-------
*,/*
''V
I-
T.
* f
"•
f
r ,f"
i
df
T*1
!"•;
S
Sis
§
cb
T*
i,.
cb
»
«
r—
^*
"»
o
5
^
o
V.
?
c
L
C
c
J
>
>
D
l>
3
O
adsorption
Thermal
-
*
ro
i
oxidation
(incineration)
3
2
c
c
o
S
3
s
3
n
o "cT
-'.0
O CO
to <2
T3 .5=
o .£=
S
i
m
Condensers
i
I
£
i
combustion
engines
o-
r£
C
..
*
C3>
.
•=>-
T3
C
CD
C3)
|
0)
C
T3 »
§ S
ve for this VO
onable effect!'
•J3 S3
O CO
CD
CO
c5
15
-------
Cost effective
range
Technology
effective range
Carbon adsorption
Catalytic oxidation
Biofiiter
Condenser
I.C engine
I 1|_|
Membrane
'1
Thermal processor
Thermal incineration
O „
u.v.
0.1 1.0 10 100 1,000 10,000 100,000
VOC concentration (ppm)
Figure 3-2. Relative cost-effectiveness for point source VOC con-
trols.
Table 3-8. Ranges of % RE for Point Source PM Controls
RyAsh
PCDD/PCDF
Acid Gases
<10 prn
>10 pirn
Metals
Baghouses —
Wet Scrubbers —
Venturl Scrubbers —
Dry Scrubbers —
ESP 994-
Quench Chambers —
HEPA Filters —
Entrained fraction removed
—
—
90-99+
98% With SDA
—
Entrained fraction removed
—
95-99+
99
95-99+
—
50%
—
99+'
Low
80-95
99+
993
99.9+4
99+'
—
80-95
99+
993
99.9+4
90-952
40-502
Variable
95-992
8S-992
_
—
1 Lower removal efficiency for mercury.
* For resistive panicles.
4 \Vilh high pressure drop.
16
-------
30.0
10.0
-§ 5.0
8 2.0
D)
1.0
0.5
0.2
104
Source: Hesketh, 1991.
Example wet FGD
Spray dry/dry injection with low
chemical requirements
High energy particle
wet scrubber
Fabric Filter
High Efficiency
ESP
10s
System capacity (acmh)
106
Figure 3-3. Typical APCD operating costs in 1988 dollars.
17
-------
Tibia 3-9. Estimated APCD Efficiencies for Controlling Toxic Metals
Air pollution control device
WS«
VS-20"
VS-60*
ESP-1
ESP-2
ESP-4
WESP"
FF": FF/WS«
PS«
SD/FF; SD/C/FF
DS/FF
ESP-1/WS; ESP-1/PS
ESP-4/WS; ESP-4/PS
VS-20/WS"
WS/IWS'
WESP/VS-20/IWSa
C/DS/ESP/FF; C/DS/C/ESP/FF
SD/C/ESP-1
Ba, Be
50
90 '
98
95
97
99
97
95
95
99
98
96
99
97
95
99
99
99
Ag ._ .
50
90
98
95
97
99
97
95
95
99
98
96
99
97
95
99 ,
99
99
Pollutant
Cr
50
90
98
95
97
99
96
95
95
99
98
96
99
97
95
98
99
98
, As,Sb,Cd,Pb,TI
40
20
40
80
85
90
95
90
95
95
98
90
95
96
95
97
99
95
Hg°
,30 •
20
40
o
o
o
fin
\j\j
50
flfl
O\J
90
en
ou
80
fl^
ou
80
RR
oo
90
Q8
30
85
* Fluo gases ore assumed to have been preceded (usually in a quench). If gases are not cooled adequately, mercury recoveries will diminish, as will cadmium and
arsenic recoveries to a lesser extent. •
APCD codes
C « Cyclone
WS . We» scrubber
including: sieve tray tower, packed tower, bubble
cap tower
PS • Proprietary wet scrubber design (high efficiency
PM and gas collection)
VS-20 . Venturi scrubber, ca. 2-30!n W.G.
VS-60 - Venturi scrubber, ca. > 60 in W.G.
ESP-1 • Electrostatic precipitate; 1 stage
ESP-2 - Electrostatic precipitatar; 2 stages
ESP-4 m Electrostatic preclpitator; 4 stages
WESP - Wet electrostatic preclpitator
IWS » Ionizing wet scrubber
OS - Dry scrubber
FF » Fabric fitter (baghouse)
SD - Spray dryer (wet/dry scrubber)
Source: Carroll, 1992.
P
P
18
-------
furans are shown in Table 3-10. The effectiveness of various
APCDs for acid gas control is shown in Table 3-11.
Area source controls also can be applied in theory to achieve
any desired removal or control efficiency. Once again, a practical
trade-off exists between removal or control efficiency and cost.
The control efficiencies of area source controls will vary more
widely than those for point source controls. This variability is
true both from site to site and over time at a given site. Control
efficiencies often may be as low as 50%. Comparative costs per
given area for various area source controls as typically applied
are shown in Table 3-12.
Table 3-10. Spray Dryer Control of Selected Organic Pollutants for Hazardous Waste Incinerators
Control system (% removal)
Compound
Dioxins
tetra ODD
penta ODD
hexa ODD
hepta ODD
oota CDD
Furans
tetra CDF
penta CDF
hexa CDF
hepta CDF
octa CDF
SD + ESP
48
51
73
83
89
65
64
82
83
85 •
SD + FF @ high temperature
<52
75
93
82
NA
98
88
86
92
NA
SD + FF @ low temperature
>97
>99.6
>99.5
>99.6
>99.8
>99.4
>99.6
>99.7
>99.8
>99.8
Source: U.S. EPA, 1987.
Table 3-11. Effectiveness of Acid Gas Controls (% Removal) for Hazardous Waste Incinerators
Control system
Dry injection + fabric filter (FF)
Dry injection + fluid-bed reactor + ESP
Spray dryer + ESP
(recycle)
Spray dryer + fabric filter
(recycle)
Spray dryer + dry injection + ESP of FF
Wet scrubber
Spray dryer + wet scrubber(s) + ESP or FF
Temperature1 °C
160-180
230
—
140-160
—
140-160
200
40-50
40-50
HCI
80
90
95+
(95+)
95+
(95+)
95+
95+
95+
Pollutant
HF
98
99
99
(99)
99
(99)
99
99
99
SO,
50
60
50-70
(70-90)
70-90
(80-95)
90+
90+
90+
1 The temperature at the exit of the control device.
Source: U.S. EPA, 1987.
19
-------
Table 3-12. Costs of Area Source Controls
Control
Clay ~
Soil
Wood ohfps. plastic net
Synthetic cover
Short-term foam
Long-term foam
Wind screen
Water spray
Additives Surfactant
Hygroscopic salt
BituVAdhes.
Material cost ($/m2
except as noted)
$4.15
1.33
0.50
4.40
0.04
0.13
40./m
$0.001 (varies)
0.65
2.58
0.02
Comments
Covers, mat, and membrane
Assume 6" deep; does not include soil transport
Chip cost vary with site
Assume 45 ml thickness
Assume 2.5" thick, $0.7/m3 foam
Assume 1 .5" thick, $3.3/m3 foam
Per linear meter
Assuming municipal water cost of $1/1 ,OOOL. Water
requires constant re-application. Water truck rental:
$500/wk.
Costs vary with chemical used
33 References
Carrol], J.P. Screening Procedures for Estimating the Air
Impacts of Incineration at Superfund Sites. EPA-450/1 -
92-003. Research Triangle Park, NC. February 1992.
Heskcth, H.E. Air Pollution Control - Traditional and Hazard-
ous Pollutants. Technomic Publishing Co., Lancaster,
PA. 1991.
U.S. EPA. Waste Incineration and Emission Control Tech-
nologies. EPA/600/D-87/147-5 (NTIS PB87-208336)
July 1987.
20
-------
Chapter 4
Air Emissions-Related Information for Various
Remediation Technologies
This chapter contains background information pertaining to air
emissions and their control for various remediation technologies.
References to general sources of information on process design,
emissions estimation, and costs are provided at the end of the
chapter. The majority of the text is taken directly from a recent
EPA publication on air emissions from the treatment of contami-
nated soil that summarizes existing information on air emissions,
controls, and cost for various remediation technologies (Eklund,
et al., 1992c). A second source of information is a recent EPA
publication giving emission factors for remediation technologies
(Thompson, et al., 1991).
For each remediation technology, a typical remediation sce-
nario for Superfund sites is given followed by a discussion of
potential air emissions. References are given for emission
estimation procedures, and applicable control technologies are
identified.
4.1 Materials Handling
Materials handling covers such activities as excavation, dump-
ing, grading, short-term storage, and sizing and feeding soil or
waste into treatment processes. Information on equipment and
operating practices is available in Church, 1981 and U.S. EPA,
199la. The discussion below primarily addresses excavation.
4.1.1 Typical Remediation Scenario for
Superfund Sites
Excavation and removal of soils contaminated with fuels is a
common practice at Superfund sites. Excavation and removal
may be the selected remediation approach or it may be a neces-
sary step in a remediation approach involving treatment. If
removal is the preferred approach, the excavated soil typically is
transported off-site for subsequent disposal at a landfill. If the
soil contains large amounts of fuel or highly toxic contaminants,
the soil may need to be treated off-site prior to final disposal.
Excavation activities are also typically part of on-site treatment
processes such as incineration, thermal desorption, batch bio-
treatment, land treatment, and certain chemical and physical
treatment methods. The soil is excavated and transported to the
process unit and the treated soil typically is put back into place on
the site.
The rate of materials handling operations at Superfund sites
tend to be controlled by factors such as safety concerns, storage
capacity or treatment capacity, rather than being limited by the
operational capacities of the equipment that is used. For these
reasons, actual materials handling rates tend to be far below
typical handling rates at construction sites (Church, 1981). Typi-
cal scenarios for excavation at Superfund sites are given in Table
4-1.
Table 4-1. Example Scenarios for Excavation of Contaminated Soil
Scenario
Parameter
Units Small Medium Large
Soil moved per scoop
No. scoops per hour
Total volume of soil moved
Excavation pit:
dimensions:
area
Storage pile:
dimensions:
area
m3
#/hr
m3/hr
m
m2
m
m2
1
50
50
10x5x1
50
5x5x2
65
2
75
150 •
10x15x1
150
5x10x3
140
• 4
60
240
10x12x2
120
8x10x3
188
Source: Eklund, etal., 1991 a.
Since digging soil and immediately transferring it directly to
transport vehicles or treatment systems is rarely feasible or
efficient, soil will be handled several times. In most cases, soil
will be excavated and placed into a temporary holding area and
then handled one to two more times on-site. Elevated levels of
VOC and PM emissions are possible each time the soil is handled.
4.1.2 Potential Air Emissions
The exchange of contaminant-laden soil-pore gas with the
atmosphere when soil is disturbed and diffusion of contaminants
through the soil both contribute to VOC emissions from excava-.
tion. Multiple potential emission points exist for each of the
various soils handling operations. For excavation, the main
emission points of concern are emissions from:
• Exposed waste in the excavation pit;
• Material as it is dumped from the excavation bucket; and
• Waste/soil in short-term storage piles.
In addition, emissions of VOC, particulate matter, nitrogen
oxides, etc. will also occur from the engines of the earth-moving
equipment. While these emissions will not require any additional
control devices (beyond those provided by the manufacturer), the
21
-------
equipment emissions should be considered when evaluating any
air monitoring data.
The magnitude of VOC emissions depends on a number of
factors, including the type of compounds present in the waste, (he
concentration and distribution of the compounds, and the poros-
ity and moisture content of the soil. The key operational param-
eters are the duration and vigorousness of the handling, and the
size of equipment used. The longer or more energetic the moving
and handling, the greater likelihood that organic compounds will
be volatilized. The equipment size influences volatilization by
affecting the mean distance a volatilized molecule has to travel to
reach the air/solid interface at the surface of the soil. In general,
the larger the volumes of material being handled per unit opera-
tion, the lower the percentage of VOCs that are stripped from the
soil. Control technologies for large area sources such as excava-
tion are relatively difficult to apply and are often much less
effective than controls for point sources.
Paniculate matter (PM) em issions will depend primarily on the
particle size distribution of the soil, its moisture content, the wind
speed, and the operating practices that are followed. The longer
or morecnergetic the moving and handling, the greater likelihood
that PM emissions will occur.
The success of excavation and removal for a given application
depends on numerous factors with the three key criteria being: 1)
the nature of the contamination; 2) the operating practices fol-
lowed; and 3) the proximity of sensitive receptors. Each of these
criteria is described below.
The magnitude of emissions from soils handling operations
will vary with the operating conditions. Add-on control technolo-
gies are available for minimizing VOC and PM emissions, but
they arc relatively ineffective and costly to implement. Control of
emissions can also be achieved by controlling the operating
conditions within preset parameters. The rate of excavation and
dumping, the drop height, the amount of exposed surface area,
the length of time that the soil is exposed, the shape of the storage
piles, and the dryness of the surface soil layers will all influence
the levels of VOC and PM emissions. Large reductions in
emissions can be achieved by identifying and operating within
acceptable ranges of operating conditions.
Since some release of volatile contaminants is inevitable
during excavation and removal unless extreme measures are
taken (e.g., enclose the remediation within a dome), the proxim-
ity of downwind receptors (i.e., people) will influence whether
excavation is an acceptable option. Excavation of contaminated
areas that abut residential areas, schoolyards, etc. may require
more extensive controls, relocation of the affected population, or
remediation only during certain periods (e.g., summertime for
school sites).
4.13 Emission Estimation Procedures
Relatively limited VOC emissions or emission rate data for
excavation are available. The process of measuring emission
rates from dynamic processes, such as excavation, is difficult and
costly, and has rarely been attempted. The factors that govern
emissions from materials handling are very complex. During
excavation, for example, the physical properties of the soil that
control the vapor transport rate (e.g., air-filled porosity) are
changing with time and the concentration of contaminants may
be rapidly decreasing.
Predictive equations for estimating emissions from excavation
and dumping are under development (Eklund, et al., 1992a).
These models are based on estimating emissions from diffusion
through the soil and from the loss of saturated pore-space gas to
the atmosphere. The predictive equations require assumptions
about the size of each scoop of soil, the dimensions of the soil
scoops and the excavation pit, and the shape of the soil after it is
dumped. Further assumptions are required about the air and soil
temperatures and the length of time that dumped soil is exposed
before it is covered with more soil or with an emissions barrier.
The equations generally predict high levels of emissions. For dry,
porous soils containing low ppb levels of contaminants, most or
all of the more volatile VOCs are assumed to be lost to the
atmosphere during soils handling. For sites with moist soils and
ppm levels of contaminants, however, only 5 to 10% of the VOCs
are assumed to be emitted to the atmosphere during each handling
step. More field measurement data are needed to validate these
assumptions. .
Soils handling operations such as excavation substantially
. increase VOC emission rates from contaminated soil over base-
line rates (Eklund, et al., 1989). The increase in emissions is
typically a factor of ten or more, and the increased emission rate
decays exponentially back to near the baseline rate over short
time periods (e.g., 4 days). A database of baseline emission rate
measurement data (Eklund, et al., 1991a) is available. Other
estimation procedures and field data are summarized in Eklund
etal., 1992c.
Particulate matter emissions can be estimated using the empiri-
cal equations in Cowherd, et al., 1988. Emissions for topsoil
removal, earth moving, and truck haulage are reported to range
from about 1 to 6 kilograms of paniculate matter per vehicle per
kilometer traveled.
4.1.4 Identification of Applicable Control
Technologies
A number of methods are available for controlling VOC and
paniculate matter emissions from soils. In general, any method
designed primarily for paniculate control will also reduce VOC
emissions and vice versa. Compared to point source controls,
VOC emission controls for excavation and. other area sources are
difficult to implement and only moderately effective. Controls
such as water sprays or foams will alter the percent moisture, bulk
density, and average heating value of the soil and may affect
treatment and disposal options.
VOC emission controls for soil area sources are discussed in
Section 7 and include:
• Covers and physical barriers;
• Temporary and long-term foam covers;
• Water sprays;
• Water sprays with additives;
22
-------
• Operational controls; •
• Complete enclosures;
• Wind screens; and ,
• Collection hoods.
4.2 Thermal Desorption Treatment
Mobile process units designed for soil remediation and the use
of asphalt kilns for soil remediation are discussed in this section.
The best currently available source of information is an engineer-
ing bulletin prepared by the U.S. EPA (U.S. EPA, 1990a).Design
and operating information for thermal desorption systems are
given in an EPA Guidance Document being prepared (Troxler, et
al., 1992). Air emissions and cost data have been summarized
(Eklund, et al., 1992c).
4.2.1 Typical Remediation Scenario for
Superfund Sites
In the thermal desorption process, volatile and semi-volatile
contaminants are removed from soils, sediments, slurries, and
filter cakes. This process typically operates at temperatures of
200°-1000°F but often is referred to as low temperature thermal
desorption to differentiate it from incineration. At these lower
temperatures, thermal desorption promotes physical separation
of the components rather than combustion. Contaminated soil is
removed from the ground and transferred to treatment units,
making this an ex situ process. Direct or indirect heat exchange
vaporizes the volatile compounds producing an off-gas that
typically is treated before being vented to the atmosphere. After
it is excavated, the waste material is screened to remove objects
greater than 1.5" in diameter (de Percin, 1991 a). In general, three
desorber designs are used: an indirectly fired rotary dryer,
internally heated screw augers, or a fluidized bed (de Percin,
1991 b). The treatment systems include both mobile process units
designed specifically for treating soil and asphalt kilns, which
can be adapted to treat soils. Typical characteristics of the
processes and off-gas streams for mobile units and rotary drum
units at asphalt plants are summarized in Table 4-2.
Because thermal desorbers, in some cases, may operate near or
above 1000°F, some pyrolysis and oxidation may occur in
addition to the vaporization of water and organic compounds.
Collection and control equipment such as afterburners, fabric
filters, activated carbon, or condensers prevent the release of the
contaminants to the atmosphere. Thermal desorbers can create up
to seven process residual streams: treated soil, oversized media
rejects, condensed contaminants, water, paniculate control dust,
clean off-gas, and spent carbon (de Percin, 1991b).
4.2.2 Potential Air Emissions
Thermal desorbers effectively treat soils, sludges and filter
cakes and remove volatile and semi-volatile organic compounds
from the material. Some higher boiling point substances such as
polychlorinated biphenyls (PCBs) and dioxins also may be
removed and thus be present in the off-gas. Inorganic compounds
are not easily removed with this process, although some rela-
tively volatile metals such as mercury may be volatilized. Tem-
peratures reached in thermal desorbers generally do not oxidize
metals (de Percin, 1991 a). VOC removal is enhanced if the soil
contains 10-15 percent moisture prior to treatment since water
vapor carries out some VOCs.
Point sources of air emissions from thermal desorption vary
widely with each process. The stack of an afterburner vents
combustion products, as does a fuel-fired heating system if the
combustion gases are not fed into the desorber. The fuel-fired
heating system typically operates with propane, natural gas or
fuel oil. If emissions controls consist of abaghouse, scrubber, and
vapor phase carbon adsorber, the stack will vent small concentra-
tions of the original contaminants, as well as products of any
chemical reactions that might occur. Relative to incineration, the
volume of off-gas from the treatment chamber may be smaller,
there is less likelihood of creating dioxins and other oxidations
products, and metals are less likely to partition to the gas-phase
(de Percin, 199 la).
Fugitive emissions from area sources may contribute signifi-
cantly to the total air emissions from a remediation site. Probably
the largest source is excavation of the contaminated soil. Other
sources may include the classifier, feed conveyor, and the feed
hopper. Fugitive emissions from the components of the thermal
desorption system and controls are possible as well. Emissions
may also emanate from the waste streams such as exhaust gases
from the heating system, treated soil, paniculate control dust,
untreated oil from the oil/water separator, spent carbon from
liquid or vapor phase carbon adsorber, treated water, and scrub-
ber sludge.
4.2.3 Emission Estimation Procedures
The volatile and semi-volatile contaminants under remedia-
tion are the species emitted if no destruction or other chemical
treatment has taken place. The sources emitting these VOC's
may include excavation, soil handling, classifier, oversize ob-
jects rejected by the classifier, feed conveyor, feed hopper,
control stack, and fugitive emissions from the entire thermal
desorption system and from waste streams. Combustion products
are emitted when a destructive control such as an afterburner is
used and also when the heating system is fuel-fired. In some
cases, pyrolysis occurs to a certain degree in the dryer so products
from these reactions may also be emitted. If scrubbers are used to
treat VOC's or combustion gases, then an additional category of
species is emitted.
Theoretical models based on fundamental principles .have
been proposed for predicting the evolution of volatile com-
pounds from soil in the thermal desorption process, but these
models are not practical for use as apredictive tool (Lighty, et al.,
1990). In practice, an assessment of the applicability of thermal
desorption for a given site will not be based on modeling
calculations, but will be based on the types of contaminants
present in the soil, the physical properties of the soil, and the
results of any bench-, pilot- or full-scale test runs. In most cases,
the process conditions such as temperature and residence time in
the desorber can be modified to yield the desired removal
efficiency, though heavier weight compounds such as those in
No. 6 fuel oil may present problems for systems with relatively
low operating temperatures. The cost to operate at these process
conditions, however, will dictate whether thermal desorption is
competitive with other remediation options.
A mass balance equation to estimate an emission rate for a
volatile compound leaving the desorber using removal efficien-
cies obtained from test runs is given in Eklund, et al., 1992c. This
23
-------
Tabla 4-2. Comparison of Features of Thermal Desorption and Off-Gas Treatment Systems
Estimated number of systems
Estimated number of contractors
Mobility
Typical site size (tons)
Soil throughput (tons/hr)
Maximum sol! feed size (inches)
Heat transfer method
Soil mixing method
Discharge soil temperature (°F)
Soil residence time (minutes)
Thermal desorber exhaust gas
temperature (°F)
Rotary dryer
40-60
20-30
Fixed and mobile
500-25,000
10-50
2-3
Direct
Shell rotation and lifters
300-600 •
600-1, 200 b
3-7
500-850 a
800-1,000"
Gas/solids flow Co-current or counter-current
Atmosphere
Afterburner temperature (°F)
Maximum thermal duty (MM Btu/hr) •
Heatup time from cold condition (hrs)
Cool down time from hot condition (hrs)
Total petroleum hydrocarbons
Initial concentration (mg/kg)
Final concentration (mg/kg)
Removal efficiency (%)
Caitoon steel materials of construction
AHoy materials of construction
Hot oil heal transfer system
Motion salt heat transfer system
Electrically heated system
Not used on all systems
Oxidative
. 1,400-1,800
15-85
0.5-1.0
1.0-2.0
800-35,000
<1 0-300
95.0-99.9
Asphalt plant
aggregate dryer
100-150
No estimate
: Fixed
0-10,000
2^-100
2-3
Direct
Shell rotation and lifters
300-600
3-7
500-850
Co-current or counter-current
Oxidative
1,400-1,800'
50-125
0.5-1.0
1.0-2.0
Not reported
Not reported
Not reported
Thermal screw
18-22
g
Mobile
500-5,000
3-15
1-2
Indirect
Auger
300-500 "
600-900 "
1 ,000-1 ,600 •
30-70
300
N/A
Inert
Generally not used
7-10
Not reported
Not reported
60-50,000
ND-5,5000
. 64-99
Conveyor furnace
•|
Mobile
500-5,000
5-10
1-2
Direct
Soil agitaters
300-800
3-10
1 ,000-1 ,2000
Counter-current
• Oxidcitivs
1,400-1,800
10
0.5-1.0
Not reported
5,000
<10.0
>99.9
Total duly of thermal desorber plus afterburner
Source: Eklund, et al., 1992c.
equation does not include emissions from excavation or other
handling of contaminated soil nor does it include fugitive emis-
sions from the desorber system or from liquid and solid phase
waste streams. Neither are'combustion gases emitted from the
heating system and exhaust gases from afterburners included in
this estimation method. However, tabulated emissions data for a
number of thermal desorption systems are included. Most of the
studies cited include data about contaminant concentrations in
the soil directly before and after treatment, data which can yield
information aboutpoint source air emissions from the desorption
process itself. These studies do not include the change in concen-
tration before and after excavation due to volatilization. Simi-
larly, little data are available on fugitive emissions from the parts
of the process that do not include the desorption chamber and
from the other waste streams.
4.2.4 Identification of Applicable Control
• Technologies
The control of volatile organic emissions is crucial to the
overall success of thermal desorption remediation of contami-
nated soils. Because the process uses physical separation driven
by heat, the vaporized contaminants would simply be transferred
from one medium (soil) to another (air) if no emission controls
were employed. The types of controls include both destruction
and separation technologies. Typically two to six controls in
series are chosen to suit the specific .VOC contaminants present
and the other pollutants of concern. Liquid phase and solid waste
streams are usually treated on site or stored for subsequent off-
site treatment.
Asphalt kilns will have similar air emission control devices as
for mobile thermal desorption units, except that no VOC controls
typically are employed and the air flowrates are higher requiring
some differences in design parameters.
24
-------
Off-gases from the desorber typically pass first through a
particulate control device. Particles that become entrained in the
off-gas stream may be removed with:
• Cyclones;
• Venturi scrubbers; or
• Fabric filters.
Collected particulates usually are returned to the incoming
waste stream and retreated with the soil.
VOC control devices include:
• Condensers;
• Fume incinerators; and
• Carbon adsorption.
Condensers serve to remove VOCs while fume incinerators
(i.e., afterburners) destroy the VOCs. Carbon adsorption is
sometimes added to either of these primary VOC control meth-
ods as a final polishing step. Exhaust gases from destruction
controls may be treated in an acid gas scrubber. Gases are first
cooled to saturation temperature then passed through a packed-
bed absorber or spray tower where acidic gases are neutralized
with caustic (sodium hydroxide) solution or dissolved into water.
Other emissions control techniques include rerouting combus-
tion off-gases to a dryer, using treated water for dust control, and
passing an inert gas such as nitrogen through the desorber as
explosion prevention. Ultraviolet rays have been used to destroy
dioxin in the condensate from the thermal desorption of contami-
nated soils (Helsel and Thomas, 1987).
4.3 On-Site Incineration
4.3.1 Typical Remediation Scenario for
Superfund Sites
A broad range of technologies fall into the category of thermal
incineration. The most common incineration technologies in-
clude liquid injection, rotary kiln, and multiple hearth. The most
common design for the remediation of contaminated soils, how-
ever, are rotary kilns. Remediation by on-site thermal destruction
using a transportable incinerator is discussed in this section.
Shipment of contaminated soils and wastes to a larger, permanent
off-site unit may also be an option for a given site, but system
design and selection of control options is not generally a consid-
eration. Although incineration is a well-established technology,
the evolution of mobile or transportable incinerators is a rela-
tively new development. The literature on incineration is very
extensive. The best sources of information on air emissions from
incineration are two recent reviews (Oppelt, 1987) and (Eklund,
etal., 1989).
In broad terms, thermal destruction of hazardous waste is an
engineered process in which controlled combustion is used to
reduce the volume of an organic waste material and render it
environmentally safe. Incineration is a flexible process capable
of being used for many waste types including solids, gases,
liquids, and sludges.
A typical system includes the waste feed system, primary and
(in most cases) secondary combustion chambers, and exhaust gas
conditioning system. The largest part of the waste destruction
usually takes place in the primary combustion chamber. As
mentioned earlier, for contaminated soils this chamber is usually
a rotating kiln. Gases formed in the primary combustion chamber
Table 4-3a. Properties of Off-Gas from Combustion Chamber from
On-Site Incineration Systems
Parameter
Units
Value
Air flowrate
Temperature
Oxygen content
Pressure drop
ACFM
°F
%
In. H20
30-50,000
1,400-1,800
3
10-15
Table 4-3b. Hazardous Waste Incinerator Emissions Estimates
EPA*
conservative
estimated
efficiencies
Particulate
matter
Hydrogen chloride
(HCI)
Sulfur dioxide (SO2)
Sulfuric Acid (H2SO4)
Arsenic
Beryllium
Cadmium
Chromium
Antimony
Barium
Lead
Mercury
Silver
Thallium
PCDD/PCDF"
99+%
• —
—
—
95
99
95
99
95
99
95
85-90
99
95
—
Typical
actual
control
efficiencies
99.9+%
99+
' 95+
99+
99.9+
99.9
99.7
99.5
99.5
99.9
99.8
40-90+
99.9+
99+
90-99+
Typical range
of emissions
rates
0.005-0.02
gr/dscf
10-50mg/NM3
30-60
2.6
1-5ng/Nm3
<0.01-0.1
0.1-5
2-10
20-50
10-25
10-100
10-200
1-10
10-100
1-5ng/NM3
* Based on spray dryer fabric filter system or 4-field electrostatic
precipitator followed by a wet scrubber.
** Total all cogeners
Source: Donnelly, 1991.
are then routed to a secondary combustion chamber, or after-
burner, where any unburned hydrocarbons or products of incom-
plete combustion can be oxidized. Typical off-gas properties for
on-site incineration systems are summarized in Table 4-3.
4.3.2 Potential Air Emissions
The air emissions associated with full-scale thermal destruc-
tion are primarily stack emissions of combustion gas. However,
some additional evaporative emissions may occur from equip-
ment leaks and waste handling. Full-scale, off-site incineration
units may vent all emissions from waste handling and transfer
activities to the combustion chamber as make-up air. The air
emissions for on-site incinerators are similar to off-site units,
except that on-site waste handling activities have a greater
likelihood of being uncontrolled. Stack heights for transportable
units may be in the range of 40-100 ft. (Good engineering practice
stack height will not apply unless large structures are present.)
25
-------
Tire fugitive emissions sources associated with thermal treatment
likely will be ground-level.
Emissions from incineratois include: undestroyed organics,
metals, paniculate matter, nitrogen oxides (NOx), carbon monox-
ide (CO), acid gases, and products of incomplete combustion
(PICs). The cause of each of these pollutants and typical levels of
emissions are discussed in Eklund, et al., 1992c. Fugitive emis-
sions associated with excavation, storage, and handling of the
feed material also must be considered when assessing potential
air impacts from incineration.
The wide variety in design and operation of incinerators makes
it difficult to predict air emissions. However, extensive research
has been done to determine the range of unburned hydrocarbon
and PIC emissions that can be expected from full-scale incinera-
tors. In general, incinerators treating wastes must achieve a
required destruction and removal efficiency of at least 99.99%
for RCRA wastes and 99.9999% forPCB- or dioxin wastes. The
remaining 0.01 % or 0.0001 % of the waste can be assumed to pass
through the system uncombusted (Eklund, et al., 1989). How-
ever, in addition to unbumed hydrocarbons there may be some
additional reactions in the combustion process that may produce
a number of simpler organic compounds, called PICs. PICs may
include dioxin, formaldehyde, and benzo(a)pyrene and other
polynuclear aromatic hydrocarbons. Possible causes of PIC
emissions include low temperatures due to quenching, residence
time short circuits due to nonplug flow and/or unswept recesses,
and locally high waste/oxygen concentration ratios due to poor
microscale mixing.
The metals introduced to the incinerator via the waste feed
stream are not destroyed. Depending on their boiling point, they
can either be volatilized or remain as solids. Volatilized metals
will exit the stack as a gas, condense or adsorb onto particles in
the stack gas stream, or be captured in the wet scrubber. Metals
associated with paniculate matter (PM) will be captured in the
PM control device. Non-volatilized metals can be fluidized and
swept up into the combustion gas or leave the incinerator in the
bottom ash.
The waste feed, auxiliary fuel, and combustion air can all serve
as sources for paniculate emissions from an incineration system.
Paniculate emissions may result from inorganic salts and metals
which cither pass through the system as solids or vaporize in the
combustion chamber and recondense as solid particles in the
stack gas. High molecular weight hydrocarbons may also con-
tribute to paniculate emissions through several possible mecha-
nisms. RCRA requirements for paniculate emissions call for a
limit of 0,08 grains/dscf corrected to 7% O2 (U.S. EPA, 1990b).
Achieving high levels of destruction of organic wastes is
directly related to combustion chamber temperature: the higher
the temperature, the greater the DRE of organics (for a given
combustion gas residence time and degree of turbulence). Unfor-
tunately, the fixation of nitrogen and oxygen to form NO also
increases with combustion temperatures above 1600°F."NO
emissions caused by this mechanism are referred to as thermal
NO,. Also if bound nitrogen atoms are present in the waste (e.g.,
amines), additional NOX emissions, called fuel N0x, will be
formed. In such cases, two stage combustion or emissions con-
trols may be needed. Carbon monoxide emissions are generally
low (<25 ppmv) in incinerators due to the high operating tem-
peratures and excess oxygen maintained in the process.
Hazardous waste incineration also will produce acid gases.
These include oxides of sulfur (SO ), and halogen acids (HC1
HF,andHBr).
4.3,3 Emission Estimation Procedures
Simple mass balance equations for estimating incineration
emissions with an assumed DRE have been published (Eklund,
et al., 1989 and IT, 1992). The equations cover VOCs, PM,
metals, and acid gases. Both of these documents summarize
typical operating rates, control efficiencies, etc.
Models have been reported for direct fired, high temperature
rotary kiln systems that predict the temperature of the solid bed
and kiln exit gas as a function of measurable physical parameters
such as kiln rotational speed, burner firing rate, soil feed rate, etc.
(Troxler, et al., 1992). These models theoretically could be
combined with thermal stability data and oxygen content of the
kiln gas data to predict the destruction efficiency of incinerators.
Emissions of PICs, both the amount and the type, will vary
greatly from unit to unit depending on design and waste feed.
Data are unavailable to generate emission factors.
The production of acid gases (HC1, SO2, HBr, and HF) is
determined by the respective chlorine, sulfur, bromine, and
fluorine contents in the waste and fuel feed streams. The concen-
trations of these elements range widely amongst different wastes;
consequently, the resulting acid gas emissions also will show
wide variability.
NOx is usually only a concern for wastes with high nitrogen
content. Typical NOx emissions for an incinerator may be on the
order of 100-200 ppmv (dry basis), or expressed on a fuel basis
0.12-0.33 Ibs per MMBtu (Eklund, et al., 1992c). CO emissions
from incinerators also are not considered a major problem since
most systems are designed to be fired with excess air (i.e. oxygen
rich) to ensure complete combustion of organic material to
carbon dioxide. Vendors typically guarantee CO emissions less
than 100 ppmv (dry basis) and actual measured CO levels are
often lower.
4.3.4 Identification of Applicable Control
Technologies
Unlike other soil remediation technologies, incineration, which
converts organics into carbon dioxide and water, does not require
additional add-on VOC controls. However, additional controls
are usually required to reduce emissions of acid gases, paniculate
matter (PM), and metals. After the combustion gases leave the
incinerator, they may be routed through a variety of air pollution
control devices including gas conditioning, paniculate removal,
and acid gas removal units. Gas conditioning is accomplished
with equipment such as waste heat boilers or quench units.
Typical paniculate matter removal devices include:
• Venturi scrubbers;
• Wet electrostatic precipitators;
26
-------
• Ionizing wet scrubbers; and
• Fabric filters.
Acid gas removal units include:
• Packed, spray, or tray tower absorbers;
• Ionizing wet scrubbers;
• Wet electrostatic precipitators; and
• Spray dryer adsorbers.
The absorbers generally will use a caustic solution. Units used
to treat soil contaminated with halogenated solvents generally
will be required to meet RCRA requirements governing HC1
emissions.
4.4 Soil Vapor Extraction
A number of reports and articles have been published recently
that provide useful information regarding soil vapor extraction
(SVE) systems. The best single source of information on SVE
design and operation is a recent EPA report (Pedersen and Curtis,
1991). Another key reference is a recent overview paper (Johnson,
et al., 1990). Air emissions from SVE systems are addressed in
Eklund, et al., 1992b and Eklund, et al., 1992c.
4.4.1 Typical Remediation Scenario for
Superfund Sites
Soil vapor extraction is one method used for the treatment of
soil contaminated with volatile hydrocarbons. The process some-
times is referred to as soil venting, vacuum extraction, aeration,
or in-situ volatilization. In general terms, soil vapor extraction
removes volatile organic constituents from contaminated soil by
creating sufficient subsurface airflow to strip contaminants from
the vadose (unsaturated) zone by volatilization. As the contami-
nant vapors are removed, they may be vented directly to the
atmosphere or controlled in a number of ways. Among the
relative advantages of SVE over other remediation approaches
are that no materials handling operations are necessary and the air
emissions are released from a point source and thus can be
controlled readily.
Soil vapor extraction has been used widely to remediate sites
contaminated with gasoline or chlorinated solvents (e.g., TCE).
It is also used sometimes to minimize migration of vapors into
structures or residential areas during other types of remediation.
By its nature, SVE is an bn-site, in-situ treatment method. SVE
often is used in conjunction with or following other remedial
measures such as excavation of subsurface waste bodies, re-
moval (pumping) of any hydrocarbon lens that is present, or air
stripping of contaminated ground water.
Typical SVE systems include extraction wells, monitoring
wells, air inlet wells, vacuum pumps, vapor treatment devices,
vapor/liquid separators and liquid phase treatment devices (if
contaminated water is extracted in the process). Example sce-
narios for SVE systems at Superfund sites are given in Table 4-
4. An option sometimes employed is to introduce the air at the air
inlet well into the saturated zone (i.e. groundwater table). This
technique, referred to as air sparging, acts to strip some of the
volatile and semi-volatile compounds from the ground water.
Table 4-4. Example Scenarios for SVE Based on Size of System
Scenario
Parameter
Exhaust gas
flowrate
Exhaust gas
velocity
Exit gas temp.
-No controls
-Carbon
-Catalytic
oxidation
Stack height
Stack diameter
Units
m 3/min
cfm
m/sec.
°C
°C
°C
m
m
Very
Small
1.4
. 50
3.0
50
25
320
3.0
0.10
Small
14
500
7.4
50
25
320
4.6
0.20
Medium
85
3,000
12.5
50
,25
320
7.6
0.38
Large
425
15,000
14.2a
50
25
320
9.1
0.46
• Assume three adjacent stacks each handling 5,000 cfm. The flow is split to
lower the velocity of the exiting gas to typical design levels to minimize
corrosion of the stack.
Source: Eklund, et al., 1992b.
Another option is to heat the air entering the inlet wells to enhance
the volatilization of less volatile, higher molecular weight con-
taminants, such as diesel fuel.
Steam-assisted SVE is another option that has been used for
improving the removal efficiency of VOCs and SVOCs. The
steam can be injected via inlet wells. A mobile treatment system
also has been demonstrated that treats blocks of soil (7 x 4 ft. x
up to 20 ft. deep) at a rate of about 3 nvVhr (U.S. EPA, 1991b).
Augers are used to stir the soil as steam is injected. The treated
area is covered by a shroud (ducted hood) and all vapors are
extracted and sent to control devices (see Section 4.4.3).
4.4.2 Potential Air Emissions
The contaminants removed from the soil by SVE systems and
hence present in the off-gas generally have vapor pressures
greater than 1.0 mm Hg at 20°F. The tendency of the organic
contaminants to partition into water or to be adsorbed onto soil
particles also affects the off-gas composition, as do the com-
pounds' water solubility, Henry's Law constant, and soil sorp-
tion coefficient. The soil temperature affects each of these
variables and hence the rate of vapor diffusion and transport. The
concentration of contaminants that are initially present will also
affect their relative partitioning between vapor and liquid phases,
and the amount that is solubilized or adsorbed. The time that the
contamination has been present is also an important factor, as
mixtures of contaminants will generally become depleted of their
more volatile components over time through volatilization. This
process, referred to as weathering, will tend to cause SVE to
become progressively less applicable as the site ages. It also
affects the operation of the SVE system, as the more volatile
components are typically removed first and the composition of
the vapors collected and treated varies over time.
27
-------
VOC loading rates in the off-gas can be 500-600 kg/day or
higher. The air emissions associated with soil vapor extraction
systems come primarily from the stack. Stack heights are typi-
cally 12-30 feet and usually only one stack is used. Additional
releases of volatile organics may occur from the treatment of any
contaminated water that is extracted. Fugitive emissions are
considered negligible due to the negative pressure throughout
most of the system.
_ Emissions include untreated volatile organics from the extrac-
tion process. Removal and emissions of semi-volatile organic
compounds will occur also, though with less efficiency than for
VOCs. Lesser amounts of air emissions associated with the
control system may occur also. Due to the variety of technologies
used, stack emissions may include products of incomplete com-
bustion, nitrogen oxides, particulate matter, carbon monoxide,
acid gases and any otherpossible products of these technologies.
Of primary concern, however, are the volatile organics emitted
from the point sources. Percent levels of carbon dioxide in the
off-gas may occur and would suggest that in-situ biodegradation
is occurring in conjunction with the S VE. The ambient air drawn
through the soil would raise the oxygen content of the soil-gas
and thus promote biodegradal;ion.
The emission rate of VOC compounds over time from continu-
ously operated SVE systems tends to show an exponential-type
decay curve. If the system is stopped and then restarted, however,
the VOC emission rate returns to near the original rate unless the
remediation is nearing completion. Apparently, shutting off the
vacuum allows the soil-gas equilibrium to become re-estab-
lished. Due to this behavior, the most efficient method of opera-
tion is to run the SVE system only for a part of each day or week,
i.e. operate in a "pulsed" mode.
4A3 Emission Estimation Procedures
The factors that govern vapor transport in the subsurface are
very complex and no practical, accurate theoretical models for
predicting emissions or recovery rates for SVE systems exist.
During operation of SVE systems, the vacuum that is applied to
the soil and the resulting pressure gradient is the dominant factor
in determining the flowrate of vapors. Subsurface vapor flow
equations based onDarcy's Law have been published thatpredict
the flowrate of vented gas, but these equations are not useful as
a predictive tool due to the large variability in air permeability
among and within sites (Johnson, et al., 1990). In practice, field
tests typically are performed to evaluate the potential effective-
ness of SVE for a given site. The field tests may be either a pilot-
scale demonstration of SVE or tests of the air permeability. This
information is used to determine the number of wells required to
remediate the site and the spacing of the wells, and also may yield
information about the off-gas stream to be treated.
A simple screening model is available based on historical
vaporextractionrates at sites where SVE systems have been used
(Eklund, et al., 1992b). The guidance given in the screening
model document encourages the user to provide site-specific
extraction rate and vapor concentration data, but conservative
default values also are provided.
4.4.4 Identification of Applicable Control
Technologies
As the vapors are removed from the soil they are either
discharged to the atmosphere or treated to reduce air emissions.
If the hydrocarbon content is high enough, direct combustion is
theoretically possible. However, because concentrations typi-
cally drop significantly during removal, natural gas or some other
fuel will be needed to maintain combustion. Also, for safety
reasons, dilution air typically is added to maintain the VOC
concentration below the lower explosive limit (LEL). In some
cases, the wells may be shut down for a period of time to allow
subsurface vapor pressures to re-equilibrate, thus yielding con-
centrations sufficient to sustain a flame. For lower levels of
hydrocarbons, catalytic oxidation may be effective. Carbon
adsorption systems are used often but they may be costly to
implement and generally are not acceptable for high-humidity
gas streams.
A recent survey indicates that the exhaust from about 50% of
SVE systems is vented directly to the atmosphere with no
controls (PES, 1989). The trend, however, is for VOC controls to
be required. For those systems with controls, the most viable
options are:
1) Activated carbon adsorption;
.2) Catalytic oxidation;
3) Thermal incineration;
4) Internal combustion engine; and
5). Miscellaneous control approaches.
The first three treatment options are the most commonly used
for large SVE systems such as those used at Superfund sites or
refineries. Internal combustion engines (ICE) are a common
choice for control of emissions for small systems such as those
used at small Leaking Underground Storage Tank (LUST) sites.
Theoretically, removal efficiencies of 95-99% for VOCs should
be achievable with any of these control options, though actual
control efficiencies in the field may be closer to 90%. Control
efficiencies for minor components of the off-gas stream may be
lower.
The miscellaneous control devices that potentially may be
applicable for controlling VOC emissions from SVE systems
include:
• Condensers;
• Packed bed thermal processors; and
• Biofilters.
Condensers using chilled water or other refrigerants can re-
move anywhere from 50 to 90% of VOCs from concentrated
streams (>5000 ppmv VOCs). Biotreatment requires time to
establish an active culture of microbes and careful control of soil
moisture, temperature, and air flow patterns to maintain the
efficiency of the microbial action.
The mobile treatment system using steam-assisted SVE has a
gas treatment system that consists of (U.S. EPA, 1991b):
28
-------
• Scrubber; '•.-'••--
• Cyclone separator; . • :
• Cooling system;
• Carbon adsorption system; and
• Compressors. . ,.
Particulate matter entrained in the process air stream is re-
moved in the scrubber. The air is then sent to the cyclone,
separator where water droplets are removed. Heat exchangers.are,.
then used to cool the gas and remove more water and organic,
compounds. The process air stream from the coojing'system is,
passed through activated carbon to remove VOCs. The cleaned
air stream is then sent to a compressor where it'is heated and
routed back to the treatment shroud. Air emissions from the
system are minimal in terms of both volume and mass loading.
