCES*
1. Development Document for Effluent Limitations Guidelines and New Source Per-
formance Standards for the Major Organic Products Segment of the Organic
Chemicals Manufacturing Point Source Category, EPA 440/1-74-009-a (April 1974).
2. Development Document for Interim Final Effluent Limitations Guidelines and New
Source Performance Standards for the Significant Organic Products Segment of the
Organic Chemical Manufacturing Point Source Category, EPA 440/1-75-045, Group I,
Phase II (November 1975).
3. Monsanto Research Corp. and Research Triangle Institute, Chapter 6. The Industrial
Organic Chemicals Industry, Part I.
4. W. D. Bruce, IT Enviroscience, Inc., Trip Report for Visit to Monsanto Textiles Co.,
Pensacola, FL, Feb. 8, 1978 (on file at EPA, ESED, Research Triangle Park, NC).
5. D. Mackay and P. J. Leinonen, "Rate of Evaporation of Low-Solubility Contaminants
from Water Bodies to Atmosphere," Environmental Science and Technology 9(13),
1178—1180 (December 1975).
6. W. L. Dilling, "Interphase Transfer Processes," Environmental Science and
Technology 11(4), 405—409 (April 1977).
7. L. J. Thibodeaux, "Air Stripping of Organics from Wastewater. A Compendium,"
pp. 358—378 in the Proceedings of the Second National Conference on Complete
Watereuse. Water's Interface with Energy, Air and Solids, Chicago, IL, May 4—8,
1975, sponsored by AIChE and EPA Technology Transfer.
8. Synthetic Organic Chemicals, 1976 United States Production and Sales, United
States International Trade Commission, GPO, Washington (1977).
*When a reference number is used at the end of a paragraph or on a heading, it
usually refers to the entire paragraph or material under the heading. When,
however, an additional reference is required for only a certain portion of the
paragraph or captioned material, the earlier reference number may not apply to
that particular portion.
-------
APPENDIX D
COMPARISON OF SECONDARY VOC EMISSIONS FOR FIVE MODEL CHEMICALS
-------
D-3
Appendix D
COMPARISON OF SECONDARY VOC EMISSIONS FOR FIVE MODEL CHEMICALS
Data on the secondary VOC emissions of chemicals in a landfill are limited.[
Because of this scarcity an assessment of the secondary VOC emissions from land-
fills and wastewater treatment was done. This assessment is a theoretical order-
of-magnitude analysis that compares the secondary VOC emissions for five model
compounds: hexachlorobenzene, paradichlorobenzene, benzene, acetone, and acetic
acid. The compounds encompass a wide range of physical properties.
The Mackay,2 Dilling,3 Thibodeaux,4 and Farmer5 models were used for estimating
the secondary VOC emissions.
The bases and calculations used in the estimation and comparison of the secondary
VOC emissions from the wastewater system with those from the landfill are as follows:
A. PHYSICAL PROPERTIES OF MODEL CHEMICALS AT 25°C
The physical properties used in the estimation of the secondary VOC emissions
from the treatment systems are given in Table D-l.
Table D-l. Physical Properties of Model Chemicals
Compound
Abbreviation
Vapor pressure
(mm Hg)
Molecular weight
Solubility in water
(mg/liter)
Hexachlorobenzene
HCB
1.9 X 10-5
285
0.0062
2- Dichlo robenz ene
DCB
1.0
147
79
Benzene
Bz
95.2
78
1790
Acetone
A
220
58
Inf.
Acetic acid
HAC
11.4
60
Inf.
B. CALCULATION OF HENRY'S LAW CONSTANTS AND K^
Henry's-law contants H. (Dilling3) and M . (Thibodeaux4) and the overall liquid
X XI
mass-transfer coefficient K.. for HCB, DCB, and Bz were calculated based on the
procedures discussed in Appendix B. The results are summarized in Table D-2.
-------
D-4
Table D-2. Calculated Properties of Model Chemicals
Constant HCB DCS Bz A HAC
H. (mg of chemical 0.047 0.100 0.224 1.5 X 10-3 4.6 X 10-5
per liter of air
•r mg of chemical
per liter of water)
Ki;L (m/hr) 0.064 0.099 0.144 0.022 7.6 X 10-4
M . (atm) 64 136 303 1.99 0.06
The M . values for acetone and acetic acid were obtained from ref. 4. The H.