4.5 Air Stripping of Contaminated Water
4.5.1 Typical Remediation Scenario for
Superfund Sites
Air strippers are used widely to remove chlorinated solvents
and other VOCs from contaminated ground water. Air stripping
is amass transfer process in which volatile contaminants in water
are evaporated (stripped) into air. The contaminated water is
introduced at the top of a packed-tower through spray nozzles and
allowed to slowly flow down through the column or tower. The
packing media acts to retard the water flow and increase the
effective surface area of the system. Air is introduced countercur-
rent to the direction of water flow. The saturated air containing
the volatiles is emitted from the top of the column or routed to a
control device. The treatment system may also contain wells,
separators, and vessels for treating inorganic contaminants. Ex-
ample scenarios for air stripping systems at Superfund sites are
given in Table 4-5.
4.5.2 Potential Air Emissions
The primary source of emissions from air stripping is the
stripper exhaust, and VOCs are the major pollutants of concern.
Table 4-5. Example Scenarios for Air Stripping
Typical Value
Parameter Units
Total influent L/min
liquid flowrate gpm
Column height m
Column diameter m
Exhaust gas rrrVmin
flowrate cfm
Stack height m
Stack diameter m
Structure m
dimensions
Exit gas velocity m/sec
Exit gas temperature °C
Ambient temperature °C
Gas/liquid ratio (vol/vol)
(G/L)
Stripping efficiency %
Small
570
150
7.6
1.2
29
1,020
8.5
.0.31
7.6x1.2x1.2
6.4
20
20
50
99+
Medium
2,840
750 •
9
3.6
140
5,000
10
0.61
9.0x3.6x3.6
8.0
2.0
20
50
99+
Large
5,700
1 ,500
14
3.6
285
10,000
15
0.91
13.0x3.6x3.6
7.3
20
20 •
50
99+
Source: Eklund, et al., 1991b.
For systems without control devices, the exhaust is vented
through a short stack, typically a (3-6 ft) pipe, at the top of the
column. For systems with control devices, the airflow from the
column usually is vented down to the control device at ground
level. A short stack (15-20 ft) is used after the control device.
In addition to the exhaust stack, other emission sources may
exist. Any place upstream of the air stripping tower where water
is in direct contact with the atmosphere, such as separators,
holding tanks, treatment tanks, or conduits, is an emission source.
Fugitive losses from pumps, valves, and flanges usually are not
significant due to the dilute nature of the water contamination.
The important parameters affecting the emission rate for a
given compound from an air stripping unit include: the concen-
tration of the contaminant in the influent to the stripper, the
influent flowrate, the stripping efficiency of the tower, and the
effectiveness of any control technologies that are in place. The
stripping efficiency will depend on a number of factors includ-
ing: the compound's Henry's Law constant, the type of packing
material in the tower, and the gas to liquid contact ratio within the
tower.
4.5.3 Emission Estimation Procedures
For a given liquid treatment rate, the magnitude of the uncon-
trolled air emissions from an air stripper are governed by the
effectiveness of the liquid-to-air mass transfer in the stripper. A
stripping efficiency of 100% for volatile organic compounds is a
reasonable, conservative assumption. A number of equations and
associated computer models are available to aid the system
designer in selecting the appropriate tower height, gas to liquid
ratio, packing material, etc. to optimize the mass transfer and
meet the performance goal in a cost effective manner (e.g., U.S.
EPA, 1990c).
A simple screening model is also available that estimates VOC
emissions using a mass balance approach, influent mass load-
ings, and the Henry's Law constants for the contaminants present
(Eklund, etal., 1991b).
4.5.4 Identification of Applicable Control
Technologies
The use of a control device can reduce emissions by one to two
orders of magnitude (i.e. 90-99% control). VOC control from air
strippers is possible by:
• Carbon adsorption; or
• Catalytic oxidation.
Thermal oxidation also could be used with air strippers if the
VOC concentration was sufficiently high, but no such use has
been found in the literature. In addition to these three VOC
control methods, flares have been used at some landfills for the
control of emissions from air strippers (Vancit, et al., 1987).
Emissions of PM, SVOC, and metals are all negligible, so no air
emission controls for these compounds are needed.
4.6 Solidification/Stabilization
General information about solidification and stabilization is
contained in Cullinane, etal., 1986. Extremely limited data exists
29
-------
about air emissions from these types of processes. The discussion
in this section was taken largely from Thompson, et al., 1991.
4.6.1 Typical Remediation Scenario for
Super/and Sites
Stabilization and solidification technologies are gaining in-
creased use as Superfund site remediation methods. The goal of
these processes is to immobilize the toxic and hazardous con-
stituents in the waste, usually contaminated soil or sludge. This
can be accomplished by several means:
1) Changing the constituents into an immobile (insoluble)
form;
2) Binding them in an immobile, insoluble matrix; or
3) Binding them in a matrix which minimizes the material
surface exposed to solvents (groundwater) which could
leach the hazardous constituents.
Several types of stabilization and solidification technologies
exist as alternatives forremedial action. A few of these processes
involve in-situ treatment; however, most generally require exca-
vation and other soil handling activities. Nearly all the commer-
cially available stabilization and solidification technologies are
proprietary.
Solidification and stabilization processes usually are batch
operations, but may be continuous and all follow the same basic
steps. Wastes are first loaded into the mix bin (wastes are
sometimes dried before addition to the bin), and other materials
for the solidification or stabilization are added. The contents of
the bin then are mixed thoroughly. After a sufficient residence
time, the treated waste is removed from the bin. Treatment rates
are 25-100 tons/hr for in-situ processes and up to 130 tons/hr for
cx-situ processes (U.S. EPA, 1990d and U.S. EPA, 1989a). The
material usually is formed into blocks and allowed to cure for up
to several days. The blocks then can be placed in lined excava-
tions on-site. Note: This description does not apply to in-situ
treatment methods, which use a variety of techniques (from
applied high voltage to injection of stabilizing agents) to immo-
bilize the contaminated waste in-place without excavation or
soils handling.
Typical raw materials used in stabilization processes include
fly ash, portland cement, cement kiln dust, lime kiln dust, or
hydrated lime. Other additives that may be used to solidify or
encapsulate wastes include asphalt, paraffin, polyethylene, or
polypropylene.
4.62 Potential Air Emissions
The primary source of air emissions from stabilization and
solidification processes is volatilization of organic contaminants
in the waste. Up to 90% of the VOCs are lost during mixing and
curing (Weitznian, et al., 1989). Volatilization can occur during
waste handling activities such as soil excavation and transport or
during the process of mixing the binding agents with the waste.
Also, some evaporative emissions will occur from waste even
after stabilization, especially during the curing period immedi-
ately after the blocks are formed. As shown in lab studies the
largest fraction of volatile loss occurs during the mixing phase
because heat may be required to assist mixing or is generated by
exothermic stabilization reactions (Weitzman, et al. 1989).
Particulate matter emissions from a full-scale system were
found to be about 2.5 Ib/hr (Ponder and Schmitt, 1991).
4.6.3 Emission Estimation Procedures
In general, VOC emissions from stabilization and solidifica-
tion processes will depend on the type and concentration of the
VOCs in the waste, the duration and thoroughness of the mixing,
the amount of heat generated in the process, and the average batch
size processed. The longer or more energetic the mixing and
processing, the greater likelihood that organic compounds will
volatilize. The volatile losses also will increase as the tempera-
ture of the waste/binder mixture increases. Binding agents with
high lime contents generally cause highly exothermic reactions.
The batch size influences volatilization by affecting the mean
distance a volatilized molecule has to travel to reach the air/solid
interface at the surface of the stabilized waste. The larger the
block of material, the lower the rate of volatilization.
In addition to volatile emissions, stabilization and solidifica-
tion processes will generate fugitive dust emissions. Possible
sources of fugitive dust emissions include storage of raw mate-
rials, preparation of the binding agents, transfer of wastes into the
mixing bin, removal of the treated material from the mixing bin,
and replacement of the material at the site after processing.
Little information exists about the fate of volatile contaminants
in wastes treated by stabilization and solidification methods. A
literature search found no available field data on air emissions at
Superfund sites using this type of remediation technology. Based
on laboratory studies, however, about 40-80% of the volatile
contaminants in the treated waste is estimated to eventually
evaporate (Weitzman, etal., 1989). Most of the loss occurs within
60 minutes of mixing the waste with binding agents. Thompson,
et al., 1991 give a simple mass balance equation for estimating
emissions.
Particulate matter emissions for stabilization and solidification
processes can be estimated using emission factors for soil han-
dling (see Section 4.1).
4.6.4 Identification of Applicable Control
Technologies
Emission controls for excavation, storage, and feeding of the
waste to the process unit were covered in Section 4.1.4. In
general, solidification/stabilization is not the remedy of choice
for wastes with high levels of VOCs and therefore VOC emis-
sions are not usually a major concern. The only reference in the
literature to emission controls from solidification/stabilization is
a solidification system processing 12 tons/hr enclosed in a
building (Ponder and Schmitt, 1991). Approximately 40,000 ft3/
min of air from the building was routed to PM and VOC control
devices. Emissions were controlled by introducing the gas stream
to a venturi scrubber, followed by a mist eliminator, an air
preheater, a disposable prefilter, and finally two parallel carbon
adsorption systems.
30
-------
For in-situ processes, several options or combination of op-
tions could be appropriate for controlling VOC or PM emissions:
• Collection hood; • ,
• Windscreens;
• Temporary foams; ,, [,
• Water sprays;
• Water sprays with additives;
• Enclosures; and ,
• Operational controls. ,
4.7 Bioremediation
4.7.1 Typical Remediation Scenario for Super/and
Sites
Bioremediation at Superfund sites may be either in-situ or ex-
situ. Ex-situ biodegradation generally refers to treatment pro-
cesses where an aqueous slurry is created by combining soil or
sludge with water and then biodegraded in a self-contained
reactor or in a lined lagoon. This is an emerging technology and
often is referred to as slurry biodegradation. Ideally, the waste is
decomposed into carbon dioxide and water. Background infor-
mation is available in U.S. EPA, 1990e and Thompson, et al.,
1991.
In-situ treatment employs the natural microbiological activity
of soil to decompose organic constituents. Systems that try to
enhance this natural biological activity typically use injection
wells to provide an oxygen source (such as air, pure oxygen, or
hydrogen peroxide) to stimulate aerobic degradation or add
nutrients to support the growth of waste-consuming microorgan-
isms. In some cases, microorganisms may be added to the soil that
have the ability to metabolize specific contaminants of interest.
In-situ bioremediation at Superfund sites also may involve
sequential isolation and treatment of waste areas using processes
that closely resemble ex-situ processes except that it may not be
necessary to excavate, pump, or otherwise transfer the waste
material prior to treatment. Ex-situ processes are more developed
and demonstrated than in-situ processes at this time.
Two main objectives behind using slurry biodegradation are:
to destroy the organic contaminants in the soil or sludge, and,
equally important, to reduce the volume of contaminated mate-
rial. Slurry biodegradation can be the sole treatment technology
in a complete cleanup system, or it can be used in conjunction
with other biological, chemical and physical treatment methods.
Systems have a number of components, all of which could be
emission sources: mix tank, bioreactor system (continuously
stirred tank reactor or CSTR), or lined lagoon. Since aerobic
treatment is the most common mode of operation for slurry
biodegradation, aeration must be provided to the bioreactors by
either floating or submerged, aerators or by compressors or
spargers. Other typical system components are a separation/
dewatering system, a clarifier for gravity separation, and waste-
water storage tanks.
Biodegradation is actually only one of several competing
mechanisms in biotreatment. For ex-situ processes, the contami-
nants may also be volatilized, undergo chemical degradation, or
be adsorbed onto the soil particles. For in-situ processes, these
same pathways exist along with leaching. The overall contami-
nant removal achieved by biotreatment processes represents the
combined effect of all of these mechanisms. Volatilization may
account for the disappearance of the majority of VOCs being
treated.
4.7.2 Potential Air Emissions
Typical emissions from biotreatment process are evaporative
losses of volatile and semi-volatile organic compounds. If the soil
is handled or mixed, though, some emissions of paniculate
matter may occur. Combustion emissions from the process
equipment are also possible.
The air emissions from slurry biodegradation processes can
either be area or point sources. For processes using open lagoons,
emissions come from the exposed surface of the lagoon. On the
other hand in systems using above-ground self-contained reac-
tors, the primary source of emissions is usually a process vent.
In bioslurry processes the emissions of concern are usually
VOCs. The soils handling steps required to deliver the contami-
nated soil to the treatment unit may also emit significant amounts
of VOCs and PM. Emissions from soils handling are addressed
elsewhere in this document.
In open lagoons, the primary environmental factors, in addi-
tion to the biodegradability and volatility of the waste, which
influence air emissions are process temperature and wind speed.
Emissions tend to increase with an increase in surface turbulence
due to wind or mechanical agitation. Temperature affects emis-
sions through its influence on microbial growth. At temperatures
outside the band for optimal microbial activity, volatilization will
increase. Emissions from self-contained reactors are also deter-
mined by reactor design parameters such as the amount of air or
oxygen used to aerate the slurry. Higher gas flow will strip more
volatiles out of solution and increase air emissions.
Little information exists on volatile losses from slurry biodeg-
radation processes. Slurry processes have only recently become
commercially available and field experience to date is limited.
However, data on air emissions from wastewaster biotreatment
processes are available. The percentage of each contaminant that
is volatilized will vary greatly depending on the physical proper-
ties of the contaminant and the design of the treatment system.
Based on field studies of an aerated impoundment treating
contaminated water, as much as 20% of each compound may be
volatilized depending on its volatility and biodegradability
(Eklund, et al., 1988). Percentage emissions for soil and waste
treatment would be expected to be higher.
4.7.3 Emission Estimation Procedures
Although no models have been developed explicitly for esti-
mating VOC or PM emissions for bioremediation processes
treating contaminated soils or waste, several public-domain PC
models are available for estimating air emissions from a variety
of other biotreatment options, principally surface impound-
ments. The two most commonly used models are CHEMDAT-7
(U.S. EPA, 1989b) and the Surface Impoundment Modeling
System (SIMS). Both CHEMDAT-7 and SIMS are based on
mass transfer and biodegradation models developed by EPA.
31
-------
While not appropriate for Superfund sites, land treatment is a
bioremcdiation process that is somewhat analogous to the types
of remediation performed atSisperfund sites. A PC-based model,
LAND? (U.S. EPA, 1989b), is recommended by the EPA to
predict the emission rates resulting from the land treatment of
wastes. Sensitivity studies doae using these models show that
under typical conditions they predict that 35-80% of the applied
volatilcs will be emitted to the air and the remainder degraded
(Coovcr, 1989).
4.7.4 Identification of Applicable Control
Technologies
When the air emissions from slurry biodegradation processes
arc released through a process vent, standard VOC air pollution
control technologies can be applied. Common alternatives for
controlling VOC vent emissions include:
• Carbon adsorption;
• Thermal incineration or oxidation; and
• Catalytic oxidation.
For the relatively low VOC levels and low gas flows from
bioreactors, carbon-based VOC emission controls are generally
the best choice. Since the vent stream will likely contain only
dilute amounts of VOCs, relatively large amounts of auxiliary
fuel must be fired in either thermal or catalytic oxidizers.
When the air emissions from slurry biodegradation processes
are area air emission sources, applying air pollution control
technologies is more difficult. The best approach is generally to
use a vapor collection hood to capture any VOC emissions and
then route those emissions to a standard control device. Other
area source control approaches (e.g., foams, covers) generally
are not applicable to in-situ bioremediation since the controls are
designed to inhibit the transfer of gases between the soil and the
atmosphere. While these approaches reduce VOC emissions,
they will also limit the replenishment of oxygen to the soil and
may cause anaerobic conditions to develop.
4.8 Separation Techniques
4.8.1 Typical Remediation Scenario for Superfund
Sites
Three remediation technologies are described below: soil
washing, solvent extraction, and soil flushing. These are all
primarily separation processes and further treatment of the col-
lected contaminants typically will be required. They have not
been used widely at Superfund sites. Soil washing is an ex situ
process in which contaminated soil is excavated and fed through
a water-based washing process. It operates on the principle that
contaminants can be dissolved or suspended in an aqueous
solution or removed by separating clay and silt particles and the
associated adhered contaminants from the bulk soil. The aqueous
solu tion containing contaminants may be treated by conventional
wastewater treatment methods (U.S. EPA, 1990f).
Most organic and inorganic contaminants bind chemically or
physically to clay or silt soil particles, which in turn adhere to
larger sand and gravel particles primarily by compaction and
adhesion. Particle size separation by washing enables the con-
taminated clay and silt particles (and the bound contaminants) to
be concentrated. Separating the sand and gravel from the small
contaminated soil particles significantly reduces the volume of
contaminated soil, making further treatment or disposal much
easier. The larger particles may be returned to the site (U.S. EPA
1990g).
Removal efficiencies range from 90-99 percent for volatile
organic compounds and 40-90 percent for semi-volatile com-
pounds, so the wastewater streams may contain high levels of
organic compounds and be an emission source.
Excavation and removal of debris and large objects precedes
the soil washing process. Sometimes water is added to the soil to
form a slurry thatcanbepumped. After the soil is prepared for soil
washing, it is mixed with wash water and sometimes with extrac-
tion agents. At this point, several separation processes occur. Soil
washing generates four waste streams:
1) Contaminated solids separated from the washwater;
2) Wastewater;
3) Wastewater treatment sludges and residual solids; and
4) Air emissions.
Solvent extraction differs from soil washing in that it employs
organic solvents rather than aqueous solutions to extract con-
taminants from the soil. The remediation process begins with
excavating the contaminated soil and feeding it through a screen
to remove large objects. In some cases, solvent or water is added
to the waste in order to pump it to the extraction unit. In the
extractor, solvent (e.g., liquefied propane and butane) is added
and mixed with the waste to promote dissolving of the contami-
nants into the solvent. Up to five waste streams may result from
the solvent extraction process:
1) Concentrated contaminants;
2) Solids;
3) Wastewater;
4) Oversized rejects; and
5) Air emissions.
Typically, solvent extraction units are designed to produce
negligible air emissions, but significant levels of emissions may
occur during waste preparation (U.S. EPA, 1990g).
Soil flushing differs from soil washing and solvent extraction
in that it is an in situ process in which the solvent is sprayed over
the contaminated area, percolates through the soil and dissolves
the contaminants. Elutriate is collected in a series of wells and
drains.
4.8.2 Potential Air Emissions
In addition to the contaminants that may volatilize, the solvents
themselves may be cause for concern. Products of aerobic and
anaerobic decomposition are also possible. No field data for
emissions from any of these processes has been identified.
In the soil washing process the greatest potential for emissions
of volatile contaminants occurs in the excavation, feed prepara-
tion, and extraction process. Collected emissions from these
32
-------
processes typically are treated by carbon adsorption or incinera-
tion (U.S. EPA, 1990f). Because soil washing occurs in liquid
and solid phases, volatile compounds emitted evolve primarily
due to their vapor pressures in these phases. The waste streams
also have the potential to be sources of VOC emissions.
Solvent extraction also may produce emissions during excava-
tion and soil transport and from contaminated oversize rejects
(U.S. EPA, 1990g). Because the solvent recovery process in-
volves vaporization of the solvent, fugitive emissions are pos-
sible from this as well as other stages of the solvent process,
including the waste streams.
Emissions from soil flushing may emanate from the soil
surface, solvent storage vessels and spray system, and from
locations where the contaminant-laden flushing solution sur-
faces.
4.8.3 Emission Estimation Procedures
No equations or models for predicting the air emissions from
these processes have been identified.
4.8.4 Identification of Applicable Control
Technologies
Carbon adsorption and fume incineration are typical controls
used to treat collected emissions. In solvent extraction, volatile
solvents are recovered and recycled.
4.9 Other Emerging Technologies
A broad range of technologies fall under this heading, with
most being some type of physical or chemical waste treatment
methods. Included in this category are emerging chemical treat-
ment methods such as:
• Chemical oxidation using ozone or hydrogen peroxide;
• Hydrolysis of alkyl halides and other organics; and
• Dechlorination (e.g., using lime);
and physical treatment methods such as:
• In-situ thermal treatment or vitrification;
• Electrokinetics; and
• Ground freezing.
These methods have in common that they are undergoing
development, little or no data are available regarding levels of air
emissions, and air emission controls have not been evaluated for
these applications. Further information is available in U.S. EPA,
1990h and U.S. EPA, 1991c. General considerations are dis-
cussed below.
In general terms, a chemical treatment method is one in which
a reactive compound (or compounds) is added to the contami-
nated groundwater or soil to react with pollutants and form less
harmful products. As the name implies, the effectiveness of this
type of treatment depends greatly on the chemical properties of
the pollutants. An example of this type of method is ozone
treatment of contaminated groundwater. In this process, con-
taminated groundwater or wastewater is mixed in a continuous
reactor with ozone and other oxidizing agents. The oxidizers
react with the organic contaminants to form CO2 and water.
Physical treatment involves the addition of energy or another
treatment agent to physically transfer the pollutants to another
state in which they are easier to dispose of or treat. The path of
physical transfer can be adsorption, absorption, dissolution, or a
change of state such as evaporation. An example of this method
is in-situ vitrification. Electrodes are placed in the ground and a
large current is applied. The soil heats and fuses. The electrodes
are removed after the ground has sufficiently cooled (e.g., after
one year).
The air emissions associated with chemical and physical waste
treatment techniques that may be used at Superfund sites have not
been characterized adequately for most methods. A broad spec-
trum of technologies are included in this category, and the types
and sources of air emissions may vary greatly. For most chemical
and physical treatment methods, however, the emissions of
primary concern are VOCs, with emissions of semi-volatile
organic compounds and particulate matter also of potential
concern. Emissions are usually from either ground level area
sources or low-level point sources. Point sources typically are
associated with the treatment method, while area sources usually
are associated with the handling of contaminated soil or water.
In general, two types of process air emission sources can be
associated with chemical and physical treatment. First, transfer
of the contaminants from the liquid- or solid-phase to air may be
an inherent consequence of the treatment method. For example,
in-situ thermal treatment volatilizes a significant fraction of the
soil contaminants. In some cases, these air emissions are con-
trolled by using a hood to collect the emissions and route them to
an add-on control device for VOCs, such as carbon adsorption
units. Second, fugitive emissions can be generated as a by-
product of the treatment method. For instance, in ozone treatment
of contaminated water, trace emissions of unreacted organic
contaminants and ozone may occur.
Additional fugitive emissions from physical and chemical
treatment methods can result from leaking valves, pumps, and
flanges in the system, as well as from transfer or handling of the
untreated contaminated material. Equipment leaks may be regu-
lated under.the CAAA regulations.
4.10 References
Church, H. Excavation Handbook. McGraw-Hill, NY, NY.
1981.
Coover, J.R. Air Emissions From Hazardous Waste Land
Farms. Presented at the Spring National AIChE Meeting,
April 2-4, 1989.
Cowherd, C., G. Muleski, and J. Kinsey. Control of Open
Fugitive Dust Sources. EPA-450/3-88-008. U.S. EPA,
RTP, NC. September 1988.
Cullinane, M., L. Jones, and P. Malone. Handbook for Stabi-
lization/Solidification of Hazardous Waste. EPA/540/2-
86/001 (NTIS PB87-116745). June 1986.
33
-------
Damlc, A.S. and T.N. Rogers. Air/Superfund National Tech-
nical GuidanceStudy Series: AirStripper Design Manual.
EPA-450/1-90-003 (NTIS PB91-125997). May 1990c.
dc Pcrcin, P.R. Thermal Desorption Attainable Remediation
Levels. In: Proceedings of the 17th Annual Hazardous
Waste Research Symposium, EPA/600/9-91/002, pp511-
520. U.S. EPA, Cincinnati, OH. April. 1991 a.
dc Pcrcin, P.R. Thermal Desorption Technologies. Presented
at the 84th Annual Meeting of AWMA Paper No. 91 -22.1,
Vancouver, BC, June 1991b.
Donnelly, J. Air Pollution Controls for Hazardous Waste
Incinerators. In: Proceedings of the 12th Annual HMCRI
Hazardous Materials Control/Superfund 1991 Confer-
ence. HMCRI, Silver Spring, Maryland. December 1991.
Eklund, B., D. Green, B. Bl aney, and L. Brown. Assessment of
Volatile Organic Air Emissions From an Industrial Aer-
ated Wastewater Treatment Tank. In: Proceedings of the
14th Annual Hazardous Waste Research Symposium,
EPA/600/9-88/021 (NTISPB89-174403).pp468-475.U.S.
EPA. July 1988.
Eklund, et al. 1989. Air/Superfund National Technical Guid-
ance Study Series, Volume HI: Estimation of Air Emis-
sions from Cleanup Activities at Superfund Sites. Report
No. EPA-450/1-89-003. U.S. EPA, Research Triangle
Park, NC, 1989.
Eklund, B., C. Petrinec, D. Ranum, and L. Hewlett. Database
of Emission Rate Measurement Projects. EPA 450/11 -91 -
003 (NTIS PB91-222059LDL). U.S. EPA, RTF, NC. June
1991a.
Eklund, B., S. Smith, and M. Hunt. Estimation Procedures For
AirStrippingof Contaminated Water.EPA-450/1-91-002
(NTIS PB91-211888). U.S. EPA, Research Triangle Park,
NC. May 1991b (revised August 1991).
Eklund, B., S. Smith, and A. Hendler. Estimation of Air
Impacts For the Excavation of Contaminated Soil. EPA-
450/1-92-004 (NTIS PB92-171925). U.S. EPA, Research
Triangle Park, NC March 1992a.
Eklund, B., S. Smith, P. Thompson, and A. Malik. Estimation
of Air Impacts For Soil Vapor Extraction (SVE) Systems
EPA-450/1-92-001 (NTIS PB92-143676). U.S. EPA,
Research Triangle Park, NC. January 1992b.
Eklund, B., P. Thompson, W. Dulaney, and A. Inglis. Air
Emissions From the Treatment of Soil Contaminated with
Petroleum Fuels and Other Substances. EPA-Control
Technology Center. EPA-600/R-92-124. July 1992.
Helsel, R.W. and R.W. Thomas,, Thermal Desorption/Ultra-
violet Photolysis Process Technology Research, Test, and
Evaluation Performed at the Naval Construction Battalion
Center, Gulfport, MS. For the US AF Installation Program
Volumes I and IV. AFESC, Tyndall Air Force Base,
Florida. Report No. ESL-TR-87-28. December 1987.
International Technology Corporation. Screening Procedures
For Estimating the Air Impacts of Incineration at Super-
fund Sites. EPA-450/1-92-003 (NTIS PB92-171917).
February 1992.
Johnson, et al. A Practical Approach to the Design, Operation,
and Monitoring of In Situ Soil-Venting Systems. Ground
Water Monitoring Review. Spring 1990.
Lighty, J.S., G.D. Silcox, D.W. Pershing, V.A. Cundy, and
D.G. Linz. Fundamentals for the Thermal Remediation of
Contaminated Soils. Particle and Bed Desorption Models.
ES&T Vol. 24, No. 5, pp750-757, May 1990.
* ; •
Oppelt, E.T. Incineration of Hazardous Waste - A Critical
Review. JAPCA, Vol. 37, No. 5, pp558-586, May 1987.
Pedersen, T.A. and J.T. Curtis. Handbook of Soil Vapor
Extraction Technology. EPA/540/2-91/003 (NTIS PB91 -
168476). February 1991.
PES Corp. Soil Vapor Extraction VOC Control Technology
Assessment. EPA-450/4-89-017 (NTIS PB90-216995).
U.S. EPA, Research Triangle Park, NC, September 1989.
Ponder, T. and D. Schmitt. Field Assessment of Air Emissions
From Hazardous Waste Stabilization Operations. In: Pro-
ceedings of the 17th Annual Hazardous Waste Research
Symposium. EPA/600/9-88/021 (NTIS PB91-233627).
July 1988.
Thompson, P., A. Inglis, and B. Eklund. Emission Factors for
Superfund Remediation Technologies. EPA-450/1-91-
002 (NTIS PB91-19075). March 1991.
Troxler, W.L., J.J. Cudahy, R.P. Zink, and S.I. Rosenthal.
Thermal Desorption Guidance Document for Treating
Petroleum Contaminated Soils. EPA Contract No. 68-C9-
0033. Report to James Yezzi, U.S. EPA, Edison, N.J.
U.S. EPA. HAZCON Solidification Process, Douglassville,
PA - Applications Analysis Report. EPA/540/A5-89/001.
U.S. EPA, Cincinnati, OH. May 1989a.
U.S. EPA. Hazardous Waste Treatment, Storage, and Disposal
Facilities (TSDF) — Air Emission Models. Report No.
EPA-450/3-87-026 (NTIS PB88-198619). U.S. EPA,
Research Triangle Park, NC, November 1989b.
U.S. EPA. Engineering Bulletin - Mobile/Transportable Incin-
eration Treatment. EPA/540/2-90/014, (NTIS PB91-
228023). September 1990a.
34
-------
U.S. EPA. 40 CFR Part 264.343. Federal Register 55, No. 82,
April 27,1990b.
U.S. EPA. International Waste Technologies/Geo-Con In Situ
Stabilization/Solidification - Applications Analysis Re-
port. EPA/540/A5-89/004. U.S. EPA, Cincinnati, OH.
August 1990d.
U.S. EPA. Engineering Bulletin - Slurry Biodegradation. EPA/
540/2-90/016 (NTIS PB91-228049). September 1990e.
U.S. EPA. Engineering Bulletin - Soil Washing Treatment.
EPA/540/2-90/017 (NTIS PB91-228056). September
1990f.
U.S. EPA. Engineering Bulletin - Solvent Extraction Treat-
ment. EPA/540/2-90/013 (NTIS PB91-228015). Septem-
ber 1990g.
U.S. EPA. Handbook on In Situ Treatment of Hazardous.
Waste-Contaminated Soils. EPA/540/2-90/002 (NTIS
PB90-155607). U.S. EPA, Cincinnati, OH. September
1990h.
U.S. EPA. Survey of Materials - Handling Technologies Used
at Hazardous Waste Sites. EPA/540/2-91/010. U.S. EPA-
ORD, Washington, B.C. June 1991a.
U.S. EPA. Toxic Treatments, In-Situ Steam/Hot-Air Stripping
Technology - Applications Analysis Report. EPA/540/
A5-90/008. U.S. EPA, Cincinnati, OH. March 1991b.
U.S. EPA. The Superfund Innovative Technology Evaluation
Program: Technology Profiles Fourth Edition. EPA/540/
5-91/008. U.S. EPA, Cincinnati, OH. November 1991c.
Vancit, M.A., et al. Air Stripping of Contaminated Water
Sources - Air Emissions and Controls. EPA-450/3-87-017
(NTIS PB88-106166). August 1987.
Weitzman,L., et al. Volatile Emissions From Stabilized Waste.
In: Proceedings of the Fifteenth Annual Research Sympo-
sium. EPA-600/9-90/006 (NTIS PB91-145524). Febru-
ary 1990.
35
-------
-------
Chapter 5
Point Source Controls for VOCs and SVOCs
Information about various control technologies whose pri-
mary use is to control air emissions of volatile organic com-
pounds (VOCs) and semi-volatile organic compounds (SVOCs)
is presented in this chapter. Control technologies addressed in
this chapter are carbon adsorption, thermal oxidation, catalytic
oxidation, condensers, internal combustion engines, biofilters,
operational controls, membranes, and emerging technologies
such as ultraviolet treatment. The discussion for each control
technology includes a process description, applicability for re-
mediation technologies, range of effectiveness, sizing criteria,
and cost information.
5.1 Carbon Adsorption
5.1.1 Process Description
Carbon adsorption systems (CAS) are one of the most com-
monly used air pollution control devices for the reduction of
VOC emissions from remediation processes. They are effective
in removing a wide range of VOCs over concentrations from low
ppbv to about 1,000 ppmv. The most common form of carbon
used in CAS is granular activated carbon (GAC), though other
adsorbents such as impregnated carbon, silica gel, or activated
alumina may be used also. These alternate adsorbents typically
cost more than GAC, but are more effective for certain corrosive
gases or pollutants that do not have ahigh affinity for pure carbon
(e.g., mercury, nickel, phosgene, or amines).
The physical principle behind any adsorption process is the
Van der Walls attractive potential between the waste stream
constituents and the GAC in the bed. The potential energy is
given off as heat during adsorption. Since adsorption efficiency
is an inverse function of temperature, the stream must be kept
relatively cool. If the carbon is to be regenerated, then heat must
be added to overcome the Van der Walls force, thus freeing the
pollutants. The carbon then must be cooled prior to re-use, and
the pollutants, once again airborne (or in some liquid solution),
must be disposed of in some acceptable way. If the system is non-
regenerable, then the carbon units themselves must be treated as
solid waste and disposed of accordingly.
Carbon has a fixed capacity or number of active adsorption
sites. As the adsorbing liquid/gas stream fills the sites in the
adsorbent, a somewhat arbitrary, empirically-determined point
is reached (called the "loading" point) where adsorption effi-
ciency is decreased significantly. This is typically around 1 g
VOC per 10 g carbon. If adsorption were continued beyond this
point, then the "break through" point would be reached, and
pollutants would no longer be controlled effectively. Eventually,
"saturation" would be reached, where all sites are filled and
virtually no adsorption occurs.
A number of CAS designs are commercially available. The
three most common basic designs are canister systems with off-
site regeneration, continuous regenerating systems, and dual bed
systems with on-site batch regeneration as shown in Figures 5-1,
5-2, and 5-3, respectively. The canister system in Figure 5-1
Clean gas out
Carbon
adsorber
VOC off-gas
Figure 5-1. Schematic diagram of canister-based granularactivated
carbon adsorption system.
37
-------
Sc
V
\
rVKr
xbent
ipper-
t"f &
' « * *
* '«
'*.*"i Cross-flow /
' 5-»l adsorbent /
Clean
gas
out
gash
Thermal
desorber
Adsorbent
feed motor
Compressed air
Figure 5-2. Schematic diagram of continuously regenerated carbon
adsorption system.
shows two side-by-side canisters of activated carbon that can be
used sequentially or in parallel. Carbon in 55-gallon drum canis-
ters is available also. In Figure 5-2, a moving-bed system is
illustrated, with adsorption occurring as the adsorbent falls
through a baffle, while the waste stream passes across the baffle.
The carbon is regenerated on its way back to the top of the baffle.
These systems are used much less commonly than fixed-bed
systems. In Figure 5-3, a standard fixed-bed system is shown,
with two beds adsorbing while the third is desorbing. Regenera-
tion typically is accomplished by passing steam through the
carbon. The high temperature and water vapor strip most organic
solvents from the carbon and the organics are captured with the
condensed water leaving the system. Subsequent treatment is
necessary to separate the organic fraction from the water before
disposal or use of the solvent. A modification to steam regenera-
tion is to use an inert gas to reactivate the carbon; an additional
step (e.g., condensation) is required to separate the VOCs from
the inert gas. Such systems are initially more expensive than
steam regeneration systems, but potentially offer savings from
reduced energy use and recovery of purer solvent.
The major components of a GAG control system include the
pretreatment devices (de-humidifiers, absorbers, particulate fil-
ter, etc.), piping to carry'the stream to the adsorbent, then the
adsorption bed or canister followed by piping to other add-on
controls or a stack. A regeneration unit is also present if the
system uses regenerable technology; it can have either multiple-
fixed beds or a moving bed. In the former, several beds usually
are used in parallel, so that while some are being regenerated,
others are in-line and adsorbing. The moving bed type is less
common and the carbon is regenerated at one point while adsorb-
ing at another.
The operational cycle for a carbon bed is adsorption, heat
regeneration, drying, and cooling. The heat to regenerate the
carbon must be greater than the heat released during adsorption.
The operational cycle for a carbon canister is adsorption, replace-
Clean gas out
Clean gas out
Clean gas out
Clean gas out
Clean gas out
Carbon
adsorber
voc
off-gas
Figure 5-3. Schematic diagram of carbon adsorption system with on-site batch regeneration.
38
Fuel
Condensate
-------
ment at the loading pqint, and, either disposal or removal and
subsequent off-site regeneration.
5.1.2 Applicability to Remediation Technologies
' GAC is a likely candidate for the control for any site remedia-
tion involving a point source having low concentrations of VOCs
emitted to the atmosphere. GAC systems are relatively cheap and
easy to install, they can be either regenerable or disposable, and
they handle many different types of contaminants. As one of the
most widely-used control technologies, much technical informa-
tion is available about these,systems and there are numerous
vendors. • • >••.'....• •
For gas streams with VOC concentrations exceeding 1,000
ppmv, condensers, incinerators, or internal combustion engines
become competitive in cost-effectiveness with GAC, systems.
GAC systems are also less efficient at higher temperatures or
pressures. Further, they require low humidity in the incoming
stream, since water binds to the active sites in the carbon.
Plugging, fouling, and some corrosive gases also pose a problem
for the adsorption of some waste stream constituents. Any of
these, drawbacks may be amended with pretreatment devices,
although such pre-treatment usually will increase the total sys-
tem cost.
The most substantial shortcoming of GAC systems is that, only
compounds with molecular weights in the 50 - 200 g/g-mol range
have the proper adsorption properties. Also, pollutants are not
destroyed, only transferred from one medium to another, inevi-
tably leaving solid or liquid waste after treatment. In industrial
applications, GAC systems often are used to capture and recycle
valuable pure VOCs, but in remediation projects these VOCs
usually are not of sufficient purity or value to warrant recycling.
Disposal is almost always the final step.
Carbon canisters generally are used for remediation projects
which are quite different from those appropriate for regenerating
beds; cans usually are used for low volume, intermittent sources.
If they are regenerated, it usually is done off-site by the carbon
supplier. One problem that may occur for such systems is that
they may be used past the effective adsorption point of the carbon
and into saturation due to lack of monitoring and the disposable
nature of canister carbon.
5.1.3 Range of Effectiveness
An unalterable limitation of adsorption is the molecular weight
of the VOCs to be adsorbed. If the molecular weight of a VO.C is
too low the compound will not adsorb very readily. If the
molecular weight is too high, the compound will be difficult to
desorb from the carbon. This limitation can be circumvented to
some degree by using different types of carbon. A typical range
of effectively adsorbed molecular weights is between 50 and 200
g/g-mol. Other factors, such as polarity and molecular shape,
may also affect adsorptivity.
A GAC system is most cost-effective when contaminant con-
centrations are low and the waste gas flowrate is low or variable.
Carbon systems also are not readily available for flow rates
exceeding 100,000 scfm. A properly operating GAC system at
moderate flow rates and hazardous air pollutant (HAP) concen-
trations (e.g., ppm level) can have a removal rate of 70-99+%,
depending on the pollutant and the operating temperature. As
with other control devices, higher efficiencies are achieved at the
cost of higher pressure drops across the adsorption unit.
Pretreatment may ameliorate other limitations of the technol-
ogy. Certain operating conditions should be met: no fouling
compounds (including solid or liquid particulates), less than
1,000 ppmv inorganics, and relative humidity below 50%. The
efficiency of VOC adsorption decreases rapidly as the relative
humidity rises above 50%. The relative humidity of the gas
stream can be lowered by raising the temperature of the gas
stream, but this can affect removal efficiencies. For adsorption to
occur readily, the waste stream must be at a moderate tempera-
ture (100 - 130°F). Outlet VOC concentrations usually are
required to be less than 10-50 ppmv. The effectiveness of CAS
for various classes of compounds is summarized in Table 5-1.
5.1.4 Sizing Criteria/Application Rates
GAC systems are capable of handling concentrations from the
ppb level to 25% of the lower explosive limit (LEL). For sizing
a system, the total carbon requirement is the most important
parameter. This will be a function of the volumetric carbon flow
rate, VOC concentration, and VOC molecular weight of the
waste strea'm, and carbon adsorption capacity, adsorption time,
the removal efficiency required, and the number of beds. (For
more detailed sizing discussions, see U.S. EPA, 1991 or U.S.
EPA, 1990). For a fixed-bed adsorption system with a specific
adsorption time, t, the following equation may be used (U.S.
EPA, 1991):
Wc = 2(6.0x10~a) • t • Q • C • -=-
E_
100
A .
(Eq.5-1)
where:
Wc = Weight of carbon required (Ib);
t = Adsorption cycle time, (hr);
Q = Emission stream flow rate (scfm);
DHAp = HAP density (gas)(lb/ft3);
A = Adsorption capacity of carbon bed, (Ib HAP/
100 Ib carbon);
C = Inlet concentration of HAPs (ppmv);
E = Removal efficiency (%)
2 = Factor for a two-bed system;' and
6.0xlO's = Conversion factor
mm
hr- ppmv
A typical default value for A is 10. Adsorption capacities for
some HAPs by a commonly-used type of activated carbon are
given in Table 5-2. Adsorption capacities also can be calculated
as a function of inlet concentration, temperature, etc. (see U.S.
EPA, 1990).
5.1.5 Cost Estimating Procedure
The U.S. EPA has published detailed cost estimation proce-
dures for GAC systems (U.S. EPA, 1990). Only a rough outline
of that discussion is given in this section, and it should only be
39
-------
Tablo 6-1. Applicability of CAS for Selected Contaminants
Contaminant class
Examples
CAS typically effective?
Comments
Aromalics
Aliphatics
Halogenated hydrocarbons
Light Hydrocarbons
(MW<50orBP<20°C)
Heavy Hydrocarbons
(MW > 200 or BP > 200"C)
Oxygenated compounds
Certain reactive organics
Bacteria
Radioisotopes
Certain inorganics
Mercury
benzene, toluene Yes
hexane, heptane Yes
chloroform Yes
methane, freon No
glycols, phenols Noa
ketones, aldehydes Nob
1,1,1 -trichloroethane, organic acids No
coliform Yes
1311 Yes
hydrogen sulfide, ammonia, hydrochloric acid Yes
— Yes0
Standard application of GAC
Standard application of GAC
Standard application of GAC
Will not adsorb
Will not desorb or will not be
adsorbed due to steric constraints
Fire hazard
Will react with and degrade GAC
Requires silver-impregnated GAC
Requires coconut-shell carbon
Requires impregnated GAC
Requires impregnated GAC
" Non-rogonorabie caibon systems may work.
» Not all oxygenated compounds aro a problem.
* High levels ol sulfur dioxide may "blind1' the charcoal and reduce Hg removal efficiencies.
used as a preliminary cost-estimating tool. The total costs of a
GAC system are broken down into three categories:
1. Equipment costs, including carbon and containers;
2. Installation, engineering, and indirect costs associated
with purchasing and installing the system; and
3. Annual expenses.
These costs will vary depending on whether canister or beds
are used, and also whether corrosive or non-corrosive gases are
present in the waste stream.
Equipment Costs
Note: The costs in this section and throughout the document
have been converted to 1992 dollars, assuming 5% per year
compounded inflation. The cost for a carbon adsorption system
is obviously a function of the total carbon requirement as well as
other factors. However, the cost can be estimated strictly on the
basis of the carbon, within some limitations. Primarily, this
approach assumes that a non-corrosive waste stream is being
used, so that a less-expensive stainless steel and inexpensive
carbon may be used. In that case, the following formulae hold
(Vatavuk, 1990):
Price (or fixed
bed regenerate: P = 173 W t848' 350 s W < 14,000 Ib (Eq. 5-2)
Modular
adsorbents: P = 6.24 Wt968' 110 14,000 :£W< 222,000 Ib
W is the weight of carbon per unit.
Installation Costs
A general rule for carbon adsorption units is that installation
costs are about 25% of the total unit for a packaged-type device,
or 61 % for a custom built device (Vatavuk, 1988). For the capital
cost factors for any buildings, power, and other general compo-
nents of the carbon adsorption system, refer to Section 5.2.5.
Annual Expenses
The VOC concentrations where GAC systems are cost effec-
tive relative to fume incineration are illustrated in Figure 5-4
(AMCEC, 1991). Systems have a return on investment in two to
three years for industrial applications (AMCEC, 1991). Since
systems have a 10-year average life-span, a salvage value is
likely after a remediation project is completed. The carbon itself
only lasts two years, and the typical cost for carbon is $1.80 to
$2.00 per pound, depending on the total weight purchased (DCI,
1991). Other grades of carbon and impregnated carbons will have
higher costs.
Relatively detailed cost estimating tables also have been devel-
oped (U.S. EPA, 1990). Estimation procedures for annualized
carbon costs for fixed bed carbon adsorbers are given in Table 5-
3. A cost-effectiveness diagram giving a rough guide for the
applicability of CAS systems for varying vapor concentrations is
presented in Figure 5-5. Non-regenerable systems are compared
with automated and manual regenerable systems in this figure.
Annual costs for canister systems containing 150 Ibs of BPL
carbon are:
Quantity
1-3
4-9
10-29
>30
Cost, each can
$920
$880
$830
$780
For the case of waste streams involving corrosive gases, the
system costs can be expected to be twice those of a regular system
which approximates the increased cost of both the special adsor-
bent, as well as all metal work which must be resistant to the
corrosive agents.
40
-------
Table 5-2. Reported Operating Capacities for Selected Organic Compounds
Compound Average inlet concentration (ppmv)
Adsorption capacity a (Ib VOC/100 Ib carbon)
Acetone .
Benzene
n-Butyl acetate
n-Butyl alcohol
Carbon tetrachloride
Cyclohexane
Ethyl acetate
Ethyl alcohol
Heptane
Hexane
Isobutyl alcohol
Isopropyl acetate
Isopropyl alcohol '
Methyl acetate
Methyl alcohol
Methylene chloride
Methyl ethyl ketone
Methyl isobutyl ketone
Perchloroethylene
Toluene
Trichloroethylene
Trichlorotriofluoroethane
Xylene
1,000
10
150
100
10
300
400
1,000
500
500
100
250
400
200
200
500
200
100
100
200
100
1 ,000
100
8
— 6
8
8
10
6
8.
8
6
6
8
8
8
7
7
10
8
7
20
7
15
8'
10
' Adsorption capacities are based ori 200 scfm of solvent-laden air at 100° F (per hour).
Source: Marzone and Oakes, 1973.
5.2 Thermal Oxidation
5.2.1 Process Description
Thermal oxidation, also known as thermal incineration, is a
commonly used approach for controlling volatile organic com-
pound (VOC) emissions in waste gases. In thermal oxidation,
contaminant-laden waste gas is heated to a high temperature
(above 1000°F) where the VOC contaminants are burned with air
in the presence of oxygen to form carbon dioxide and water.
Figure 5-6 is a simplified schematic of a thermal oxidation
system. This type of system, which is designed only for handling
waste gases and not liquids or solids, often is referred to as a fume
incinerator. The three key design parameters for fume incinera-
tors are commonly called the "Three T's": temperature, resi-
dence time (also referred to as "retention time" or "dwell time")
and turbulence. The "Three T's" have an interrelated effect on
combustion performance. To achieve good combustion, the
waste gas must be held for a sufficient time (usually 0.3-1.0
seconds) at combustion temperatures 100°F or more above the
auto-ignition temperatures of the contaminants in the waste gas.
Additionally, turbulent flow conditions must be maintained in
the incinerator to ensure good mixing and complete combustion
of the waste contaminants.
In a typical fume incinerator, waste gas is introduced into the
combustion chamber as shown in Figure 5-6. In the combustion
chamber the waste gas temperature is raised to the appropriate
combustion range by burning auxiliary fuel. Because of the high
combustion temperatures (1,000 to 1,600°F for most VOCs),
refractoryrlined chambers are required. At these temperatures,
95 to 99 percent of the VOCs in the waste gas are combusted
(Katari, et al., 1987a).
In most cases, the flue gas from the combustor then passes
through a heat exchanger where a portion of its sensible heat is
used to preheat the incoming waste gas. The flue gas then is
vented to the atmosphere through a stack downstream of the heat
exchanger.
One way in which fume incineration systems differ from one
another is in the type of heat recovery used. The heat exchanger
design is important because it determines the amount of heat
recovery. In turn, the fraction of heat which can be recovered
from the flue gas will affect directly the amount of fuel required
to operate the incinerator. Typically, two types of heat exchange
systems are used: recuperative heat exchange and regenerative
heat exchange. In a recuperative heat exchanger, hot gas travels
on one side of a partition while cold gas passes on the other. Heat
is transferred directly from the hot side to the cold side through;
the partition. This is the most common type of heat exchanger.
Both counter-flow and cross-flow exchanger designs are used for.
this purpose. For a recuperative exchanger, heat recovery typi-
cally varies from 30-75%.
In a regenerative heat exchange system, energy is transferred
indirectly from the hot stream to the cold stream. First, the hot
flue gas is passed through a ceramic matrix to recapture as much
of the energy as possible. The heated ceramic then is used to
preheat the contaminated flue gas, which in turn is run through a
second matrix to recapture its energy before being exhausted to
the atmosphere. Vendors of regenerative systems typically guar-
41
-------
g
1
a-
40
30
S 20
10
I
II
-10
-20
7 • 1
Adsorption/oxidizer
Incineration
100
1
L
Source: Amcec, 1991.
200 300
Contaminant concentration (ppmv)
400
500
Figure 5-4. Fuel cost/gain vs. concentration of carbon and incineration systems at 50,000 scfm of solvent-laden air.
antcc8Q-95%heatrccovery.Asaresultofthehighheatrecovery,
fuel costs tend to be low compared to traditional thermal incin-
erators using recuperative gas-gas heat exchange for energy
recovery. However, because the technology is relatively new,
equipment and other capital costs tend to be high. For Superfund
rcmcdiations that are short in duration (e.g., <36 months), the
high capital and installed costs often make regenerative incinera-
tion unattractive since the period over which the equipment is
depreciated is brief. The economics for regenerative thermal
incineration are most favorable for the treatment of a dilute, large
volume waste gas, since it would not require large amounts of
auxiliary fuel. In industrial applications, regenerative thermal
incineration is commonly used for controlling VOC emissions
from process point sources such as paint spray booths or solvent
degreasers.
An alternative method of heat recovery, which may be practi-
cal in some cases, is to produce low-pressure steam in a waste
heat boiler. This alternative is only used in cases where low
pressure steam is needed at or near the remediation site.
5.2.2 Applicability to Remediation Technologies
The applicability of thermal incineration depends on the con-
centration of oxygen and contaminants in the waste gas. The
waste gas composition will determine the auxiliary air and fuel
requirements. These requirements in turn will have a strong
influence on whether thermal oxidation is an economical ap-
proach for controlling air emissions.
For most remediation technologies used at Superfund sites, the
off-gases that require control are dilute mixtures of VOCs and air.
The VOC concentration of these gases tends to be very low, while
their oxygen content is very high. In this case auxiliary fuel is
required but no auxiliary air is needed. However, if the waste gas
has a VOC content greater than 25 percent of its LEL (e.g., some
SVE-based clean-ups), auxiliary air must be used to dilute the
contaminant to below 25 percent of its LEL prior to incineration.
If the remediation activity generates an off-gas that has a low
oxygen content (below 13 to 16 percent), ambient air must be
used to raise the oxygen level to ensure the burnerflame stability.
In the rare case when the waste gas is very rich in VOCs, using
it directly as a fuel may be possible.
Information is presented in Table 5-4 for determining the
suitability of a waste gas for incineration and establishing its
auxiliary fuel and oxygen requirements. This same information
is shown in Figure 5-7 in an alternative format.
42
-------
Table 5-3. Equations for Carbon Adsorption Annualized Cost Estimate
Cost item
Equation
I. Direct costs
a. Steam costs, Cs
b. Cooling water cost,
c. Electricity
1. Pressure drop, Pb, for regenerative systems
(based upon superficial velocity of 60 ft/min)
2. Pressure drop, P0, for canister systems
3. System fan horsepower, h ^
4. Bed cooling/drying fan, h M
5. Cooling water horsepower, h
6. Required electricity usage per year, kWh
d. Carbon replacement cost, CRC
C5=3.5x10-3Mvoc(HRS)Ps
where: Mvoc = Inlet VOC loading, Ibs/hr
MRS = Operating hours per year
P = Steam price, $/103lbs
=3.43C3Pcw/Pa
Cooling water price, $/103 gal
(assumed to equal $0.225/103 gal)
where:
Pb = tb (2.606)
where: t b = Bed thickness, ft. carbon
0.0166Creq a
*b = LD~
Creq = carbon required
L = vessel length
D = vessel diameter
Po = 0.0471 Qc +9.29x10-" Qg
where: Q,. = Emission stream flowrate, ft 3/min
h^ = 2.5x10-"(PborP0+1)Q0
hdd= 1.86x10 -4(FRdcf)(Pb+ 1) (BM)
where: FRdcf = (^00) (C req) /Q^^o,
6 M = cooling/drying cycle time, hr
6 d=, . = 0.4(ereg)(NA)(HRS)/ead
8 reg = regeneration cycle time, hr
6 ^ = adsorption cycle time, hr
li= (2.52 x 10 -"q^, HS)/n
where: qcm = Cooling water flowrate, gal/min
H = Required head (usually 100 ft H2O)
S = Specific gravity of fluid
n = Pump and motor efficiency
kWh = 0.746 (h•„„,+ h ^ ) MRS + h^,
CRC = CRFC(1.08CC + COI)
where: CRFC = Capital recovery factor for carbon
C0 = Carbon cost, $/lb
COI = Replacement labor cost, $/lb
(typically about $0.05/lb) ,
a Assumes a two-bed system.
Source: U.S. EPA, 1991.
43
-------
7000
6000
5000
•I* 4000
3000
2000
1000
Replaceable Cannisters
Dual Bed, On-Site Regeneration
Automatic Regenerate Beds
Note: Capital costs amortized over five years.
10 15 20
VOC Recovery (Ibs/day)
25
30
35
Source: U.S. EPA, 1991.
Figure 5-5. Activated carbon systems cost comparison.
If halogenated VOCs are present in the influent gas stream,
then hydrochloric acid (HC1) may be produced in the thermal
incinerator. HC1 emissions are regulated and off-gas controls
such as packed-column gas absorbers for HC1 and other acid
gases may be required.
533 Range of Effectiveness
Thermal incineration is a well-established method for control-
ling VOC emissions in waste gases. The control efficiency (also
referred to as destruction and removal efficiency or DRE) for
thermal incineration is typically 98% or higher. Factors which
affect DRE include the three "T's" (temperature, residence time,
and turbulence) as well as the lype of contaminants in the waste
gas. With a 0.75-second residence time, the suggested thermal
incinerator combustion temperatures for waste gases containing
nonhalogenated VOCs are 1,6CO°Fand 1,800°F, respectively for
98 and 99 percent VOC destruction efficiencies. Higher tempera-
tures (about2,000°F) and longerresidence times (approximately
1 second) are required for achieving DRE's of 98% or more with
halogenated VOCs (Katari, etal., 1987a, U.S. EPA, 1991).
In this discussion the term Itowrate implies the flowrate at standard conditions
wWch aro assumed to be 60BF and 1 aim, following standard engineering
practices.
5.2.4 Sizing Criteria
To size a thermal incinerator with a given residence time and
estimate the capital and anntialized costs, three pieces of data are
required:
1) The flue gas flow rate;-
2) The auxiliary fuel and air requirements;
3) Inlet VOC concentration (or heat content);
4) Inlet temperature; and
5) Combustion temperature.
Flue Gas Flowrate
For dilute waste gases the flue gas flowrate1 is approximately
equal to the waste gas flowrate. In cases where auxiliary air is
required, the flue gas flowrate is roughly equal to the sum of the
waste gas flowrate and the auxiliary air flowrate. The flue gas
flowrate can be used in many correlations to size the incinerator
and estimate equipment costs.
Auxiliary Fuel Requirement
The auxiliary fuel is usually the largest operating expense for
a thermal incineration system. The fuel requirement can be
estimated by making a heat balance around the incinerator
system. The approach described below assumes heat losses to be
negligible. In many cases this assumption is not valid and the fuel
44
-------
Auxiliary air
(if required)
Auxiliary
fuel
Combustion
chamber
(Refractory-lined
for temperatures to
2000+ °F)
Q
heat loss
Flue gas
Preheated waste gas
Waste gas from
remediation process
Heat
exchanger
Flue gas
to stack
Figure 5-6. Schematic of thermal Incineration system with recuperative heat exchanger.
Table 5-4. Categorization of Waste Gas Streams
Category
5
6
Waste gas
Composition
VOC
Heat content
Auxiliaries and other requirements
Mixture of VOC, air, and inert gas >16% <25% LEL <13 Btu/ft3 Auxiliary fuel is required. No auxiliary air
is required.
Mixture of VOC, air, and inert gas 16% 25-50% LEL 13-26 Btu/ft3 Dilution air is required to lower the heat
content to <13 Btu/ft3. (Alternative to
dilution air is installation of LEL
monitors.)
Mixture of VOC, air and inert gas <16%
Mixture of VOC and inert gas 0-neglibible
Mixture of VOC and inert gas 0-neglibible
Mixture of VOC and inert gas 0-neglibible
— — . Treat this waste stream the same as
categories 1 and 2, except augment the
portions of the waste gas used for fuel
burning with outside air to bring its O2
content to above 16%.
— <100 Btu/scf Oxidize it directly with a sufficient a
mount of air.
— >100 Btu/scf Premix and use it as a fuel.
— Insufficient to raise gas Auxiliary fuel and combustion air for
temperature to the both the waste gas VOC and fuel are
combustion required.
temperature
Source: Adapted from Katari, et al., 1987a.
45
-------
Dilute
waste gas
to
25% LEL
Incinerate, but must
use LEL controls
and monitors
CATEGORY 3
Auxiliary air
is required
CATEGORY2
YES
No auxiliary
air Is required
CATEGORY 13
i UEL - Upper Explosive Limit
2 LEL - Lower Explosive Limit
3 The majority of waste gases generated during Superfund
remediations fall into Category 1.
Source: Adapted from V. Katari, et al., 1987a.
Inappropriate for
incineration.
Disposition depends
on composition.
(Not covered in this
section.)
Could be explosion
hazard depending
on O2 content.
Hazardous, should
not be encountered.
For majority of VOCs,
waste gas is outside
the flammability limits
if 02 <10%.
Do not use
incineration.
Use as fuel or premix
with additional fuel.
CATEGORY5
Auxiliary air may
be required for
incineration.
Auxiliary fuel
is required for
incineration.
CATEGORY 6
No auxiliary fuel is
required for
incineration.
CATEGORY 4
Figure 5-7. Flow chart for categorization of a waste gas to determine Its suitability for Incineration and need for auxiliary fuel and air.
46
-------
requirements may be 10% higher to account for heat losses. The
heat balance requires the following data:
1) The waste gas and flue gas flowrates (Ib-mol/hr);
2) The incinerator combustion temperature (typically 1,600-
2,000°F);
3) The waste gas temperature as it comes from the remedia-
tion system before it goes through a heat exchanger;
4) The concentration of VOCs in the waste gas;
5) The approximate heat capacity of the flue gas (Btu/lb-
mol/°F); and
6) The fraction of the total heat release which is recovered in
the heat exchange system.
Figure 5-8 shows a simple heat balance around an incinerator
system. From the heat balance the fuel requirement (in MMBtu/
hr) can be estimated as shown below:
d-TlV(Qscns-Qrcl)
(Eq.5-5)
where:
Qfuc,
T|
Fuel heat required, MM Btu/hr;
Fraction of heat recovered in the heat
exchanger;2
Total sensible heat required to bring
waste gas and auxiliary air to combus-
tion temperature, MMBtu/hr; and
Heat release from complete oxidation
of VOCs in the waste gas stream,
MMBtu/hr.
Equations 5-6 and 5-7 can be used to estimate the terms
required for Equation 5-5. Qscns is calculated as indicated below:
where:
m
Cp
comb
amb
(Eq.5-6)
Total sensible heat required to the bring
waste gas and auxiliary air to combus-
tion temperatures, MM Btu/hr;
Mass of flue gas (waste gas plus auxil-
iary air), Ibmol/hr,
Heat capacity of gas, Btu/lbmol/0F3;
Combustion temperature, °F; for non-
halogenated volatiles, default tempera-
ture is 1,600°F; for halogenated vola-
tiles use 2,000°F; and
Ambient air temperature, °F.
2 For recuperative heat exchange, T| is typically 0.35-0.70. For regenerative
heat exchange, T) may be as high as 0.80-0.92.
3 A rough estimate for Cp is to use the Cp of air which is approximately 6.91
Btu/lb-mol/°F.
Qrel, which is the heat release from combusting the VOCs in the
waste gas, can be estimated as follows:
m-C.
H ./10s
comb
(Eq.5-7)
v T /
where:
m
H
106
Heat release from complete oxidation
of VOCs in waste gas stream, MM Btu/
hr,
Flowrate of flue gas (waste gas plus
auxiliary air), Ibmol/hr;
Concentration of VOCs, ppmv;
Heat of combustion of VOCs in the
waste gas, MM Btu/lbmol.4; and
Conversion Factor (ppmv).
Table 5-5. Typical Pressure Drops a-b for an Incineration System
Equipment type
Pressure drop
(in. H20)
Thermal incinerator
Heat exchanger 35% efficiency
Heat exchanger 50% efficiency
Heat exchanger 70% efficiency
4"
4"
8"
'15"
Total system pressure drop equals the sum of pressure drops across all
pieces of equipment in the system.
This table is taken from V. Katari, et al., 1987a.
System Pressure Drop
The total pressure drop for an incinerator depends on the type
of equipment included in the system as well as other design
considerations. The total pressure drop across an incinerator
system determines the waste gas fan size and horsepower re-
quirements, which in turn determine the fan capital cost and
electricity consumption (Katari, et al., 1987a).
An accurate estimate of system pressure drop would require
complex calculations. A preliminary estimate can be made using
the approximate values listed in Table 5-5. The system pressure
drop is the sum of the pressure drops across the incinerator and
the heat exchanger plus the pressure drop through the duct work.
The pressure drop can then be used to estimate the power
requirement for the waste, gas fan using the empirical correlation
given below (U.S. EPA, 1990):
Power =1.17-10-4-V' P/e
(Eq.5-8)
where:
Power = Fan power requirement, kW-hr;
V = Waste gas flowrate, scfm;
P =» System pressure drop, inches of water
column; and
e = Combinedmotorfanefficiency.dimen-
sionless (approximately 60%).
A rough estimate for H ^ is to use the heat of combustion of benzene which
is approximately 1.42 MMBtu/lb-mol. Values for Ho for various VOCs are also
available in standard chemical engineering reference books.
47
-------
Heat in
Heat from auxiliary fuel
Ore.
Heat from combustion of
VOCs In waste gas
Incinerator system
Heat out
Qsens
Heat required to raise
waste gas temperature from
100°Fto700°F
I
w recovery
Heat recovered in heat exchanger
recovery = TlQams, where 11= efficiency of heat exchanger)
Figure 5-8. Incinerator heat balance.
1000
No HE
35% HE
50% HE
70% HE
10 20
Volume Flow Rate (1000 scfm)
Source: Adapted from V. Katari, et al., 19875.
Figure 5-9. Thermal Incinerator equipment cost estimates.
(Source: Adapted from V. Katari, et al., 1987b.)
48
-------
5.2.5 Cost Estimating Procedure
The process of estimating capital and annual expenses for an
incineration system can be divided into three parts. Estimates
must be made for the following:
1) Equipment costs including incinerator, stack, and con-
trols;
2) Installation, engineering, and indirect costs associated
with purchasing and installing the control equipment; and
3) Direct and indirect annual expenses.
Equipment Costs
A typical incinerator system may include the following com-
ponents: (1) a waste gas fan; (2) a refractory chamber with
burner; (3) heat recovery equipment; (4) controls, instrumenta-
tion, and control panel; and (5) a stack. In addition, other
equipment such as ductwork may be required to integrate the
incinerator with the remediation process. The equipment costs
for an incineration system generally can be estimated two ways:
1) by obtaining quotations from vendors, or 2) by using general-
ized cost correlations available in the literature.
The purchased cost of a typical incinerator system will vary
widely depending on several design factors. Consequently, cau-
tion is required when using generalized costcorrelations. Among
the factors that influence the purchased cost of a thermal incin-
eration system are the supplier's design experience, materials of
construction, instrumentation, the type of heat exchanger used,
and the nature of the installation (i.e., Do any factors exist that
make installing the equipment unusually difficult?)
Thermal incinerator equipment costs are presented in Figure 5-
9 as a function of flue gas flowrate at standard conditions of 60°
and 1 atm (absolute). This figure is adapted from an article by
Katari, et al., 1987b which used cost information from incinerator
manufacturers to develop costcorrelations. The equipment costs
given represent the cost for a complete incineration system
including acombustion chamber with burner, waste gas fan, inlet
and outlet plenums, prepiping, prewiring, instrumentation and
controls, a 10-ft stack, and in the case of heat recovery, a primary
heat exchanger. Additional cost information is available in
Vatavuk, 1990 and U.S..EPA, 1990. Cost estimates for systems
to treat less than 500 scfm should be obtained directly from
vendors.
Installation, Engineering, and Indirect Costs
The total capital investment (TCI — equipment costs plus
installation, engineering, and indirect costs) for an incineration
system can vary widely as a function of the total equipment cost.
The TCI for a small skid-mounted unit to be placed at a preprepared
site may be only 150-200% of the equipment cost. On the other
hand, for a custom installation requiring extensive site-work
(e.g., a typical Superfund site), the TCI may run as high as 300-
400% of the purchased equipment cost.
One method for generating an estimate of installation, engi-
neering, and indirect costs is to use the factor approach presented
in Table 5-6 (Katari, et al., 1987b). Based on the approach given
in this table, the TCI is approximately 160% of equipment costs,
plus any costs for site preparation and construction.
Annual Expenses
Annual costs for incinerators can be estimated from factors
given in Table 5-7. Determining these expenses requires an
extensive amount of site-specific data. Fuel costs are typically the
major direct annual cost. The system capital recovery is typically
the largest indirect expense. Additional costs may be incurred
due to monitoring requirements and permit activities.
5.3 Catalytic Oxidation
5.3.1 Process Description
Catalytic oxidation (also known as catalytic incineration) is a
commonly applied combustion technology for controlling VOC
emissions in waste gases. In catalytic oxidation a contaminant-
laden waste gas is heated with auxiliary fuel to between 600 and
900°F. The waste gas is then passed across a catalyst where the
VOC contaminants react with oxygen to form carbon dioxide and
water.
Except for the addition of a noble or base metal catalyst,
catalytic oxidizers are similar to thermal oxidation systems in
their basic design and operation (see Section 5.2). For catalytic
oxidation, the "Three T's" (temperature, residence time, and
turbulence) are also important design variables. In addition, the
catalyst type has significant effect on the system performance
and cost.
A typical catalytic oxidation system is shown in Figure 5-10.
As the figure shows, the waste gas stream is usually first passed
through a primary heat exchanger to recover heat from the
exhaust gases. Additional heat is then added to the waste gas by
a natural-gas-fired or electric preheater. From the preheater the
waste gas then passes into the catalyst bed.
The catalyst bed (or matrix) is generally a metal-mesh mat,
ceramic honeycomb, or other ceramic matrix structure designed
to maximize catalyst surface area. Catalysts may also be in the
form of spheres or pellets which may operate in either a fixed-bed
or fluidized-bed configuration. It is important that the preheat
temperature not be too high regardless of the type of catalyst. The
preheat temperature and the temperature rise across the catalyst
due to combustion must not produce temperatures which are
outside the recommended operating range for the catalyst. This
could cause the catalyst bed to lose activity.
Downstream of the catalyst bed the hot exhaust gas passes
through a heat exchanger where it gives up heat to the inlet gas
streams. In a catalytic oxidation system recuperative heat ex-
change is used. Catalytic systems using regenerative heat ex-
change are in the developmental stage. In some systems a
secondary heat recovery system such as a waste heat boiler may
also be used.
5.3.2 Applicability to Remediation Technologies
The applicability of catalytic oxidation depends primarily on
waste gas composition. As described in the preceding section on
thermal incineration, waste gas composition will determine the
.49
-------
Tabla 5-6. Cost Factors for Thermal Incinerator Capital Costs
Cost Item
Cost factor (fraction of indicated cost)
Purchased equipment
Incinerator and auxiliary equipment
Instrumentation and controls.
Taxes
Freight
Installation
Foundations and supports
Erection and handling
Electrical
Piping
Painting
Insulation
Site preparation
Building/construction
Engineering and supervision
Construction/field expenses
Construction fee
Start-up
Performance test
Contingency
Direct costs
Total equipment costs (TEC):
A
0.10 A
0.03 A
0.05 A
Total Installation Costs (TIC):
Total Direct Cost (TEC + TIC):
Indirect costs
B = 1.18A
0.08 B
0.14 B
0.04 B
0.02 B
0.01 B
0.01 B
SP
Bldg
0.30 B + SP + Bldg
1.30B
Total Indirect Costs:
Total Capital Investment
0.10 B
0.05 B
0.1 OB
0.02 B
0.01 B
0.03 B
0.21 B
1.61 B + SP + Bldg
e! V. Katrf, et al.. 1987b.
Auxiliary air
(if required)
Waste gas from
Q remediation process
Heat loss
Catalyst bed
Preheated waste gas
Flue gas
to stack
Figure 5-10. Schematic of cataljrtio oxidation system with recuperative heat exchanger.
50
-------
Table 5-7. Cost Factors for Thermal Incinerator Annual Costs •
Itemized expenditures
Cost factor
Direct costs
Labor
Operating labor
Supervision
Maintenance
Maintenance materials.
Utilities
Electricity
• Fuel
Overhead
Administrative charges
Property tax
Insurance
Capital recovery
0.5 h/shift
15% of operating labor
0.5 h/shift
100% of maintenance labor
See note b
See note c
Indirect costs
60% of sum of operating, supervisory, maintenance labor and
maintenance materials
2% x TCI
1%xTCI
1%xTCI
CRF x TCI (see note d)
power requirement the annua, operating hours and .the per kV^hr cost of electrify.
Annual Electricity Cost = Power Requirement (kW) • Operating Hours (hrs) • Electricity Cost ($/kW-hr)
c Annual fuel costs can be estimated using the system fuel requirement, the operating hours, and the per MM Btu cost of Juel.
Annual Fuel Cost = Fuel Requirement (MMBtu/hr) • Operating hours (Hrs) • Fuel Cost ($/MMBtu)
d The capital recovery factor is a function of the equipment life (typically about 10 years) and the interest rate.
auxiliary air and fuel requirements for combustion controls.
These requirements in turn will have strong influence on whether
catalytic oxidation is an economical approach for controlling air
emissions. The waste gas composition is also important in that for
catalytic oxidation to be effective the waste gas cannot contain
catalyst poisons which would limit system performance.
A table and a flow chart for determining the suitability of a
waste gas for catalytic oxidation and establishing its auxiliary
fuel and oxygen requirements were presented in Section 5.2.2 for
thermal oxidation; this same information is applicable to cata-
lytic oxidation. While catalytic oxidation has traditionally not
been widely used to control halogenated hydrocarbons, im-
proved catalysts make this application more feasible (Kittrell, et
al., 1991).
Table 5-8 presents a list of poisons/inhibitors which can
significantly degrade the catalyst activity. The presence of any of
these species in the waste gas stream would make catalytic
incineration unfavorable.
If halogenated VOCs are present in the influent gas stream,
then hydrochloric acid (HC1) may be produced in the catalytic
oxidizer. HC1 emissions are regulated and off-gas controls for
HC1 and other acid gases may be required.
Table 5-8. Common Catalyst Poisons
Sulfur
Chlorine
Chloride salts
Heavy metals (e.g., lead, arsenic)
Particulate matter
5.3 3 Range of Effectiveness
Catalytic oxidation is a well-established method for control-
ling VOC emissions in waste gases. The control efficiency (also
referred to as destruction efficiency or DE) for catalytic oxidation
is typically 90-95 percent. In some cases the efficiency can be
significantly lower, particularly when the waste stream being
controlled contains halogenated VOCs.
Factors which affect the performance of a catalytic oxidation
system include the following:
1) Operating temperature;
2) Space velocity (the reciprocal of residence time);5
3) VOC composition and concentration;
4) Catalyst properties;
Space velocity is defined as the volumetric flowrate of the flue gas entering
the catalyst bed divided by the volume of the catalyst bed. Space velocity is
the inverse of residence time and for a fixed bed catalytic oxidizer is in the
range of 30,000-100,000 hr1.
51
-------
5) Presence of poisons/inhibitors in the waste gas stream-
and '
6) Surface area of the catalyst.
The operating temperature of a catalytic incineration system is
dependent on the concentrati on and composition of the VOCs in
the waste gas stream as well as the type of catalyst used. In most
cases, the temperature at the inlet of the catalyst bed is at least
600°F while the temperature atthe outlet is less than 1200°F. The
temperature, together with catalyst space velocity, has signifi-
cant affect on system performance. At a given space velocity,
increasing the operating temperature at the inlet of the catalyst
bed increases the destruction efficiency. At a given operating
temperature,decreasingspacevelocity(i.e.,increasingresidence
time in the catalyst bed), increases destruction efficiency. How-
ever, as is the case with thermal incinerators, it is not possible to
predict beforehand the exact temperature and residence time
needed to obtain a given DRE for a VOC mixture. Rough
estimates can be made using simple models (Cooper and Alley,
1986). For example, temperatures reported for 80% DRE of
1,1,1 -trichlorethane vary from 382°F to 661 °F depending on the
catalyst used.
_ The influence of temperature and space velocity on the effec-
tiveness of a catalytic oxidation system are shown in Figures 5-
11 and 5-12, respectively. The data shown in these figures are for
a fluidized-bed catalytic oxidation system. The waste gas treated
by this unit contained 10-200 ppmv of mixed VOCs, including
aliphatic, aromatic, and halogenated compounds.
In designing a catalytic oxidation system temperature and
space velocity are not the only variables which must be consid-
ered. The waste gas composition and catalyst type must be
evaluated simultaneously since the type of catalyst chosen for a
systemplaces practical limits on the types of compounds thatcan
be treated. For example, waste gases containing chlorine and
sulfur can deactivate noble metal catalysts such as platinum.
However, chlorinated VOCs can be treated by certain metal
oxide catalysts.
The control efficiencies of some common VOC contaminants
are shown in Table 5-9 at two different operating temperatures
for the fluidized bed catalytic combustor discussed previously.
As the data show, the destruction efficiency of a catalytic oxida-
tion systcmcan vary greatly for different contaminant types. The
lowest destruction efficiencies typically are seen for chlorinated
compounds.
5.3.4 Sizing Criteria
To size a catalytic oxidizer correctly and estimate the capital
and annualizcd costs, three pieces of data are required:
1) The flue gas flow rate;
2) The auxiliary fuel and air requirements; and
3) The pressure drop across the system and waste gas fan
powerrequirements.
Table 5-9. Destruction Efficiencies of Common VOC Contami-
nants in a Fluidized Bed Combustor
Cyclohexane
Ethylbenzene
Pentane
Vinyl chloride
Dichloroethylene
Trichloroethylene
Dichloroethane
Trichloroethane
Tetrachloroethylene
Destruction
efficiency at
650° F
mean
99
98
96
93
85
83
81
79
52
Destruction
efficiency at
950° F
mean
99+
99+
99+
99
98
98
99
99
92
Flue Gas Flowrate
For dilute waste gases the flue gas flowrate6 is approximately
equal to the waste gas flowrate. In cases where auxiliary air is
required, the flue gas flowrate is roughly equal to the sum of the
waste gas flowrate and the auxiliary air flowrate. For catalytic
oxidation systems, the flue gas flowrate can be used in many
correlations to size the catalyst and the overall system.
Auxiliary Fuel Requirement
The auxiliary fuel required for a catalytic oxidizer is signifi-
cantly less than for a thermal incineration unit. In many cases
auxiliary fuel requirements will be minimal for catalytic oxida-
tion systems. However, the process of estimating fuel require-
ments is the same for both catalytic and thermal systems.
As described in Section 5.2.4, the fuel requirement can be
estimated by making a rough heat balance around the oxidation
system. The approach described below assumes heat losses to be
negligible. In many cases this is not a valid assumption and the
fuel requirements will be significantly higher than calculated by
this simple approach. The heat balance requires the following
data:
1)
2)
3)
4)
5)
6)
The waste gas and flue gas flowrates (Ibmol/hr);
The average temperature across the catalyst bed (700-
900°F);
The waste gas temperature as it comes from the remedia-
tion system before it goes through a heat exchanger;
The concentration of VOCs in the waste gas;
The approximate heat capacity of the flue gas (Btu/lbmol/
°F); and
The fraction of the total heat release which is recovered in
the heat exchange system.
A simple heat balance around a catalytic oxidation system is
shown inFigure5-13.Fromtheheatbalance the fuel requirement
(in MMBtu/hr) can be estimated using the equations shown for
thermal oxidation in Section 5.2.4.
In this discussion the term flowrate implies the flowrate at standard conditions
which are assumed to be 60°F and 1 atm, following standard engineering
52
-------
100
95
t 90
o
o
i
c
o
I
OJ
85
80
75
70
Mixture 1
• -— Mixture 2
Mixtures
- Mixture 4
Notes
CombUstor:
Mixture 1:
Mixture 2:
Mixture 3:
Mixture 4:
Fluidized bed catalytic oxidizer
Trichloroethylene/1,2-dichloroethylene
Trichloroethylene/benzene/ethylbenzene/
pentane/cyclohexane
Vinyl chloride/trichloroethylene
1,2-dichlorpethane/trichloroethylene/
1,1,2-trichloroethane/tetrachloroethylene
_L
_L
_L
600
650
700
750
800
850
900
950
1000 '
1050
Catalyst inlet temperature (°F)
Figure 5-11. Effect of temperature on destruction efficiency for catalytic oxidation at 10,500 hr -1 space velocity.
System Pressure Drop
The total pressure drop for an catalytic oxidizer depends on the
type of equipment included in the system as well as other design
considerations. The total pressure drop required across a catalytic
oxidation system determines the waste gas fan size and horse-
power requirements, which in turn determine the fan capital cost
and electricity consumption (Katari, et al., 1987a).
An accurate estimate of system pressure drop would require
complex calculations. A preliminary estimate can be made using
the approximate values listed in Table 5-10. The system pressure
drop is the sum of the pressure drops across the oxidizer and the
heat exchanger.
As for thermal oxidation, the pressure drop can then be used to
estimate the power requirement for the waste gas fan using the
empirical correlation given below (U.S. EPA, 1990):
Power = 1.17 • 10"» • V- AP/e (Eq. 5-9)
where:
Power
V
= Fan power requirement, kW-hr;
= Waste gas flowrate, scfm7;
AP = System pressure drop, inches of water
column; and
£ = Combined motor fan efficiency, dimen-
sionless (approximately 60%).
5.3.5 Cost Estimating Procedure
The process of estimating capital and annual expenses for a
catalytic oxidizer can be divided into three parts. Estimates must
be made for the following:
1) Equipment costs including combustor, catalyst, stack, and
'controls;
2) Installation, engineering, and indirect costs associated
with purchasing and installing the control equipment; and
3) Direct and indirect annual expenses.
Table 5-10. Typical Pressure Drops a'b for a Catalytic Oxidation
System
Equipment type Pressure drop (in. H2O)
Catalytic oxidizer
Heat exchanger 35% efficiency
Heat exchanger 50% efficiency
Heat exchanger 70% efficiency
6"
4"
8"
15"
1 Ib-mol at 60° F and 1 atm equals 379 scfm.
Total system pressure drop equals the sum of pressure drops across all
pieces of equipment in the system.
This table is taken from V. Katari, et al., 1987a.
53
-------
100
95
rs- 90
85
80
75
70
65
Notes
Combustor: Fluldized bed catalytic oxidizer
Mixture 1: Trichloroethylene/1,2-dichloroethylene '
Mixture 2: Trichloroethylene/benzene/ethylbenzene/
pentane/cyclohexane
Mixtures: Vinyl chloride/trichlorqethylene
Mixture 4: 1,2-dichIoroethane/trichloroethylene/
1,1,2-trichloroethane/tetrachloroethylene
JL
1 - 1 - 1 - . _ ' • '
_L
6500 7030 7500 8000 8500 9000 9500
Space velocity (hr -1)
Figure 5-12. Effect of space velocity on destruction efficiency for catalytic oxidation at 720° F.
10000
10500
11000
Equipment Costs
A typical catalytic oxidizer will include the following compo-
nents: (1) a waste gas fan; (2) combustion chamber and pre-
heater, (3) catalyst bed; (4) heat recovery equipment; (5) con-
trols, instrumentation, and control panel; and (6) a stack. In
addition, other equipment such as ductwork may be required to
integrate the oxidizer with the remediation process. The equip-
ment costs for an catalytic oxidation system can generally be
estimated two ways: 1) by obtaining quotations from vendors, or
2) by using generalized cost correlations available in the litera-
ture.
The purchased cost of a typical catalytic oxidation system will
yary widely depending on several design factors. Consequently,
caution is required when using generalized cost correlations.
Among thefactors that influence the purchased cost of a catalytic
oxidation system are the supplier's design experience, materials
of construction, instrumentation, the type of catalyst used, the
type of heat exchanger used, and the nature of the installation
(i.e., Do any factors exist which make installing the equipment
unusually difficult?)
Catalytic oxidizer equipment costs as a function of flue gas
flowrate are shown in Figure 5-14 at standard conditions of 60°F
and 1 atm (absolute). This figure is adapted from an article by
Katari, et al., 1987b which used cost information from catalytic
oxidizer manufacturers to develop cost correlations. The equip-
ment costs given represent the cost for a complete oxidation
system including a combustion chamber with burner, catalyst,
waste gas fan, inlet and outlet plenums, prepiping, prewiring,
instrumentation and controls, a 10-ft stack, and in the case of heat
recovery, a primary heat exchanger.
Another set of cost correlations for catalytic oxidation systems
have been published recently (van der Vaart, et al., 1991). Using
data provided by several vendors, the authors developed relations
between the equipment cost and the flue gas flowrate for different
levels of heat recovery. Based on regressions the following
correlations were established in 1988 dollars for fixed bed
catalytic oxidizers having flowrates between 2,000 and 50,000
scfm (Q is the flowrate in scfm, EC is the equipment cost in
dollars, and HR is the percent heat recovered):
54
-------
Heat in
Heat out
Qfuel
Heat from auxiliary fuel
Qrel
Heat from combustion of
VOCs in waste gas
Catalytic oxidation
Qsens
Heat required to raise
waste gas temperature from
100°Fto700°F
I
^ recovery
Heat recovered in heat exchanger
(Q recovery = iQsens, where 11= efficiency of heat exchanger)
Figure 5-13. Heat balance for catalytic oxidation.
EC - 1100 ' Q ?0f47 HR = 0% (Eq. 5-10)
EC - 3620 ' Q 0.419 HR - 35% (Eq. 5-11)
tot
EC = 1220-Q 0.558 HR = 50% (Eq.5-12)
EC = 1440 ' Q 0.553 HR = 70% (Eq. 5-13)
For fluidized-bed systems a second set of correlations, also in
1988 dollars, covering the range of 2,000-25,000 scfm were
developed based on data from vendors:
EC - 8.48 x!04 + 13.2 ' Qtot HR = 0% (Eq.5-14)
EC = 8.84xl04+14.6'Qtot HR = 35% (Eq.5-15)
EC = 8.66 x!04+15.8-Q(t HR = 50% (Eq.5-16)
EC = 8.39xl04+19.2-Q(M HR = 70% (Eq.5-17)
The costs for fluidized-bed systems are higher, since these
units aredesigned to handle waste streams with (1) higher heating
values, (2) higher particulate contents, and (3) chlorinated spe-
cies.
Cost estimates for systems to treat less than 2000 scfm should
be obtained directly from vendors.
Installation, Engineering, and Indirect Costs
The total capital investment (TCI—equipment costs plus in-
stallation, engineering, and indirect costs) for a catalytic oxida-
tion system can vary widely as a function of the total equipment
cost. The TCI for a small skid-mounted unit to be placed at a
preprepared site may be only 150-200% of the equipment cost.
On the other hand, for a custom installation, requiring extensive
site-work, the TCI may run as high as 300-400% of the purchased
equipment cost.
One method for generating an estimate of installation, engi-
neering, and indirect costs is to use the factor approach presented
in Table 5-6 in Section 5.2.5. Based on the approach given in this
table, the TCI is approximately 160% of equipment costs, plus
any costs for site preparation and construction.
Annual Expenses
Estimating the annual expenses associated with using catalytic
oxidation requires an extensive amount of site-specific data.
Suggested factors for estimating thermal oxidizer annual costs
are presented in Table 5-7 in Section 5.2.5. The only additional
cost consideration for catalytic oxidizers is capital recovery for
the catalyst material, This is a function of the catalyst life (e.g.,
3 years) and the interest expense. Overall utilities, the annualized
catalyst cost, and the system capital recovery are typically the
largest expenses.
5.4 Condensers
5.4.1 Process Description
Condensers are primarily used to remove VOCs from gas
streams prior to other controls such as incinerators or absorbers
but can also be used alone to control emissions of high VOC
concentration gas streams. Condensation is a recovery technique
where the volatile components of a vapor mixture are separated
from the remaining gas by a phase change. Condensation occurs
when the partial pressure of the volatile components is greater
than or equal to its vapor pressure. This situation can be achieved
by lowering the temperature or increasing the pressure of the gas
stream.
Figure 5-15a is a simple process flow diagram for condensa-
tion. A typical condensation system consists of the condenser,
refrigeration system, storage tanks, and pumps. Figure 5-15b is
55
-------
1000
No HE
35% HE
50% HE
70% HE
i ! . . . .
HE = Heat Exchanger1 •
10 20
Volume Flow Rate (1000 scfm)
50
Source: Adapted from V. Katari, et al., 1987b.
FIguro 5-14. Catalytic Incinerator equipment cost estimates.
a more detailed process diagram of an entire condensation and
recovery process. VOC off-gas is compressed as it passes through
a blower. The exiting hot gas is routed to an aftercooler com-
monly constructed of copper tubes with external aluminum fins.
Air is passed over the fins to maximize the cooling effect. Some
condensation occurs in the aftercooler. The gas stream is cooled
further in an air-to-air heat exchanger. The condenser cools the
gas to below the condensing temperature in an air-to-refrigerant
heat exchanger. The cold gas then is routed to a centrifugal
separator where the liquid is removed to a collecting vessel. The
aftercooler and heat exchanger may not be necessary for all
condensing systems. Typically, further treatment of the gas
stream isrequired for final polishing, such as acarbon adsorption
unit, before the stream can be vented to the atmosphere.
Condensing systems usually contain either a contact con-
denser or a surface condenser. Contact condensing systems cool
the gas stream by spraying ambient or chilled liquid directly into
the gas stream. The spraying is usually accomplished in a packed
column where surface area and contact time are maximized.
Somecontactcondensers aresirnple spray chambers with baffles,
while others have high velocity jets designed to produce a
vacuum. Since the coolant comes into direct contact with the
recovered contaminants, the contaminants and coolant must be
separated or extracted before either can be reused. This separa-
tion process may lead to a disposal problem or secondary emis-
sions. Contact condensers are more flexible, simpler, and less
expensive to operate than surface condensers. Contact condens-
ers usually remove more air contaminants due to greater conden-
sate dilution.
In surface condensing systems, the coolant does not make
contact with the gas stream. These condensers are usually the
shell and tube or plate/fin type. Condensed vapor forms a film on
, the cooled surface and drains into a collection vessel for storage,
reuse or disposal. Condensation can occur in the tubes or on the
shell outside of the tubes. Typically, condensers are the shell and
tube type with the coolant flowing on the inside of the tubes
counter-currently to the gas stream. Condensation occurs on the
outside of the tubes in this arrangement. The condenser is usually
horizontal but vertical condensers also exist. Surface condensers
require less water and produce 10 to 20 times less condensate than
contact condensers. Surface condensers are more likely to pro-
duce a salable product. These type of condensers have a greater
amount of maintenance due to the auxiliary equipment required!
56
-------
Emission Stream
Outlet
Emission
stream
inlet
i
Condenser
Condensed
VOC
1
i
i
Coolant
Refrigeration unit
Source: Taken from U.S. EPA, 1991
Figure 5-15a. Process flow diagram for condensation.
5.4.2 Applicability to Remediation Technologies
Condensation generally is used to remove and recover VOCs
prior to other control technologies. Condensation can be used
alone to control emissions at high VOC concentrations, i.e.,
greater than 5000 ppmv. This type of VOC control is not suited
for gas streams that contain organics with low boiling points (i.e.,
very low condensation temperatures <32°F) or gas streams with
large quantities of inert or noncondensible gases (air, nitrogen, or
methane). Condensation is a very efficient removal process for
high concentration streams.
In most control applications, the emission stream will contain
large quantities of noncondensible gases and small quantities of
condensible compounds. Care should be taken,in design and
operation to ensure limited emissions of VOCs from discharged
condensate, i.e., secondary emissions. Further subcooling of the
condensate may be required to correct this situation. If
uncondensible air contaminants are in the gas stream, these
contaminants must be either dissolved in the condensate or
vented to other control equipment. Since gas streams at Super-
fund sites will usually contain a variety of contaminants, the
recovered stream may not be salable due to purity requirements.
If so, the stream must be disposed of by incineration or some other
method. Another consideration is the moisture content of the gas
stream; any water in the stream will condense with the organic
vapors creating a dilute solvent stream. If the gas is not treated
below emission standards, the off-gas from the condenser must
be treated further, usually with activated carbon. Disposal prob-
lems and high power costs are some of the disadvantages associ-
ated with condensation.