XI 1
and K., values for acetone were estimated from the following relationships:
73 P*M .
,, *
(1)
xi S .
i
and
16.04 P*M .
H =
TS .
Dividing Eq. (1) by Eq. (2) results in the following approximate relationship
between H. and M . :
i xi
73 TH.
M = - -
xi 16.04
Then, for T = 298 K, M . = 1356 H. . (3)
xi i
C. ESTIMATION OF SECONDARY VOC EMISSIONS FOR A CLARIFIER (SEc)
The following bases were used for calculation of the secondary VOC emissions
for the model wastewater treatment plant:
The clarifier is 3 m deep, has a surface area of 700 m2, and has a retention
time of 3 hr.
The wastewater flow is 17,000 m3/day.
-------
D-5
For the HCB, DCB, and Bz calculations the wastewater will be saturated at the
solubility limit of each chemical at 25°C.
For the acetone and acetic acid calculations the concentration used is 1360 mg/liter
(see Table IV-3).
The percent desorption (D ) of the chemical from the wastewater in the clarifier
is calculated from the Mackay model [Appendix A, Eq. (7)].
After the percent desorption (D ) of each chemical was obtained, the secondary
VOC emissions from the clarifier (SE ) were calculated. For example, for acetone:
_ 17,000 m3 of WW 365 days 1 Mg of WW 1360 g of acetone in WW
a " day X yr X m3 of WW X Mg of WW *
y °-°22 g of acetone as secondary emission
g of acetone in WW
= 186 Mg of acetone as secondary emissions per year.
D. ESTIMATE OF SECONDARY VOC EMISSIONS FOR AN ACTIVATED-SLUDGE WASTEWATER TREATMENT
SYSTEM (SE )
3
The following bases were used for calculating the secondary VOC emissions (SE )
from the model activated-sludge system:
The activated-sludge system in 6 m deep and has a surface area of 6300 m2.
The wastewater flow is 17,000 m3/day.
The wastewater chemical concentrations into the activated-sludge system are
based on the clarifier inlet concentrations less the amount of secondary emis-
sions of that chemical from the clarifier.
The percent desorption (D ) of a chemical from the wastewater in the activated-
cl
sludge system is estimated from the Thibodeaux model4 discussed in Appendix A.
The estimation of the SE for the model chemicals is based on HCB being diffi-
cl
cult to biodegrade and on Bz, HAC, and A being easy to biodegrade. The SEa for
DCB was estimated with both types of biodegradability to show the relative import-
ance of the biodegradability parameter.
-------
D-6
After the percent desorption (D ) was obtained for each chemical [Eqs. (8) and
3
(9) in Appendix A] SE was calculated. For example, for benzene
cl
op = 17,000 m3 of WW 365 days 1 Mq of WW 1550 g of benzene in WW
a day X yr * ma of WW x Mg of WW *
0.21 g of benzene as secondary emission
g of benzene in WW
= 2020 Mg of benzene as secondary emissions per year.
E. ESTIMATE OF SECONDARY VOC EMISSIONS FOR A LANDFILL (SE,)
The following bases were used in the estimations:
The soil in the landfill is dry, having no moisture content.
There is no water movement through soil.
The soil bulk density is 1.2 g/cm3.
The soil particle density is 2.65 g/cm3.
The depth of soil cover is 0.6 m.
There is no biological decomposition of the chemical in soil.
There are no cracks or other small openings in the soil cover.
The secondary VOC emission, in kg/(m2)(yr), is calculated from the Farmer model.5
The landfill usage rate is 4047 m2 per year.
The estimation of the organic vapor from a landfill is obtained from the following
equation:
la
D.C ...
(4)
-------
D-7
where
SE, = the secondary VOC emission from the landfill [kg/(yr)(m2)],
D. = the chemical's vapor diffusion coeffient in air (m2/yr),
C = the concentration of volatilizing material in air or the vapor
s
density at the bottom of the soil cover (kg/m3),
Pfc = the total soil porosity (m3/m3),
L = the soil cover depth (m).
D., C , and P are estimated from Eqs. (5a), (5b), and (5c):
IS U
/M \1/2
»i ' °HCB HP) • <
\ „!/
where
D = the experimentally derived vapor diffusion coefficient of hexa
chlorobenzene;
MwHCB = tne m°lecular weight of HCB;
M . = the molecular weight of the compound to be estimated;
Wi
P*M
r = * wi '
s RT
where
P* = the pure-component vapor pressure;
M . = the molecular weight;
R = the gas constant;
T = the temperature;
Pt = 1 - | , <5c)
where B is the soil bulk density and P is the soil particle density.