Air-Ccooled
aftercooler
Centrifugal
separator
Refrigerant in
Source: Redrawn from APC, 1991 a.
Figure 5-15b. Schematic diagram of a vapor condensation system.
57
-------
5.43 Range of Effectiveness
Condensation has the capability of removing 50-95% of the
condcnsible VOCs. The removal efficiency is dependent on the
characteristics of the vapor stream and the condenser operating
parameters. The efficiency depends on the nature and concentra-
tion of emission stream components. For example, compounds
with high boiling points (low volatility) condense more readily
compared to those with low boiling points. The temperature
required to attain a given removal efficiency depends on the
vapor pressure of the VOC at: the vapor/liquid equilibrium. The
condensation temperature can be determined from data relating
vapor pressure and temperature. The coolant selection is based on
the required condensation temperature. Some practical limits for
coolant selection are presented in Table 5-11.
Table 5-11. Condensation Temperature Limits for Various
Categories of Coo'ants
Required condensation
temperature (°F)
Coolant
80-
60-80
45-60
-30-45
-90 to -30
Air
Water
Chilled water
Brine solutions
Freons
The effect of volatility on the condensation temperature and
removal efficiency is shown in Figure 5-16. The two components
have varying atmospheric boiling points, as shown on the figure.
For a given inlet concentration of contaminant, the removal
efficiency increases for a given condensing temperature as the
boiling point increases. ,.'.!.-.,'.
5.4.4 Sizing Criteria
Sizing the condenser involves several steps to determine the
surface area of the condenser. For a condenser system containing
a shell and tube heat exchanger, with condensate forming on the
shell side, the following design procedure can be followed (U.S.
EPA, 1991). The waste gas stream is assumed to be a two
component mixture: a condensible component (VOC) and a
noncondensible component (air). For estimating the condensa-
tion temperature, the gas stream consists of air saturated with a
VOC component. For a given removal efficiency, the partial
pressure in mm Hg for the contaminant in the exiting stream,
Pparti;i,, can be calculated:
Ppartial=760.
(1-0.01 RE)
[l-(RExlO-8HAPe)|
•HAPC x 10"6
(Eq. 5-18)
Source: Adapted from U.S. EPA, 1991.
100
80
60
40
20
Toluene (bp = 231 °F)
Xylene (bp = 292°F)
Basis: 20,000 ppmv Dry Waste Gas
10
20
30 40 50
Condensing Temperature (°F)
60
70
80
Figure 5-16. Example of condenser performance.
58
-------
where:
RE = Removal efficiency, %; and
HAPe = Contaminantconcentration in entering gas
stream, ppmv.
The condenser is assumed to operate at a constant pressure of
1 atmosphere. The condensing temperature can then be deter-
mined from equilibrium data where the calculated partial pres-
sure of the hazardous air pollutant (HAP) is equal to its vapor
pressure at that temperature. After the condensing temperature is
determined, the appropriate coolant can be selected using Table
5-11 as a guide. •'.•.<••'.
The heat load of the condenser can be determined from an
energy balance:
Hload = 1.1 x60(Hon + Honcon + H nron) (Eq.5-19)
where:
H,
'bad
Condenser heat load, Btu/hr;
H = Enthalpy of the condensed HAP; and
H
Enthalpy of the noncondensible vapors.
These parameters are defined in Table 5-12. Equations for
determining these enthalpies are also given in this table. The
factor 1.1 is included as a safety factor in the design.
Condenser systems are typically sized based on the total heat
load and the overall heat transfer coefficient. The overall heat
transfer coefficient is estimated from individual heat transfer
coefficients and from coefficients of the gas stream and coolant.
An accurate measurement of the individual coefficients can be
made from physical/ chemical data for the gas stream, the
coolant, and the specific shell and tube that is to be used. Some
typical heat transfer coefficients for condensing systems are
presented in Table 5-13. Condensers are sized from their area by
the following equation:
where:
U
Acon = Hload/UAT,
LM
(Eq.5-20)
Condenser surface area, ft2;
Overall heat transfer coefficient, Btu/
hr-ft2-°F; and
Table 5-12. Design Equations for Condensing Systems
Equations
Hc0n = HAPcon [AH + CPHAP (Te - Tcon ) ]
Huncon = HAP0,m CPHAP (Te - TCOn )
Hnoncon - [(-^-) - HAPe,mJ CPair (Te - Tcon
HAPcon = HAPe,m-HAPo,ra
= -^-)[l - HAPe X 1 O'6] [ Pvapor
^392/ [(P.- Pvapor.
HAPe m = -^-
' V3921
CP
HAP,
HAP,
HAP,
AH
Nomenclature
Average specific heat of compound, Btu/lb-mol-°F
Entering concentration of HAP, ppmv
Molar flow of HAP, inlet Ib-mol/min
Molar flow of HAP, outlet, Ib-mol/nim
Heat of evaporation, Btu/lb-mol
System pressure, mm Hg
partial
Maximum flow rate, scfm at 77° F and 1 atm
Condensing temperature, °F
Entering emission stream temperature, °F
Source: U.S. EPA, 1991.
59
-------
Table 5-13. Typical Overall Heat Transfer Coefficients in Shell and
Tube Heat Exchangers for Condensing Vapor-Liquid
Media
Shell side
High-boiling
hydrocarbons, V
Low-boiling
hydrocarbons, A
Organic solvents, A
Organic solvents
Mgh.NC.A
Organic solvents
tow NC, A
Tube side
Water
Water
Water
Water or brine
Water or brine
Design U
(Btu/ °F-
ft*-hr)
20-50
80-200
100-200
20-60
50-120
Fouling factor
(hr-ft2-°F/Btu)
0.003
0.003
0.003
0.003
0.003
V - Vacuum
A - Atmospheric pressure
NC » Non-condensiMe gas present
Source: Adapted from Perry, 1973.
AT,
LM
Logarithmic mean temperature difference,
op
(Tc-TCMl,0)-(Tcon-TCM|.)
lnKTc-TCOOI>0)-(Tcon-Tcoo,.)]
Emission stream temperature, °F;
Coolant outlet temperature, °F -
Coolant inlet temperature, °F = (T - 1 5)
where:
The coolant flow rate can be determined from a simple heat
balance:
H,«/[CPcMlM(Tcool_o -T^,.)] (Eq. 5-21)
Coolant flow rate, Ib/hr; and
Average specific heat of the coolant, Btu/
The design procedures are more complicated for a mixture of
condensible gases in a noncondensible gas. More information for
determining the condensation temperature for these type of
mixtures can be found in Lud wig, 1965 and Kern, 1950. Physical/
chemical data can be found in Perry's Chemical Engineers'
Handbook, Smith and Van Ness, 1959, and CRC, 1992. Walas,
1988 and Danielson, 1967 have more information on condenser
design.
If a refrigerant is chosen for the coolant, then a refrigeration
system also must be designed for the condensing system. The
refrigeration capacity, Ref, in units of tons is determined from:
Ref - H10M/12,000 Btu/hr-ton
(Eq. 5-22)
5.4.5 Cost Estimating Procedure
The following cost equations are for refrigerated surface
condenser systems. These cost correlations are in 1 990 dollars; a
factor of 1.1025 should be applied to these figures to convert to
1992 dollars. The refrigeration unit equipment cost (ECr) for
packaged solvent vapor recovery systems can be determined by
the following equations (Shareef, et al., 1991):
• Single stage refrigeration units (less than 10 tons)
ECr = exp(9.83 - 0.014Tcon + 0.341nR) (Eq. 5-23)
• Single stage refrigeration units (greater than or equal to
10 tons)
ECr = exp(9.26 - 0.007Tcon + 0.6271nR) (Eq. 5-24)
• Multistage refrigeration units
ECr = exp(9.73 - 0.012Tcon + 0.5841nR) (Eq: 5-25)
These costs were determined from vendor-supplied informa-
tion. The equipment cost for packaged solvent vapor recovery
systems, ECp, is estimated to be 25% greater than the refrigera-
tion unit cost. The purchased equipment cost, PECP, includes the
packaged equipment costs and factors for sales tax and freight:
PECp - 1 .08 ECp = 1 .08 (1 .25 ECr) (Eq. 5-26)
Equipment costs for custom solvent vapor recovery systems
were developed by (Shareef, et al., 1991). The following equa-
tions can be used to determine the capital costs (limitations are
noted):
EC
condcnsCr
34Acondcnscr + ^775' (Eq. 5-27)
38 to 80° ft2 of 304 stainless steel tubes
ECU = 2.72^ + 1960; (Eq. 5-28)
Vtank = 50 to 5000 gallons, 3 1 6 stainless steel
The estimated cost of the precooler can be determined from the
refrigeration unit costs by using Equations 5-23 to 5-25. The cost
of auxiliary equipment such as ductwork, piping, fans, or pumps
also needs to be determined by procedures outlined in (Vatavuk,
1990). The total equipment cost for custom systems and the
purchased equipment cost can be described by:
EC,
PEC,
(Eq.5-29)
1.18EC
~c ~ I.JLO^V.C , (Eq. 5-30)
The total capital investment can then be determined for pack-
aged and custom systems by the following:
TCI - 1.15PECp or TCI = 1.74PECC (Eq. 5-31)
The total annual operating cost correlations are summarized in
Table 5-14. This table contains the basis for calculating direct
60
-------
Table 5-14. Annual Cost Factors for Refrigerated Condenser Systems
Cost items
Factor
Direct Annual Costs, DC
Operating labor
Operator
Supervisor
Operating materials
Maintenance
Labor
Material
Electricity
at 40° F
at 20° F
at -20° F
at -50° F
at-100° F
Indirect Annual Costs, 1C
Overhead
Administrative charges
Property tax
Insurance
Capital recovery "
Recovery Credits, RC
Recovered VOC
Total annual costs:
1/2 hour per shift
15% of operator
1/2 hour per shift
100% of maintenance labor
1.3kW/ton
2.2 kW/ton
4.7 kW/ton
5.0 kW/ton
11.7kW/ton
60% of total labor and maintenance material costs
2% of total capital investment
1% of total capital investment
1 % of total capital investment
13.15% x total capital investment
Quantity recovered x operating hours
DC + 1C - RC
• Assuming a 15-year life at 10%.
Source: Reprinted from Shareef, et al., 1991.
annual costs, indirect annual costs, and recovery credits. The
recovery credit, RC, may not be applicable if the product purity
is not high enough for resale. The recovery credit can be deter-
mined from the quantity of VOC recovered:
RC = W,
VOC.con s ^ VOC
(Eq.5-32)
where:
W,
9,
Pvoc
voc,con
'
= Quantity of VOC recovered (Ib/hr);
= System operating time (hr/yr); and
- Resale value of recovered VOC ($/lb).
Other cost correlations can be found in U.S. EPA, 1991 and
Walas, 1988.
5.5 Internal Combustion Engines
5.5.1 Process Description
The principle of operation of acontrol device that incorporates
an internal combustion engine (ICE) is to use a conventional
automobile or truck ICE as a thermal incinerator. The physical
difference between ICE units and incinerators is primarily in the
geometry of the combustion chamber. A simplified schematic of
a typical ICE-based system is shown in Figure 5-17. The major
components include the engine itself (standard automobile or
truck engine), supplemental fuel supply (usually propane or
natural gas), carburetor, off-gas lines from remediation system,
and additional air emission control devices (adsorbent bed,
catalytic converter, etc.). Some pretreatment device also may be
required, since ICE units require a "clean" waste stream contain-
ing no acids and low levels of particulate matter.
Supplemental fuel is required when the VOCs in the waste
stream are at insufficient concentrations to support combustion.
This requirement is especially common for system start-ups,
remediation projects with low VOC extraction rates, and sources
such as SVEs that produce changing VOC concentrations over
time. Concentration ranges of 60,000 to 100,000 ppm at flo wrates
of 1.7 - 2.0 m3/min (60 - 70 acfrn) are possible (RSI, 1991a).
However, additional oxygen may be required to dilute the gas
stream if the VOC level exceeds 25% of the lower explosive
limit. The carburetor must be modified to include two input
valves (in addition to the change allowing gaseous rather than
liquid fuel).
A major advantage to the technology is the mobility of an ICE
unit. This advantage may be enhanced if all power needs are met
by the ICE, and no external power source is needed to drive the
remediation equipment. The ready availability of automobile
parts and wide knowledge of their operation are other advan-
tages; even the catalytic converter as an add-on control is inex-
pensive if an automobile manufacturer's unit is used.
5.5.2 Applicability to Remediation Process
ICEs may be used for VOC control from any point source
where the air stream meets certain criteria. To be economically
attractive the stream should be of relatively small flow rate
61
-------
Clean air out
Supplemental
fuel and air
Muffler
VOC
olf-gas-
r~Fr
Six cylinder engine
Catalytic
converter
-cr
Blower
Carburetor
o
Figure 5-17. Internal combustion engine-based VOC control system.
(< 1,000 cfrn) since the largest ICE is only capable of a few
hundred cfm, and it should contain high concentrations of VOCs
(> 1,000 ppm) or else supplemental fuel costs would become
excessive. For applications involving large flow rates of dilute
waste gas, however, the technology still may be potentially cost-
effective if used in conjunction with a condenser, membrane, or
other pre-treatment concentrator.
ICEs have been used for years to control landfill gases, but they
have been applied to hydrocarbon destruction only since 1986,
primarily forS VE and air stripping. Their use is most common in
California, where die majority of ICE system manufacturers are
located (Pedersen and Curtis, 1991). Their use at Superfund sites
is expected to be limited to the control of VOC emissions from
small-scale S VE systems and perhaps to small-scale air strippers.
In general, arelative lack of information exists concerning details
of ICE technology for site remediation purposes. Their use to
date, is limited due to the relatively small flow rates (hundreds of
cfms) that these units are able to handle. To date, the limited
available literature has focused on ICE use for SVE, landfill
capping, and air stripping applications.
The use of ICEs is especially attractive when it replaces the
need for electrical power to the site by using the engine to run
vacuum fans, etc. This use saves not only utility costs, but
equipment costs as well. DREs of 99+% may be achieved,
usually with a catalytic converter in place. Other advantages of
ICEs over other VOC destruction systems are in mobility and
size. The potential problems of excessive noise and labor require-
ments (for monitoring fuel intake) may be avoided by the use of
computer-controlled air-to-fuel ratios and sophisticated muffler
systems. ICEsystems with automated controls are recommended
for Superfund applications.
5.5.3 Range of Effectiveness
ICE systems typically can achieve destruction efficiencies of
99% or greater. A recent report (Pedersen and Curtis, 1991)
contains the results of several studies listing removal efficiencies
of different VOCs by ICEs for various SVE and air stripping
systems. These results are presented in Table 5-15. Additional
case study information for specific ICE units is summarized
below.
One vendor, VR Systems, has a series of portable ICEs that are
designed for use with SVE systems and also can perform tank de-
gassing. These units burn up to 100 kg/hr (220 Ib/hr) of hydrocar-
bons. They use liquid propane or natural gas as a supplementary
fuel, and are computer-controlled for higher DREs and less labor
requirements.
The Soil-Scrub(R) process (K-V Associates, Falmouth, MA)
was used with a heat-assisted SVE system (HWC, 1988). An ICE
was the primary control for this system, followed by a catalytic
converter and a GAC bed. Gasoline-soaked soil was first encap-
sulated in plastic sheets, then the soil was heated to 100°C. The
final DRE was 99.9% upon exiting all controls. The entire
remediation process took 36 hours, and the treated soil contained
no detectable benzene, toluene, and xylenes and only 82 ppm oil.
A thermal vacuum spray aeration/compressive thermal oxida-
tion system (Robert Elbert & Associates, Santa Barbara, CA)
incorporating an ICE has been used for ground water remediation
(HWC, 1988). Heat (110°F) and vacuum (12" Hg) were used to
preferentially evaporate gasoline from water, and the vapor was
sent to an ICE. The system can strip and oxidize 120 Ibs
hydrocarbons/ day; the treated water had 32 ppm contaminants
and the waste gas had 70 ppm. It required approximately 0.75 gal
fuel/hr. Two limitations to the process were noted: over-rich
62
-------
Table 5-15. Destruction Efficiencies of ICEs for SVE Systems
Parameter
THC
THC
Benzene
Ethylbenzene
Toluene
Xylenes
TPH (non-methane)
Methane HCs
Benzene
Toluene
Xylenes
Ethylbenzene
TPH
Benzene
THC
Benzene
Toluene
Ethylbenzene
Total xylenes
THC
Benzene
Toluene
Ethylbenzene
Total xylenes
Initial concentration
(ppm)
38,000
200,000 .
318,832
995
125
1,005
1,550
49,625
741
380
400
114
18
65,450
34,042
30,500
39,000
• 1,094
470
785
730
58,000
1,400
720
77
320
26,000
960
840
91
360
After catalytic
converter (ppm)
89
39
16 ppm
ND1 (<10ppb)
ND (<10ppb)
0.014
<11.5ppb
225
109
0.8
1.1
0.7
<0.5
30
14.5
1.4
4.7
67
1.6
0.63
,0.056
160
0.13
0.024
0.062
0.13
140
0.024
0.020
ND (0.02)
0.080
Removal efficiency
(%)
99.76
99.98
99.99
99.99
99.99
99.99
99.99
99.56
85.29
99.79
99.73
99.39
—
99.95
99.96
99.99
99.99
93.88
99.66
99.92
99.99
99.72
99.99
99.99
99.92
99.96
99.46
99.99
99.99
100.00
99.98
Reference
Millican,1989
Millican, 1989
Wayne Perry, 1989
Wayne Perry, 1989
Wayne Perry, 1989
Wayne Perry, 1989
Wayne Perry, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
Rippberger, 1989
Rippberger, 1989
Rippberger, 1989
Rippberger, 1989
Rippberger, 1989
Rippberger, 1989
Rippberger, 1989
, Rippberger, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
RSI, 1989
'Nondetectable
Source: Pedersen and Curtis, 1991.
combustion conditions may be met if the remediation process
occurs in a well, and the system can be smothered with excessive
water vapor.
5.5.4 Sizing Criteria/Application Rates
The sizing of an ICE device is based on the volumetric flow rate
of the waste stream to be treated. Information from several
vendors is summarized below (VR Systems, 1991 and RSI,
1991b).
VR Systems has various SVE-ICE systems with controllers
ranging in sizes from 25 to 1,000 scfm. This largest system is
actually several engines in parallel, and can destroy about 20 Ib7
hr of hydrocarbons. RENMAR, on the other hand, reports that
their SVE system is able to accommodate 100 to 200 scfm of
input gas (depending on the loading) for every 300 cubic inches
of engine capacity (Pedersen and Curtis, 1991). RSI manufac-
tures a system that can accommodate either SVE or air stripping.
Their ICE unit can handle up to 80 cfm of VOC-laden air.
5.5.5 Cost Estimating Procedure
Only limited cost data are available. A recent listing of several
commercially available systems and their prices is given in Table
5-16.
One vendor (RSI, 1991 a) compares this technology to a carbon
adsorption system; consider an example case of a $2 carbon cost
per pound of hydrocarbon adsorbed. If an ICE system leases for
$200 per day, burns $100 per day of supplemental fuel, and
destroys 15 Ibs/hr of hydrocarbon, then the cost per Ib is $3007(24
x!5) = $0.83. For this example, the cost-effectiveness cross-over
occurs at a VOC extraction rate of around 8 pounds/hour, at
which point the carbon becomes preferable.
One ICE manufacturer, RENMAR, claims an SVE system
uses only 25% of useful power so that excess power is available
for site lighting, etc., and thereby increasing the cost-effective-
ness of an ICE system. In general, however, the excess power
produced will depend on the difference between the work needed
to extract the vapors and the energy available in the vapors.
63
-------
Table 5-16. Costs for Some Commercially Available ICE Systems
Vendor
System
size
(scfm)
Cost
capital
Cost
lease per
month
Op.
expense
Comments
RSI
RENMAR
VR Systems:
Model V-3
Model V-4
Model V-5
Environmental techniques
30-60
100-200
0-250
0-500
0-1.000
100
$59,500
$52,100
$73,450
$98,880
custom
$40,250
bydistrib. $1,000 For SVE and groundwater
systems. Includes 8 hours
start-up labor.
$4,630 Prices updated from 1989
levels.
Includes operating cost: lease
$6,980 based on 3 months. Custom
$9,775 built only.
Includes operating and
maintenance. Training for
oxygen recording system:
$2,700.
Source: Vendor-supplied data.
5.6 Soil Beds/Biofflters
5.6.1 Process Description
Biofil tration is an emerging technology for controlling volatile
organic compound (VOC) emissions in waste gas streams. Bio-
filtration has been extensively used inEurope, especially for odor
control. Biofiltration has been demonstrated at full-scale (Leson
and Winer, 1991). In the biofiltration process, the waste gas is
vented through a biologically active material where the biode-
gradable VOCs are oxidized into carbon dioxide and water.
Physical sorption and chemical degradation may also occur and
contribute to the overall removal efficiency. Figure 5-18 is a
schematic of a typical single-bed biofilter system. Since the
biofilter system is biologically sensitive, the temperature and
moisture of the gas and filter bed are extremely important in
design considerations. Radial blowers are used to transport the
waste gas to the humidifier. The humidifier saturates the gas
stream to 95% relative humidity, which prevents drying out of the
filter material. The effect of the filter drying out is death of the
microorganisms and a resultant loss of control efficiency.
Some systems have automatic irrigation from the top of the
filter (soil) bed. The gas stream then enters the gas distribution
system below the filter. As the gas diffuses through the filter, air
contaminants will diffuse into the wet, biologically active layer
(biofilm) where degradation occurs. Clean gas diffuses out the
top of'the filter. Excess drainage from the filter bed is the only
potential source of wastewater discharge. In particular, where
drainage contains organic contaminants that are regulated, the
drainage is recycled to the humidifier to minimize wastewater
discharge. Since particulates in the waste stream may clog the
humidifier and the biofilter, a pre-filter may be required. A heat
exchanger may also be required to heat or cool the waste gas
stream if temper-atures are not within the optimum range (i.e.,
20-40°C).
Clean Gas
Biofilter
Humidifier
Drainage
Raw Gas
Figure 5-18. Schematic of an open, single-bed biofiltration system.
64
-------
Typically, the filter material is compost, peat, wood chips, or
soil with an inert material such as polystyrene particles or porous
clay. As the VOCs are degraded, water, carbon dioxide, mineral
salts, and biomass are generated. Mineralization leads to com-
paction of the filter material which causes an increase in back
pressure. Typically the filter material is turned over after 2 years
of operation and usually replaced 1 -2 years after turning over the
filter to prevent back pressure problems (Leson and Winer,
1991).
The most common biofilter system is an open, single-bed
system. The clean gas is vented directly to the atmosphere in an
open biofilter. Enclosed, multiple-bed systems can be stacked
and have been employed for low maintenance and space con-
straint situations.
5.6,2 Applicability to Remediation Technologies
The applicability of biofiltration is dependent on the character-
istics of the waste gas. Typical biodegradable contaminants
include: alcohols, ethers, aldehydes, ketones, amines, sulfides,
and certain monocyclic aromatics (xylene, benzene, toluene and
phenol). Waste streams containing chlorinated solvents are not
readily biodegradable and are not appropriate for emissions
control by biofiltration.
Biofiltration, as a VOC control technology, results in the
complete degradation of the biodegradable contaminants and
avoids the cross media transfer of pollutants. A major require-
ment, and thus limitation, of biofiltration is the absence of
biologically toxic substances in the waste gas, such as heavy
metals. The technology is limited to biodegradable components.
Since biofiltration is biologically sensitive, the potential sys-
tem failures represent areas that should be considered when
evaluating this technology. An undersized filter can result in
VQC air emissions due to insufficient treatment. Since the filter
is sized by off-gas flow rate and concentration, the off-gas should
remain within these design parameters during operation to pre-
vent the loss of control efficiency. Inadequate preconditioning of
the off-gas for temperature, moisture, particulates, or toxic con-
stituents can also result in the complete loss of control efficiency.
Intermittent off-gas streams can be treated with a biofilter
assuming the flow rate and concentration of the gas stream are
within the design values. Filter beds can survive shut down
periods of at least two weeks without any significant reduction in
biological activity. Shut down periods up to two months are
feasible with nutrient addition and aeration of the filter (Leson
and Winer, 1991).
Biofiltration is not known to be used as a control technology at
any Superfund sites, but this technology would be an appropriate
VOC control for large volume gas streams with low concentra-
tions (e.g. certain soil vapor extraction systems). One potential
use of biofiltration is odor control at Superfund sites assuming the
odor constituents are biodegradable. Since odor problems usu-
ally are caused by compounds with low odor thresholds, off-gas
concentrations often will be relatively low (Leson and Winer,
1991). Biofiltration also may be an appropriate treatment for
VOCs that have already been reduced -by a primary control
device. The lower concentration off-gas stream would require a
smaller filter size, which results in lower capital costs.
5.6.3 Range of Effectiveness
Biofiltration usually is cost effective for large volume gas
streams with relatively low concentrations (< 1000 ppm as
methane) of easily biodegradable contaminants (Leson and Winer,
1991). Maximum influent VOC concentrations have been found
to be 3000-5000 mg/m3 (Leson and Winer, 1991). For optimum
efficiency, the waste gas should be 20-40°C and 95% relative
humidity. The filter material should remain at 40-60% moisture
by weight and have a pH between 7 and 8 (Leson and Winer,
1991). For most easily biodegradable constituents, control effi-
ciencies greater than 90% are achievable (APC, 1991b). Degra-
dation rates for common air pollutants are typically from 10 to
100 g/m3-hr (Leson and Winer, 1991).
The key parameters affecting the control efficiency of a biofil-
tration system include the environmental conditions in the filter
material, biofilter design, filter size, and waste gas composition.
The filter must also have a large reactive area and low pressure
drops; therefore, compaction must be kept to a minimum.
5.6.4 Sizing Criteria
Typical biofilter systems have been designed to treat 1,000-
150,000 m3/hr waste gas with the systems having 10-2,000 m2 of
filter area (Leson and Winer, 1991). The depth of biofilter
material is typically three to four feet. The size of a biofilter
system is dependent on the following parameters:
1) The loading rate of waste gas;
2) The concentration of compounds in the waste gas; and
3) The rate of degradation of the compounds per unit volume.
Surface loads up to 300 m3/hr of waste gas per m2 filter area are
feasible without excessively high back pressures (Leson and
Winer, 1991). The type of filter material affects the pressure drop
across the filter. The effect of filter material on pressure drop is
shown in Figure 5-19 as a function of the surface loading rate.
3000
2 2000
1000
Conventional Filter
Compost
Optimized Filter
Compost
Mixed with Coarse
Bark
12
10 g.
8 s_
6 Q
4 I
2 ol
100 200 300 400
Surface Load (m3m-2rr1)
500
Figure 5-19. Pressure drop for two filter materials as a function of
surface loading rate.
65
-------
Higher surface loads render the biofilter more susceptible to
dehydration and heat losses caused by insufficient raw gas
conditioning. The filter's large mass often provides sufficient
buffer capacity to prevent breakthroughs during peak loadings,
which allows sizing based on average hourly peak loads (Leson
and Winer, 1991).
The removal process in biofilters has been postulated to be
controlled initially by a first-order-type biodegradation rate, but
to be limited by transport properties at low inlet air flow rates
(Miller and Canter, 1991). Pilot testing of industrial waste gas
streams with multiple contaminants is usually required, rather
than modelling, to accurately size the full scale system.
5.6".5 Cost Estimating Procedure
Capital costs have been estimated at $60-95 per ft2 filter area
for installed open, single-bed biofilter systems. Gosts of open,
multiple-bed systems are approximately two times these costs.
Enclosed systems have been estimated to cost between $95-525
per ft2 filter area, depending on the size of the biofilter and the
degree of process control (Leson and Winer, 1991). Operating
costs are $0.35-1.60 /100,000 SCF, not including filter replace-
mentcosts (APC, 1991b; Leson and Winer, 1991). Maintenance
costs are abou t one labor hour per square meter of filter per year.
5.7 Operational Controls
Operational controls are thoseprocedures or practices inherent
to the operation (and design) of control systems that can be
followed to minimize the overall long-term emissions. Among
these arc:
Adequate system design and installation;
Startup testing;
Preparation of standard operating procedures for opera-
tors;
Control of operating variables to minimize emissions;
Monitoring of system performance;
Minimization of process upsets and startups; and
Preventative and routine maintenance.
Obviously, a properly designed and operated control system is
necessary to achieve the required emission control efficiency or
emission limits. The use of experienced contractors and vendors
will help ensure that the system design and installation are done
correctly. Startup testing is advisable, with as many test condi-
tions examined, aspossible, and all meaningful datarecorded and
evaluated. Systematic checks of wiring, direction of fan and
pump rotation, integrity (leak, tightness), etc. should be made.
The startup testing results should be incorporated into the formal
standard operating procedures (SOPs) prepared for and followed
by the operators of the equipment.
Supcrfund remedial actions tend to present special problems
that affect control system operation arid effectiveness. The waste
or soil to be treated tends to be highly heterogeneous which
results in off-gas streams with variable composition. Further, the
remediation activities themselves tend to start and stop due to
.problems with equipment, weather, schedule, etc., therefore the
control systems frequently may encounter non-steady state con-
ditions (e.g. startup). The control efficiencies reported in this
document are based on steady-state operating conditions and
emissions will be significantly higher during startup, process
upsets and excursions, or when the off-gas stream is out of design
specifications.
Operating variables can be controlled to minimize emissions.
The most obvious variable to control is the treatment rate; e.g.,
the lower the feedrate to an incinerator, the lower the mass of
potential emissions. Other variables such as the aeration rate for
biodegradation systems, also directly influence emissions. Con-
trolling operating variables to minimize emissions is not always
straightforward. There may be a number of competing variables
that must be balanced for optimal control system performance.
For example, reheating of off-gas streams prior to carbon adsorp-
tion systems frequently is done to lower the relative humidity of
the gas stream and improve performance. However, the higher
temperature can affect performance negatively. These two com-
peting requirements must be offset. Similarly, pressure drop
versus control efficiency and operating cost is often an issue.
To properly operate control devices, the system design and
performance must be understood. Performance data can be
generated by routine monitoring of influent and effluent emis-
sion levels, pressure drops, operating temperatures, and so on.
Operators should maintain the monitoring system so thatplugged
lines, water in the lines, etc. don't result in misleading readings.
Proper maintenance is another obvious requirement for suc-
cessful control system operation, including routine inspection of
the equipment and implementation of corrective action when
needed.
5.8 Membrane Technology
5.8.1 Process Description
Membrane technology is an emerging control process for
volatile organic compound emissions in waste gas streams. The
membrane module acts to concentrate the organic solvent by
. being more permeable to organic constituents than air. The
imposed pressure difference across a selective membrane drives
the separation of the solvent from the gas stream.
A schematic of a typical membrane separation process is
shown in Figure 5-20. The pressure difference is caused by either
a vacuum pump on the permeate side of the membrane module
(Figure 5-20a) or a compressor before the membrane separator
(Figure 5-20b). Collected VOC emissions are transported to the
membrane module by either a blower or compressor. The im-
posed pressure difference across the membrane drives the sepa-
ration of the feed gas into a concentrated stream (permeate) and
a depleted residue gas stream. Most of the organic contaminant
is transferred through the membrane with some gas permeating
the membrane. The stripped off-gas is either vented or recycled
to the VOC source.
The concentrated permeate stream must be treated further to
either recover or dispose of the contaminants. In membrane,
systems, the membrane functions to concentrate the VOCs in the
stream. .The permeate may be treated in various ways. The
process configurations for recovering the contaminants are
shown in Figure 5-21.
66
-------
Blower
Stripped
off-gas
>• Permeate
Figure 5-20a. Membrane separation system with vacuum pump.
Compressor
Figure 5-20b. Membrane separation system with compressor.
Vacuum pump
Stripped
off-gas
Membrane
module
Permeate
In Figure 5-2la, the recovery system consists of a carbon
adsorption system to collect the solvent. The solvent is recovered
during the steam regeneration process. The vapor stream from the
regeneration process is condensed and then decanted to separate
the water from the recovered solvent. Membrane Technology
and Research, Inc. (MTR) has developed and patented aprocess
(Baker, et al., 1984) to recover the solvent by condensation after
concentration by the membrane (Figure 5-21b). Figure 5-21c
represents a membrane system that collects the solvent by direct
condensation of the permeate and polishes the stripped gas
stream with activated carbon to remove any residual VOCs.
Incineration also can be used to destroy the contaminants in the
concentrated stream.
The membrane module itself consists of an ultrathin layer of a
selective polymer supported on a porous sublayer. The polymer
layer acts as the selective barrier; the microporous substructure
provides mechanical strength for the module. Typical membrane
materials include rubber (silicone and nitrile), PVC, neoprene,
silicone polycarbonate, and other polymer compounds. In a
spiral-wound membrane module, layers of the polymer are
supported on a mocroporous structure. The module can also be in
the hollow tube form. Either a blower and vacuum pump or a
compressor is required to supply the pressure differential re-
quired for separation across the membrane. Other equipment
requirements depend on the process configuration.
5.8.2 Applicability to Remediation Technology
Membrane technology as a control device for VOC emissions
is an emerging technology. Some of the theoretical aspects of the
technology are being developed; at this time, only very limited
practical applications exist. Typically, a membrane separation
system would be used as a concentrator prior to other VOC
control devices. The concentrated waste stream generated from
the membrane module could be used to reduce the size and,
therefore, the capital and operating requirements of the primary
VOC control device.
The industrial applications that are best suited for membrane
technology are situations that require a high quality recovered
product and situations where carbon adsorption will not work
67
-------
Stripped off-gas
Condensed
VOCs
Steam
Figure 5-21a. Membrane concentrator with carbon bed adsorption recovery system.
Water
Clean gas to vent
Waste
sream
Gas
recycle
Condenser
Condensed VOCs
Figure 5*21b. MTR single-stage membrane system.
Membrane
module
Vacuum pump
Clean gas to vent -*-
Carbon adsorption unit
Blower
Waste
stream
Gas
recycle
Condenser
Membrane
module
Condensed VOCs
Figure 5-21c. Single-stage membrane separation system with carbon bed adsorber polishing.
Vacuum pump
68
-------
(e.g. recovery of ketones or aldehydes due to fire hazard or 1 ,1 , 1
- trichloroethane due to its reactivity with the carbon). Several
options are available for further treatment of the permeate. Direct
condensation of the solvent is feasible at higher concentrations,
especially for solvents that are expensive or where recovery is
necessary. Incineration would be appropriate if the concentrated
stream has a high heating value and the solvent is inexpensive.
Carbon adsorption may also be an alternative for treating the
permeate stream. The solvent can be collected in a carbon
adsorption unit and recovered by a steam regeneration process.
Overall, a membrane concentrator can result in cost savings,
improvement in the reduction of emissions and the reduction of
energy requirements for incineration (Hummel and Nelson,
1990).
5.5.3 Range of Effectiveness
Membrane technology has been reported to be applicable for
low volume, high concentration off-gases (APC, 1991 a). Gas
streams containing 0.05 - 20% organics are suitable for mem-
brane processes. Membrane separation processes are very effi-
cient bulk concentrators. The permeate stream concentration
may be 10 to 50 times the VOCs concentration of the inlet gas
stream.
The control efficiency of membrane separation technology is
influenced by the following factors:
1 ) The solvent permeability; and
2) The separation factor.
The permeability is the solvent flux across the membrane. The
permeability of a solvent is related to its diffusivity and solubil-
ity. For organic vapors, the permeabilities usually increase with
concentration and at high pressures (Baker, et al., 1987).
The separation factor is defined as the degree of concentration
the membrane can achieve and is related to the selectivity of the
membrane. The separation factor refers to the relative perme-
abilities of the solvent and gas. A higher separation factor results
in a more efficient separation process. Both of these parameters
are dependent on operating conditions such as the pressure ratio
(permeate-side pressure/inlet pressure) and the membrane mate-
rial.
For high removal efficiency, the membrane material should
exhibit high permeability and good selectivity for the solvents to
be recovered. The membrane should be durable and stable to
withstand normal wear during operation.
.
Membrane performance is determined by the selectivity and
the pressure (Peinemann, et al., 1 986). The relationship between
these parameters can be described by the following equation:
2 7
where:
a-l
C2 + 7 + •
a-l / a-l
(Eq. 5-33)
c;
7
a
Permeate concentration of solvent gas;
Feed concentration of solvent gas;
Pressure ratio = total permeate pressure (p") /
total feed pressure (p); and
Selectivity = permeability to solvent / perme-
ability to air.
A graphical representation of the relationship of pressure ratio
and permeate concentration is shown in Figure 5-22. This rela-
tionship is described by Equation 5-33. The relationship simpli-
fies when a »1/7 where the concentration is dependent only on
the pressure ratio. When a « 1/y , the permeate concentration
is determined by the selectivity (Peinemann, et al., 1986).
Permeability data can be found as a function of pressure and
selectivity for various membrane materials and contaminants in
Baker, et al., 1987. Other references with test data include:
Strathman, et al., 1986 and Peinemann, et al., 1986.
5.8.4 Sizing Criteria
The optimum membrane selectivity is chosen to balance the
capital costs of the membrane area and the cost of pumping
energy. Since the solvent flux decreases as the membrane selec-
tivity, increases, the membrane area required to treat a given
amountof solvent increases. The optimum membrane selectivity
is the lowest selectivity that will produce the desired permeate
concentration. The energy requirement for a low selectivity
membrane, however, is greater since a higher volume of gas must
be pumped to meet the permeate requirements (at a fixed perme-
ate pressure). The selection of the membrane must therefore
balance the membrane area and energy requirements (Peinemann,
etal, 1986).
The fundamental mass and energy balance equations govern-
ing the design and performance of a single-stage gas permeation
system is presented by Weller and Steiner, 1950. Further analysis
was performed by Pan and Habgood for the cross-flow pattern,
which applies to the spiral-wound membrane. The simplifying
assumptions for their analyses are: permeabilities of both com-
ponents are constant, negligible pressure drop across flow paths,
and negligible mass transfer resistances except for permeation
through the membrane. The error introduced by assuming con-
stant permeabilities was not found to be excessive in test studies
for most cases (Hummel and Nelson, 1990).
The equations describing membrane performance simplify
considerably as the feed concentration approaches zero. The
equations are outlined below (Pan and Habgood, 1974):
-1
(Eq. 5-34)
xf
xf
(Eq. 5-35)
69
-------
Selectivity Controlled
r" Approximation
Pressure Controlled
Approximation
0.0001
0.001
0.01
Pressure Ratio
Source: Redrawn from Peinemann. etal., 1986.
Figure 5-22.
where:
F
xf
x
y
7
a
Calculated permeate solvent concentrations produced by a membrane with a selectivity of 200 and a feed organic vapor
concentration of 0.5%. The three different operating regions for this type of membrane are shown.
(i-r)
(Eq.5-36)
R,
Fraction permeated (stage cut);
Mole fraction of solvent in feed gas;
Mole fraction of solvent in residue off-gas;
Mole fraction of solvent in permeate;
Pressure ratio;
Selectivity (permeability of solvent/perme-
ability of nitrogen); and
DimensioBless membrane area.
These equations are valid for finite l/xf provided that l/xf is
grcaterthan both a and 1/7. An outline of the design procedures
for a spiral-wound membrane module is shown in Figure 5-23.
The membrane area can be calculated from the following equa-
tion (Hummel and Nelson, 1990):
where:
S
d
(Eq.5-37)
Membrane area, ft2;
Membrane, thickness, ft;
Inlet molar flowrate, Ib-mol/hr;
Feed side pressure, psia; and
Qa = Permeability of sol vent, Ib-molft/hrpsi ft2=
(flux)-aYS' (partial pressure).
The permeability of the solvent, Qa, is determined experimen-
tally from testing at the appropriate operating conditions.
One consideration that should be taken into account for the
design of amembrane separation unit is the lower explosion limit
(LEL). A membrane preconcentrator handling flammable vapors
in the presence of oxygen (i.e., air) could result in shifting the
concentration within the explosive range (higher than the LEL
value).
5.5.5 Cost Estimating Procedure
The cost of implementing membrane technology as a control
mechanism for VOCs must be considered with the additional
components needed to comprise the complete system. The mem-
brane usually acts as a concentrator with the permeate stream
requiring further treatment to collect or destroy the contaminants.
As stated before, further treatment may consist of carbon adsorp-
tion (residue off-gas or permeate stream), condensation, or
incineration. A more concentrated stream that is to be incinerated
has less of an energy requirement than a diluted stream. In this
case, incineration will be more economical. S&lvent recovery by
70
-------
*Given parameters: Inlet solvent concentration
Inlet gas flow rate
Membrane selectivity
Desired pressure ratio
Select x,
concentration
in residual gas
J .
Calculate F
from Equation
5-33
J
Calculate y
from Equation
5-34
Not
adequate
Check with
desired removal
fficiency
Adequate
Valid for 1/x, finite,
where 1/x, > a and 1/y
Design area
Source: Adapted from Hummel and Nelson, 1990.
Figure 5-23. Design procedureforaspiral-wound membrane, based
on Pan and Habgood principles.
condensation may be more feasible for a more concentrated
stream. In the case of recovering the solvent by carbon adsorption
and steam regeneration, the addition of the membrane
preconcentrator was not found to be cost effective even with the
decreased gas flowrate since the carbon required to collect the
solvent remained the same (Hummel and Nelson, 1990).
The capital cost of the membrane is directly related to its
surface area. The costs of the membrane module and other system
costs associated with the membrane unit are presented in Table
5-17. These costs do not represent any permeate treatment costs.
The system costs were not clearly defined and may include
vacuum pump or compressor costs. Some cost estimates for
designs found in the literature are presented in Table 5-18. Any
specifics on the estimate are provided in the table. The cost of the
system will be very dependent on the treatment requirements of
the permeate and off-gas streams.
Cost information for capital and operating costs are presented
in Table 5-19. The effects of plant size, membrane flux, and inlet
feed concentration are considered.
An extensive cost estimate was prepared for the system shown
inFigure 5-21abyHummelandNelson, 1990. The base case was
considered to be recovery of the solvent with a carbon adsorption
unit alone. The cost estimate evaluated the effect of membrane
selectivity, control efficiency, origination of pressure differential
(compressor or vacuum pump), inlet gas flow rate, and solvent
concentration for toluene and CFC-113. The results for a 100
ppmv CFC-113 gas stream at an inlet of 250 acfm are shown in
Figure 5-24. The estimate includes the cost of solvent recovery,
by carbon adsorption and steam regeneration, which as stated
before, does not seem to be the most economical approach.
Further data can be found in the reference.
5.9 Emerging/Miscellaneous Controls
A number of emerging technologies for VOC control have
received attention in recent years. The two emerging technolo-
gies that have been best demonstrated, biofilters and membranes,
have been discussed in detail in earlier subsections. A third
emerging VOC control technology, ultraviolet (UV) photolysis,
is discussed below.
5.9.1 Process Description
Ultraviolet light technologies have been used for the destruc-
tion of toxic organics in aqueous solutions since about 1988. In
some cases, UV light has been used alone for treatment; in others,
UV light has been used in conjunction with ozone and hydrogen
peroxide, which serve as oxidants (Roy, 1990). Figure 5-23.
Design procedure for a spiral-wound membrane, based on Pan
and Habgood principles.
Recently, direct UV photolysis of organics has been achieved
experimentally using a broad spectrum of high intensity ultravio-
let light. Experiments have included treatment of water, air, and
soil. The authors have claimed that the direct UV photolysis
process can disintegrate toxic organics into non-toxic byproducts.
Ultraviolet Energy Generators, Inc. (UVERG) have claimed the
Wekhof Direct UV Photolysis Process to be "both the most
efficient and clean method of organics destruction in water, gas,
and in soil" (Wekhof, 1991).
Purus, Inc. has also developed a direct UV photolysis process
for on-site cleanup of organic contaminants. The company claims
its systems, which use xenon UV flashlamps, convert organic
contaminants into harmless byproducts. Purus, Inc. has adver-
Table 5-17. Costs for the Membrane Module and Other System
Costs
Membrane module Other system costs Source of cost information
$231 /m2
54/m2
201 /m2
Best cost estimate
1607m2" •
$231 /m2 Nitto Denko (Japan)
54/m2 (Peinemann, et al., 1986)
252/m2 (Stattman, etal., 1986)
60/m2b
• Cost basis for membrane, an approximate average of the cost data.
" Cost basis for other system costs was reduced to S60/m2 since cost data from
other sources included equipment not within the membrane system.
Adapted from Hummel and Nelson, 1990.
71
-------
Tabla 5-18. Design Specifications and Costs for Components of Membrane Control Systems
Case
System
Specifics
Cost ($)
Source
1 Vacuum pump and blower
2 Vacuum pump and blower; carbon
adsorption recovery unit
3 Compressor; carbon adsorption
4 Vacuum pump
Inlet cone.
Permeate cone..
Residue cone.
Feed flow
Volume recovered
Mem area
Mem. selectivity
Pressure ratio
Inlet cone.
Removal eff.
Feed flow
Inlet cone.
Removal eff
Feed flow
Permeate cone.
Inlet cone.
Feed flow
Volume recovered
Mem. area
Mem. solvent flow
0.5 vol%
4.4 vol%
0.1 vol%
1400 scfm
1000 I/day
1020m2
150
0.05
• 100ppm
CFC-113
• 57%
2500 CFM
1 000 ppm
CFC-113
57%
2500 CFM,
50 Wt%
0.5 sol%
1 0,000 sofm
6 l/min
1700 m*
5 l/m2 day
Membrane
Other system
Blower
Pump
Total
Membrane
System
Pump
Install &
other
Total
54,700
54,700
40,200
295,000
444,600
736,500
437,500
91,100
91,100
13,400
20,000
215,600
Peinemann, et al., 1986
Hummel & Nelson, 1990
Hummel & Nelson, 1990
Baker, etal., 1985'
tiscd a commercial system for treatment of contaminated air
emissions by UV photolysis.
A flow diagram for a direct UV photolysis system is shown in
Figure 5-25. In a typical direct UV photolysis system, contami-
nated air enters one or more processing chambers in which the
contaminants are subjected to a broad spectrum of UV light
emitted by UV flashlamps. The organic contaminants absorb
varying wavelengths of UV light. The absorbed energy causes
the bonds of the organic molecules to break apart. Under ideal
conditions, the carbon atoms of the molecules, along with oxygen
present in the air, may simply form carbon dioxide. If analysis
indicates sufficient removal of contaminants, the treated air
stream may be released to the atmosphere (Purus, 1991).
5.93 Applicability to Remediation Technologies
UV photolysis may be effective in destroying volatile organic
compounds (VOCs) in contaminated air streams. Contaminants
that could be removed from air streams by UV photolysis may
include volatile chlorinated organic compounds (e.g., trichloro-
cthylenc (TCE) and methylene chloride) and volatile organic
compounds present in gasoline and petroleum products (e.g.,
benzene and toluene).
UV photolysis has not been reported in the literature as a
control technology for air emissions at Superfund sites. This
technology may be appropriate to control emissions of toxic
organic compounds released by wastewater and groundwater
treatment technologies such as biological treatment and air
stripping. It also may be appropriate in treating air emissions
Table 5-19. Effect of Plant Size, Membrane Flux, and Feed Concen-
tration on Capital and Operating Costs for Membrane
Systems ,
Parameter
Plant size (scfm)
10,000°
5,000
2,000
500
Area (m 2)
1,700
850
340
85
Capital
costs ($) •
215,800
31,500
71 ,400
30,700
Operating
costs ($) b
67,100
33,600
13,400
3,350
Membrane flux
(l/m2 day)
10.0
5.0c
2.5
1.0
Area (m 2)
850
1,700
3,800
8,500
131,500
215,800
362,600
759,200
Feed concentration Membrane flux
52,000
67,100
97,600
188,800
(vol %)
0.5"
0.25
0.1
0.05
(l/m2 day)
5.0 215,800 '
2.5 215,800
1.0 215,800
0.5 215,800
67,100
48,800
37,800
34,000
• Membrane module costs assumed to increase directly in proportion to
membrane area. Others increase in proportion to the square root of plant size.
" Operating costs are assumed to be proportional to the solvent (low through
the plant. Includes module replacements costs (3-year lifetime).
c Base case.
Source: Adapted from Baker, et al, 1985.
72
-------
1.0
0.8
I 0.6
O 0.4
I
0.2
0.0
Selectivity-200, Vacuum Pump
Selectivity-200, Compressor
Selectivity-20, Vacuum Pump
Selectivity-20, Compressor
Selectivity-5, Vacuum Pump
Selectivity-5, Compressor
Carbon Adsorption
50
60
70 80
Overall Control Efficiency (%)
90
100
Figure 5-24. Capital cost comparison (250 acfm, 100 ppm, CFC-113 feed).
UV Flashlamp
Treated Air
UV Processing
Chamber
UV Processing
Chamber
Contaminated Air
Figure 5-25. Schematic of direct UV photolysis.
73
-------
from in situ remediation of soil by vacuum extraction. UV
photolysis would have the advantage of destroy ing toxic organic
compounds rather than transferring the compounds to another
medium (e.g., activated carbon).
5.93 Range of Effectiveness
Since studies examining the, destruction of organics through
treatment of contaminated ajr streams by UV photolysis have
been limited, the range of conditions over which this treatment
may be effective is unclear. One study indicated that the concen-
tration of TCE in an ajr stream was reduced by UV photolysis
from 300,000 ppb to 100 ppb for a residence time of approxi-
mately 3 seconds (Purus, 1991). Without additional data, the
general effectiveness of this technology for Superfund applica-
tions can not be determined.
5.9 A Sizing Criteria
No specific sizing criteria are available for treating air emis-
sions by UV photolysis. The size of a UV photolysis system will
depend primarily on the following parameters:
1 ) The flow rate of the contaminated air stream;
2) The concentrations of destructible compounds in the air
stream; and
3) The refractoriness of the compounds.
Pilot UV photolysis tests would be performed on a sample air
stream to determine the appropriate size for a full-scale system.
Cost Estimating Procedure
No cost estimates have been reported in the literature for
vapor-phase treatment using full-scale direct UV photolysis
units. Although Wekhof (1991) estimates costs for treatment of
wastewater and soil by UV photolysis, the costs for treatment of
air streams is not estimated. Costs for treatment of contami-
nated air emissions by the Purus, Inc. direct UV photolysis
process are not available.
5.10 References
The Air Pollution Consultant. VOC Emission Control During
Site Remediation. McCoy and Associates, Lakewood,
CO. September/October 1 99 1 a.
The AirPollution Consultant. Biofiltration Shows Potential as
an AirPollution Control Technology. McCoy and Asso-
ciates, Lakewood, CO. November/December 1991b.
Amcec Corp. Productbrochurefor activated carbon. Oakbrook,
EL. 1991.
Baker, R.W., I. Blume, V. Helm, A. Kahn, J. Maguire, and N.
Yoshioka. Membrane Research in Energy and Solvent
Recovery from Industrial Effluent Streams. DOE Report
DE84016819. DOE-INEL, Idaho Falls, ID. 1984.
Baker, R.W., K. Hibino, J. Mohr and T. Kuroda. Membrane
Research in Energy and Sol vent Recovery from Industrial
Effluent Streams. DOE/ID/12379-T2, report for the pe-
riod 5/1 1/83 - 31/1/85, prepared by Membrane Technol-
ogy & Research , Inc. for the U.S. DOE. DOE-INEL,
Idaho Falls, ID. 1985.
Baker, R.W., N. Yoshioka, J.M. Mohr and A. J. Khan. Sepa-
ration of Organic Vapors from Air. J. Membrane Science',
Vol.31,pp259-271.1987.
Chu, RJ. VOC Emission Control Technologies for Site
Remediation. Proceedings of National Research & De-
velopment Conference on the Control of Hazardous
Materials, Anaheim, CA. HMCRI Publications Dept,,
Greenbelt, MD. 1991.
Cooper, C. and F. Alley. Air Pollution Control: A Design
Approach. Waveland Press, Prospect Heights, IL. 1986.
CRC Handbook of Chemistry and Physics, CRC Press, Boca
Raton, FL. 1992.
Danielson, J.A., ed. Air Pollution Engineering Manual. Na-
tional Center for Air Pollution Control, U.S. Department
of Health, Education and Welfare, Cincinnati, OH. 1967.
DCI. Product brochure for activated carbon. Indianapolis, IN.
1991.
Eklund, B.M., P. Thompson, A. Ihglis, and W. Dulaney. Air
Emissions from the Treatment of Soils Contaminated with
Petroleum Fuels and Other Substances. EFA-600/R-92-
124. July 1992.
The Hazardous Waste Consultant. McCoy and Associates,
Lakewood, CO. Vol. 6, No. 5.1988.
Hesketh, H.E. AirPollution Control: Traditional and Hazard-
ous Pollutants. Technomic Publishing Co., Lancaster, PA.
1991.
Hummel, K.E. and T.P. Nelson. Test and Evaluation of a
Polymer Membrane Preconcentrator. EPA 600/2-90-016.
April 1990.
Katari, V.W., W. Vatavuk and A.H. Wehe. Incineration Tech-
niques for Control of Volatile Organic Compound Emis-
sions, Part I: Fundamentals and Process Design Consider-
ations. JAPCA Vol. 37, No. 1, Pittsburgh, PA. 1987a.
Katari, V.W., W. Vatavuk and A.H. Wehe. Incineration Tech-
niques for Control of Volatile Organic Compound Emis-
sions, Part II: Capital and Annual Operating Costs. JAPCA
Vol. 37, No. 2, Pittsburgh, PA. 1987b.
Kern, D.Q. Process Heat Transfer. McGraw-Hill, New York,
NY. 1950.
Kittrell, J., C. Quinlan, and J. Eldridge. Direct Catalytic
Oxidation of Halogenated Hydrocarbons. JAWMA, Vol.
41, No. 8, ppl 129-1133.1991.
74
-------
Leson, G. and A. Winer. Biofiltration: An Innovative Air
Pollution Control Technology for VOC Emissions.
JAWMA Vol. 41, No. 8, Pittsburgh, PA. August 1991.
Ludwig, E.E. Applied Process Design for Chemical and Petro-
chemical Plants, Vol. III. Gulf Pub., Houston, TX. 1965.
Marzone, R.R. and D.W. Oakes, Profitably Recycling Sol-
vents from Process Systems. Pollution Eng., Vol. 5, No.
10, pp23-24. New York, NY. 1973.
Miller, D. and L. Canter. Control of Aromatic Waste Air
Streams by Soil Bioreactors. Env. Progress, Vol. 10, No.
4, pp300-306. New York, NY. November 1991.
Millican, R. of SCAQMD; Pers. comm. with J. Curtis, Camp,
Dresser & McKee. Boston, MA. 1989.
Pan, C. Y. and H.W. Habgood. An Analysis of the Single Stage
Gaseous Permeation Process. Ind. Eng. Chem. Fundam.,
Vol. 13,pp323-331.1974.
Pedersen, T. and J. Curtis. Soil Vapor Extraction Technology:
Reference Handbook. EPA/540/2-91/003 (NTIS PB91-
168476). February 1991.
Peinemann, K.V., J.M. Mohr and R.W. Baker. Separation of
Organic Vapors from Air. AIChE Symp. Ser., Vol. 82, No.
250, p!9. New York, NY. 1986.
Perry, R.H. and C.H. Chilton, eds. Chemical Engineering
Handbook, 5th Ed. McGraw-Hill, New York, NY. 1973.
Purus, Inc. On-Site Organic Contaminant Destruction with
Advanced Ultraviolet Flashlamps. Product brochure for
UV lamps. Purus, San Jose, CA. 1991.
Roy, K.A. Hazmat World. Vol. 3, No. 5. From a series on UV-
oxidation technologies. Tower-Borner Publishing, Glen
. Ellyn,IL. 1990.
RSI. Personal communication from Jim Sadler to Charles
Albert of Radian Corporation. Remediation Services In-
ternational, Oxnard, CA. 199 la.
RSI. Product brochure - the S.A.V.E. System. Remediation
Services International, Oxnard, CA. 1991b.
Shareef, G.S. Personal communication from Gunseli Shareef
to Bart Eklund of Radian Corporation. 1991.
Smith, J.M. and M.C. VanNess. Introduction to Chemical
Engineering Thermodynamics, 3rd ed. McGraw-Hill, New
York, NY. 1975.
Strathman, H., C. Bell and K. Kimmerle. Development of a
Synthetic Membrane for Gas and Vapor Separation. Pure
and Applied Chem. Vol. 50, No. 12. Blackwell Scientific
Publications, Ltd. Oxford, England. 1986.
U.S. EPA. Handbook: Control Technologies for Hazardous
Air Pollutants EPA/625/6-91/014. Cincinnati, OH. June
1991.
U.S. EPA. Destruction of Chlorinated Hydrocarbons by Cata-
, lytic Oxidation. EPA/600/2-86/079. RTP, NC. 1986.
U.S. EPA. Soil Vapor Extraction Technology VOC Control
Technology Assessment. EPA/450/4-89/017 (NTIS PB90-
216995). RTP, NC. September 1989.
U.S. EPA (W.M. Vatavuk). OAQPS Control CostManual (4th
Edition) EPA/450/3-90/006 (NTIS PB90-169954). Janu-
ary 1990.
van der Vaart, D.R., W.M. Vatavuk, A.H. Wehe. The Control
Efficiencies of Thermal and Catalytic Incineration for the
Control of Volatile Organic Compounds. JAWMA Vol.
41, No. 4. Pittsburgh, PA. 1991.
Vatavuk, W.M. Estimating Costs of Air Pollution Control.
Lewis Publishers. Chelsea, MI. 1990.
VR Systems. Product brochure - Soil Venting Cost Compari-
son. Anaheim, CA. 1991.
Walas, S.M. Chemical Process Equipment Selection and De-
sign. Butterworth Publishers, Boston, MA. 1988.
Wekhof, A. Treatment of Contaminated Water, Air, and Soil
with UV Flashlamps. Env. Progress, Vol. 10, No. 4. New
York, NY. 1991.
Weller, S. and W.A. Steiner. Fractional Permeation Through
Membranes. Chem. Eng. Progress, Vol. 46, pp585-591.
New York, NY. 1950.
75
-------
-------
Chapter 6
Point Source Controls for Particulate Matter, Metals,
Acid Gases, and Dioxins and Furans
Information is presented in this chapter about various control
technologies whose primary use is to control emissions of par-
ticulate matter (PM), metals, acid gases, and dioxins and furans.
Control technologies addressed in this chapter are fabric filters
(i.e., baghouses), wet and dry electrostatic precipitators (ESPs),
wet scrubbers, dry scrubbers, operational controls, and miscella-
neous technologies such as HEPA filters. Quench chambers,
cyclones, and venturi scrubbers are covered under wet and dry
scrubbing. The discussion for each control technology includes
a process description, applicability for remediation technologies,
range of effectiveness, sizing criteria, and cost information.
The Superfund applications of the control technologies cov-
ered in this section are limited almost exclusively to incineration
and thermal desorption. The controls may each be somewhat
effective at removing paniculate matter, metals, acid gases, and
dioxins. They tend to be used in series so that the overall removal
efficiency for the train of air pollution controls (APCs) meets
design specifications. For example, an on-site incinerator may
have a series of control devices: 1) cyclone, quench tower,
baghouse, and wet scrubber; or 2) spray tower (quencher) and
baghouse; or 3) cyclone, water quench, and packed tower (HMCRI,
1991). Many other combinations also may be used.
6.1 Fabric Filters
6J.I Process Description
Fabric filters are a type of air pollution control device designed
for controlling particulate matter emissions from point sources.
A typical fabric filter consists of one or more isolated compart-
ments containing rows of fabric bags or tubes. In a fabric filter,
particle-laden gas passes up along the surface of the bags then
radially through the fabric. Particles are retained on the upstream
face of the bags, while the clean gas stream is vented to the
atmosphere. The filter is operated cyclically so that it alternates
between long periods of filtering and short periods of cleaning.
During cleaning, dust that has accumulated on the bags is
removed from the fabric surface and deposited in a hopper for
subsequent disposal.
Fabric filters will collectparticle sizes ranging from submicron
to several hundred microns in diameter at efficiencies generally
in excess of 99 percent. Gas temperatures up to about 500°F, with
surges to approximately 550°F can be accommodated routinely.
Most of the energy used to operate a fabric filter system appears
as pressure drop across the bags and associated hardware and
ducting. Typical values of pressure drop range from about 5 to 20
inches of water column.
Important process variables in a fabric filter system include
particle characteristics, gas characteristics, and fabric properties.
The most important design parameter is the air-to-cloth ratio, and
the usual operating parameter of interest is the pressure drop
across the filter system. Baghouses can be operated on an
intermittent or continuous basis. In the latter case the system is
usually divided into sections, so bags can be taken off-line for
cleaning. When cleaning must be done off-line the system tends
to be more expensive since extra cloth area (capacity) is required.
Fabric filters often are categorized by the method used to clean
the dust-cake off of the filter. Using this method of categoriza-
tion, three common types of filters are: 1) shaker filters, 2)
reverse-air filters, and 3) pulse-jet filters. The process flow
diagram for a typical shaker filter is shown in Figure 6-1.
In a shaker filter, the bags are hung on a framework that is
oscillated by a motor controlled timer. In this type of system the
baghouse usually is divided into several compartments. The flow
of gas to each compartment periodically is interrupted and the
bags are shaken to remove the collected dust. The shaking action
produces more wear on the bags than other cleaning methods. For
this reason, the bags used in this type of filter are usually heavier
and made from durable fabrics.
In the second type of fabric filter, reverse air filters, gas flow
to the bags is stopped in the compartment being cleaned, and a
reverse flow of air is directed through the bags. The advantage of
this approach is that is "gentler" than shaking, which allows the
use of more fragile or lightweight bags.
The third type of baghouse, pulse-jet fabric filters, are by far
the most common type for Superfund applications. In this type of
system a blast of compressed air is used to expand the bag and
dislodge the collected particles. One advantage of pulse jetfabric
filters is that bags can be cleaned on line, which means that fewer
bags (less capacity) are required for a given application.
6.1.2 Applicability to Remediation Technologies
Three factors that affect the feasibility of using a baghouse to
control particulate emissions are the flue gas temperature, the gas
stream composition, and the particle characteristics. The tem-
77
-------
Shaker
mechanism
Clean
air out
Dirty
air in
Figure 6-1. Fabric filter process flow diagram.
78
-------
perature of the waste gas stream to be cleaned must be above the
dewpoint of any condensibles in the stream, but below the
maximum temperature for the fabric. Condensibles will wet the
filter cake and make cleaning very difficult, as well as increase
the pressure'drop across the filters. Other gas stream and particle
characteristics also must be considered in since fabric filters may
not be suitable for certain types of gas streams or particles. For
example, a fabric filter may not be suitable for an application in
which the waste gas particulate matter contains a significant
fraction of acid mist. Also, "sticky" or adhering particles might
preclude the use of a baghouse. For baghouses operated in cold
climates that use compressed air, shelters should be constructed
around the equipment to prevent freezing of condensed moisture
in the air compressor lines.
Since baghouses are used only as particulate controls on dry
waste gases, their use as controls for Superfund site remediations
are limited to cases where incineration or thermal desorption are
being used to remediate the site. Baghouses frequently are used
in conjunction with dry scrubbers. Baghouses are also highly
effective for the removal of heavy metals.
6.1.3 Range of Effectiveness
A well designed fabric filter can achieve collection efficiencies
in excess of 99 percent, although optimal performance of the
system may not occur for a number of cleaning cycles as the new
filter material is "broken in." The fabric filter collection effi-
ciency is related to the pressure drop across the system, compo-
nent life, filter fabric, cleaning method and frequency, and the
air-to-cloth (A/C) ratio.
Modifications to improve performance include changing the
A/C ratio, using a different fabric, or replacing worn or leaking
filter bags. Collection efficiency can also be improved by de-
creasing the frequency of cleaning or allowing the system to
operate over a greater pressure drop before cleaning is initiated.
6.1.4 Sizing Criteria
The key parameter in fabric filter design is the air-to-cloth
ratio. The A/C ratio, or filtration velocity, is a defined as the
actual volumetric flow rate (acfm) divided by the total active, or
net, fabric area. Selection of an appropriate range of A/C ratios
is not based on any theoretical or empirical relationship, but
rather is based on industry and fabric filter vendor experience
from actual installations. A ratio is usually recommended for a
specific dust and a specific cleaning method.
The ranges of recommended A/C ratios for many different
dusts and fumes are summarized in Table 6-1. A conservative
estimate for the A/C ratio of particulate matter generated in
Superfund remediations would be 3.0 for woven fabric and 10.0
for a felt fabric. The A/C ratio and the emission stream flowrate
(Q) can be used to calculate the net cloth area (Anc) as shown
below in Equation 6-1:
Table 6-1. Air-To-Cloth Ratios
A/C ratio
(Eq.6-1)
Shaker/woven Pulse jet/felt
Dust ' reverse-air/woven ' reverse-air felt
Alumina
Asbestos
Bauxite
Carbon black
Coal
Cement
Fly ash
Graphite
Gypsum
Lime
Limestone
Source: U.S. EPA, 1 990.
• A
Q"- =
A/C ratio =
2.5
3.0
2.5
1.5
2.5
2.0
2.5
2.0
2.0
2.5
2.7
Net cloth area, ft2;
8
10
8
5
8
8
5
5
10
10
8
Waste gas flowrate, acfm; and
Air-to-cloth ratio, acfm/ft2.
The net cloth area is the area that must be active at any point in
time; it is not the total required cloth area. The gross cloth area
(Atc) is the total cloth area in the fabric filter, including that which
is out of service at any point for cleaning or maintenance. Given
the net cloth area, an estimate of the gross cloth area can be made
using factors given in Table 6-2 and Equation 6-2 shown below:
A -C
(Eq.6-2)
where:
£
Total cloth area, ft2;
Net cloth area, ft2; and
Design factor based on size (Table 6-2),
dimensionless.
6.1.5 Cost Estimating Procedure
The equipment costs for a fabric filter system can generally be
estimated two ways: 1) by obtaining quotations from vendors, or
2) by using generalized cost correlations available in the litera-
ture.
Table 6-2. Approximate Guide to Estimate Gross Cloth Area
Net cloth area (ft2)
Gross cloth area (ft2)
1-4,000
4,001-12,000
12,001-24,000
24,001-36,000
36,001-48,000
48,001-60,000
60,001-72,000
72,001-84,000
84,001-96,000
96,001-108,000
108,001-132,000
132,001-180,000
above 180,001
Multiply by
Multiply by
Multiply by
Multiply by
Multiply by
Multiply by
Multiply by
Multiply by
Multiply by
Multiply by
Multiply by
Multiply by
Multiply by
2.0
1.5
1.25
1.17
1.125
1.11
1.10
1.09
1.08
1.07
1.06
1.05
1.04
where:
Source: U.S. EPA, 1990.
79
-------
The purchased cost of a fabric filter system will vary widely
depending on several design factors. Consequently, caution is
required when using generalized cost correlations. Among the
factors that influence the purchased cost of a baghouse are the
supplier's design experience, materials of construction, instru-
mentation, the method of cleaning, and the nature of the applica-
tion (i.e., Are there any factors, such as "sticky" particles, that
make the application difficult?).
Baghouse equipment costs as a function of gross cloth area for
cleaning systems are presented in Figure 6-2. This figure is
adapted from U.S. EPA, 1990. The equipment costs given
represent the cost for a fabric filter system without bags. A rough
estimate of the bag cost can be made by assuming $1.00/ft2. A
moreaccurateestimaterequiresknowledgeof the fabric type and
the cleaning technique (see U.S. EPA, 1991).
The installation and engineering costs for a fabric filter system
can be estimated using the factor method presented in Table 6-3.
6.2 Electrostatic Precipitators
62.1 Process Description
Electrostatic precipitators (ESPs) are point-source particulate
matter control devices that use an electrostatic field to charge
particulate matter contained in a gas stream. The charged par-
ticles then migrate to a grounded collecting surface. The col-
lected particles are dislodged from the collector surface periodi-
cally by vibrating or rapping the collector surface, and subse-
quently are collected in a hopper at the bottom of the ESP.
A typical dry electrostatic precipitator is shown in Figure 6-3.
The major components of the ESP include: gas inlet, discharge
electrodes, collecting electrodes, rappers, cable from rectifier,
wire-tensioning weights, hopper baffles, hopper, shell, and sup-
port frame. The gas enters the ESP and passes through a series of
discharge electrodes. The discharge electrode is usually a small
diameter wire and a plate or cylinder, which together create a
nonuniform electric field. The electrodes typically are negatively
charged and create a corona around the electrode. A negative
charge is induced in the particle matter as it passes through the
corona. A grounded surface, or collectorelectrode, surrounds the
discharge electrode. The charged particle collects on the collect-
ing electrode, which is typically aplate. The charged particles are
neutralized by the collecting electrode. Common types of collect-
ing electrodes used in ESPs are presented in Figure 6-4. The
particulate matter is removed from the plate by rappers. This
device strikes the collecting electrode to dislodge the collected
particles, which then fall by gravity into a hopper. The intensity,
frequency and number of blows is determined as part of the
design. Removal of the particles is essential to ensure that the
particulates collected do not act as an insulator, thereby decreas-
ing the ability of the ESP to function. Reentrainment of the
particles must be minimized to ensure adequate control effi-
ciency. The particles are collected in the hopper and can be
disposed when necessary.
900
800
700
5 600
V"
1 500
a.
I 40°
300
200
100
Note: Costs are without! bags, insulation, or
stainless steel upgrade.
Pulse Jet
Reverse Air Row
Mechanical Shaker
_L
20 30 40 50 60
Gross Cloth Area, Ato (1000 ft2)
70
80
Source: Adapted from U.S. EPA, 1990.
Figure e-2. Fabric filter equipment cost
80
-------
Table 6-3. Cost Factors for Fabric Filter Installation and Engineering
Cost item
Cost factor (fraction of indicated cost)
Direct costs
Purchased equipment
Fabric filter w/bags
Instrumentation
Taxes
Freight
Installation
Foundations and supports
Erection and handling
Electrical '
Piping
Painting
Insulation
Site preparation
Building/construction
Indirect costs
Engineering and supervision
Construction/field expenses
Construction fee
Start-up
Performance test
Contingency
Total Equipment Costs:
Total Installation Costs:
Total Direct Costs (TEC + TIC):
Total Indirect Costs:
Total Direct + Total Indirect Costs = Total Capital Investment:
A
0.10 A
0.03 A
0.05 A
B = 1.18A
0.04 B
0.50 B
0.08 B
0.01 B
0.02 B
0.07 B
SP
Bldg.
0.72 B + SP + Bldg.
1.72 B + SP + Bldg.
0.1 OB
0.20 B
0.1 OB
0.01 B
0.01 B
0.03 B
0.45 B
2.17 B + SP + Bldg.
Source: U.S. EPA, 1990.
Several types of ESPs are commonly used. Plate-wire precipi-
tators are used to treat high volumes of gases. For example, flat
plate ESPs can handle flow rates of 100,000 to 200,000 cfm. At
Superfund sites, weighted-wire ESPs are the most commonly
used type of ESP (Donnelly, 1991). Another commonly used
type of ESP is a tubular ESP.
An alternative to "rapping" for removing the particles is
washing the sides of the ESP with water, either intermittently or
continuously, i.e., a wet electrostatic precipitator. In wet ESPs,
water is sprayed on the incoming gas stream to achieve a
saturated condition.
The electric charge is transferred to liquid droplets. The liquid
containing the particles becomes charged, is collected, and
thereby washes away from the gas stream. A potential disadvan-
tage to wet ESPs is that the collected waste stream may present
a solids and liquid handling problem.
Another operating arrangement is the two-stage ESP. In this
system, the gas stream passes through a corona discharge prior to
entering a separate collection area. The two-stage ESPs are
usually used for gas flow rates less than 50,000 cfm.
The most important variable to be considered in the design of
an ESP is the collection plate area. Collection plate area is a
function of the desired collection efficiency, gas stream flowrate,
and particle drift velocity. The particle drift velocity is a compli-
cated function of particle size, gas velocity, gas temperature,
particle resistivity, particle agglomerization, and the physical
and chemical properties of the paniculate matter. Unfortunately,
there are no easy empirical approaches to calculate drift velocity
from these variables. Vendors typically rely on experience to
estimate pressure drop.
6.2.2 Applicability to Remediation Technology
Electrostatic precipitators are very efficient paniculate matter
control devices. Efficiencies of 99 percent or more are attainable.
ESPs are capable of removing very small particulates (0.01
micron up to 70 micron diameter particles) and can treat dry or
wet particles (Brunner, 1984). Wet ESPs typically are not af-
fected by the insulation effect and are able to collect gaseous or
condensible contaminants along with the particulate matter. The
use of ESP technology incurs a relatively high capital cost and the
control efficiency is sensitive to variable gas stream conditions,
such as dust loading, flow rate, and temperature. Another limita-
tion for ESPs is the type of particles that can be removed, which
81
-------
Collector
Plate
Dirty
Gas In
Particle
Layer on
Collector
Surface
Clean
Gas Out
Collected
Particles Out
Figure 6-3. Electrostatic preclpltator process flow diagram.
••<>••••••
Gas I-K '
Flow H'
Rod Curtain
r r "] i r
L L J J L
Gas pK
Common Plate
AAAAA
Gas r-K j '
Flow '-v'
AAAAA
Zig-Zag Plate
Gas p\
Flow Lv? .
Dual Plates
!f : !f
^^
Gas
Flow
Vertical Gas
Flow Plates
Source: Redrawn from Danielson, 1973.
Figure 6-4. Special collecting eleictrodes used In electrostatic precipltators.
82
-------
is dependent on the resistivity of the particles. Wet ESPs have
some added disadvantages: corrosion potential and wastewater
treatment.
Since ESPs are used only as paniculate controls on point
sources, their use as controls for Superfund site remediations is
usually limited to cases where incineration or thermal desorption
is used to clean up the site.
6.2.3 Range of Effectiveness
ESPs control particulates effectively down to the sub-micron
range. The key parameters affecting the control efficiency of
ESPs include the following:
• Particulate composition, density, and resistivity;
• Gas stream temperature and pressure;
• Gas stream velocity;
• Power; and
• Plate area.
The particulate composition, density, and resistivity affect the
particle drift velocity. In turn, the drift velocity, along with the
control efficiency and gas flow rate, are used to determine the
collection plate area, which ultimately determines the cost of the
ESP. The relationship of the specific collection area (collector
area normalized by volumetric flow rate) and control efficiency
is shown in Figure 6-5.
The resistivity is important for determining the ability of an
ESP to collect a specific material. The optimum resistivity range
for adequate control efficiency is 104 to 1010 ohm-cm. Particles
with low resistivities impose special considerations on ESP
design. These type of particles (resistivities from 104to 107 ohm-
cm) are difficult to collect in an ESP because the particles tend to
lose their charge and drop off the collector plate and become
reentrained in the gas stream. In such cases, specially designed
collecting plates or coatings may be used to reduce reentrainment.
Particles with high resistivities can also cause ESP operating
difficulties. High resistivity particles accumulate on the collec-
tion plates and insulate the collection plate, thus reducing the
attraction between particles and collecting plate. In these cases,
using an oversized ESP and more frequent cleaning or rapping of
the collector plates may be necessary.
99.9
99
CD
I
5
90
w = Drift Velocity
X
w = 10 cm/sec
~f
L-
-e—e-
-G-
-e-
-e-
Q
o
O
-o
w = 4 cm/sec
X
A Design
O Test
0.1
0.2 0.3 0.4
Specific Collection Surface Area, A
0.5
0.6
Source: Oglesby and Nichols, 1970.
Figure 6-5. Relationship between collection efficiency and specific collection area for municipal incinerators.
83
-------
I
CD
65
55
45
35
25
80
Temperature
Relative Humidity
84
88 92
Precipitator Efficiency (%)
96
145
140
IT
O
¥
135 o
2
130
125
100
Source: Redrawn from Brunner, 1984.
Figure 6-6. Effect of gas stream temperature and humidity on collection efficiency for a specific ESP.
An alternative to oversizing the ESP is the use of conditioning
agents to reduce theresisitivity of the particles. The resistivity is
dependent on temperature, moisture conditions in the gas, and the
concentration of electronegative gases (e.g., SO2). The effect of
temperature and humidity on the precipitator efficiency for a
given ESP installation is shown in Figure 6-6.
The gasvelocityshouldbe maintained within an optimal range
to ensure that continuous reenlrainment, which usually occurs at
high gas velocities, is not a large risk. The optimum range is
usually 2-4 ft/sec (Brunner, 1984). The amount of delivered
power also affects the control efficiency, since thebestcollection
usually occurs at the highest electric field, i.e., the highest
voltage. Too high a voltage, however, may result in discharging
from the wires to the plates and thereby decrease RE and increase
powercosts. The relationship of control efficiency and delivered
power for municipal incinerators is shown in Figure 6-7.
62 A Sizing Criteria
Typical ESP design criteria are presented in Table 6-4. As
indicated, the key design parameters include:
• Paniculate composition, density, and resistivity;
• Flue gas temperature and moisture;
* Inlet particulate loading and collection efficiency;
* Specific collection area;
Number of fields;
Flue gas velocity;
Collector plate spacing;
Rapping frequency and intensity; and
Transformer rectifier power levels.
The most important front end design parameter is the specific
collection area (SCA). The SCA can be estimated from the
following equation (Vatavuk, 1990):
-In i - i*
SCA= ^ 100)
W.
(Eq.6-3)
Table 6-4. Typical Design Parameters for Electrostatic Precipita-
tors
Parameters
Value
Particulate loading (gr/acf)
Required efficiency (%)
Number of sections
SCA(ft2/1000acfm)
Average secondary voltage (kv)
Average secondary current
(mA/IOOOft3)
Gas velocity (ft/sec)
0.5-5.0
98.0-99.9
2-4
350-500
35-55
30-50
3.0-3.5
Source: U.S. EPA, 1990.
84
-------
99.9
99
z
-------
Tablo 6-5. ESP Drift Velocities for Incinerator Fly Ash in Units of
cm/sec
Particle source
Design efficiency
ESP unit
95
99
99.5 99.9
Incinerator fly ash • Plate-wire 15.3 11.4 10.6 9.4
Incinerator fly ash" Flat-plate 252. 16.9 21.1 18.3
• 200* F; no back corona
* 250* F; no back corona
Source: Adapted from U.S. EPA, 1930.
systems, which could be typical at Superfund sites. These sys-
tems are usually packaged modules that are sized and sold on the
basis of waste gas volumetric flow rate. Further design informa-
tion can be found in Vatavuk, 1990 and Oglesby and Nichols,
1970. A secondary consideration in the design of an ESP is the
material of construction. For example, stainless steel must be
used for corrosive applications.
Often the ESP design is inadequate and once the system is
installed optimization in the field may be required. Basic design
problems include undersized equipment, reentrainment, or high
resistivity particles. Agents can be injected to the system to alter
the resistivity to achieve higher removal efficiencies. The ESP
can also be manipulated effectively if performance is monitored
closely. The following measurements should be taken periodi-
cally to ensure the collection efficiency is within design param-
eters: dust loading at ESP inlet and outlet, gas velocity distribu-
tion, electrical voltage and current input, gas composition, and
dust resistivity (Oglesby and Nichols, 1970).
6.2.5 Cost Estimating Procedure
A cost analysis was performed by Vatavuk, 1990 for various
ESP systems ranging from 10,000 -1,000,000 ft2 in size. A cost
correlation forESPs based on plate area based on this analysis is
presented in Figure 6-8. The cost of material can have a signifi-
cant affect on the cost of the ESP. The cost factors for upgrading
from carbon steel to another more specialized material are shown
in Table 6-6. The costs for two-stage ESPs arepresented in Figure
6-9. This figure contains costs for the basic system and for
packaged systems. , ,
The total capital investment is determined from direct and
indirect costs. The capital cost factors for ESPs are presented in
Table 6-7. Since a packaged two-stage system includes some
installation costs, the total direct costs for installation for two-
stage ESPs is approximately 0.20B to 0.30B (Vatavuk, 1990).
The annual operating cost estimates forESPs are given in Table
6-8. Vatavuk, 1990 contains further detailed cost analysis infor-
mation for ESPs.
10000
200
100
Collection Plate Area, Ap (1000 ft2)
1000
Figure 6-8. ESP equipment cost
86
-------
Table 6-6. Cost Factors for Upgrading ESP Construction Material
Material Factor
Stainless Steel, 316
Carpenter 20 CB-3
Monel-400
Nickel-200
Titanium
1.3
1.9
2.3
3.2
4.5
Source: U.S. EPA, 1990.
6.3 Operational Controls
Operational controls are those procedures or practices inherent
to the operation (and design) of control systems that can be
followed to minimize the overall long-term emissions. Among
these are:
• Adequate system design and installation;
• Startup testing;
• Preparation of standard operating procedures for
operators;
• Control of operating variables to minimize emissions;
• Monitoring of system performance;
• Minimization of process upsets and startups; and
• Preventative and routine maintenance.
Operational controls for particulate matter and acid gas con-
trols follow the same philosophy as those for VOC controls (see
Section 5.7).
6.4 Wet Scrubbers
6.4.1 Process Description
Wet scrubbing is one of the most widely used methods of flue
gas treatment for the control of acid gases, particulate matter
(PM), heavy metals, and trace organics. The use of two or three
different types of scrubbers in sequence can result in high
removal efficiencies, particularly for acid gases. The absorption
may be either physical or chemical. Physical absorption occurs if
the pollutant is merely trapped by the liquid, e.g. particulate
matter impingement on water. Chemical absorption occurs when
a reaction takes place between the pollutant and the liquid; e.g.,
HC1 reacting with a lime-based slurry to form CaCl2. Note: dry
scrubbing, the injection of an alkaline reagent into the gas stream
120
100
* 60
8
CD
40
20
Packaged Systems
I
System without Precooler,
Installed Cell Washer, or Fan
I
6 8
How Rate (1000 scfm)
10
12
14
Figure 6-9. Purchase costs for two-stage precipitators.
87
-------
Tablo 6-7. Capita) Cost Factors for ESPs
Cost item
Factor
Direct Costs
Purchased equipment costs
ESP + auxiliary equipment
Instrumentation
Sales taxes
Freight
Purchased equipment cost, PEC
Direct installation costs
Foundations and supports
Handling and erection
Electrical
Piping
Insulation for ductwork
Painting
Direct installation costs
Site preparation
Buildings
Total Direct Costs, DC °
Indirect Costs (installation)
Engineering
Construction and field expenses
Contractor fees
Start-up
Performance test
Model study
Contingencies
Total Indirect Costs, 1C
Total Capital Investment - DC + 1C
For two-ataga pfeclpitators, total Installation direct costs are more nearly 0.20 to 0.30 B + SP + Bldg.
As estimated, A
0.10 A
0.03 A
0.05 A
B=1.18A
0.04 B
0.50 B
0.08 B
0.01 B
0.02 B
0.02 B
0.67 B
As required, SP
As required, Bldg.
1.67B + SP + Bldg.
0.20 B
0.20 B
0.10 B
0.01 B
0.01 B
0.02 B
0.03 B
0.57 B
2.24 B + SP + Bldg.
SOIKCO; Valavuk, 1890.
while not allowing the gas to be saturated with water vapor, is
discussed in Section 6.5.
The physical concepts involved in wet scrubbing are quite
simple: use of a liquid to absorb pollutants from a waste gas
stream, enhanced through a large liquid/gas contact surface area.
Gaseous matter is removed by diffusion and absorption of the
pollutant into the liquid. Quenchers work by condensing the
gases into a liquid. The absorber removal rate is a function of the
concentration of the vapor, and the equilibrium concentration of
the I iquid phase of the pollutant with the scrubber liquid. Particu-
late scrubbers (Venturis) capture particles by impingement and
agglomeration with the liquid droplets.
The use of a scrubber, which introduces a liquid into the waste
gas stream, requires a liquid separator downstream of the ab-
sorber. Separators can be cyclones, mist eliminators, or swirl
vanes, and use impaction or centrifugal force to remove the liquid
droplets from the exhaust stream. Mist eliminators can be either
the chevron or mesh pad type.
The pH of the scrubbing liquor is an important process vari-
able. Although even an acidic liquor can remove some HC1, HF,
PM, and metals, a more neutral liquor is required for high
removals of other pollutants. A pH of 5.0 or higher is required for
SO2 removal. Trace organics removal also is enhanced by an
alkaline pH. The most common alkaline scrubbing reagents are
lime, limestone, sodium hydroxide, and sodium carbonate.
Sodium-based scrubbing liquors have the advantage over
calcium reagents of being less prone to causing scale formation
on scrubber internal surfaces. Gypsum scale, which is formed in
calcium-based processes, can be particularly difficult to remove.
However, gypsum scale formation can be avoided through proper
design and operation. Sodium reagents also have the advantage
that they can provide higher removal rates, but usually they are
more expensive than lime or limestone. Calcium-based scrubber
waste products, however, are typically easier to dispose than
sodium-based solids. The higher solubility of sodium salts makes
leaching more of a problem from such waste streams. Therefore,
the planned disposition of scrubber byproduct solids will be a
factor in the selection of a reagent.
Wet scrubbing can be accomplished by many methods. Four
such processes are illustrated in Figure 6-10. A discussion of
these four and others is found below.
High Efficiency Venturi Scrubber-Venturi scrubbers often
are used as a primary control device, operating at a low pH to
remove particulates and hydrogen chloride. The particulate re-
moval efficiency in a venturi is normally in trie range of 80 to 95%
for particles larger than 0.2 microns (Brna, 1987).
88
-------
This device uses the venturi principle to scrub the waste
stream: as gas enters the venturi throat, its velocity increases, and
scrubber liquor is introduced as a spray perpendicular to the gas
stream. The high gas velocity atomizes the liquid, creating a large
surface area for absorption. The small droplets agglomerate as
the gas velocity decreases downstream.
Prefabricated units can handle waste streams of up to 80,000
cfm. Although these units are considerably less expensive than
ESPs or fabric filters, the savings may be offset by their high
pressure drop. Removal efficiency increases with increasing
particle size and increasing venturi pressure drop.
Jet Venturi Scrubber-These systems use energy from pressur-
ized liquid to induce a draft which entrains the flue gas. The jet
scrubber is capable of PM collection efficiencies of 90% or
better, and will also remove acid gases. The jet scrubber can
process gas directly from the combustion chamber, but it is
usually preceded by a quencher.
Packed Tower Scrubber-Packed towers are primarily used for
gas absorption. In this system, waste gas enters at the bottom of
the unit, and the scrubber liquid is sprayed onto the top of a
packed bed. As the gas passes upward, it contacts falling liquid
that absorbs the pollutant. A mist eliminator above the packing
removes liquid droplets from the gas stream, and cleaned gas
exits the top. This type of scrubber can typically achieve higher
removal efficiency than other scrubber devices, due to its high
liquid/gas contact surface area.
Spray Tower Scrubber-This system is also primarily used for
gas absorption. Three configurations can be used. The gas can
flow upward or countercurrent to liquid spray; downward or
cocurrent to spray; or, horizontally through a vessel with spray
perpendicular to the gas flow direction. The main design vari-
ables affecting spray tower efficiencies are tower height, liquid
to gas ratio, gas velocity, droplet size, and liquid chemistry/pH.
The spray tower must operate with a higher liquid to gas flow
ratio than the packed tower, to achieve equivalent removal rates.
Tray Scrubber—These units typically are in the form of a
vertical cylindrical tower with many levels of trays inside. The
scrubber liquor is recirculated through the absorber, with a layer
of liquor being held on each tray. The flue gas bubbles up through
holes in each tray, ensuring a high surface contact area. These
systems can be bubble-cap, perforated-tray, or valve-tray scrub-
bers. Often, one or more trays can be added to enhance the mass
transfer performance of open spray tower absorbers.
Quencher—This type of device typically is not used on its own,
but instead is used as the first step in a wet off-gas treatment
system. Quenchers are similar to spray tower scrubbers, but they
are used for pre-treatment and they are designed for temperature
control and humidification rather than pollutant removal. A
quencher can cool the off-gas from incineration temperatures to
saturation or near saturation temperature, increase the humidity
to or near saturation, and reduce gas volume. The degree of
approach to saturation temperature depends mainly on the liquid
rate, droplet surface area, and gas residence time. This increased
Table 6-8. Annual Operating Maintenance Costs for ESP System
Cost item
Calculation
Direct Annual Costs, DC
Operating labor
Operator
Supervisor
Coordinator
Operating Materials
Maintenance
Labor
Material
Utilities
Electricity-fan
Electricity-operating
Waste disposal
Total DC
Indirect Annual Costs, 1C
Overhead
Administrative charges
Property Tax
Insurance
Total 1C
Total Annual Cost (rounded)
3 hr/day x 6 (day/yr) x hourly rate
15% of operator
1/3 of operator
0.0825A (minimum of $5,265)
1% of purchased equipment cost
0.000181 x Q x AP x 9 x electricity cost ($/kwh)
1.94x10-3xAx0x electricity cost
DD=4.29x10-6xGx9 xQx [T+ (TM x D)]
60% of sum of operating, supv., coord., and maint. labor and maintenance
materials
2% of TCI
1% of TCI
1% of TCI
DC + 1C
where: 8 = Operating time (hr/yr);
A - ESP plate area (ft2)
Q = Flow rate (acfm);
AP m System pressure drop (in H2O);
Source: Adapted from Vatavuk, 1990.
G = ESP inlet grain loading (gr/ft3);
T - Tipping fee ($/ton);
TM = Mileage rate ($/ton mile); and
D = Hauling distance (mile).