Combining Eq. (4) with Eqs. (5a), (5b), and (5c) results in the following
estimation equation at 25°C:
-------
("v
D-8
a/2
SE, = 0.148 P* v * ' (6)
J-3 1 Li
where
P* is in mm Hg at 25°C,
L is in meters,
SEla is in kg/(yr)(m2).
Equation (6) gives a conservatively high estimate of emissions because of the
bases used in the estimate.
The secondary VOC emission rate is calculated by multiplying Eq. (6) by the
annual landfill usage rate of 4047 m2, dividing by the soil cover depth of
0.6 m, and converting to Mg per year:
0.148 X 4047 P*(Mwi)1/2
SE1 0.6 (1000)
= 1 P?(Mwi)1/2,
where SE, is in Mg/yr.
The landfill secondary emission for DCB, for example, is calculated as follows:
SE1 =
= 12 Mg of DCB secondary emissions per year.
G. COMPARISON OF SECONDARY VOC EMISSIONS
Comparisons are made in Table IV-6 for both the secondary VOC emissions for 5
model chemicals and for the emissions from a wastewater treatment plant vs those
from a landfill.
-------
D-9
G. REFERENCES*
1. R. Markle, R. B. Iden, and F. A. Sliemers, "Preliminary Examination of Vinyl
Chloride Emissions from Polymerization Sludges During Handling and Land Disposal,"
pp. 186—194 in Residual Management by Land Disposal. Proceedings of the
Hazardous Waste Research Symposium, edited by W. H. Fuller, EPA-600/9-76-015
(July 1976).
2. D. Mackay and P. J. Leinonen, "Rate of Evaporation of Low-Solubility Contaminants
from Water Bodies to Atmosphere," Environmental Science and Technology 9(13),
1178—1180 (December 1975).
3. W. L. Dilling, "Interphase Transfer Processes," Environmental Science and
Technology 11_(4), 405—409 (April 1977).
4. L. J. Thibodeaux, "Air Stripping of Organics from Wastewater. A Compendium,"
pp. 358—378 in the Proceedings of the Second National Conference on Complete
Watereuse. Water's Interface with Energy, Air and Solids, Chicago, IL, May 4—8,
1975, sponsored by AIChE and EPA Technology Transfer.
5. W. J. Farmer and M. H. Roulier, "Land Disposal of Hexachlorobenzene Wastes:
Controlling Vapor Movement in the Soil," p. 182 in the Fourth Annual Research
Symposium on Land Disposal of Hazardous Waste, San Antonio, TX, March 6—8, 1978,
sponsored by the Southwest Research Institute and the EPA.
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
APPENDIX E
COST CALCULATIONS FOR VOC SECONDARY EMISSION CONTROL OPTIONS
-------
E-3
Appendix E
COST CALCULATIONS FOR VOC SECONDARY EMISSION CONTROL OPTIONS
The calculations for the estimated costs and cost effectiveness for the control
of secondary VOC emission are presented in this Appendix. The cost estimates
are based on a model plant. Details of the model plant are given in Tables IV-2,
IV-3, IV-5, and E-l.
Table E-l. Parameters for Model-Plant Surface Impoundment
Surface impoundment area (m2) 4047
Chemical A solubility (Mg/liter) at 25°C 0.0062
Chemical B solubility {Mg/liter) at 25°C 79
Uncontrolled secondary VOC emissions
Chemical A (Mg/yr) 0.014
Chemical B (Mg/yr) 277
Capital cost estimates represent the total investment required for purchase and
installation of all equipment and material needed for a complete emission control
system performing as defined for a new plant at a typical location. These estimates
do not include the cost of research and development, land acquisition, or production
lost during installation or startup. The primary difficulty in retrofitting
may be in finding space to fit the system into the existing plant layout. Because
of these associated costs the cost of retrofitting emission control systems in
existing plants may be appreciably greater than that for a new installation.
Bases for the annual cost estimates for the control alternatives include utilities,
waste disposal, raw materials, operating labor, recovery and disposal credits,
capital-related charges, and miscellaneous recurring costs such as taxes, insurance.
and administrative overhead. The cost factors used are itemized in Table E-2.
The capital recovery factor of 0.18 is based on an assumed 10-year life and a
12% annual interest rate. Annual costs are for a 1-year period beginning in
December 1979.
-------
E-4
Table E-2. Cost Factors Used for Computing Annual Costs
Utilities
Natural gas $1.90/GJ
Electricity $0.00833/MJ ($0.03/kWh)
Maintenance material and labor 0.06 X capital cost
Capital charges
Capital recovery 0.18 X capital cost
Misc. (taxes, insurance, administration) 0.05 X capital cost
Recovery credits
Chemical $0 to 0.36 per kg
Energy $1.90/GJ
Chemical disposal charges $0.12/kg
A summary of the control option cost estimates is shown in Table E-3. These
net annual costs were used in calculating the cost effectiveness of each option.
A. CONTROL OPTION 1 RESOURCE RECOVERY BY CARBON ADSORPTION OF THE MODEL-PLANT
WASTEWATER
1. Bases Used in the Calculations
The wastewater parameters are those described in Table IV-3.
A carbon adsorption system is used to remove 99 wt % of the VOC from the waste-
water; the remainder (84.4 Mg/yr) enters the model-plant wastewater treatment
system (see Table IV-2 and Fig. VI-1).
The carbon adsorption parameters are as follows:
Carbon loading 0.1 g of VOC per g of carbon
Steam for regeneration 10 g/g of organic absorbed
Cooling water for the steam/VOC condenser 1.4 M m3/yr
Granular activated carbon replacement $2.58/kg
every 2 yr
Carbon loading cycle time 10 hr
-------
Table E-3. Cost Estimates for Control of Secondary VOC Emissions for Model Plant
Annual Operating Cost, Decenber 1979
Cost Item
Wastewater
c
Carbon adsorption (CA)
d
Cover for clarifier
Cover and CA with chemical
e
recovery
Cover and CA without
chemical recovery
Cover and fume incineration
(FI) without energy recovery
Cover and FI with energy
recovery^
Landfill
Cover/chemical Ag
Cover/chemical B
Surface impoundment
Cover/chemical A"
Cover/chemical B
Total
Installed Utilities
Capital and Paw
Cost Materials Manpower
$3,900,000 647,000 39,400
75,320
1,363,200 25,700 39,400
1,363,200 25,700 j9,400
1,583,200 909,300 27,000
2,153,200 910,100 27,000
435,600
435,600
435,600
435,600
Contract
Capital Waste
Related Disposal
Costs Costs
1,131,000
21, 840
395,300
395,300 $17,500
459,100
624,400
126,300
126,300
126,300
126,300
Net
Recovery Annual
(Credits)3 Costs
$1,817,400
21,840
(26,300) 434,100
477,900
1,395,400
(598,700) 962,800
126,300
126,300
126,300
126,300
Emission
Reduction
(Hg/yr) (%)
1049 99
903 85
1049 99
1049 99
1049 99
1049 99
-4
3 X 10 99
12 99
0.014 99
274 99
b
Effectiveness
(per Mg)
$ 1,733
24
414
450.
1,330
918
W
1
(J\
420,000,000
10,500
9,000,000
4GO
"not annual cost and cost effectiveness depend strongly on recovery credits.
'Cost per Mg of VOC emission reduction.
CSee Fig. VI-1.
dSee Fig. VI-2.
GSee Fig. VI-3.
fSee Fig. VI-4.
gLow-vapor-pressure solid.
High-vapor-pressure solid.
-------
E-6
The chemical recovery credit was estimated as follows: The production-weighted
average selling price for chemicals shown in Table C-l, based on the Oct. 9,
1978, issue of the Chemical Marketing Reporter, was $0.36/kg. The recovered
value of the chemical from the model industry was then varied from zero to $0.36
per kg of recovered chemical.
2. Calculation of Costs and Cost Effectiveness
The net annual costs were calculated to be $1,817,400 (see Table E-3).
The uncontrolled VOC secondary emissions from the wastewater were calculated to
be 1060 Mg/yr. The controlled VOC secondary emissions with the use of control
option 1, resource recovery, were calculated to be 11 Mg/yr. The total VOC
reduction was calculated as 99%.
The cost effectiveness was calculated to be $1733/Mg ($1,817,400 4- 1049 Mg of
VOC emission reduction).