89
-------
Dirty
gas in
Clean gas
out
solvent
in
o o o o
°
Bubble caps or
perforated
trays
Slurry
out
Clean gas
put
Mist
eliminator
Packing
DirtV
gas in
'••*••*••*••'••'•***•*••'•'
0 0
0 . 0
Clear liquid
wash
Scrubbing
liquid in
Slurry
out
a. Tray Scrubber
b. Packed Tower
Dirty,
gas in
Scrubbing
liquid
recycle
&. f»J
<;'
/
Tank
Clean
gas
out
Hot gas
from
incinerator
Quench
tower
Water
recycle
Quenched
gas to
scrubber
c. Venturl Scrubber
d. Quencher (spray tower)
Figure 6-10. Four common types of wet scrubbing systems.
90
-------
humidity inhibits evaporation and contributes to absorption
efficiency in downstream devices.
Quenchers are capable of some scrubbing, and can remove up
to 50% of the acid gas present in a waste stream, (if the scrubber
liquor contains an alkaline reagent). The quencher can be fol-
lowed by a higher efficiency scrubber unit, such as a venturi or
packed tower. •
6.4.2 Applicability to Remediation Technologies
Incineration commonly generates the pollutants that wet scrub-
bers remove most cost-effectively: acid gases, heavy metals, and
particulate matter. Wet scrubbing is also applicable to other
thermal treatment methods such as thermal desorption. The
particulate removal efficiency of a wet scrubber usually is not as
high as that of a baghouse or ESP. One possible drawback to the
use of wet scrubbers is that the flue gas temperature is cooled to
its saturation temperature. This will lower the dispersion charac-
teristics of the released flue gas and can result in a visible plume.
Hue gas reheating may be required to eliminate the plume.
Some attributes of wet scrubbers that make them a popular
choice among pollution control devices is their simplicity of
operation, as compared with some other control devices. How-
ever, there are drawbacks: for VOC control, solutions other than
water are usually required, since most VOCs are not water-
soluble. Therefore, proprietary solvents are required, which
raises the cost. Also, these devices are not efficient for low VOC
concentrations, and must be used in conjunction with other
controls for these applications.
Wet scrubbers achieve high efficiencies for the removal of
heavy metals, since most of the volatile metals (except mercury)
will condense at the temperatures reached in wet scrubbers.
Mercury has a higher vapor pressure than the other heavy metals,
and will not condense a"s readily. The collection efficiency of
mercury vapors in wet scrubbers is not well known.
Physical scrubbing does not destroy pollutants, it only trans-
fers them from the gaseous medium to a solid or liquid medium.
Chemical absorption may well neutralize but will not destroy the
pollutant, as when acid gases are chemically absorbed by an
alkaline solution and form salts. Therefore, some waste usually
is generated, which must be treated or disposed. This additional
cost will affect the economics of the wet scrubber option.
Because of the relatively specialized nature of wet scrubbers,
there is not one standard configuration used for all jobs; rather,
the scrubber sy stem(s) chosen will depend on the pollutants to be
removed and on the physical characteristics of the waste stream.
Typical wet scrubbing systems used with thermal treatment of
hazardous materials and wastes are described below.
Thermal Desorption—One combination of wet scrubbers used
for this application is a quench chamber followed by a venturi
scrubber. This system would primarily be used to control particu-
late emissions, but could also control acid gases if the hazardous
waste feed contains halogenated compounds.
Incineration—Wet scrubbers typically are used for cases not
requiring very high total and fine (<10 micron) particulate
removal or when inlet loading is not too high. All types of
scrubbers may be used, depending on the waste stream param-
eters. Sometimes, wet scrubbers are the only controls used for
removal of pollutants. Their ability to remove heavy metals, trace
organics, and acid gases makes them especially well-suited to
incinerator control.
One typical combination would be a quencher for cooling and
condensing the hot waste gas as well as initial acid gas collection,
followed by a venturi for primary particulate removal, and then
a packed or spray tower for final acid gas absorption.
A wet scrubber also may be used downstream of an ESP or
fabric filter, in situations that require higher PM removal than a
venturi alone can achieve, to control acid gases, and to control
some fraction of the heavy metals and trace organics that make
it through any prior APCDs.
6.4.3 Range of Effectiveness
Absorption efficiency depends on many factors, including
viscosity, diffusivity, density, temperature, liquid surface area,
system chemistry, and flow rates. Absorption efficiency is en-
hanced by increasing liquid-gas interface surface area, reducing
temperature, and maintaining a high liquid-gas ratio. Wet scrub-
bing does not remove pollutants efficiently from low volumetric
flows; however, increased turbulence enhances removal rates.
Typical liquids used to date include water, non-volatile organics,
and alkaline solutions. The latter is especially common as the
liquid in a second scrubber, since absorption of acid gases is
enhanced if the pH of the solution is maintained in the range 6.5
to 9.0.
The use of common types of wet scrubbers for PM control is
limited by the particle size distribution and the removal require-
ments. The wet scrubber systems described above may reach a
maximum of about 99.5% removal. Newer hybrid types of
scrubbers are being developed that combine wet scrubbing with
other processes, such as ionization, increasing the equipment's
range of effectiveness and improving its economics.
Wet scrubbers are often the technology of choice for high
removal rates of acid gases. HC1 removal efficiency will usually
be greater than 99%. SO2 is more difficult to remove than HC1,
but removal rates greater than 95% can be achieved.
6.4.4 Sizing Criteria
Sizing wet scrubbers for remediation applications is a difficult
process, requiring the determination of many design variables,
including flue gas saturation temperature,.tower heights, packing
requirements, etc. The interested reader is referred to U.S. EPA,
1991 or Vatavuk, 1990, which asserts "...in reality, column
design is so complex that nearly all towers are custom fabricated,
making itextremely difficultto postulate a general costing/sizing
procedure."
Some of the costing equations given in the next subsection
require an estimate of the diameter of the scrubber unit. The
91
-------
diameter (ft.) can be calculated from the flow rate Q (cfm) and
velocity V (ft/min) using the following equation:
(Eq.6-4)
A typical design velocity for non-venturi scrubbers is 10 ft/sec.
In a wet scrubber, the gas will be saturated with water vapor, and
the gas volumetric flow rate, Q, must be based on this saturation
temperature. Thescrubber process should be optimized by choos-
ing the best diameter and gas velocity.
6.43 Cost Estimating Procedure
Capital Coste—Equipment costs will vary with the system. A
representation of some equipment costs are presented below.
A venturi system consisting of a mist eliminator, recirculation
liquid pump, and sump in addition to the venturi, all in carbon
steel, would cost (Vatavuk, 1990):
P-S9018 +1.55 Q for 600 <: Q < 19,000 acfm (Eq. 6-5)
P-S92.8 Q0612 for 19,000 < Q < 59,000 acfm (Eq. 6-6)
For applications with corrosive streams, carbon steel is not
appropriate. To obtain the cost of this venturi system with a
rubber lining, or fabricated of fiber-reinforced plastic, multiply
the above equations by 1.6. For epoxy coated carbon steel,
multiply by 1.1. Instruments and controls cost approximately
10% of the equipment cost.
A costing equation was developed (Vatavuk, 1990) for im-
pingement scrubbers of 304 stainless with 1, 2, or 3 stages
(levels), internal sprays and piping. This estimate does not
include fan.pumps, orother auxiliary equipment. In the presence
of chlorides, type 304 stainless is not appropriate and higher
grade alloys or linings are required. The following equation is
valid for parameters listed in the table below:
P - $ a Qb with 900 < Q < 77,000 acfm
(Eq.6-7)
ThefollowingtwocostingguidelinesareadaptedfromHesketh,
1991:
Stages
Effective height, ft.
1
2
3
8
16
24
58.9
68.8
69.5
0.570
0.586
0.610
Purchase cost of packed tower absorbers, Pa, in 1992
dollars can be estimated by:
Pa-1337 D0-75
(Eq.6-8)
where:
D - Column diameter, in inches, from 10 to 200.
Packing cost per cubic foot of material Pp is:
Pp =
32.2
(Sp1-05)
(Eq. 6-9)
where:
Sp = Packing size from 1 to 3 in.
Hesketh's estimate for mist eliminators is $121/ft2 of cross-
sectional area. Vatavuk, 1990, estimates mist eliminator cost for
mesh pad designs by:
86.4 D1-66
(Eq. 6-10)
where, again, D is the diameter of the unit and typically is
between 2 to 10 feet.
Installed capital costs are 2.2 times equipment cosii. Installed
absorber costs are in the range of $9.1 -18/scfm for systems with
a flow greater than 10,000 scfm (Hesketh, 1991).
Operation and Maintenance Costs-A. conservative estimate
(Vatavuk, 1990) is that operation and maintenance must be
performed 1/8 of the annual running time. Corrosion and scaling
are the main sources of maintenance. Mist eliminators will need
periodic replacement. The major operation costs are for chemical
reagent, makeup water, and operating labor. Additional cost
considerations include:
• Absorbers usually operate automatically, needing little
labor or maintenance;
• Where Venturis are used, most power is to regain pressure
after scrubbing;
• Unlike fume destruction, absorbers incur fees for disposal
of waste solids; and
• The wastewater may be disposable in an existing-on-site
treatment system or used to cool heated soil.
6.5 Dry Scrubbers
6.5.1 Process Description
There are two principal types of dry absorption systems: dry-
dry and semi-dry absorption. Dry-dry systems inject the alkali
absorbent as a dry powder, and semi-dry systems inject the alkali
in a concentrated slurry, then evaporate the liquid. Both types of
systems remove any unreacted alkali and solid wastes via ESP?
or fabric filters. A dry scrubbing device includes a chemical
injection zone, a reaction zone where the pollutants react with the
alkali, and a particle removal device where the solids are re-
moved from the waste stream. Figure 6-11 illustrates three
commonly used dry scrubbing systems.
The most common use of this technology with remediation
systems is for incinerators. Wherever halogenated compounds
are thermally destroyed, some type of scrubber (wet or dry) will
usually be required to remove the resulting acid gas. Dry scrub-
bers operate on absorption principles similar to wet scrubbers,
but produce lower pressure drops and require less power. An-
other difference is that the waste gas is not saturated with
92
-------
Flue gas in
Quench
tower
(optional)
I Water
(optional)
Figure 6-11 a. Dry sorbent injection process.
Hydrated
lime silo
-Air
Blower
Dust
collector
Stack
Fan
Dry waste
Hydrated
lime silo
Air.
Blower
Precollector
Circulating
fluid bed
reactor
Recycle
Water
Flue gas in
Dust
collector
Periodic
recycle
Stack
Dry waste
Figure 6-11b. Circulating fluid bed reactor process.
Pump
Particle recycle
(optional)
Dry waste
Figure 6-11c. Spray dryer absorption (semi-dry) process.
93
-------
moisture, and the waste material is collected in dry form. With
wet scrubbing the plume must usually be reheated, and waste is
typically aslurry. A wet waste can be a liability, if alarger volume
of waste is generated for disposal. These advantages of dry
scrubbers are sometimes offset by their need for a separate
downstream paniculate collection device. High sulfur removal
rates (>90%) are generally more difficult and expensive to
achieveinadryscrubberthaninawetscrubber.Thestoichiomet-
ric ratio (i.e. moles of absorbent per mole of pollutant absorbed)
is higher with dry scrubbers than with wet scrubbers.
Practically all dry chemical absorbers use alkalis, in particular
calcium and sodium, which have good absorbent properties for
many of the acid gases and for some organic pollutants. In dry
scrubbers the calcium-based absorbent is usually in the form of
slaked lime, Ca(OH)2. The solids, including both used and
unused absorbentparticles, and particulate matter from the waste
stream, are collected at the bottom of the absorber vessel, and in
a baghousc or other particulate separator. Removal efficiency is
mainly dependent on the acid-alkali ratio and the outlet gas
temperature. Fabric filters and ESPs are the most common
removal devices downstream of dry scrubbers. Acid gas removal
efficiency is enhanced as the gas passes through the particle
separator. However, a fabric filter provides greater removal of
acid gases than does an ESP. The collected solids often can be
recycled, to achieve greater utilization.
The use of dry scrubbers for power plant applications is
relatively common, and standard design and operating param-
eters have been developed. The use of dry scrubbers for hazard-
ous waste incinerators and remediation projects, however, is less
well-defined. There are many control possibilities, and no single
combination of scrubbers, fabric filters, ESPs, etc. is best for
every application. Some typical systems are described below.
Dry-Dry Systems—Dry sorbentinjection (DSI) involves inject-
ing hydrated lime or limestone directly into the furnace or into the
ductwork downstream of the furnace (see Figure 6-1 la). Sulfur
dioxide removal efficiencies of about50% have beenreported for
this technology. The process has a high alkali requirement, and
erosion of mechanical components can be aproblem. Dry sorbent
injection uses pneumatic equipment to introduce alkali into the
hot gas stream. The gas stream may be humidified between the
furnace and the particulate control equipment to improve re-
moval efficiency. In general, SO2 removal efficiencies increase
with more alkali, greater contact time, higher pollutant concen-
tration, higher moisture content, and lower temperature at the
particulate control device inlet (to about 20-30°F above satura-
tion temperature). On the other hand, if the gas stream is not
humidified, higher temperatures may make separation of other
pollutants difficult, and also may allow formation of polychlori-
natcddibcnzodioxins(PODDs),polychlorinateddibenzodifurans
(PCDFs), and other products of incomplete combustion (PICs).
Dry-dry systems therefore are not used as the only device for
control of hazardous air pollutants (HAPs).
Semi-Dry Systems—This process consists of conditioning the
gas to a temperature above its saturation point by the adiabatic
evaporation of water. An alkali is injected into the gas either as
a slurry with the water, or separately. As with wet scrubbing, this
causes cooling and a decrease in gas volume, as well as chemical
reaction with the acid gases. A fabric filter or ESP then collects
the particles and also may serve as a secondary reaction bed.
Semi-dry systems have the advantage of no liquid waste, but by
cooling the waste gas avoid the dry-dry problems with PICs
discussed above. Further, the presence of liquid droplets in the
semi-dry process produces higher acid gas removal rates than do
dry-dry systems.
In spray dryer absorber (SDA) systems, hydrated lime is the
most common alkali sorbent, and is mixed with water to form a
slurry of approximately 15% Ca(OH)2. The dosage of lime is
regulated according to the acid gas concentration in the flue gas
and.the desired removal rate. This process often is used for power
plants and municipal solid waste (MSW) incinerators, and as
such has been well studied. The slurry can be introduced to the
SDA by single or multiple rotary atomizers, or multiple dual fluid
nozzles. Slurry droplet size is about 50 to 90 urn. With respect to
gas flow, the SDA vessel .can be downflow, upflow, or upflow
with a cyclone pre-collector, and can have single- or multiple-gas
inlets. Acid gas removal can be as high as 95% for SO2,994-% for
HC1, 99+% for SO3, and 95% for HF. The temperature of
incoming gas can be up to 1000°C since the evaporating liquid
will cool it back to 100 to 180°C. Flue gas residence time is J 0 to
18 seconds, and up to 25% of the reaction products and ash can
be collected in the SDA vessel.
6.5.2 Applicability to Remediation Technologies
Dry absorption has two features that make it relatively less
attractive as a control for remediation purposes. First, it is
difficult to achieve the very high removal efficiencies of wet
scrubbers or other technologies. Second, dry absorption adds to
the volume of waste to be disposed of because of the absorbent
in the slurry. On the other hand, dry absorption has some aspects
whicrrare useful for some remediation applications; it is able to
handle heavy metals, PM, and acid gases, as well as trace
organics and some PICs. PCDD and PCDF removal rates for
SDA/baghouse systems can reach 90 to 99+% (Donnelly, 1991).
Dry activated carbon can be added to the SDA to increase the
removal of heavy metals and trace organics. SDA systems also
show promise for mercury removal. The applications where dry
absorption is used are described below.
Incinerators—Dry absorption is effective for every type of
pollutant in the incinerator exhaust gas streams, to varying
degrees; sticky or corrosive particles are not a problem. Spray
dryer absorption often is used, with a spray dryer coupled with a
fabric filter. Reagent addition, flue gas humidification, and most
of the absorption take place in the dryer; additional absorption
occurs in the collector as dust is removed from the gas stream.
The acid gases, trace metals, and trace organic compounds
present in the waste stream of an incinerator are the pollutant
types that SDAs control best. High electrical power costs can
result from maintaining air pressure to dual fluid nozzles. Rotary
atomizers consume less electrical power than dual fluid nozzles.
Reagent and disposal costs are generally higher than for wet
absorbers; however, the dry system itself has a lower capital
investment cost.
The end product from an SDA is hygroscopic with a significant
soluble fraction, stickier than fly ash and more difficult to handle.
94
-------
End.product constituents include fly ash, calcium compounds,
trace metals, and trace organics. Due to its origin, this material
must be disposed of as hazardous waste. If the fabric filter would
be clogged by lime residue or unable to withstand the high
temperatures, another post-SDA control, such as an ESP, can be
used. , .
Thermal Desorption—Systems used as controls for thermal
desorbers typically include an SDA for acid gases with a bag-
house filter forparticulate matter. This combination can meetthe
desired removal rates for acid gases, heavy metals, and trace
organics.
6.53 Range of Effectiveness
Removal efficiencies for an SDA/baghouse system can range
as high as,99%+ for most incinerator pollutants: acid gases, and
heavy metals. More commonly, removal rates will be near 70 to
80%. Spray dryers with ESPs can remove 98% of dioxins and
furans (PCDD/PCDF). The achievable removal efficiencies for
an SDA/baghouse operating on the waste stream from an incin-
erator are indicated in Table 6-9. . -...-,
SDA systems can accommodate input temperatures up to
1000°C. -They are capable of handling a wide range of flue gas
flow rates, although they are not very effective at removing low
concentrations of pollutants.
6.5.4 Sizing Criteria
The major design variables for an SDA system are gas resi-
dence, time and reagent slurry flow rate. Residence time is a
function of SDA volume, flue gas flow, and gas inlet and exit
temperatures. SDA volume determines the size of the vessel.
Reagent slurry flow rate determines the size of slaking, pumping,
and atomization equipment. The required slurry flow rate de-
pends on many factors including gas temperatures, concentra-
tions and types of pollutants, use of solids recycle, and type of
reagent. .
For both SDA and dry injection, the capital cost of a system of
one size may be approximated from the known cost of a system
of another size by the sizing exponent:
n = 0.73 (Eq.6-11)
This parameter is used in the sizing equation:
Ib - la (Cb/Ca)n (Eq.6-12)
where Ib is the cost of a system of size Cb, and la, Ca are the
respective cost and size of a reference system. An estimate of the
installed-cost-to-purchase ratio is 2.17 (Hesketh, 1991).
-------
6.6 HEPA Filters
6.6.1 Process Description
High efficiency paniculate air (HEPA) filters are commonly
used in medical, research, and manufacturing facilities requiring
99.9% or greater paniculate removal. Although their use during
remediation activities at Superfund sites has not been wide-
spread, they could be used as a PM polishing step in ventilation
systems for buildings undergoing asbestos removal, for enclo-
sures, or with solidification/stabilization mixing bins.
The major components of a PM control system employing
HEPA filters include the following:
• HEPA filters;
• Filter housing;
• Duct work; and
• Fan.
Such a system is shown in Figure 6-12.
The HEPA filter housing unit required is dependent on the
nature of the PM collected and on the number/arrangement of
filters required. For example,, PM consisting pf asbestos or PM
laden with dioxins/furans will require a bag-out housing unit be
installed. This type of housing unit is designed so that personnel
removing the HEPA filters are never in direct contact with the
filters. Such a unit is shown in Figure 6-13.
HEPA filters can be arranged in parallel, in series, or in a
combination of these arrangements depending on the degree of
PM control desired and the allowable pressure drop across the
filters. Generally, parallel filter arrangement will lower the
pressure drop across the filters, but will increase the size of the
housing unit. Serial filter arrangement generally will increase the
PM collection efficiency and the total pressure drop.
6.6.2 Applicability to Remediation Technology
The advantages/disadvantages of using HEPA filters to con-
trol PM emissions are given in Table 6-10. Remediation tech-
nologies with which HEPA filters are compatible are listed in
Table 6-11.
6.6.3 Range of Effectiveness
Parameters that will affect the efficiency and/or useful lifetime
of HEPA filters are outlined in Table 6-12. Vendors report HEPA
filter PM control efficiencies to be 99.9% and up for paniculate
diameters of 0.3 microns.
Airflow
Direction s
Hepa
Filters
Hepa Adsorbers
Filters
Downstream
Plenum
Prefilters
Airflow'
Direction
Upstream
Plenum
Source: Flanders, 1984.
Figure 6-12. PM control system employing HEPA filters.
96
-------
Door hand
knowb stud
Filter
removal
rod
Shaft seal
Hand
knob
Locking
mechanism
s- Type FE
adsorber
Door
«— Sample port
2" pre-filter
track Filter-locking
mechanistm shaft
Source: Barneby-Cheney, 1987.
Figure 6-13. Bag-out HEPA filter housing unit.
Table 6-10. Advantages/Disadvantages of HEPA Filters
, Advantages
Disadvantages
Easy to operate
99.9% or greater PM removal efficiencies are achievable
May require prefilter for exhaust with high PM
concentrations
Required housing units are expensive and may be
subject to corrosion
Filters are subject to fouling by high humidity
exhaust gases
Filters must be replaced periodically due to
plugging caused by PM
High power costs due to pressure drop across
filter ,' , __
97
-------
Table 6-11. Remediation Technologies Compatible with HEPA Filters
Emission Source
Qualifications
Asbestos removal from buildings
Enclosure ventilation system
Hoods or enclosures of solidification/stabilization mixing bins
HEPA filters must either be installed in building
ventilation system or negative air pressure system used
during asbestos removal
HEPA filters will require bag-out housing units and must
be disposed of properly
Pre-filters may be required for high PM concentrations
Depending on the nature of the PM (e.g., heavy metal
or SVOCs contamination), bag-out housing units may
be required
High humidity within the enclosure will limit filter lifetime
HEPA housing material may be subject to corrosion
due to lime
Pre-filters may be required for high PM concentrations
Depending on the nature of the PM (e.g., heavy metals
or SVOCs) bag-out housing units may be required
High humidity exhaust gases will limit filter lifetime
Table &-12. Parameters Affecting HEPA Filter Efficiency/Lifetime
Parameter
Comments
Moisture
PM loading
Moisture will bind filter resulting in increased pressure drop across filter, eventually leading to
filter failure due to excessive resistance.
Higher the PM loading the shorter the useful life of the filters. Also, the change in pressure drop
across the filters will be accelerated.
Higher the velocity, the lower the PM control efficiency, higher the pressure drop across the
filter, and diminished filter life.
6.6.4 Sizing Criteria/Application Rates
Sizing of HEPA filters is based on pressure drop vs. face
velocity curves which are developed by the manufacturer for
each type of filter design. If the maximum allowable pressure
drop across the filter and the air flow rate are specified, then the
type of filter and the filter arrangement can be determined. A
family of pressure drop vs facet velocity curves is depicted in
Figure 6-14.
If, for example, HEPA filters are to be used to control PM
emissions in an exhaust gas flowing at 9000 acfm (2250 fpm for
a 2 ft x 2 ft HEPA filter) and the maximum allowable pressure
drop across the filter is 0.8 inches H2O gauge, then ten-H2424B,
nine-H2430B, eSghteen-H2424A HEPA filters must be used in
parallel (see Figure 6-14).
6.6.5 Cost Estimating Procedure
HEPA filter costs are dependent on the specific filter charac-
teristics:
• PM removal efficiency achievable; and
• Maximum face velocity allowable across filter.
Also, the useful filter lifetime is dependent on face velocity
across the filter, PM loading rate, and the moisture loading rate
onto the filter. The useful lifetime will determine the frequency
of filter replacement. Generally the range of HEPA filter costs is
$20 - 100/ft2 filter area. The costs.of housing units is a function
of the type of housing unit required (e.g. regular vs. bag-out) and
ranges in price from $ 150-500/ft2 filter area.
98
-------
1.8
1.6
1.4
q
I „
1.0
co
s
0.8
0.6
0.4
H2424A
I
I I i
I
I
0 4
Face Velocity (1pm) 125
8 12 16
250 375
Capacity x 100
20
500
24 (cfm)
Source: HEFCO, 1989. '
Figure 6-14. Pressure drop vs. face velocity curves for specific HEP A filter designs.
6.7 References
Barneby and Cheney. Product Brochure. Barneby-Cheney,
Columbus, OH. 1987.
Brna, T.G. and C.B. Sedman. Waste Incineration and Emission
Control Technologies. EPA/600/D-87/147 (NTIS PB87-
191623). RTF, NC. May 1987.
Brna, T.G. Controlling PCDD/PCDF Emissions from Incin-
erators by Flue Gas Cleaning. EPA/600/D-90/239. U.S.
EPA, Research Triangle Park, NC. September 1990.
Brunner, C.R. Incineration Systems: Selection and Design.
Van Nostrand Reinhold, New York, NY. 1984.
Danielson,J.A.,ed. Air Pollution Engineering Manual Second
Edition (AP-40). U.S. EPA, RTF, NC. May 1973.
Donnelly, J.R. Air Pollution Controls for Hazardous Waste
Incinerators. Thermal Treatment and Air Pollution Con-
trol In: Proc. of the 12th National Conference on Hazard-
ous Materials Control/Superfund '91. HMCRI, Silver
Spring, Maryland. December 1991.
Flanders. Product Brochure. Flanders Equipment, Washing-
ton, DC. 1984.
HEFCO. Product Brochure. HEFCO, Eatontown, NJ. 1989.
Hesketh, H.E. Air Pollution Control: Traditional and Hazard-
ous Pollutants. TechnomicPublishing Co., Lancaster, PA.
1991.
HMCRI. Incineration Monograph. Hazardous Materials Con-
trol Research Institute, Silver Spring, Maryland. 1991.
Lawless, P.A. and L.E. Sparks. A Review of Mathematical
Models for ESPs and Comparison of their Successes.'
Proceedings of the Secondlnternational Conf. onElectro-
static Precipitation, pp513-522. S. Masuda, ed., Kyoto.
1984.
Oglesby, S. and G.B. Nichols. A Manual of Electrostatic
Precipitator Technology. Prepared by the Southern Re-
search Institute for the National Air Pollution Control
Admin. APTD-0610 (NTIS PB-196380). 1970.
99
-------
U.S.EPA(W.M.Vatavuk).OAQPSControlCostManual(4th
Edition) EPA/450/3-90/006 (OTIS PB90-169954). RTP,
NC. January 1990.
U.S. EPA. Handbook: Control Technologies for Hazardous
Air Pollutants EPA/625/6-91/014. Cincinnati, OH. June
1991.
Vatavuk, W.M. Estimating Costs of Air Pollution Control.
Lewis Publishers Chelsea, MI. 1990.
100
-------
Chapter 7
Area Source Controls for VOCs, SVOCs, PM, and Metals
Information about various control technologies used to control
emissions from area sources is presented in this chapter. The
control technologies generally are applicable to the control of all
classes of air contaminants, including volatile organic com-
pounds (VOCs), semi-volatile organic compounds (SVOCs),
particulate matter (PM), and metals associated with PM. The
specific control technologies addressed in this section are covers
and physical barriers, foams, wind barriers, water sprays, water
sprays with additives, operational controls, enclosures, collec-
tion hoods, and miscellaneous controls. The discussion for each
control technology includes a process description, applicability
for remediation technologies, range of effectiveness, sizing cri-
teria, and cost information.
Emissions from area sources are more difficult to measure,
model, and control than emissions from point sources. The
sources may be several acres in size and the concentration of
emissions in the source/atmospheric boundary layer is generally
very low. Therefore, the types of controls suitable for point
sources are not applicable to area sources. Two general control
approaches exist for area sources: 1) Collect the emissions in a
hood or enclosure and route the air stream to a point source
control device; and 2) Prevent the emissions from occurring. The
first approach is merely a conversion of the area source to a point
source and is the most suitable for batch or in-situ remediation
processes such as solidification/stabilization and bioremedia-
tion. The second approach is primarily suited for materials
handling operations such as excavation.
Relative to point source controls, the capture efficiency and
control efficiency of area source controls tends to be low. Overall
control efficiencies may be only 50% or lower. Also, many of the
area source controls have short-lived effects and require frequent
reapplication. The controls are generally simpler in design than
point source controls and do not require as highly a trained
operator for their operation and maintenance. Finally, opera-
tional controls may be the single most cost-effective method of
minimizing emissions from area sources; their consideration is
strongly recommended.
7.1 Covers and Physical Barriers
Cover materials used to control VOC and/or particulate matter
(PM) emissions include the following: soils, organic solids such
as mulch, asphalt/concrete (paving), gravel/slag with road car-
pet, and synthetic covers (e.g. tarps). Cover materials are used
extensively at controlled landfills and construction sites for
vapor and dust suppression. The mostcommonuses of covers for
Superfund sites are soil covers for inactive sites; asphalt, con-
crete, or gravel covers for roadways; and thin polymer liners
(e.g., 45 mil HDP) for storage piles of contaminated soil.
Soil material can range from top-soil to clays. However, sand
generally is not used because of its porous nature and tendency
to erode.
Organic solids include such materials as wood chips, sawdust,
sludges, mulch, straw, corn stalks, etc. Because some of these
materials are prone to wind erosion (e.g., straw, cornstalks), they
must be anchored with a net. The availability of these materials
may limit their use.
Synthetic liners (polymer sheeting) are relatively new and are
used widely at landfills to minimize leachate migration. They
also serve as a barrier to vapor transport. Liner thickness varies
from 2-125 mil; however, 30-60 mil is common. Some common
geomembranes are
• Polyethylene
- High density—HOPE
- Low density—LDPE - should not be used
- Very low density—VLDPE
- Linear low density—LLDPE
• Poly vinyl chloride (PVC)
• Chlorosulfonated polyethylene (CSPE)
• Ethylene interpolymer alloy (EIA)
Note: Others will not sell to our industry due to potential liability.
Road carpets are water permeable polyester fabrics that are
placed between the road bed and the coarse aggregate road ballast
(e.g., gravel, slag).
The effectiveness of cover materials to control VOC and PM
emissions is outlined in Table 7-1.
7.1.1 Process Description
Covers control emissions of contaminated particulate matter
and VOCs by physically isolating the contaminated media from
the atmosphere. Physical isolation is the primary means of
controlling particulate matter emissions, while increasing the
resistance to diffusion is the primary means of controlling VOC
101
-------
Table 7-1. Cover Material Effectiveness for Controlling VOC and
Paniculate Emissions
Contaminated
Cover material media
Soil
Organic solids
Goomembranes
Asphalt/concrete
Road carpet and gravel/slag
Soil
Soil
Soil
Sludge
Liquid
Soil
Soil
Control
effectiveness
VOCs
Low '
Low1
High1
High1
High1
None2
None2
Particulates
High
High
High
N/A
N/A
High
High
1 VOC emission control dependent on dilfusivity coefficient of individual VOCs
through cover material and on cover material depth.
* Assumes cover material is applied to unpaved roads only.
emissions. Mass emission rates for VOCs, assuming diffusion
only, can be determined from Equation 7-1:
where:
(Eq.7-1)
Mass emission rate of species i (g/sec);
A - Area of emitting source (cm2);
D, - Diffusivity coefficient of species i through
cover material (cm2/sec);
X - Cover thickness (cm); and
Cy-Cy - Concentration gradient of species i(g/cm3).
Therefore, covers reduce VOC emission rates by:
1. Decreasing diffusivity coefficientrelativetoair(viachemi-
cal interactions and soil temperature reductions); and
2. Increasing required VOC diffusion path (x).
Some cover materials (e.g., sawdust, straw) are tilled into the
contaminated soil as an anchoring mechanism. This practice
results in a soil/cover layer with a higher porosity than the soil
alone, resulting in increased VOC emission rates. The major
components typically required to apply the various cover mate-
rials are listed in Table 7-2.
7.13 Applicability to Remediation Technologies
The applicability of the various cover materials is dependent
on site characteristics (terrain, vegetation, access, and contami-
nated media) and on the desired PM/VOC control efficiencies
required. The advantages/ disadvantages and applicable reme-
diation technologies of various cover materials are given in
Tables 7-3 and 7-4, respectively. Also, some specific character-
istics of various geomembranes are given in Table 7-5.
7.1.3 Range of Effectiveness
Parameters that influence the effectiveness of cover materials
to control VOC/particulate matter emissions are presented in
Table 7-6. Reported PM/VOC control efficiencies for various
cover materials are presented in Table 7-7. These control effi-
ciency ranges should be used only as a guide since methods used
in determining these efficiencies, site characteristics, and cover
application procedures vary from site to site. Information about
the permeability of various polymeric materials to specific liquid
VOCs is available for gloves and other personal protective
equipment (Radian, 1992); these data can be extrapolated for
selecting soil covers.
7.1.4 Sizing Criteria/Application Rates
The amount (depth, thickness, etc.) of cover material required
to achieve a given control efficiency is not well defined in the
literature. However, there are general sizing guidelines reported
in the literature that are presented in Table 7-8.
7.1.5 Cost Estimating Procedure
Costestimates of implementing cover-based VOC/PMcontrol
measures are presented in Table 7-9. Caution should be exercised
when using these cost estimates, since costs are highly dependent
on the site characteristics, labor costs, weather conditions, and
the availability of specific cover materials at each site.
7.2 Foams
7.2.1 Process Description
Modified fire-fighting foams are commonly used to control
PM/VOC emissions during the remediation of hazardous waste
sites. Suppression of PM/VOC is accomplished by blanketing the
emitting source (liquid, sludge, or soil) with foam, thus forming
a physical barrier to those emissions. Foams also act to insulate
the emitting source from the wind and the sun, further reducing
PM/VOC emissions. Some foams are "sacrificial", meaning that
the chemicals compromising the foam will react with specific
VOCs thus further suppressing their emissions. Rusmar and 3M
are the two primary manufacturers of foams for use at Superfund
sites.
Table 7-2. Major Components for Cover Material Applications
Cover material
Major components required
Soil
Organic solids
Polymer sheeting
Asphalt/concrete
(paving)
Road carpet and
gravel/slag
Front end loader
Grader
Water wagon (optional)
Compaction equipment (optional)
Front end loader
Grader (dependent on material)
Tilling equipment (optional)
Water wagon (optional)
Anchoring net
None
Grader
Front end loader
Paving application equipment
Compaction equipment
Base material
Grader
Front end loader
Base material
Compaction equipment
102
-------
Table 7-3. Advantages/Disadvantages of Cover Materials
Cover material Advantages
Disadvantages
Soils
Organic solids
Asphalt/concrete
Road carpet and gravel/slag
Polymer sheeting
Inexpensive
Easy to apply
Equipment readily available
Inexpensive
Easy to apply
Equipment readily available
Long working life
Equipment readily available
Cheaper than asphalt/concrete
Long working life
Applicable to low traffic volume areas
Allows contaminated soil from spillage
and "track-out" to be washed beneath
road carpet
100% particulate emission control
No erosion potential
Carbon black addition can limit
photodegradation
Conversion of area source to a
point source
Creates more contaminated soil
Subject to erosion -
No particulate or VOC control for working face
Material availability location dependent
Often requires anchoring to reduce erosion
No particulate or VOC control for working face
Combustible
Creates more contaminated soil
Expensive
High maintenance
Extensive preparation required
Limited to high traffic volume or permanent traffic pattern
areas
Higher maintenance than paved roads
Extensive preparation required
Subject to erosion
Limited tear resistance
Limited chemical resistance
Photodegrades easily
Must be anchored to surface
No particulate or VOC control for working face
Table 7-4. Cover Materials vs. Applicable Remediation Technologies
Cover material
Applicable remediation
technology
Qualifications
Soils
Organic solids
1. Materials handling
- Storage piles
- Unpaved roads
- Inactive sites*
2. Bioremediation
3. Solidification/stabilization
4. On-site incineration
1. Materials handling
- Storage piles
- Inactive sites*
2. Bioremediation
3. Solidification/stabilization
4. On-site incineration
Daily cover may be applied to active storage piles,
however soils more applicable to inactive storage piles. •
High erosion potential. Best to use "tight1 soils (e.g., clay).
Soils must be uniformly applied and deep enough to prevent
erosion from exposing contaminated soil.
Well compacted soils may limit biological growth by limiting
moisture loading.
Increases soil to be treated.
Increases soil to be treated. Also decreases Btu content of
contaminated soil.
Impractical for active storage piles unless "anchoring" is not
required.
Some materials require anchoring via a net or tilling into soil
to prevent erosion.
May add nutrients to soil (e.g., organic sludges).
Applicable only for small grained organic solids (e.g.,
sawdust, sludges).
May increase Btu content of contaminated soil.
(continued)
103
-------
Tablo 7-4. (continued)
Cover material
Applicable remediation
technology
Qualifications
Asphalt/Concrete
Road Carpet and Gravel/Slag
Polymer Sheeting
1. Materials Handling
- Unpaved Roads*
1. Materials Handling
- Unpaved Roads*
1. Materials Handling
- Storage Piles*
- Inactive sites*
2. In-S!tu Thermal Treatment
3. Sol| Vapor Extraction
4. Bioremediation
" Typical use of cover materials. Other listed uses are less common.
Applicable for permanent traffic pattern areas (>100 passes/
day).
Applicable to permanent and temporary traffic pattern areas.
Applicable to inactive storage piles.
Inactive areas may require extensive pre-applicatlon
procedures (I.e., grading, base material, weed control).
May be used to convert area source to a point source if
liners can withstand the heat.
May be used to minimize/control infiltration of ambient air.
May reduce biological activity dues to decreased moisture •
loading and oxygen transfer impairment.
Tablo 7-5. Synthetic Cover C haracterlstlcs1
Synthetic material
Chemical resistance
Weather Gas permeability
resistance resistance Tear resistance
Polyethylene
High density polyethylene (HOPE)
Low density polyethylene (LDPE)
Very low density polyethylene (VLDPE)
Linear Low Density Polyethylene (LLDPE)
Polyvlnyl chloride (PVC)
Chlorosulfonated polyethylene (CSPE)
Ethyleno Inlerpolymer alloy (EIA)
Inorganics
Organlcs
Inorganics
Organics
Inorganics
Organics
Inorganics
Organics
Inorganics
Organics
Inorganics
Organics
Inorganics
Organics
Good
Good
Good
Poor
Good
Good
Good
Good
Good
Poor
Good
Poor
Good
Good
Excellent
Poor
Excellent
.{
Excellent
Poor
Good
Good
Excellent
Poor
Excellent
Excellent
Good
Good
Good
• Good •,.
Poor
Good
Good
Good
Good
Good
' Source: Landrelh, 1988 ~ ~~ —
Foams are generally classified as either long-term (stabi-
lized) or temporary foams. Temporary foams are effective at
controlling PMTVOC emissions from 1-24 hours, at which time
25% or more of the water incorporated in the foam will have
been released ("quarter drainage time"). Long-term foams ei-
ther contain a stabilizing agent incorporated into a short-term
foam (3M products) or are comprised entirely of proprietary
agents (Rusmar Products). Long-term foams will form an elas-
tomeric membrane upon setting (1-2 min)' which is the primary
mechanism of PM/VOC control. Generally the useful life of a
long-term foam is from several days to several months.
Foams generally are produced by pressurizing a mixture of
proprietary foam concentrate/water solution through an air-aspi-
rated or air-injected foam nozzle; a schematic of this system for
3M products is shown in Figure 7-1. However, the long-term
foams produced by Rusmar do not require dilution with water.
Two important indicators of foam quality in relation to vapor
control are the "expansion ratio" and the "quarter drainage time".
The expansion ratio is adimensionless number that expresses the
ratio of the volume of foam to the volume of foam concentrate that
104
-------
Table 7-6. Parameters Influencing Cover Material Effectiveness
Cover material Parameter Comments
Soil
Organic solids
Asphalt/gravel
Polymer sheeting
- Porosity
- Moisture content
- Depth
- Porosity
- Moisture content
- Depth
- Road base
- Maintenance
- Surface cleaning
- Liner material
- Liner thickness
Decreased porosity -» increased VOC control.
Increased moisture -»increased VOC/PM control.
Increased depth ->• increased VOC/PM control (to a maximum).
Decreased porosity -»increased VOC control.
.Increased moisture content -* increased particulate/VOC control and decreased
erosion due to wind (only for small grained solids).
Increased depth -»increased VOC/PM control (to a maximum).
Stable road base -> decreased maintenance required.
Well maintained road will limit spillage.
Frequent cleaning will limit PM emissions due to spillage and "track-out."
Liner material influences diffusivity coefficient of various VOCs with respect to liner.
Also determines susceptibility to photodegradation.
Thicker liner -+ increased resistance to VOC diffusion, puncture, and tear.
Source: Landreth, 1983.
Table 7-7. Cover Material Control Efficiency Ranges
Cover material
Soils
Organic soils
Asphalt/concrete
Road carpet
Gravel/slag
Polymer sheeting
Pollutant
PM
VOCs
PM
VOCs
PM
PM
PM
PM
VOCs4
Control
efficiency
(%)
-100
92-99.8
852
NR
85-99
90
45
30-50
100
90
Notes
Theoretical1
1" and 40" soil layer on an inactive site
Inactive storage piles2
99% with frequent cleaning
Calculated3
—
Gravel/slag on unpaved roads
Theoretical
Polyethylene liner for hexachlorobenzene
emission control
Source
U.S. EPA, 1988a'
Vogel 1985
U.S. EPA, 1987a
~
U.S. EPA, 1987a
U.S. EPA, 1987b
U.S. EPA, 1987a
U.S. EPA, 1987a
Vogel, 1985
NR = Not Reported
1 Theoretical based on zero wind/water erosion of soil layer.
2 Paniculate control efficiencies for inactive storage piles/sites are equivalent (EPA, 1989).
3 Calculated based on AP-42 emission factors for paved and unpaved roads (U.S. EPA, 1985b).
4 Some VOC diffusivity coefficients for a 20 ml PVC liner are reported by Springer, et al., 1986.
Table 7-8. Cover Material Sizing/Application Guidelines
Cover material Pollutant
Sizing Guidelines
Soil
Organic solids
Asphalt/concrete
Road carpet and gravel/slag
Polymer sheeting
PM Apply enough soil to insure even application and to prevent exposure of contaminated soil due to
erosion of clean soil layer.
VOCs Sizing is dependent on moisture content and porosity of soil. Assume linear relationship from
data presented in Table 7-7.
PM Apply organic solids to the point that no emissions are measured.
VOCs None currently available.
PM Follow accepted civil engineering guidelines for road construction using asphalt or concrete.
PM Follow accepted civil engineering guidelines for road construction using gravel or slag.
PM Surface roughness will determine liner thickness required.
VOCs VOCs to be controlled will determine liner material to be used. Contact manufacturer for specific
information.
105
-------
Table 7-9. Cost of Implementing Cover Based Area Control Measures
Cover material
Cost (1992$)
Equipment1 Labor/materials
Backfill dirt 2.0 m3
Clay 1.0m3
Rood baso, road carpet, and grave-!' 3-6/m
AsphaH, road base8 6-12/m
Wood fibers with plastic netting3 0.5/m2
Polymer sheeting 1.0/m2
15/m3
15/m3
4-10/m
200-300/m
0.5/m2
1.0/m2
Means, 1991; Vendor
Means, 1991; Vendor
Means, 1991; Vendor
Means, 1991; Vendor
Means, 1991; U.S. EPA, 1988b
Vendor
1 7.5 m wkte and 0.15 m gravel.
J 7.5 mwkfo and 0.10m asphalt.
1 Wood liber depth not stated.
4 Assumes material not already on site.
Temporary IFoam
Stabilized Foam
Pump
pressurized
Air- aspirating
noxzle
Temporary foam
Educt or
pressure inject
Stabilized foam
(gels in 1-2min.)
Source: Aim, etal., 1987
Figure 7-1. Production of temporary and long-term 3M foams.
produced the foam. Expansion ratios are generally defined as
follows:
• High-expansion: greater than 250:1;
• Medium-expansion: between than 20-250:1; and
• Low-expansion: less than 20:1.
High-expansion foams can be generated only by using high-
expansion surfactant foam concentrates in combination with
special foam-generating equipment. However, low- and me-
dium-expansion foams can be generated using various combina-
tions of foam types and foam nozzles. For example, a given brand
of foam concentrate may produce a medium-expansion foam
with one nozzle and a low-expansion foam using a different
nozzle. Another brand of foam may produce a medium-expan-
sion foam with both nozzles.
The term "quarter drainage time" refers to the time it takes for
a foam to release 25 percent of the total liquid incorporated into
the foam. Long quarter drainage times are indicative of stable
foams, which are capable of suppressing vapors for long time
periods before reapplication is necessary. High-expansion foams
exhibit the longest quarter drainage times, often exceeding one
hour. However, these foams may be blown away easily under
windy conditions. Medium-expansion foams exhibit quarter
drainage times exceeding 15 minutes, but usually less than 30
106
-------
minutes. Quarter drainage times for low-expansion foams gener-
ally range between 3 and 12 minutes.