B. CONTROL OPTION 2 FLOATING PLASTIC COVER FOR MODEL CLARIFIER
1. Bases Used in the Calculations
The parameters for the model plant and those for the model-plant wastewater
treatment system are described in Tables IV-2 and IV-3.
The control option consists of a plastic cover on the clarifier but not on the
activated-sludge system.
No secondary VOC control system is used for the activated-sludge system.
The total installed capital cost of the floating gas-tight plastic cover is
$107.60/m2 (ref. 1).
The life of the plastic covers is 10 years.
Uncontrolled secondary VOC emissions from the model-plant wastewater treatment
system are 1060 Mg/yr.
-------
E-7
The secondary VOC emissions from the clarifier are reduced 99% by the floating
gas-tight plastic cover.
2. Calculation of Costs and Cost Effectiveness
The net annual costs were calculated to be $21,840 (see Table E-3).
The uncontrolled secondary VOC emissions of 1060 Mg were reduced by 85% to 157 Mg
of secondary VOC emissions per year.
The cost effectiveness was calculated to be $24/Mg ($21,840 -f 903 Mg) of VOC emis-
sion reduced).
C. CONTROL OPTION 3 CARBON ADSORPTION WITH AND WITHOUT CHEMICAL RECOVERY OF THE
VOC SECONDARY EMISSIONS FROM A CONVENTIONAL ACTIVATED-SLUDGE (CAS) SYSTEM
1. Bases Used in the Calculations
The parameters for the model plant and those for the model wastewater treatment
system are as described in Tables IV-2 and IV-3.
Gas-tight floating plastic covers are used on the clarifier and CAS system.
The cover on the clarifier results in a VOC reduction of 99%. The CAS off-gas
is vented to a carbon adsorption system as shown in Fig. VI-3.
The CAS off-gas is saturated with water vapor at 32.2°C and has the following
composition and flow rate:
Carbon dioxide
Oxygen
Nitrogen
Water
VOC
The actual volumetric flow rate at 32.2°C is 657 m3/min.
Amount
(vol %)
1.6
17.4
76.4
4.6
185 (ppm)
100.0
Rate
(kq/hr)
1,130
8,751
33,640
1,306
17
44 , 844
-------
E-8
Uncontrolled secondary VOC emissions from the clarifier and CAS system are calcu-
lated to be 1060 Mg/yr.
The following carbon adsorption parameters were used:
Carbon loading 0.05 g of VOC per g of carbon
Steam for regeneration 20 g/g of organic absorbed
Cooling water for the steam/VOC condenser 69 km3/yr
Cost of granular activated carbon $2.57/kg ($1.17/lb)
replacement every 2 yr
Carbon loading cycle time 8.5 hr
The chemical recovery credit was estimated as follows: The production-weighted
average selling price for the chemicals shown in Table IV-3 was $0.36/kg; the
recovered value of the chemical from the model was then varied from zero to
$0.36 per kg.
For the option without chemical recovery the contract waste disposal charge for
the recovered chemical was $0.12/kg.
2. Calculation of Costs and Cost Effectiveness
The net annual costs for the control option with chemical recovery at $0.18/kg
were calculated to be $434,200 (see Table E-3).
The net annual costs for the control option without chemical recovery were
calculated to be $478,000 (see Table E-3).
The annual uncontrolled VOC secondary emissions from the clarifier and CAS system
amount to 1060 Mg. The annual controlled VOC secondary emissions are 11 Mg
(see Fig. VI-3). The total annual VOC reduction is 1049 Mg.
The cost effectiveness with chemical recovery was calculated to be $414/Mg
($430,000 T 1049 Mg of VOC emission reduced) and without chemical recovery to
be $456/Mg ($478,000 -r 1049 Mg of VOC emission reduced).
-------
E-9
D. CONTROL OPTION 4 FUME INCINERATION WITH AND WITHOUT ENERGY RECOVERY OF
THE VOC SECONDARY EMISSIONS FROM A CAS SYSTEM
1. Bases Used in the Calculations
The parameters for the model plant and those for the model wastewater treatment
system (CAS) are as described in Tables IV-2 and IV-3.
Gas-tight floating plastic covers are used on the clarifier and on the CAS system.
The clarifier cover reduces the VOC emissions by 99%. The CAS off-gas is vented
to a fume incineration system.
The CAS off-gas is saturated with water vapor at 32.2°C and has the following
composition and flow rate:
Amount
(vol %)
1.6
17.4
76.4
4.6
185 (ppm)
100.0
Rate
(kg/hr)
1,130
8,751
33,640
1,306
17
44,844
Carbon dioxide
Oxygen
Nitrogen
Water
VOC
The actual volumetric flow rate at 32.2°C is 657 m3/min.
The total uncontrolled VOC secondary emissions from the clarifier and CAS system
are 1060 Mg/yr.
The VOC has a heat of combustion of 23.3 MJ/kg.
The following fume incineration parameters were used:
Incineration temperature 871°C
Incineration residence time 0.5 sec
Fume energy content 0.1 MJ/m3
Auxiliary fuel Natural gas
-------
E-10
Oxygen out stack
Energy recovery
Boiler outlet temperature
VOC
VOC destruction efficiency
^3 mole %
As steam
260°C
Essentially all C, H, and 0
>99 wt %
An energy credit of $1.90/GJ was used.
2. Calculation of Costs and Cost Effectiveness
The net annual costs for the control option with energy recovery were calculated
to be $962,800 (see Table E-3).
The net annual costs for the control option without energy recovery were calcu-
lated to be $1,395,400 (see Table E-3).
The annual uncontrolled VOC secondary emissions from the clarifier and CAS system
are 1060 Mg. The annual controlled VOC secondary emissions are 11 Mg (see
Fig. VI-4). The total VOC reduction is 99.0%.
The cost effectiveness with energy recovered was calculated to be $918/Mg
($962,800 -i- 1049 Mg of VOC emission reduced) and without energy recovery to be
$1330/Mg ($1,395,400 •=- 1049 Mg of VOC emission reduced).
E. USE OF PLASTIC COVER TO REDUCE VOC SECONDARY EMISSIONS FROM A LANDFILL
1. Bases Used in the Calculations
Two model chemicals are evaluated, chemical A, with physical properties similar
to those of hexachlorobenzene, and chemical B, with physical properties similar
to those of j)-dichlorobenzene, with a 0.6-m-deep soil cover (see Table IV-5).
The secondary VOC emissions for these chemicals are 3.2 X 10-4 Mg of VOC emissions
per year for chemical A and 12 Mg of VOC emissions per year for chemical B.
The capital cost of the plastic cover is $107.60/m2 (ref. 1).
-------
E-ll
The secondary VOC emissions are reduced 99% by the plastic cover.
The area of the landfill cell to be covered is 4047 m2 per year.
2. Calculation of Costs and Cost Effectiveness
The net annual costs were calculated to be $126,300 for each chemical (see Table E-3)
The reductions in VOC secondary emissions for chemicals A and B are as follows:
Uncontrolled secondary VOC emissions (Mg/yr)
Chemical A 3.2 X 10-4
Chemical B 12
Controlled secondary VOC emissions (Mg/yr)
Chemical A 3.2 X 10-6
Chemical B 0.12
Secondary VOC emission reduction (Mg/yr)
Chemical A 3.17 X 10-4
Chemical B 11.88
The cost effectiveness for chemical A (see Table E-3) was calculated to be
$420,000,000/Mg ($126,300 -r 3 X 10-4 Mg of VOC emission reduced) and for chemical B
(see Table E-3) was calculated to be $10,500/Mg ($126,300 T 12 Mg of VOC emission
reduced).
F. USE OF PLASTIC COVER TO REDUCE VOC SECONDARY EMISSIONS FROM A SURFACE
IMPOUNDMENT (SI)
1- Bases Used in the Calculations
Two chemicals are evaluated: model chemical A with physical properties similar
to those of hexachlorobenzene and model chemical B with physical properties
similar to those of p_-dichlorobenzene.
The secondary VOC emissions for these chemicals were calculated using the following
bases:
(a) The model-plant surface impoundment system is a basin containing an organic
liquid or solid that is covered by an aqueous layer (see Table E-l).
-------
E-12
(b) A steady-state situation exists. The rate of solution of the organic material
into the aqueous cover equals the rate of desorption of the organic material
into the atmosphere from the aqueous cover.
(c) Sufficient atmospheric dispersion exists above the surface impoundment so
that the partial pressure (P*) of the organic material above the aqueous
cover is negligible.
(d) The concentrations of the organic materials in the aqueous cover are the
solubility limits for model chemical A (0.0062 mg/liter) and model chemical B
(79 mg/liter) at 25°C.