The major components of a foam PM/VOC suppression sys-
tem are as follows:
Foam concentrate;
Water (subject to foam requirements);
Foam concentrate bulk storage tank; and
Foam generating/distributing unit
Manifold distribution
Hand-line distribution.
A list of foams by type, brand name, manufacturer, and useful
lifetime is presented in Table 7-10.
7.2.2 Applicability to Remediation Technology
The advantages and disadvantages of using foams to control
PM/VOC emissions are outlined in Table 7-11. Remediation
technologies compatible with foam PM/VOC suppression sys-
tems are listed in Table 7-12. Foams are most commonly used as
temporary covers to control emissions during excavation, dredg-
ing, and other materials handling operations.
7.2.3 Range of Effectiveness
Several parameters that influence the effectiveness of foam
systems to control PM/VOC emissions are given in Table 7-13.
Reported control efficiencies for foam based systems as a func-
tion of time, foam type, and contaminant to be controlled (PM or
VOC) are given in Table 7-14.
7.2.4 Application Rates
Prior to applying foam, the concentrate must be diluted with
water. Dilution ratios (watenconcentrate) can range from 0 to
16:1. The manufacturer should be contacted to determine the
appropriate dilution ratio required.
Long-term foams are generally applied to depths of one-two
inches, while temporary foams are generally applied to depths of
two to three inches. The method of application, hand-line or
manifold distribution systems, will depend on the tppography
and media characteristics (i.e. liquid, solid, sludge). Hand-line
systems are capable of shooting foam 30 to 200 feet. Manifold
distribution systems are capable of delivering foam at a rate of
roughly 100 mVmin.
Currently, no empirical relationships exist to predict the PM/
VOC control efficiency of foam suppression systems. Therefore
it is suggested to either use the data presented in Table 7-14 or
conduct pilot scale testing at the site to be treated.
Table 7-10. Commercially Available Foams for PM/VOC Emissions Control
Foam type Brand name ; Manufacturer
Useful lifetime
Temporary
Long-term
FX9162
AC645
FX9161/9162 mixture
AC900 series
3M
Rusmar
3M
Rusmar
0.5-1.0 hours
12-24 hours
1-7 days
up to 5 months
Table 7-11. Advantages/Disadvantages of Foam Systems to
Advantages
Control PM/VOC Emissions
Disadvantages
100% Control of PM emissions achievable
Effective control of VOC emissions
Easy to apply ,
Specialized foam application units available
Allow control of working face - temporary foams
Applicable to liquid, sludge, and solid media
Reduced amount of material to be decontaminated
relative to soil covers .
Long-term foams forming elastomeric membrane
reduce water infiltration-hence minimize leaching
Ca and Mg hardness of dilution water will adversely affect useful
lifetime
May not work well on steep slopes
Temporary foams are easily blown/washed away by wind/water.
Materials handling problems due to water in foam
Possible that foams will react adversely with VOC's to be controlled
Difficult to apply on windy days
Foam itself may off-gas
Moderately expensive
Reapplication required to maintain PM/VOC suppression
107
-------
Table 7-12. Applicable Uses of Foams by Remediation Technology
Remediation technology Foam type
Qualifications
- Inactive sites/storage piles
- Active storage piles
- Dump cycle
- Excavation
- Incineration
- Solidification/stabilization
- Bforemediation
Long-term
Temporary
Temporary
Temporary
Both
Both
Both
Foam must be compatible with VOCs to be controlled.
May use manifold distribution systems on open, flat, and solid terrain,
otherwise must use hand-line distribution systems
Slope of storage piles must be gradual so foam will not run off.
Place foam in truck to be loaded then add contaminated material. Because
operators can't see foam level, they must know how much material they can add
before foam overflows.
To maintain continuous control of PM/VOC emissions, foam must be reapplied to
areas as it is removed during excavation.
Water content of foam may preclude incineration of contaminated material.
Foam may be incompatible with process.
Possible limits on biological activity due to impairment of water and oxygen
transfer to active biological zone.
Tabta 7-13. Parameters Influencing Effectiveness of Foam-based PM/VOC Suppression Systems
Parameter Comment
Quarter drainage time
Wind Speed
Precipitation
Surface roughness
Expansion ratio
Temperature
Surface activity
Contaminant VOC characteristics
The longer the quarter drainage time the longer the foam will be useful.
Wind speeds greater than 10 mph preclude the application of foams (U.S. EPA, 1986a)
High wind speeds will blow away foams already applied, except for long-term foams which have formed
an elastomeric membrane.
Rain will tend to wash foams away, except for long-term foams which have formed an elastomeric
membrane.
For some surfaces (i.e., areas covered with shrubs), foams can not be applied evenly.
The higher the foam expansion ratio the longer the quarter drainage time. Also the higher the foam
expansion ratio, the more susceptible it is to being blown away.
Increased temperatures result in decreased quarter drainage times (U.S. EPA, 1989).
Increased surface activity (i.e. travel) will decrease effectiveness of foam systems.
VOCs that are reactive with chemicals comprising foam material will degrade foam effectiveness and
may produce undesirable compounds.
Table 7-14. Reported PM/VOC Control Efficiencies Using Foam Suppression Systems
Foam type
NR
NR
Long-term
Temporary
Long-term
NR - Not reported.
Contaminant
PM
PM
VOC
Paraffins
Olefins
Aromatics
VOC
Paraffins
Olefins
Aromatics
VOC
VOC
VOC
VOC
Control
efficiency (%)
92
74
100
100
99
95
73
90
64 '
99
79
99
91-97
100
100
Time since
application
Continuous
Continuous
24 hours
24 hours
24 hours
20 minutes
2 hours
20 minutes
2 hours
20 minutes
2 hours
7 days
7 days
7 days
7 days
Comments
10.5 ft3 /ton material
8.4 ft3 /ton material
6:1 Expansion ratio 1 inch foam
6:1 Expansion ratio 1 inch foam
„
2-3 inches
2-3 inches
1 inch
t
Source
U.S. EPA, 1989
U.S. EPA, 1989
Aim, 1987
Aim, 1987
Radian, 1991
Schmidt, 1992
Schmidt, 1992
U.S. EPA, 1991
108
-------
7.2.5 Cost Estimating Procedure
Costs for various foam types are given in Table 7-15. These
costs are a function of the area to be treated at the application
depths recommended by the manufacturer.
Table 7-15. Foam Costs
Foam type
Temporary
Long-term
Brand name
FX9162
AC645
FX91 62/91 61
AC904
AC912 -
AC918
AC930
Costs ($/m3)
1.10
0.54-0.86
3.80
1;.30-1.94
1.94-2.70
2.70-3.77
3.77-5.28
Source
Vendor (3M)
Vendor (Rusmar)
Vendor (3M)
• Vendor (Rusmar)
Vendor (Rusmar)
Vendor (Rusmar)
Vendor (Rusmar)
Costs for foam application units range from $8,000 to $12,000
per month for manifold application units (including bulk storage
tanks) and $3,250 to $7,750 for hand-line application units
(Rusmar, 1992). Small 3M application units can be rented for
about $660 per week; about $500 of ancillary equipment is also
required.
7.3 Wind Screens
7.3 .1 Process Description
Wind screens can be used to reduce PM emissions from storage
piles, excavation sites, and other area sources. The principle is to
provide an area of reduced wind velocity that allows settling of
the large particles and reduces the particle flux from the exposed
surfaces on the leeward side of the screen. Wind screens also
provide limited control of VOC emissions by increasing the
thickness of the laminar film layer (stagnant boundary layer) on
the leeward side of the screen. In addition, wind screens reduce
soil moisture loss due to the wind, resulting in decreased VOC
and particulate matter emissions.
Generally wind screens are porous and constructed of plastic
materials. Solid wind screens are not as effective, and are thus
rarely used, because they create a turbulent zone immediately
behind the fence which increases PM/VOC emissions in this
area. A diagram of the effect of wind screens on wind speeds is
provided in Figure 7-2.
7.3.2 Applicability to Remediation Technology
The advantages and disadvantages of wind screen systems to
reduce PM/VOC emissions are presented in Table 7-16.
Wind screens are compatible with all remediation technolo-
gies involving contaminated materials handling operations (e.g.,
excavation, storage piles, and inactive sites). The qualifications
for using wind screens with specific materials handling opera-
tions are outlined in Table 7-17.
7.3.3 Range of Effectiveness
Parameters influencing the effectiveness of wind screens to
control particulate/VOC emissions are listed in Table 7-18.
Reported PM/VOC control efficiencies achieved with wind
screens are given in Table 7-19. In general, wind screens provide
approximately 15% control of total suspendedparticulates (TSP)
than PM10 (U.S. EPA, 1987b). Also, particulate matter control
efficiency for relatively flat active surfaces is expected to be
similar to that observed for active storage piles.
7.3.4 Sizing Criteria/Application Rates
Wind screen sizing for various materials handling operations
is outlined in Table 7-20. In addition, crude empirical models
have been developed to describe particulate and VOC emissions
reductions. To predict VOC control efficiency, the following
equation applies:
Efficiency (%) =
xlOO
(Eq.7-2)
where:
SR = Surface Roughness (m); and
SH = Screen Height (m).
To predict PM control efficiency, use the following model:
Efficiency (%) = (l - (1 - WSR)U) x 100 (Eq. 7-3)
where:
WSR = Mean wind speed reduction due to wind screen
(fraction). .
V
Unfortunately this model requires that the mean wind speed
reduction due to the wind screen be known. This can either be
measured or based on estimates provided by the vendor.
i
7.3.5 Cost Estimating Procedure
Capital costs for wind screens vary with the type of control
desired (VOC or PM) and the operation requiring control (e.g.,
inactive sites, excavation, etc.). Costs as a function of pollutant
to be controlled and operation requiring control are outlined in
Table 7-21.
7.4 Water Sprays
7.4.1 Process Description
Water sprays are used primarily to control PM emissions. The
control mechanism is the agglomeration of small particles with
larger particles or with water droplets. Also, water added to the
soil will cool the surface soil and will decrease the air-filled
porosity of the soil. These actions resultin an initial displacement
of VOCs followed by a decrease in VOC emissions until the
water evaporates.
Typically, water is applied with mobile water wagons; how-
ever, it also may be applied via fixed perforated pipes. Fixed
systems for water application are limited to long-term fixed
emission sources (e.g. conveyor belts, long-term storage piles,
fixed loading areas, "track-out" elimination sites). The major
components of water spray systems are listed in Table 7-22.
7.4.2 Applicability to Remediation Technology
The advantages/disadvantages of using water as a particulate
control technique are outlined in Table 7-23. The applicability of
water spray systems to control area sources of particulate matter
emissions for various remediation technologies is outlined in
Table 7-24.
109
-------
JC
I 3
(a)
Wind direction
Vertical section
along £ of fence
Fence top
% of urfstream velocft
-10H 0 10H 20H 30H 40H
Distance In fence heights
50H
60H
&
Ground plan
Wind readings at
1 1/3 ft, above ground
-10
Source: U.S. EPA, 1986b.
10H 20H SOU 40H
Distance in Fence Heights
50H
60H
Figure 7-2. Wind velocity pattern above a mown field during a 17 m/sec wind blowing at right angles to a 4.9 m high wood fence 122 m lonq
of 50% porosity, (a) Side view profile, (b) Plan view profile.
7A3 Range of Effectiveness
Water spray systems are an effective control measure for
particulate matter emissions. However, they are not recom-
mended for VOC emissions control, unless they are applied with
soil cover materials followed by compaction.
Parameters that influence the performance of water spray
systems are outlined in Table 7-25. Particulate matter control
efficiencies for various water application rates and intensities
reported in the literature are presented in Table 7-26. Particulate
matter control efficiencies for relatively flat active surfaces are
expected to be similar to that observed for active storage piles.
7.4.4 Sizing Criteria/Application Rates
In general, the water loading and application rates presented in
Table 7-26 can be used to obtain the reported particulate matter
control efficiencies. However, predictive equations have been
developed to estimate particulate matter control efficiencies
obtainable for water spray systems applied to unpaved roads.
Table 7-16. Advantages/Disadvantages of Wind Screen Systems
Advantages
Disadvantages
Inexpensive
Easy to install
Limits site access
Obstructs view of site
Easy to relocate
Limited VOC control.
Limited effective control area.
Limited practical screen height due to
construction and stability problems.
Easily damaged.
Maximum achievable particulate control
efficiency, approximately 90%.
110
-------
Table 7-17. Applicable Remediation Technologies
Materials handling operation
Qualifications
Storage piles
Inactive sites
Excavation areas
Dump cycles
Pile height must be controlled. Can completely enclose inactive piles, while
active piles can only be surrounded on three sides, one side open for
vehicle access.
Large inactive areas require either taller wind screens or many parallel wind
screens to control emissions effectively.
Can only enclose three sides, one side open for excavation equipment.
Screens must be moved as excavation site moves. VOC control not
possible.
Screen height must be greater than dump height. VOC control not possible.
Table 7-18. Parameters Influencing Wind Screen Effectiveness
Parameter
Influence
Wind screen porosity
Wind direction with respect to wind screen
Wind screen height
Soil silt content
Solid wind screens form turbulent boundary layer on leeward side thus increasing
emissions. Most effective wind screens have 50% porosity.
Wind direction influences the size of the protected area. Area of protection is greatest
for perpendicular winds to the screen, length and least for parallel winds.
Zone of wind velocity reduction is directly proportional to wind screen height.
As soil silt content increases, wind screen particulate control efficiencies decrease.
Table 7-19. Reported PM/VOC Control Efficiencies Using Wind Screens
Control efficiencies
Area source VOCs
Storage pile 80
Inactive sites 80
Particulates
48-97
75-80
30-80
75 0"SP)
60 (IP)
0-92 (TP)
64-88
—
Comments
Theoretical
Measured
Theoretical
Measured
Measured
Measured
Measured
Measured
Measured
Source
Vogel, 1985
U.S. EPA, 1989
U.S. EPA, 1987b
U.S. EPA 1987b
U.S. EPA, 1987b
U.S. EPA, 1987b
U.S. EPA 1986b
U.S. EPA, 1988a
Springer et al., 1986
TSP = Total suspended particulates (<39 urn);
IP = Inhalable particulates (<1 5 pm); and
TP = Total particulates.
111
-------
Tabta 7-20. Wind Screen Sizing
Materials handling operation
Wind screen sizing
Storage Piles
Dump cycles
Excavation
Inactive sites
Screen length = five times pile diameter.
Screen/pile distance = two times pile height.
Screen height = pile height. '
Screen height > 1 ft. above bucket drop height.
Screen/site distance = two times screen height.
Wind screen should be placed perpendicular to prevailing wind direction.
Distance between parallel wind screens = four to ten times screen height.
Table 7-21. Wind Screen System Costs
Type of control
Operation requiring
Control
Cost
Source
voc
Paniculate matter
Inactive surface impoundment
Inactive sites'
Storage piles2
Excavation 3
0.7-1.4$/m 2 impoundment area
0.6-13 $/m 2 of inactive area
721 $/m2 of pile area
11 $/m 2 of excavation area
U.S. EPA, 1991
Vendor data/sizing guidelines
Vendor data/sizing guidelines
Vendor data/sizing guidelines
IrKSrt^Sre181" "** k""8 '" avai{able to S8Cure wind screen a9alnst (valid for sma" areas)- Cost Per "™* meter for wind screen is about $40,
not
1 Assumes conically-shaped storago pile of roughly 10m diameter.
* Assumes 60m diameter excavation site and 1.8m high wind screen around 2/3 of site.
CAVO - 100 -
0.8 pdt
(Eq.7-4)
where:
C*
P
P
P
d
i
t
Average control efficiency (%);
Potential average hourly daytime evaporation
rate (mm/hr); also
0.0049 x (value in Fig. 7-3) for annual condi-
tions; or
0.0065 x (value inFig. 7-3) for summer condi-
tions;
Average hourly daytime traffic rate (hr1);
Water loading rate (L/m2); and
Time between applications (hr).
For storage piles and inactive sites, the particulate matter
control efficiency achieved can be determined from the soil
moisture content before and after water application as shown in
Equation 7-5.
xlOO
where:
sii
A,B
(Eq.7-5)
Instantaneous control efficiency (%);
Soil moisture content (weight %); and
After, before water application, respectively.
7.4.5 Cost Estimating Procedure
For mobile water spray systems, capital costs are estimated to
be $23,000/water wagon per year while operating and mainte-
nance (O & M) costs (fuel, water, labor, truck maintenance) are
estimated to be $44,000/water wagon per year. Furthermore, the
number of water wagons required can be estimated by assuming
that a single truck applying 1 L/m2 can treat roughly one square
mile per hour (approximately 11,000 m2). Capital and O&M
costs for fixed water systems will vary with the type of emission
source to be controlled (e.g., "track-out", excavation, loading
operations) and the amount of plumbing required.
7.5 Water Sprays with Additives
7.5.1 Process Description
Water additives can be classified as hygroscopic salts, bitu-
mens, adhesives, or surf actants and are primarily used to reduce
particulate matter emissions. The processes by which these
additives act to reduce particulate matter emissions are outlined
in Table 7-27. Some common water additives and their classifi-
cation are listed in Table 7-28.
Water additives generally are applied topically; however,
some additives can also be tilled into the soil. Components
required for topical application include those used for water
spray systems (see Section 7.4) and a storage tank for the
undiluted chemical. Tilling of the water additives into the soil
also requires tilling equipment.
112
-------
7.5.2 Applicability to Remediation Technology
The advantages/disadvantages of using water additives as a
participate matter control technique are outlined in Table 7-29.
Remediation technologies that are compatible with water addi-
tives are listed in Table 7-30.
7.5.3 Range of Effectiveness
Parameters that influence the performance of water additives
are presented in Table 7-31. Paniculate matter control efficien-
cies for various water/additive mixtures reported in the literature
are reproduced in Table 7-32. Paniculate matter control efficien-
cies for relatively flat active surfaces are expected to be similar
to those observed for active storage piles.
7.5.4 Application Rates
In general, the water/additive dilution, application intensities,
and frequencies presented in Table 7-32 can be used to obtain the
reported paniculate matter control efficiencies. Paniculate mat-
ter control efficiencies for hygroscopic salt/water mixtures can
be estimated from the predictive equations presented earlier for
water spray systems (Section 7.4).
For bitumens and adhesives, time averaged PM10 control
efficiencies as a function of additive "ground inventory" can be
determined directly from Figure 7-4. The term "ground inven-
tory" is a measure of residual effects from previous applications.
Ground inventory is found by adding together the total volume of
additive concentrate (not solution) since the start of additive
application per surface area treated. Also, AP-42 emission fac-
tors for paved roads can be used to conservatively estimate PM
Table 7-22. Major Components of Water Spray Systems
emissions (1.5 to 2 times actual emissions) from unpaved road
surfaces treated with bitumens/adhesives.
Typical dilution ratios and application rates (bitumens and
adhesives) used in the iron/steel industry for treatment of un-
paved roads are applicable and are presented below.
Paved road "housekeeping" techniques can be used on roads
treated with bitumens/adhesives after the ground inventory ex-
ceeds approximately 0.9 L/m2 (U.S. EPA, 1989). These "house-
keeping" techniques will limit emissions due to spillage and
"track-out" and will reduce required application frequencies.
Tilling of bitumens/adhesives into the top 7.6 cm (3 in.) of the
soil followed by compaction resulted in paniculate matter con-
trol efficiencies of 33 to 95% five mqnths after application (U.S.
EPA, 1987b).
7.5.5 Cost Estimating Procedure
Water additives costs include the costs associated with water
spray systems (Section 7.4.5) and also include the cost of addi-
tives and storage tanks for the additives. Storage tank costs will
vary depending on the size of the operation, the water/additive
application rate and the time between deliveries of additive.
Some additive costs, by product name and classification are
presented in Table 7-33. The dilution ratio, application rate and
frequency must be determined to predict the cost/ft2.
7.6 Operational Controls
7.6.1 Process Description
Operational controls are those procedures/practices inherent to
most site remediation projects that can be instituted to reduce
VOC/particulate matter emissions. These procedures/practices
include:
Capital equipment
Comment
Water spray system
Supply pumps, nozzles, and plumbing
Flat spray
Hollow cone
Water wagon
Plumbing (plus winterization)
Control system
Filtering units
Primarily used for fixed spray systems.
Used for water screens that control particulate emissions from dump cycle.
Used for all other particulate emission sources.
Used for mobile spray systems.
Used for stationary spray systems.
Controls water application rate. .
Prevents fouling of spray nozzles.
Table 7-23. Advantages/Disadvantages of Using Water to Control PM Emissions
Advantages
Disadvantages
Inexpensive
Easy to apply
Well defined models for determining particulate control efficiency
Equipment availability.
Frequent application is required. . • • .
May create groundwater contamination via mobilization of contaminant.
Creates material handling problems.
Increases "track-out".
Results in VOC emission spikes.
Availability of water at some sites may preclude its use.
Possible runoff of contaminated water.
113
-------
Table 7-24. Remediation Technologies Compatible with Water Spray PM Control Systems
1.
2.
3.
4.
5.
Remediation technology
Materials handling
• Storage piles
- Active
- Inactive
• Excavation
• Inactive sites
• Conveyor belt systems
• Unpavcd roads
• Fixed loading areas
• Track-out" elimination sites
Bioremediation
Solidification/stabilization
On-site Incineration
Thermal desorption
Water spray
system
Mobile
Fixed
Mobile
Mobile
Fixed
Fixed
Fixed
Fixed
All
All
All
All
Qualifications - • . , «
Water application rate must be adjusted for exposed area and to
prevent erosion.
Use of hoses. Fixing hoses to excavation equipment is impractical.
Water application rate must be adjusted to present materials handling
problems.
Use of water wagons. Water application must be uniform and
controlled to insure a given application rate.
Water should be applied to underside of conveyor belt.
Unpaved roads should be well graded to insure uniform water
application. Water applied to unpavecl roads will increase "track-out".
Water spray system designed to apply a flat spray, forming a water
curtain around loading area which is two to three feet above material
drop height.
Vehicles pass over sump covered with a metal grate located at site
exit. Water spray directed upwards to wash vehicle undercarriage and
wheels of contaminated soil.
Water may dissolve organic compounds to be biodegraded and
transport them past biological area. Water may also limit oxygen
transport from atmosphere into the soil.
Water spray system should be compatible with waste or stabilization
process used unless soil moisture is >95%.
Water may lower Btu content of soil to the point that incineration is not
possible.
Water may lower Btu content of soil. However, soils having 1 0-15%
moisture exhibit enhanced VOC removal.
Table 7-25. Parameters Influencing Water Spray Systems Performance
Parameter Influence
Application rate
Application frequency
Meteorological conditions
Traffic rate
For fixed meteorological conditions and traffic rates the particulate matter emissions are inversely proportional
to the square of the soil moisture content (U.S. EPA, 1988a).
Particulate matter emissions are minimum after water application and rise steadily thereafter until the next
application of water.
Wind, temperature, and humidity influence evaporation rate of water, hence particulate matter control efficien-
cies.
Higher traffic volumes results in higher particulate matter emissions due to increased fines produced.
Read cleaning practices;
Seasonal scheduling;
Vehicle speed control;
Storage pile geometry/orientation;
Excavation practices;
Dumping practices; and
Soil handling practices.
Of the above procedures/practices, only road cleaning requires
additional equipment (i.e., broom sweepers, vacuum sweepers,
or water wagons). Road cleaning practices are aimed at reducing
particulate matter emissions from spillage and "track-out," and
can be applied to paved roads and some chemically treated
unpaved roads (see Section 7.5).
Planning site remediations according to the time of year can
reduce overall PM/VOC emissions by1 taking advantage of lower
temperatures and wind speeds and avoiding excessively dry
weather. However, since site remediation is generally a relatively
continual process, seasonal scheduling is only advantageous for
sites which can be remediated within a season or two.
For unpaved roads, particulate matter emissions increase as the
speed of the vehicle increases, all other factors remaining con-
114
-------
Table 7-26. Reported Particulate Matter Emission Control Efficiencies for Watering Systems
Application
Area source intensity (Urn2)
Unpaved roads
Storage piles
Loading operations
Above conveyor belt operations
Below conveyor belt operations
Excavation
Dump cycle (area spray)
NR
NR
NR
NR
NR
NR
NR
NR
NR
2.3
0.2
0.2
0.6
1.9
NR
NR
9.51/min
9.51 /min
4.1
4.1
Time since
application (hr)
0.5-4.5
0-1.0
0-0.5
0-0.25
0.3-1.0
1.0-4.8
0.5-2.0
1.0-2.0
0.5-2.0
4.0
1.8
2.0
4.5
2.8
NR
NR
Continuous
Continuous
NR
NR
Particle
size1
TP
IP
FP
TP
IP
FP
TSP
TSP
TSP
TP
IP
PM10
FP
TSP
TSP
TSP
NR
TSP/FP
TSP/FP
TSP/FP
TSP/FP
NR
NR
IP
TP
IP
TP
FP
TSP
FP
TSP
Control
efficiency (%)
96-55
98-50
98-61
69-59
73-61
58-54
88
97
75-25
98-61
98-78
98-79
96-67
77-12
66-31
60-15
50-30
59
69
77
88
25-50
70-90
56
59
81
87
64
42
66
69
Source
U.S. EPA, 1985a
U.S. EPA, 1985a
U.S. EPA, 1985a
U.S. EPA, 1985a
U.S. EPA, 1985a
U.S. EPA, 1985a
U.S. EPA, 1985a
U.S. EPA, 1985a
U.S. EPA, 1985a
U.S. EPA, 1985b
U.S. EPA, 1987a
U.S. EPA, 1987a
.-U.S. EPA, 1987a
U.S. EPA, 1987a
U.S. EPA, 1987b
U.S. EPA, 1987b
U.S. EPA, 1989a
U.S. EPA, 1989a
U.S. EPA, 1989a
U.S. EPA, 1985a
U.S. EPA, 19853
NR = Not reported;
TP = Total particulates;
TSP = Total suspended particulates (<39um); ..
IP m Inhalable particulates (<15Mm);
PM10 = Particulate matter (<10 urn); and
FP » Fine particulates (<2.5um).
slant. Speed reduction reduces the turbulence and energy im-
parted to fine particles, which reduces particulate matter entrain-
ment.
PM and VOC emissions from storage piles can be minimized
by controlling the placement and shape of piles. When feasible,
the piles should be placed in areas shielded from the prevailing
winds at the site. The pile surface area can be minimized for the
given volume of soil by shaping the pile. The orientation of the
pile will affect the wind velocity across the pile, hence PM/VOC
emissions will be affected.
Since PM/VOC emissions are proportional to the surface area
exposed, a reduced surface area/volume ratio will minimize
emissions per unit volume of soil excavated. This reduction can
be accomplished by utilizing larger excavation equipment (i.e.,
larger bucket volumes for front end.loaders, bulldozers, and
backhoes).
Dumping practices which can be employed to reduce particu-
late matter emissions include drop height reduction and loading/
unloading of material on the leeward side of storage piles. By
minimizing the soil drop height, the energy and turbulence
caused by the falling soil are reduced thus reducing particulate
matter entrainment. VOC emissions will also be reduced.
115
-------
Table 7-27. Additive Processes
Additive
Process description
Hygroscopic salts
Bitumens/Adhesives
Surfactant
These compounds adsorb moisture from the air, thereby increasing the soil moisture content.
Act to agglomerate surface soil particles to form a surface "crust".
Act to reduce water surface tension, thereby increasing "wetting" capacity of the water.
Table 7-28. Common Water Additives
Product
Manufacturer
Calcium chloride
Dowllake, Liquid Dow®
DP-IO®
Dust Ban 8806®
Dustgard®
Sodium silicate
A. Hygroscopic salts
Allied Chemical Corporation
Dow Chemical
Wen-Don Corporation
Nalco Chemical Company
G.S.L Minerals and Chemicals Corporation
The PQ Corporation
AMS 2200, 2300®
Coherex®
Docal 1002®
Peneprime®
Pelro Tac P®
Resinex®
Retain®
Acrylic DLR-MS®
Bfo Cat 300-1®
CPB-12®
Curasol AK®
DCL-40A. 1801.1803®
DC-859, 873®
Dust Ban®
Flambinder®
Ltgnosite®
Norlig A, 12®
Orzan Series®
Soil Gard®
B. Bitumens
Arco Mine Sciences
Witco Chemical
Douglas Oil Company
Utah Emulsions
Syntech Products Corporation
Neyra Industries, Inc.
Dubois Chemical Company
C. Adhesives
Rohm and Haas Company
Applied Natural Systems, Inc.
Wen-Don Corporation
American Hoechst Corporation
Calgon Corporation
Betz Laboratoires, Inc.
Nalco Chemical Company
Flambeau Paper Company
Georgia Pacific Corporation
Reed Lignin, Inc.
Crown Zellerbach Corporation
Walsh Chemical
M070E
Sterox
Sowco: Paao 45 of U.S. EPA, 1987t>.
D. Surfactants
Mona Industries Inc.
Monsanto Company
116
-------
Table 7-29. Advantages/Disadvantages of Using Water Additives to Control Paniculate Matter Emissions
Advantages Disadvantages
Easy to apply
Equipment availability.
Surfactants reduce water required by 75% for same
level of control (U.S.. EPA, 1989)
Less frequent reapplication required than water spray systems.
Additives may be washed away by rain, thus negating their effects (e.g., salt).
Additives may react with contaminants present in the soil.
Additives often contain organic compounds which may vaporize, leading to
ozone production (U.S. EPA, 1988a).
Not applicable to unpaved roads that require frequent grading.
Expensive.
Application of chemical additives in cool weather may be inadvisable for traffic
safety reasons (U.S. EPA, 1988a).
Additives may contribute to water contamination (surface water and groundwa-
ter).
Some salts (e.g., CaCL,) are corrosive to vehicles.
Table 7-30. Remediation Technologies Compatible with Water Additives
Remediation technology
Qualifications
1. Materials handling
• Storage piles (inactive
• Inactive sites
• Unpaved Roads
2. Bioremediation
3. Solidification/stabilization
4. On-site incineration
Same as water spray (see Section 7.4).
Same as water spray systems.
Same as water spray systems. Also can use paved road cleaning techniques for
unpaved roads treated with adhesives and bitumens. This practice reduces particulate
matter emissions due to spillage and "track-out".
Same as water spray systems. Also additives may inhibit biological activity.
Same as water spray systems. Additives may be incompatible with waste or stabiliza-
tion process used.
Water may lower Btu content of soil to be treated. However, certain additives may
increase Btu content of soil to be teated (e.g., bitumens and adhesives).
Table 7-31. Parameters Influencing Performance of Water Additives
Parameter
Influence
Dilution ratio
Application rate
Application frequency
Meteorological conditions
Vehicle weight, speed, and passes
Unpaved roads base and subgrade bearing strength
Higher dilution ratio results in decreased additive applied for a given water/additive
loading rate. This decreases long-term control efficiency (e.g., 2-3 weeks after
application).
Higher loading rate results in higher particulate matter control. However, this relation-
ship applies only to a point, because too intense an application will produce run-off.
Same as for water spray systems.
Affect required application frequency for a given control efficiency. For example,
freeze-thaw cycles break up crust formed by chemical binding agents; heavy
precipitation washes away water soluble additives like hygroscopic'salts; and intense
solar radiation dries out treated surfaces. However, light precipitation or high humidity
might improve the efficiency of hygroscopic salts.
Acts to break up crust formed by chemical binding agents.
Low base and subgrade bearing strength will result in road deformation which will
destroy crust formed by chemical binding agents.
117
-------
Table 7-32, Reported Partlculate Matter Emission Control Efficiencies for Water Additives
Area source
Additive classification
Unpaved Roads Hygroscopic salts
Storage piles
Inactive sites
Excavation
Dump cycle
Bitumens/adhesives
Surfactant
Bitumens/adhesives
Bitumens/adhesives
Surfactant 1
Surfactant 1
Dilution
ratio
NR
1:2
1:1
NR
NR
NR
1:8
1:6
1:5
1:0
NR
1:5
1:35
1:2
1:2
1:2
1:12
1:4
1:4
1:4
1:19
1:1
1:1
1:1
1:7
1:3
1:3
1:3
1:11
:1000
:1000
Application
Intensity (L/m 2)
2.3
2.7
0.9
NR
NR
NR
NR
0.9
NR
16.0
NR
3.4
6.8
0.81
0.41
0.20
0.31
1.13
0.57
0.28
1.13
1.13
0.57
0.28
1.13
1.12
0.75
0.37
1.12
3.4
4.1
Time since
application (days)
3-60
90
30-270
RP
93
1-49
14
30
<7
1-2
14
30-270
1-42
60
4
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Continuous
Continuous
Particle
size2
TSP
IP
TSP
IP
FP
TSP
42
TSP
IP
FP
TSP
IP
FP
TP
PM10
TP
PM
TSP
TP
TSP
FP
TP
TSP
IP
TSP
IP
FP
TP
IP
FP
TP
IP
TP
TP
TP
TP
TP
TP
TP
TP
TP
TP
TP
TP
TP
TP
TP
TP
TSP
FP
TSP
FP
Control
efficiency (5)
48
24
95
95
95
46
95
98
88
0-83
0-74
0-80
47-96
60-96 ..
47-90
60-94
91
92-98
91-96
90-97
99
96
57
0-87
0-68
0-85
90
62
62
44
50
94
• 68
36
99
99
96
84
99
99
93
85
100
99
98
87
100
63
70
77
62
Source
U.S. EPA, 1987a
,
U.S. EPA, 1987a
U.S. EPA, 1987C
U.S. EPA, 1987a
U.S. EPA, 1989
U.S. EPA, 1987b
U.S. EPA, 1985a
1 Additfvo concentrate: Water
« TP
TSP
IP
FP
PMM
NR
Total particulate matter
Total suspended pmliculate matter (<38pm);
Inhalable partteulata matter (<15Mm);
Fine particulate matter(<2.5(jm)
Particulate matter (<10 vm)
Not reported
118
-------
100 ' 95 90_
I ' 1
Mean Annual Class A Pan Evaporation
(In Inches) i
" ' 35 30 Cl.
-^-T—V-H J'Tntemalional Fate
- sL \ ""^
\ ^
Based on period 1946 - 1965
Source: Buonicore, et al., 1992.
Figure 7-3. Annual evaporation data for the contiguous United States.
Spillage of soil during transport is a consequence of loading
practices and is a primary cause of particulate matter emissions
from paved roads. Minimizing spillage can be accomplished by
covering or enclosing tracks transporting soils increasing free-
board requirements, and repairing trucks exhibiting spillage due
to leaks.
7.6.2 Applicability to Remediation Technology
The relative advantages/disadvantages of each operational
practice/procedure are outlined in Table 7-34. Road cleaning
practices, seasonal scheduling, and vehicle speed control are
applicable to all remediation technologies, while the other prac-
tices/procedures are applicable to materials handling operations.
7.6.3 Range of Effectiveness
Reported control efficiencies for some operational practices/
procedures are given in Table 7-35.
7.6.4 Sizing Criteria/Application Rates
For a target control efficiency, the operational practices/proce-
dures required can generally be determined by using AP-42
emission factors presented in U.S. EPA, 1985b. Operational
practices/procedures which are amenable to this approach are:
• Road cleaning practices;
• Seasonal scheduling;
• Vehicle speed control;
• Excavation practices; and
• Dumping practices.
Quantification of particulate matter emission controls achiev-
able for soil loading practices and storage pile geometry/orienta-
tion are not possible. However, guidelines are available for each
of these operational control measures.
7.6.5 Cost Estimating Procedure
For the majority of the operational control measures presented
in this section, the cost is negligible with the exception of road
cleaning equipment and possibly seasonal scheduling. The cost
of seasonal scheduling will vary with season primarily due to
labor costs and equipment availability. Cost for street cleaning
practices are estimated to be $140 per day per street cleaner and
$66 per day per crew (Means, 1991). The use of larger excavation
equipment to minimize emissions will increase costs to some
extent (Means, 1991).
119
-------
100
0.05 0.1 0.15 0.2 0.25
Ground inventory (gal/sq. yd.)
0.3
Source: U.S. EPA. 1987c.
Figure 7-4. Average PM10 control efficiency for bitumen/adhesive
additives.
7.7 Enclosures
7.7.7 Process Description
Enclosures are used to provide near 100% PM/VOC emission
control for area sources undergoing excavation and PM/VOC
emissions control for storage piles. Enclosures used during
excavation are either self-supported or air-supported structures,
while those used for storage piles are self-supported structures
similar to the "beehives" used, to store road salt.
Enclosures provide a physical barrier between the emitting
area and the atmosphere and in essence convert an area source
into apoint source. Prior to releasing the air entrapped within the
enclosure, conventional point source controls are employed to
control PM/VOC emissions.
Self-supported structures consist of a rigid frame covered with
an all-weather outer skin. The frame is generally constructed of
light weight aluminum which may require concrete footings,
depending on the size of the structure. The outer skin generally
is constructed of corrugated steel or textile materials. Since self-
supported structures do nol: rely on interior air pressure for
Table 7-33. Additive Costs
Additive classification
Typical material cost ($/ft2)
Hygroscopic salt
Bitumens/add esives
Surfactant
0.02-0.10'
0.15-0.32 '
0.002 2
1 Source: U.S. EPA. 1888a.
* Vendor data at recommended dilution and application rates.
support, they usually are operated under negative pressure.
Negative interior pressure prevents PM/VOC emissions via
entrances/exits and leaks in the structure.
Air-supported structures consist of an all weather membrane
that is supported by positive pressure within the enclosure.
Because the enclosure is under positive pressure any leaks or
openings will result in PM/VOC emissions. To counter this, air-
supported structures often are equipped with air lock systems.
7.7.2 Applicability to Remediation Technology
The advantages/disadvantages of using enclosures to control:
PM/VOC emissions are outlined in Table, 7-36.
Because enclosures do not alter the physical properties of the
material to be treated, their compatibility with various remedia,-
tion technologies is only a function of the enclosure properties
(size and materials of construction) and the chemical properties
of the VOCs to be controlled (i.e., reactivity with enclosure
materials). In general, all materials handling processes (e.g.,
excavation, loading, storage piles, stabilization processes) emit-
ting PM/VOC can be controlled using enclosures. The key,
consideration as to whether or not enclosures are appropriate for
a given site is how expensive will it be to maintain the tempera-
ture and air quality within the dome at acceptable levels for
workers.
7.7.3 Range of Effectiveness
The effectiveness of enclosures to limitPM/VOC emissions is
a function of the enclosure "capture" efficiency and the point
source control system efficiency. Capture efficiency is a measure
of the ability of the enclosure to capture the emitting PM/VOC.
For example, if an air supported structure has a leak due to an
incomplete seal between the structure and the surface or if the
skin is torn, then PM/VOC will be emitted to the atmosphere, thus
lowering the capture efficiency. Point source control system
efficiency is dependent on the particular control system but
generally is in the range of 95-100%. Reported enclosure control
efficiencies are presented in Table 7-37. *!.
7.7A Sizing Criteria/Application Rates
Enclosures range in size from 30 feet in diameter to 130 feet
wide x 62 feet tall x unlimited length. For self-supported struc-
tures wider than 60 feet, footings may be required. Prior to
erecting an enclosure, the site may require grading so that the
slope is less than three percent (Sprung Instant Structures, Inc.,
1992).
7.7.5 Cost Estimating Procedure
The costs of air supported and self-supported enclosures are
given in Table 7-38. Note: The costs presented in Table 7-38 do
not include the costs of gas collection/treatment systems.
7.8 Collection Hoods
7.8.1 Process Description
Hoods are commonly used to capture PM/VOC emitted from
small area sources (e.g., waste stabilization/solidification mixing
silos, bioremediation reactors) and route those emissions to,
appropriate air pollution control devices. In practice, hoods are
120
-------
designed using the capture velocity principle which involves the
creation of an air flow after the emitting source that is sufficient
to remove the contaminated air.
Three hood designs that are commonly used are depicted in
Figure 7-5. The selection of hood type will be dependent on the
emitting source characteristics (e.g., source area and accessibil-
ity, emitting air velocity, surrounding air currents) and the
required capture efficiency. Major components of a hood exhaust
system are depicted in Figure 7-6.
7.8.2 Applicability to Remediation Technology
Hoods can be used to capture PM/VOC emissions from ex-situ
waste stabilization/solidification mixing silos and bioremedia-
tion reactors. The use of a hood will be contingent upon access to
the emitting source and upon the area of the emitting source. As
the distance between a source and hood increases, so does the
required total volumetric flow rate of air into the hood to maintain
a given capture efficiency. Since the cost of most air pollution
control equipment is proportional to the volumetric flow rate, a
point is reached where it is not economically feasible to use a
hood. The emitting source area will impact the hood size required
and for canopy and capturing hoods will impact the air flow rate
required to maintain a given capture efficiency. The advantages/
disadvantages of using a hood to capture PM/VOC emissions are
outlined in Table 7-39.
7.8.3 Range of Effectiveness
Parameters which influence the capture efficiency of hood
exhaust systems are given in Table 7-40.
Hood PM/VOC capturing efficiencies can be as high as 90 to
100%. However, PM/VOC control efficiencies will be a function
of both the hood capture efficiency and the air pollution control
equipment removal efficiency.
7.8.4 Application Rates
Hood exhaust systems designs are based on the hood aspect
ratio (width/length of hood), the required capture velocity (v),
and the distance of the furthest point of the emitting source from
the hood centerline (x). Ranges of capture velocities required as
a function of surrounding air turbulence and the emitting source
are listed in Table 7-41 .The velocities obtained from Table 7-41
can then be used in the design, equations presented in Table 7-42.
For a more thorough presentation on hood designs, see "Indus-
trial Ventilation" (ACGIH, 1980).
7.8.5 Cost Estimating Procedure
The costs of hood exhaust systems are highly dependent on the
volumetric flow rate, the length of ducting required, the hood/
ducting materials of construction required (e.g., carbon steel,
stainless steel), hood size, and fan size required to move the air.
An example of a hood exhaust system cost breakdown is pre-
sented in Table 7-43.
Table 7-34. Advantages/Disadvantages of Operational Practices/Procedures
Operational practices/procedures
Advantage
Disadvantage
Road cleaning practices
Seasonal scheduling
Vehicle speed control
Storage pile geometry/orientation
Excavation practices
Dumping practices
Soil loading practices
Equipment readily available
Easy to operate
No equipment required
Potentially high control of PM/VOC emissions
Inexpensive
Easy to implement
Reduces road maintenance required
Decreases incidence of accidents
Inexpensive
Easy to implement
Inexpensive
Easy to implement
Decreases time to excavate a given volume
Easy to implement
Inexpensive
Easy to implement
Inexpensive
Requires additional equipment.
Broom sweeping may increase particulate
matter emissions.
Stagnant wind conditions may lead to
unacceptable ambient air concentrations at
the work site.
Rigorous timing constraints.
Increases haul time.
May require more vehicles.
Difficult to maintain optimum pile geometry.
Optimum pile geometry may not be possible
due to space limitations.
Requires larger equipment which may
increase excavation cost.
Increased equipment size may not be
practical due to site size constraints.
Increased equipment size may damage
paved roads and increase particulate matter
emissions from unpaved roads.
May increase unloading time.
May decrease volume of soil per haul.
121
-------
Tablo 7-35. Reported PM/VOC Control Efficiencies for Operational Practices/Procedures
Operational practice/procedure
Road cleaning practice
• Vacuum sweeping
• Water flushing
• Water flushing/broom sweeping
Vehicle speed control
Storage pile geometry/orientation
Excavation practices
Dumping practice
Pollutant
PM
PM
PM
PM
PM4
PM4
PM4
Reported control
efficiency (%)
0-58
0-69
0-96
0-80
0-80/60 1
20 2
50 3
Comments
Measured
Measured
Measured
Estimated
Estimated
Theoretical
Theoretical
Source
U.S. EPA, 1989
U.S. EPA, 1987b
U.S. EPA, 1991/Vogel, 1985
U.S. EPA, 1985a
U.S. EPA, 1985a
1 Pilo length Is perpendicular to prevailing wind direction.
* Calculated from AP-42 Emission Factors, assuming doubling of bucket capacity.
* Calculated from AP-42 Emission Factors, assuming a 50% reduction in material drop height.
4 Somo VOC control would also occur, but no control efficiency data are available.
For further guidance in obtaining cost estimates for hood
exhaust systems, consult Vatavuk, 1990.
7.9 Miscellaneous Controls
A number of miscellaneous controls for area sources could
theoretically be used at Sur.erfund sites. VOC emissions from
lagoons could be controlled using floating solid objects such as
hollow plastic spheres or rafts, or a floating layer of immiscible
oil (Springer, et al, 1986). Blankets of nitrogen or other inert
gases are another option. The use of these types of controls at
Superfund sites has not been documented and feasibility testing
certainly would be advisable prior to any full-scale use.