The secondary VOC emission from the aqueous cover into the atmosphere is obtained
from the following equation:2
P*
SE = K., C - rji
sa il s H.
Since P^ is negligible, this equation reduces to
SE = K.,C
sa il s
where
SE = the secondary VOC emission rate in g of VOC per m2 per hr,
So
K., = the overall liquid mass transfer coefficient in m/hr,
C = the chemical concentration in the aqueous cover of the surface
s impoundment in mg of chemical per liter of aqueous cover (mg/liter
is equivalent to g of chemical per m3 of aqueous cover).
Conversion of SE to Mg of VOC emission per year (SE )is done in the following
example for chemical B:
SE = (0.099 m/hr)(79 g of B/m3 of WW)(1 Mg of B/106 g of B)(4047 m2)(8760 hr/yr)
S
= 277 Mg of secondary VOC emission per year.
In the same manner SE for model chemical A is estimated as 0.014 Mg of secondary
5
VOC emission per year.
-------
E-13
The capital cost of the plastic cover is $107.60/m2 (ref. 1).
The secondary VOC emissions are reduced by 99% by the plastic cover.
The area of the SI to be covered is 4047 m2.
2. Calculation of Costs and Cost Effectiveness
The net annual costs were calculated to be $126,300 for each chemical (see Table E-3).
The secondary VOC emissions were calculated to be the following:
Uncontrolled secondary VOC emissions (Mg/hr)
Chemical A 0.014
Chemical B 277
Controlled secondary VOC emissions (Mg/yr)
Chemical A 1.4 X 10-4
Chemical B 2.77
Secondary VOC emisson reduction (Mg/yr)
Chemical A 0.0139
Chemical B 274
The cost effectiveness for chemical A (see Table E-3) was calculated to be
$9,000,000/Mg ($126,300 -=- 0.014 Mg of VOC emission reduced) and for chemical B
(see Table E-2) was calculated to be $460/Mg ($126,300 4- 274 Mg of VOC emission
reduced).
-------
E-14
G. REFERENCES*
1. Letter from L. P. Hughes, CMA, to D. C. Beck, EPA, Mar. 13, 1980.
2. D. Mackay and P. J. Leinonen, "Rate of Evaporation of Low-Solubility Contaminants
from Water Bodies to Atmosphere," Environmental Science and Technology 9(13),
1178—1180 (December 1975).
*When a reference number is used at the end of a paragraph or on a heading,
it usually refers to the entire paragraph or material under the heading.
When, however, an additional reference is required for only a certain portion
of the paragraph or captioned material, the earlier reference number may not
apply to that particular portion.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing!
1. REPORT NO.
EPA-450/3-80-025
3. RECIPIENT'S ACCESSION NO.
4. TITLE AMDSUBTITLE
Organic Chemical Manufacturing
Volume 3: Storage, Fugitive, and Secondary Sources
_REPORT DATE
December 1980
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
D. G. Erikson J. J. Cudahy
V. Kalcevic R. L. Standifer
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFQRMIJ^G ORGANIZATION NAME AND ADDRESS
IT tnviroscience, inc.
9041 Executive Park Drive
Suite 226
Knoxville, Tennessee 37923
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-2577
12. SPONSORING AGENCY NAME AND ADDRESS
DM for Air Quality Planning and Standards
Office of Air, Noise, and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
EPA is developing new source performance standards under Section 111 of
the Clean Air Act and national emission standards for hazardous air pollutants
under Section 112 for volatile organic compound emissions (VOC) from organic
chemical manufacturing facilities. In support of this effort, data were gathered
on chemical processing routes, VOC emissions, control techniques, control costs,
and environmental impacts resulting from control. These data have been analyzed
and assimilated into the ten volumes comprising this report.
This volume covers emissions from storage tanks, fugitive sources (pump seals,
valve seals, etc.), and secondary sources (emissions arising from treatment or
disposal of process wastes).
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI F;ield/Group
13B
it. D'STRIBUTIOM STATEMENT
Unlimited Distribution
19. SECURITY CLASS (This Report!
Unclassified
20,.S£CURITY CLASS (This page)
Unclassifie
21. NO. OF PAGES
344
22. PRICE
EP* Form 2220-1 (R.v. 4-77} PREVIOUS EDITION
------- |