Table 7-36. Advantages/Disadvantages of Enclosures to Control PM/VOC Emissions
Advantages
Disadvantages
Near 100% control of PM/VOC emissions
Conversion of area source Into point source for easier control of PM/
VOC emissions.
Limits access to working area.
Compatible with remediation of soils, sludges, and liquids.
Does not create additional material to be treated.
Reduces groundwater and surface water contamination due to
precipitation.
Assist thermal treatment of solids by excluding moisture loading rate
onto soil.
Expensive.
Increased temperatures and PM/VOC concentrations limit the ability to
work inside enclosure to short-time periods and may require workers
to use protective apparel. Also specific VOC concentrations must
not exceed OSHA prescribed IDLH values.
Air supported structures can be damaged by wind.
Some VOCs may damage polymeric skin materials.
Building permits may be required by local municipalities.
Limits to structure size may require structure relocation as excavation
site moves.
Requires point source PM/VOC controls.
Enclosure may require decontamination following its use.
122
-------
Table 7-37. Reported Enclosure PM/VOC Control Efficiencies
Contaminant .
Reported control efficiency (%)
Comment
Source
PM/VOC
PM
up to 100
70-99
Material handling operations
Active storage piles
U.S. EPA, 1991
U.S. EPA, 1987b
Table 7-38. Enclosure Costs
Type of enclosure
Enclosure rental cost a'b
($/month-m 2)
O&M cost.($/m2).'
Source
Air-supported
Self-supported
Self-supported
5.5
19
4.4-8.5
NR U.S. EPA, 1988b
48 ' Aul, 1992
1.2-6.5 Sprung Instant Structures, 1992
The cost of grading and footings are not included.
If the structure is needed for more than 2 years, purchase of the structure may be more economical than renting.
O&M costs include the cost of erecting/dismantling the enclosure.
To fan
(a) Enclosures—contain
contaminants released
inside the hood
(b) Canopy hoods—catch
contaminants that rise
into them
To fan
Source: Cooper and Alley, 1990.
(c) Capturing hoods—reach
out to draw in contaminants
Figure 7-5. Three commonly used hood designs.
123
-------
Entry •—
Source: NIOSH, 1973.
Figure 7-6, Components of a hood exhaust system.
Hood
Cleaner
Table 7-39. Advantages/Disadvantages of Hoods to Capture PM/VOC Emissions
Advantages Disadvantages
90-100% PMA/OC capture efficiencies are possible
Much data available regarding selection and design of hoods
Conversion of an area source into a point source. Most air
pollution control equipment is designed for point sources
Emission source must be accessible to hood.
Contaminant diluted by air flow into hood. This can effect air pollution
control efficiencies (e.g., carbon adsorption, incineration).
Power cost may be high due to required capture velocity and headless
through ductwork. Use of hoods is practical for small area sources only.
Hoods subject to corrosion (e.g., acid gases, lime).
Tablo 7-40. Parameters That Affect Hood Capture Efficiencies
Parameter
Comment
Distance between hood and farthest point of emitting source.
Volumetric flow rate into the hood
Surrounding air turbulence
Hood design
As this distance increases, for a given volumetric flow rate into the hood, the
capture efficiency decreases.
As the volumetric flow rate into the hood increases, the capture efficiency
increases.
As the surrounding air turbulence increases, the required volumetric flow
rate into the hood increases to maintain a given capture efficiency.
Hood designs are tailored to specific types of emitting sources. For
example, canopy hoods are designed to collect emissions from heated
open-top tanks.
124
-------
Table 7-41. Range of Capture Velocities
Condition of dispersion of contaminant
Examples Capture velocity (fpm)
Released with practically no velocity into quiet air
Released at low velocity into moderately still air
Active generation into zone of rapid air motion ,
Released at high initial velocity into zone of
very rapid air motion.
Evaporation from tanks; degreasing, etc.
Spray booths; intermittent container filling;
low speed conveyor transfers; welding;
plating; pickling.
Spray painting in shallow booths; barrel
filling; conveyor loading; crushers.
Grinding; abrasive blasting.
50-100
100-200
200-500
500-2000
Note: In each category above, a range of capture velocity Is shown. The proper choice of value depends on several factors:
Lower end of range
1. Room air currents minimal or favorable to capture
2. Contaminants of low toxlcity or of nuisance value only.
3. Intermittent, low production.
4. Large hood, large air mass in motion.
Upper end of range
1. Disturbing room air current.
2. Contaminants of high toxicity.
3. High production, heavy use.
4. Small hood—local control only.
125
-------
Table 7-42. Hood Design Equations
Description
Slot
Flanged slot
Plain opening
Flanged opening
Booth
Aspect ratio (W/L)
0.2 or less
0.2 or less
0.2 or greater and round
0.2 or greater and round
To suit work
Air volume
Q = 3.7 LVX
Q = 2.8 LVX
V(10X2
Q = 0.75V (10X2 +A)
Q = VA = VWH
Canopy
To suit work
Q = 1.4PVD
Kay.
W
Conledine distance to point x in emissions plume (ft)
Length (It)
Width (II)
Height (ft)
Distance between hood and source (ft)
Area (sq. It.)
Ftowrate (It * /min)
Perimeter of hood (ft)
Velocity at point x (ll/min)
Source: NIOSH, 1973.
126
-------
Table 7-43. Hood Exhaust System Cost Estimate
Equipment item
Applicable dimensions
Cost ($)
Canopy hood 3/16 in. thick carbon steel $2,400
10 ft. diameter.
Ductwork 100 ft. of 1 -foot 1,300
diameter, 16-gauge carbon
steel straight duct.
Four 1-foot diameter 16 1,750
gauge carbon steel 90°
elbows.
Radial tip fan Moves 11,000 acfm at 10 7,700
in. H2O with a 45 1/2 in.
wheel diameter.
Total cost: $13,150
7.10 References
ACGIH. Industrial ventilation, 16th Edition American Confer-
ence of Government Industrial Hygienists, Lansing, MI,
1980.
Aim, R.R., K.A. Olson, and R.C. Peterson. Using Foam to
Maintain Air Quality During Remediation of Hazardous
Waste Sites. Presented at the 80th Annual AWMA Meet-
ing (Paper 87-18.3), New York City, June 21-26,1987.
. Aul, Ed (Ed Aul Engineering). Personal communication from
Ed Aul to Barry Walker of Radian Corporation. 1992.
Buonicore, A.J. and W.T. Davis, ed., Air Pollution Engineer-
ing Manual, Air and Waste Management Assoc.; Van
Nostrand Reinhold. NY, NY. 1992.
Cooper, C.D. and F.C. Alley. Air Pollution Control A Design
Approach. Waveland Press Inc. (ISBN 0-88133-521-5).
Prospect Heights, IL. 1990.
Landreth, R.E. et al., Lining of Waste Containment and Other
Impoundment Facilities. U.S. EPA 600/2-88/052 (NTIS
PB89-129670).
Means Site Work Cost Data, K. Smit Sr. Editor. Published by
RJ. Grant. 1991
NIOSH. The Industrial Environment - its Evaluation and
Control. HSM-99-71-45. Washington, D.C. 1973.
Radian Corp. Preliminary Assessment of Potential Organic
Emissions from Dredging Operations. EPA Contract No.
63-02-4288, WA 35. Report to Dennis Timberlake, U.S.
EPA, Cincinnati, OH. September 1991.
Radian Corp. GLOVES Software, Product Brochure. Radian,
Austin, TX. 1992.
Rusmar Corporation. Personal communication fromPaul Russo
to Barry Walker of Radian Corporation. Rusmar, West
Chester, PA. 1992.
Schmidt, C. (Independent Consultant). Personal communica-
tion from Chuck Schmidt to Barry Walker of Radian
Corporation. January 1992.
Springer, C., K.T. Valsaraj, and L.J. Thibodeaux. In Situ
Methods to Control Emissions from' Surface Impound-
ments and Landfills. JAPCA Vol. 36, No. 12, pp!371-
1374, December 1986.
Sprung Instant Structures Inc. Personal communication from
Grant Cleverley to Barry Walker of Radian Corporation.
Sprung, Allentown, PA. 1992.
3M Corporation. Personal Communication from Stu Wagner
to Barry Walker of Radian Corporation. 3M, St. Paul, MN.
1992.
U.S. EPA. Dust Control at Hazardous Waste Sites. EPA/540/
2-85/003. U.S. EPA-HWERL, Cincinnati, OH. Novem-
ber 1985a.
U.S. EPA. Compilation of Air Pollution Emission Factors, AP-
42,4th Ed. U.S. EPA, Research Triangle Park, NC. 1985b.
(Supplement A, October 1986; Supplement B, September
1988; Supplement C, September 1990).
U.S. EPA. Handbook For Using Foams to Control Vapors
From Hazardous Spills. EPA/600/8-86/019 (NTTS PB87-
145660). U.S. EPA, Cincinnati, OH. July 1986a.
U.S. EPA. Field Evaluation of Windscreens as aFugitiveDust
Control Measure for Material Storage Piles. EPA/600/7-
86/027 (NTIS PB-86-231289). U.S. EPA, Research Tri-
angle Park, NC. July 1986b.
U.S. EPA. Emission Control Technologies and Emission Fac-
tors for Unpaved Road Fugitive Emissions. User's Guide
EPA/625/5-87/022. U.S. EPA, Cincinnati, OH. Septem-
ber 1987a.
U.S. EPA. Method for Estimating Fugitive Particulate Emis-
sions from Hazardous Waste Sites. EPA/600/2-87/066
(NTIS PB87-232203). U.S. EPA, Cincinnati.OH. August
1987b.
U.S. EPA. Evaluation of the Effectiveness of Chemical Dust
Suppressants on UnpavedRoads.EPA/600/2-87/112. U.S.
EPA, Research Triangle Park, NC. 'November 1987c.
U.S. EPA. Control of Open Fugitive Dust Sources. EPA-450/
3-88-008. U.S. EPA-OAQPS, Research Triangle Park,
NC. September 1988a.
U.S. EPA. Dust and Vapor Suppression Technologies for Use
During the Excavation of Contaminated Soils, Sludges, or
Sediments. Land Disposal, Remediation Action, Incin-
eration, and Treatment of Hazardous Waste - Proceedings
of the Fourteenth Annual Research Symposium (pp.53-
64). EPA/600/9-88/021. U.S. EPA, Cincinnati, OH. July
1988b.
127
-------
U.S. EPA. Engineering Bulletin - Control of Air Emissions
from Material Handling. EPA/540/2-91/023. U.S. EPA,
Cincinnati, OH. Octoljer 1991.
Vatavuk, W. Estimating Costs of Air Pollution Control. Lewis
Publishers, Chelsea, MI. 1990.
Vogel, G.A. Air Emission Control at Hazardous Waste Man-
agementFacilities. JAPCA Vol. 35, No. 5, p558-566, May
1985.
128
-------
Appendix A
Remediation Control Vendors
The names of remediation control vendors that appear in this appendix were obtained from the March 1991 Journal of Air
and Waste Management Association Buyer's Guide, the Thomas Register, the Pollution Engineering Yellow Pages, and
various references in the literature. Many of these vendors assisted in the preparation of this document by providing techni-
cal and costing information. These vendors are identified by an "*."
This list does not represent endorsement of any of the following companies by the U.S. EPA. Furthermore, while the list is
as complete as possible, it does not necessarily include all vendors of all control devices that could be used at Superfund sites.
Carbon Adsorption
Amcec Corp., Solvent Recovery Div.*
Oakbrook, IL
(708)954-1515
American Environmental Int'l Inc.*
Northbrook, IL
(708)272-8635
Barnebey & Sutcliff Corp.*
Columbia, MD
(301)381-5870
Calgon Carbon Corp.*
Pittsburgh, PA
(412)787-6700
Cameron-Yakima*
Yakima, WA
(509)452-6609
DCI Corp.*
Indianapolis, IN
(317)872-6743
Dedert Corporation
Olympia Fields, IL
(708) 747-7000
Environmental Instruments, Inc.
Concord, CA
(800) 648-9355
Envirotrol, Inc.*
Sewickley, PA
(412)741-2030
Extraction Systems, Inc.
Woonsocket, RI
(401)769-1113
RaySolv.lnc.
Piscataway, NJ
(908)981-0500
Vic Manufacturing Co.*
Minneapolis, MN
(612)781-6601
Zink Co., John*
Tulsa, OK
(918)747-1371
Catalytic Oxidation
Advanced Catalyst Systems Inc.*
South Plainfield, NJ
(908)753-9670
The Air Preheater Co. Inc.*
Wellsville, NY
(800)828-0444
Allied Signal*
Tulsa, OK
(708)450-3900
American Environmental International, Inc.
Northbrook, IL
(708) 272-8635
129
-------
Anguil Environmental Systems*
Milwaukee, WI
(414)332-0230
Branch Environmental Corp.*
Somerville.NJ
(908)526-1114
CametCo.
Hiram, OH
(216)569-3245
Dedcrt Corporation
Olympia Fields, IL
(708)747-7000
EPCON Industrial Systems Inc.*
The Woodlands, TX
(409)273-1774
Johnson Matthey*
Wayne, PA
(215)971-3000
ORS Environmental Equipment*
ChaddsFord.PA
(215)558-1750
Saxton Air Systems Inc.
Harrisburg, PA
(717)545-3784
Condensers
Airco Industrial Gases
Murray Hill, NJ
(908)464-8100
Amcec Corp.*
Oak Brook, IL
(708)954-1515
American Environmental Int'l, Inc.*
Northbrook, IL
(708)272-8635
Internal Combustion Engines
Remediation Services Int'l
Oxnard, CA
(805)644-5892
VR Systems Inc.
Anaheim, CA
(714)826-0483
Environmental Techniques Inc.
Huntington Beach, CA
(714)962-5025
SoilBeds/Biofilters
Ambient Engineering Inc.
Matawan, NJ
(908) 566-7722
Fabric Filters
AirPol, Inc.
Teterboro.NJ
(201)288-7070
American Air Filter
Louisville, KY
(502)637-0011
Ducon Environmental Technology, Inc.
Mineola, NY
(516)420-4900
Dustex Corp.
Charlotte, NC
(704)588-2030
Fair Co.
Los Angeles, CA
(800)333-7320
George A. Rolfes Co.*
Boone, IA
(515)432-3300
Lodge-Cottrell Systems*
Houston, TX
(713)297-2092
MIDWESCO Inc.*
Winchester, VA
(800)336-7300
MiKroPul Environmental Systems*
Morris Plains, NJ
(201)606-5900
Ogden Environmental Services, Inc*
San Diego, CA
(619)455-3045
P & S Filtration*
Skaneateles Falls, NY
(315)685-3466
130
-------
Sealed Air Technologies*
Fort Thomas, KY
(606)7881-4330
SteelcraftCorp.*
Memphis, TN
(901)452-5200
Wheelabrator Air Pollution Control
Pittsburgh, PA
(412)562-7300
Smidth, F.L. & Co.
Cresskill.NJ
(201)871-3300
United McGill
Columbus, OH
(614)443-0192
Wheelabrator Air Pollution Control
Pittsburgh, PA
(412)562-7300
Electrostatic Precipitators
American Air Filter
Louisville, KY
(502)637-0011
Beltran Associates Inc.
Brooklyn, NY
(718)338-3311
CE Environmental Systems Div.*
Birmingham, AL
(205)991-2832
Ducon Environmental Technology, Inc.
Mineola,NY
(516)420-4900
Flakt Inc. Environmental Systems Div.
Vancouver, BC
(615)693-7550
Fuller Co.
Bethlehem, PA
(215)264-6011
Lodge-Cottrell Systems*
Houston, TX
(713)297-2092
MiKroPul Environmental Systems*
Morris Plains, NJ
(201)606-5900
Niro Atomizer Inc.
Columbia, MD
(301)997-8700
Sealed Air Technologies*
Fort Thomas, KY
(606)7881-4330
Scrubbers
Advanced Air Technology*
Arlington Heights, IL
(703) 394-9553
AirPol Inc.*
Teterboro, NJ
(201)288-7070
American Air Filter
Louisville, KY
(502)637-0011
Anderson 2000
Peachtree City, GA
(414)332-0230
Astec Industries*
Chattanooga, TN
(615)867-4210
B ACT Engineering, Inc.
Arlington Heights, IL
(708)577-0950
Bayco Industries of California*
San Leandro, CA
(415)562-6700
Branch Environmental Corp.*
Somerville, NJ
(908)526-1114
The Ceilicote Co.
Berea, OH
(216)243-0700
Croll-Reynolds Co., Inc.
Westfield, NJ
(908)232-4200
131
-------
Dow Chemical Gas Spec Div.*
Houston, TX
(713)978-3894
Ducon Environmental Technology, Inc.
Mineola.NY
(516)420-4900
Environmental Elements Corp.*
Baltimore, MD
(301)368-7000
Fisher-Klosterman*
Louisville, KY
(502)776-1505
Lodge-Cottrell Systems*
Houston, TX
(713)297-2092
MetPro Corp. Duall Div.*
Owosso, MI
(517)725-8184
NaTec Environmental Systems Div.*
Houston, TX
(214)824-2910
Niro Atomizer Inc.
Columbia, MD
(301)997-8700
Ogden Environmental Services, Lie*
San Diego, CA
(619)455-3045
Quad Environmental Technologies Corp.*
Northbrook.IL
(708)564-5070
Wheelabrator Air Pollution Control
Pittsburgh, PA
(412)562-7300
Zink Co., John*
Tulsa.OK
(918)747-1371
Thermal Oxidation
The Air Preheater Co. Inc.*
Wellsville,NY
(800)828-0444
AmcecCorp.*
Oakbrook, IL
(708)954-1515
American Environmental Int'l Inc.*
Northbrook, IL
(708)272-8635
Astec Industries*
Chattanooga, TN
(615)867-4210
Bayco Industries of California*
San Leandro, CA
(415)562-6700
Branch Environmental Corp. PCT Div.*
Somerville, NJ
(908)526-1114
Brule Incinerators*
Blue Island, EL
(708)388-7900
CEL Incinerator System Inc.*
Collegeville, PA
(215)287-8037
Conversion Technology, Inc.*
Norcross, GA
(404)263-6330
DedertCorp.*
Olympia Fields, IL
(708)747-7000
EPCON Industrial Systems Inc.*
The Woodlands, TX
(409)273-1774
Eutherengy Systems, Inc.
Sanford.MI
(512)687-2899
FECO Engineered Systems Inc., Environmental Div.*
Cleveland, OH
(216)441-2400
In-Process Technology*
Sunny vale, CA
(408)745-1066
Johnson Matthey*
Wayne, PA
(215)971-3000
132
-------
M & S Engineering and Manufacturing Co. Inc.*
Broad Brook, CT
(203)627-9396
MetPro Corp. Duall Div.*
Owassa, MI
(517)725-8184
NAOInc.*
Philadelphia, PA
(215)743-5300
Precision Quincy Corp.
Woodstock, IL
(815)338-2675
REECOInc.*
Morris Plain, NJ
(201)538-8585
Salem Industries*
South Lyon, MI
(313)437-4188
Saxton Air Systems Inc.
Harrisburg, PA
(717)545-3784
Somerset Technologies Inc. Ross-Waldron Div.
New Brunswick, NJ
(908)356-6000
Smith Engineering Co.*
Duarte, CA
(714)923-3331
Surface Combustion
Toledo, OH
(800)537-8980
Vulcan Iron Works Inc.*
Wilkes-Barre.PA
(717)822-2161
WBR Engineering Inc.
Cherry Hill, NJ
(609)354-9372
Williams Environmental*
Stone Mountain, GA
(800)247-4030
Zink Co., John*
Tulsa,OK
(918)747-1371
Venturi Scrubbers
Croll-Reynolds Co., Inc.
Westfield.NJ
(908)232-4200
Fairchildlnt'l
Glen Lyn, VA
(703)726-2380
Miscellaneous (HEPA Filters)
Covers/Barriers
see local contractors
Foams
3M: Industrial Chemicals Div.: Environmental Protection
Group
St. Paul, MN
(612)733-3493
Rusmar
West Chester, PA
(215)436-4314
Water Sprays
see local contractors
Enclosures
Sprung Instant Structures Inc.
Allentown,PA
(800)677-7864
Houston, TX
(713)520-6888
see also local contractors
Wind Barriers
see local contractors
133
-------
Region
Appendix B
Regional Air/Superfund Coordinators
(as of September 1992)
Name
I RoseToscano
H Alison Devine
HI Patricia Flores
IV Lee Page
V Charles Hall
VI MarkHansen
VII Wayne Kaiser
VIE NormHuey
K KathyDiehl
X Chris Hall
(617)565-3280
(212)264-9893
(215)597-9134
(404)347-2864
(312)886-6043
' (214)655-7223
(913)551-7603
(303)293-0969
(415)744-1133
(206)553-1949
Bibliography of Air/Superfund Documents
AS-1 Stoner, R., et al. Procedures for Conducting Air Pathway Analyses for Superfund Activities, Interim Final
Document: Volume 1 -Application of Air Pathway Analyses for Superfund Activities. EPA-450/1-89-001
(NTIS PB90-113374/AS). July 1989.
Eklund, B., et al. Procedures for Conducting Air Pathway Analyses for Superfund Activities, Interim Final
Document Volume 2 - Estimation of Baseline Air Emissions at Superfund Sites (Revised). EPA-450/1-89-
002a (NTIS PB90-270588). August 1990.
Eklund, B., el: al. Procedures for Conducting Air Pathway Analyses for Superfund Activities, Interim Final
Document: Volume 3 - Estimation of Air Emissions From Clean-up Activities at Superfund Sites EPA-450/1-
89-003 (NTIS PB89-180061/AS). January 1989.
Stoner, R., et al. Procedures for Conducting Air Pathway Analyses for Superfund Activities, Interim Final
Document: Volume 4 - Procedures for Dispersion Modeling and Air Monitoring for Superfund Air Pathway
Analyses. EPA-450/l-89-004(NTlSPB90-113382/AS). July 1989.
TRC Environmental Consultants. A Workbook of Screening Techniques For Assessing Impacts of Toxic Air
Pollutants. EPA-450/4-88-009. September 1988.
Salmons, C., F. Smith, and M. Messner. Guidance on Applying the Data Quality Objectives For Ambient Air
Monitoring Around Superfund Sites (Stages I & II). EPA-450/4-89-015 (NTIS PB90-204603/AS). August
1989.
AS-2
AS-3
AS-4
AS-5
AS-6
AS-7
Pacific Environmental Services. Soil Vapor Extraction VOC Control Technology Assessment EPA-450/4-89-
017. September 1989.
134
-------
AS-8 TRC Environmental Consultants. Review and Evaluation of Area Source Dispersion Algorithms for Emission
Sources at Superfund Sites. EPA-450/4-89-020. November 1989.
AS-9 Letkeman, J. Superfund Air Pathway Analysis Review Criteria Checklists. EPA-450/1-90-001 (NTIS PB90-
1825447AS). January 1990. • , /,,
AS-10 Smith F C Salmons/M. Ressner, and R. Shores. Guidance on Applying the Data Quality Objectives For
Ambient Air Monitoring Around Superfund Sites (Stage ffl). EPA-450/4-90-005 (NTIS PB90-20461 I/AS).
March 1990.
AS-11 Saunders, G. Comparisons of Air Stripper Simulations and Field Performance Data. EPA-450/1-90-002.
.March 1990.
AS-12 Damle, A.S., and T.N. Rogers. Air/Superfund National Technical Guidance Study Series: Air Stripper Design
', Manual. EPA-450/1-90-003. May 1990.
AS-13 Saunders, G. Development of Example Procedures for Evaluating the Air Impacts of Soil Excavation Associ-
ated with Superfund Remedial Actions. EPA-450/4-90-014(NTISPB90-255662/AS). July 1990.
AS-14 Paul, R. Contingency Plans at Superfund Sites Using Air Monitoring. EPA-450/1-90-005 (NTIS PB91-
,10212,9). September 1990.
AS-15 Stroupe, K, S. Boone, and C. Thames. User's Guide to TSCREEN - A Model For Screening Toxic Air
Pollutant Concentrations. EPA-450/4-90-013. December 1990.
AS-16 Winges, K.D. User's Guide for the Fugitive Dust Model (FDM) (Revised), User's Instructions. EPA-910/9-88-
202R (NTIS PB90-215203.PB90-502410). January 1991.
AS-17 ' Thompson, P., A.Ingles,and B.Eklund. Emission Factors For Superfund Remediation Technologies. EPA-
'•'-:!"; 450/1-91-001 (NTIS PB91-190-975), March 1991.
AS-18 Eklund, B., C. Petrinec, D. Ranum, and L. Howlett. Database of Emission Rate Measurement Projects -Draft
;; Technical Note. EPA-450/1-91-003'(NTISPB91-222059). June 1991.
AS-19 Eklund, B., S. Smith, and M. Hunt. Estimation of Air Impacts For Air Stripping of Contaminated Water. EPA-
450/1-91-002 (NTIS PB91-211888), May 1991 (Revised August 1991).
AS-20 -Mann, G. and J: Carroll. Guideline For Predictive Baseline Emissions Estimation Procedures For Superfund
Sites. 'EPA-450/1-92-002 (NTIS PB92-171909). January 1992.
AS-21 Eklund, B., S. Smith, P. Thompson, and A. Malik. Estimation of Air Impacts For Soil Vapor Extraction (SVE)
• •-•-•••: Systems. EPA-450/l-92-001.(NTISPB92-143676/AS), January 1992. :
AS-22 Carroll, J. Screening Procedures For Estimating the Air Impacts of Incineration at Superfund Sites. EPA-450/
. 1-92-003 (NTIS PB92-171917). February 1992.
AS-23 Eklund, B., S. Smith, and A. Hendler. Estimation of Air Impacts For the Excavation of Contaminated Soil.
. • EPA 450/1-92-004 (NTIS PB92-171925), March 1992.
AS-24 Draves, J. and B. Eklund. Applicability of Open Path Monitors for Superfund Site Cleanup. EPA-45 l/R-92-
001. May 1992.
135
-------
vvEFA
United States
Environmental
Protection Agency
Office of
Air Quality Planning
and Standards
September 1993
AIR/SUPERFUND COORDINATION
PROGRAM
offices
PURPOSE
rPOSB °f thS Air/Superfund Coordination program is to assist EPA Regional Superfund
Evaluate the impact of air emissions from Superfund sites prior to and during remediation
3(10
Develop and implement site cleanup measures to mitigate these impacts to ensure
protection of public health and the environment.
REGIONAL ACTIVITIES
An Air/Superfund Coordinator in each Regional Air office is responsible for ensurina that air
program support Is provided to Superfund. Air offices provide routine site support s "rtices such as
S!±SE and T?6",? Pr°PO,SalS' Plans' and Studies' TheV ParticiP*e in deoSrs relaS to
preremedtel, remedial, and removal actions that may have significant air impacts. They help to ensure
that Superfund site decisions involving air pollution issues are consistent with Federa State and
Sn1rHand-Pt° CieSrf,Th,ey alS° may Perf0rm Special field evaluations du™9 remova and
— Jn ^ SUCh 3S ^ m0delin9' m
PROGRAM SUPPORT ACTIVITIES
The program includes four types of activities to support the Regional Offices. They are:
0 Coordination o Technical assistance
Tracing ° National Technical Guidance Studies
Coordination:
Coordination program facilities the exchange of information on Air/Superfund issues
Sat^tS9 ufnalf ™T ^ betW6en Regi°ns and EPA He^q£^oSTa
updated technical information and periodic reports to these offices on ongoing studies. Coordination
136
-------
meetings are held at four month intervals to exchange information, coordinate the overall program,
participate in miniworkshops, and receive briefings on pertinent technical and administrative subjects.
Training:
Training involves briefing Regional Air office staff on Superfund program issues, priorities,
methods, and procedures; and briefing Regional Superfund staff and contractor personal on air issues,
and guidance for analyzing and resolving them.
Technical Assistance:
Technical assistance is offered to Regional Air offices to assist them in analyzing air issues
associated with specific sites, reviewing analyses prepared by Superfund contractors, and preparing
recommendations on remedial actions proposed to minimize air impacts.
National Technical Guidance Studies:
National Technical Guidance Studies (NTGS) provide Regional Air and Superfund staffs and State
and local agency staffs with technical support, data, and guidance to improve the quality of the data base
and the analysis of air issues associates with Superfund sites.
/•*
o
o
0
o
o
1-
AIR/SUPERFUND COORDINATORS
Region 1
Region II
Region III
Region IV
Region V
Abdi Mohamoud
(617) 565-4044
Alison Devine
(212) 264-9868
Patricia Flores
(215)597-9134'
Lee Page
(404) 347-2864
Dan Meyer
(312) 886-9401
0 Region VI
0 Region VII
0 Region VIII
0 Region IX
0 Region X
Mark Hansen
(214) 655-6582
Wayne Kaiser
(913) 551-7603
Norm Huey
(303) 293-0969
Kathy Diehl
(415)744-1133
Chris Hall
(206)553-1949
WHERE CAN I OBTAIN INFORMATION ON THE AIR/SUPERFUND
PROGRAM
Regional Air/Superfund Coordinators
or
Office of Air Quality Planning and Standards
Joseph Padgett
(919)541-5589
137
-------
Appendix C
Bibliography
VOC and PM Emission Controls
Point Sources
The Air Pollution Consultant. VOC Emission Control During Site Remediation. McCoy and Associates, Lakewood CO
September/October 199 la. . '
The Air Pollution Consultant. Biofiltration Shows Potential as 'an Air Pollution Control Technology. McCoy and Associates
Lakewood, CO. November/December 1991b.
Baker, R.W., K. Hibino, J. Mohr, and T. Kuroda. Membrane Research in Energy and Solvent Recovery from Industrial
Effluent Streams. DOE/ID/12379-T2, report for the period 5/11/83 - 31/1/85. Prepared by Membrane Technology &
Research, Inc. for the U.S. DOE. 1985. •
Baker, R.W., N. Yoshioka, J.M. Mohr, and A. J. Khan. Separation of Organic Vapors Air. J. Membrane Science, Vol. 31,
pp25y-27 ] * 1987.
Brkkwell, M., M. de Zeeuw, and F. Sapienze. Flue Gas Treatment Technologies in Western Europe: Regulations and
Practice International Waste Management. In: Proceedings of the 12th National Conference on Hazardous Materials
Control/Superfund '91. HMCRI, Silver Spring, Maryland. 1991. ' '
Brna, T. G. and C.B. Sedman. Waste Incineration and Emission Control Technologies. EPA/600/D-87/147-S. 1987.
Brna, T.G. Controlling PCDD/PCDF Emissions from Incinerators by Flue Gas Cleaning. EPA/600/D-90/239. 1990.
Brunner, C.R. Incineration Systems: Selection and Design. Van Nostrand Reinhold, New York. 1984.
Buck, F.A. and D.E. Hasselmann. Control of Air Emissions From Soil Venting Systems. Presented at the Symposium on
New Technologies for Waste Incineration (Paper No. 94a). Aprils, 1989.
Byers, W.D. Control of Emissions From an Air Stripper Treating Contaminated Groundwater. Environmental Progress Vol
7, No. 1. February 1988. . , '
Chu, RJ. VOC Emission Control Technologies for Site Remediation. In: Proceedings of National Research & Development
Conference on the Control of Hazardous Materials, Anaheim, CA. HMCRI Publications Dept., Greenbelt, MD. 1991.
Cooper, C. and F. Alley. Air Pollution Control - A Design Approach. Waveland Press, Prospect Heights, IL. 1986.
Danielson,J.A.,ed. Air Pollution Engineering Manual-Second Edition. (AP-40) U.S. EPA. RTP, NC. May 1973.
Donnelly, J.R. and K.S. Felvang. Low Outlet Temperature Operation Resource Recovery SDA Emission Control Systems
In: Proceedings of the 82nd Annual Meeting and Exhibition of A WMA. Anaheim, CA. June 1989.
Donnelly, J. Air Pollution Controls for Hazardous Waste Incinerators. Thermal Treatment and Air Pollution Control In-
Proceedings of the 12th National Conference onHazardous Materials Control/Superfund'91. December 1991.
138
-------
Frame, G.B. Air Pollution Control System for Municipal Solid Waste Incinerators. JAPCA, Vol. 38, No. 8. 1988.
HWC. Internal Combustion Engines Destroy Vapors From Gasoline Cleanup Projects. The Haz. Waste Consultant, ppl .4-
1.7. McCoy and Associates, Lakewood, CO. Sept/Oct 1988.
Hesketh, H.E. Air Pollution Control: Traditional and Hazardous Pollutants. Technomic Publishing Co., Lancaster, PA.
1991.
Hummel, K.E. and T.P. Nelson. Test and Evaluation of a Polymer Membrane Preconcentrator. Prepared by Radian Corp.
EPA 600/2-90-016. April 1990.
Katari, V.W., W. Vatavuk, and A.H. Wehe. Incineration Techniques for Control of Volatile Organic Compound Emissions,
Part I: Fundamentals and Process Design Considerations. JAPCA, Vol. 37, No. 1. 1987a.
Katari, V.W., W. Vatavuk, and A.H. Wehe. Incineration Techniques for Control of Volatile. Organic Compound Emissions,
Part'll: Capital and Annual Operating Costs. JAPCA Vol. 37, No. 2. 1987b.
Keener, T.C. and D, Zhou. Prediction of Activated Carbon Adsorption Performance Under High Relative Humidity Condi-
tions. Env. Progress, Vol. 9, No. 1, pp40-46. February 1990.
Kittrell, J., C. Quinlan, and J. Eldridge. Direct Catalytic Oxidation of Halogenated Hydrocarbons. JAWMA, Vol. 41, No. 8,
ppll29-'ll33. 1991. •
Kosusko, M., M.E. Muffins, K. Ramanathan, and T.N. Rogers. Catalytic Oxidation of Groundwater Stripping Emissions.
Env. Progress, Vol. 7, No. 2, pp!36-142,May; 1988.
Kosusko, M. and C.M. Nunez. Destruction of Volatile Organic Compounds Using Catalytic Oxidation. JAPCA, Vol. 40,
No. 2, pp254-259, February 1990. '
Kosusko, M. and G.M. Ramsey. Destruction of Air Emissions Using Catalytic Oxidation. EPA/600/D-88/017. (NTIS
PB88-214952). May 1988.
Larsen, E.S. and M J. Pilat. Design and Testing of a Moving Bed VOC Adsorption System. Env. Progress, Vol. 10, No. 1,
pp75-82, February 1991.
Lawless, P.A. and L.E. Sparks. A Review of Mathematical Models for ESPs and Comparison of their Successes. In:
Proceedings of the Second International Conf. on Electrostatic Precipitation, pp513-522. S. Masuda, ed., Kyoto. 1984.
Leson, G. and A. Winer. Biofiltration: An Innovative Air Pollution Control Technology for VOC Emissions. JAWMA, Vol.
41, No. 8. August 1991.
Marzone, R.R. and D.W. Oakes. Profitably Recycling Solvents from Process Systems. Pollution Eng., Vol. 5, No. 10, pp23-
24. 1973.
Miller, D. and L. Canter. Control of Aromatic Waste Air Streams by Soil Bipreactors. Env. Progress, Vol. 10, No. 4, ppSOO-
306. November 1991.
Nothrup Services Inc. Fabric Filter Workshop Reference Materials. Air Pollution Training Institute. RTP,NC. 1977.
Oglesby, S. and G.B. Nichols. A Manual of Electrostatic Precipitator Technology. Prepared by the Southern Research
Institute for the National Air Pollution Control Admin. APTD-0610 (NTIS PB-196380). 1970.
Pan, C.Y. and H.W. Habgood. An Analysis of the Single Stage Gaseous Permeation Process.. Ind. Eng. Chem. Fundam., Vol.
13,pp323-331. 1974.
Pederson, T. and J. Curtis. Soil Vapor Extraction. Technology: Reference Handbook. EPA/540/2-91/003 (NTIS PB91 -
168476). 1991.
139
-------
Separatl°n °f °rganic Vap°rs from Ain AICHE SymP- Ser- Vol. 82, No.
Purus.Inc. On-Site Organic Contaminant Destruction with Advanced Ultraviolet Flashlamps. 1991.
Roy K.A Hazmat World, Vol. 3, No. 5. From a series on UV-oxidation technologies. Tower-Bonier Publishing, Glen
miyn, UL. lyyo.
Singh, S.P. and R.M. Counce. Removal of Volatile Organic Compounds From Groundwater: A Survey of Technologies
Oak Ridge National Laboratory. (NTIS DE89-015653). May 1989.
S^SS ™™ ve"i *?n M ^r1^ D^eelo?ment of a Synthetic Membrane for Gas and Vapor Separation. Pure and
Applied Chem., Vol. 60, No. 12. Blackwell Scientific Publications, Ltd. Oxford, England. 1986.
U.S.EPA. Air Stripping of Contaminated Sources: Air Emissions and Controls. EPA/540/3-87/017. 1987.
U.S.EPA. Soil Vapor Extraction VOC Control Technology Assessment. EPA 450/4-89-017 (NTIS PB90-216995) Re
search Triangle Park, NC. 1989. , -^o^;. K.B
U.S. EPA. OAQPS Control Cost Manual - 4th Edition. EPA 450/3-90-006. January 1990.
U.S.EPA Handbook: Control Technologies for Hazardous Air Pollutants. EPA/625/6-91/014. Cincinnati, OH. June 1991.
nnnvM **?*' ^^ ""* COn^1 Efflciendes °f Thermal and Catalytic Incineration for the
Control of Volatile Organic Compounds. JAWMA, Vol. 41 , No. 4. Pittsburgh, PA. 1 991a.
nnrnvnr ™ W<*e> ^ COSt Estimation of T**** ™« Catalytic Incinerators for the
Control of VOCs. JA^VMA, Vol. 41 , No. 4, pp497-501 . 1 99 1 b.
Vatavuk, W. Estimating Costs 'of Air Pollution Control. Lewis Publishers. Chelsea, MI. 1 990.
Vatavuk, W. and R. Neverfl. Estimating Costs of Air Pollution Control Systems, Part XI: Estimate the Size and Cost of
Baghouses. Chem. Eng.,ppl53-158. March 22, 1982.
Walas, S.M. Chemical Process Equipment Selection and Design. Butterworth Publishers, Boston, MA. 1988.
Vtv ^f Cnt °f Contaminated Water' Air> ^d Soil with UV Flashlamps. Env. Progress, Vol. 1 0, No. 4. New
\« 1991.
Area Sources
A1^'R^:A; ??!!,' "S* f • Pe^^^Sing Foam t0 Maintain Air Quality DurinS ^mediation of Hazardous Waste
Sites. Presented at the 80th Annual AWMA Meeting (Paper 87-18.3), New York City, June 21-26, 1987.
Cooper, CD. and F.C. Alley. Air Pollution Control: A Design Approach. Waveland Press. Prospect Heights, IL. 1990.
C°A ' CAi' ASAatSn a"d C°ntro1 °f Air Contaminants Du™g Hazardous Waste Site Remediation. Presented at the 80th
Annual AWMA Meeting (Paper 87-18.1), New York City, June 21-26, 1987.
NIOSH. The Industrial Environment - its Evaluation and Control. HSM-99-71-45. Washington, DC. 1973.
Radian Corp. Air Quality Engineering Manual for Hazardous Waste Site Mitigation Activities - Revision #2. NJ Depart-
ment of Environmental Protection. May 1989.
> Preli™na? Assessment of Potential Organic Emissions from Dredging Operations. EPA Contract No.63-02-
4288. Report to Dennis Timberlake, U.S. EPA, Cincinnati, OH. September 1991 .
140
-------
Schmidt, C.E. and R. Stephens. Case Study: Control and Monitoring of Air Contaminants During Site Mitigation. Presented
at the 80th Annual AWMA Meeting (Paper 87-18.2), New York City, June 21-26,1987.
Shen, T.T., T.P. Nelson, and C.E. Schmidt. Assessment and Control of VOC Emissions from Waste Disposal Facilities. In:
Air Pollution Engineering Manual - Buonicore and Dvis Eds. AWMA, Pittsburg, PA. 1992.
Shen, T.T. and G.H. Sewell. Control of VOC Emissions From Waste Management Facilities. J. Env. Eng., Vol. 114, No. 6,
pp!392-1400. 1988.
Springer, C., K.T. Valsaraj, and L J. Thibodeaux, In-Situ Methods to Control Emissions from Surface Impoundments and
Landfills/U.S. EPA Project Summary EPA/600/S2-85/124. July 1986.
Springer, C., K.T. Valsaraj, and L.J. Thibodeaux. In-Situ Methods to Control Emissions from Surface Impoundments and
Landfills.'JAPCA Vol. 36,No. 12,ppl371-1374. December 1986.
Todd, Q.R., W. Beers, W. Celenza, and P. Puglionesi. Dust and Vapor Suppression Technologies for Use During the
Excavation of Contaminated Soils, Sludges, or Sediments. In: Proceedings of the 14th Annual Hazardous Waste Research
Symposium. EPA/600/9-88/021, pp53-64. July 1988. . •
U.S. EPA. Dust Control at Hazardous Waste Sites. EPA/540/2-85/003. November 1985,
U.S. EPA. Handbook For Using Foams to Control Vapors From Hazardous Spills. EPA/600/8-86/019 (NTIS PB87-
145660). July 1986. " .
U.S. EPA. Field Evaluation of Windscreens as a Fugitive Dust Control Measure for Material Storage Piles. EPA/600/7-86/
027. July 1986.
U.S. EPA. User's Guide - Emission Control Technologies and Emission Factors for Unpaved Road Fugitive Emissions.
EPA/625/5-87/022. September 1987.
U.S. EPA. Method for Estimating Fugitive Paniculate Emissions from Hazardous Waste Sites. EPA/600/2-87/066 (NTIS
PB87-232203). Cincinnati, OH. August 1987. . . . .
U.S. EPA. Evaluation of the Effectiveness of Chemical Dust Suppressants on Unpaved Roads. EPA/600/2-87/112. Re-
search Triangle Park, NC. November 1987. , ' ,' ,
U.S. EPA. Control of Open Fugitive Dust Sources. EPA-450/3-88-008. U.S. EPA-OAQPS, Research Triangle Park, NC.
September 1988.
U.S. EPA. Land Disposal, Remediation Action, Incineration, and Treatment of Hazardous Waste - Proceedings of the
Fourteenth Annual Research Symposium. EPA/600/9-88/021. July 1988.
U.S. EPA. Hazardous Waste TSDF - Fugitive Paniculate Matter Air Emissions Guidance Document. EPA 450/3-89-019
(NTIS PB90-103250). Research Triangle Park, NC. May 1989.
U.S. EPA. Engineering Bulletin - Control of Air Emissions from Material Handling. February 1991.
Vatavuk.W. Estimating Costs of Air Pollution Control Equipment. Lewis Publishers, Chelsea, MI. 1990.
Vogel, G.A. Air Emission Control at Hazardous Waste Management Facilities. JAPCA Vol. 35, No. 5, pp558-566. May
1985.
Wetherold, R.G., B.M. Eklund, and T.P. Nelson. A Case Study of Direct Control of Emissions from a Surface Impoundment.
Presented at the 11 th Annual EPA Symposium on Land Disposal, Remedial Action, Incineration and Treatment of Hazard-
ous Waste. EPA/600/9-85/013. April 1985. ..'.-"
141
-------
Overviews of Soil Cleanup Techniques
lnTl ' T' DUlan%'n ^ A- IngliS' ^ Emissions From the Treatment of Soil Contaminated with Petro-
leum Fuels and Other iSubstances. EPA-600/R-92-124. July 1992.
' EmiS$i°n FaCt°r$ ** Superftlnd Remediation Technologies. EPA-450/ 1-9 1-001.
U.S.EPA. Underground Storage Tank Corrective Action Technologies. EPA/625/6-87-015. January 1987.
U.S EPA. Air/Superfund National Technical Guidance Study Series, Volume III: Estimation of Air Emissions from
Cleanup Activities at Superfund Sites. EPA-450/1-89-003. (NTISPB89- 18006 I/AS). January 1989.
U.S.EPA. Handbook on In-Situ Treatment of Hazardous Waste-Contaminated Soils. EPA/540-2-90/002. January 1990.
U.S.EPA. Summary of Treatment Technology Effectiveness for Contaminated Soil. EPA/540/2-89/053. June 1990.
Cost of Remediation
h f t> r ' ST' and A' P^™- Costs of Remedial Acti<™ at Uncontrolled Hazardous Waste Sites: Worker
Health and Safety Considerations. U.S. EPA Project Summary EPA/600/S2-86/037. September 1 986.
Smit, Sr., K., Ed. Means Site Work Cost Data. R.J. Grant. 1 991 .
U.S.EPA. Remedial Action Costing Procedures Manual. EPA/600/S8-87/049. 1987.
U.S.EPA. Compendium of Costs of Remedial Technologies at Hazardous Waste Sites. EPA/600/S2-87/089. 1987.
U.S. EPA. OAQPS Control Cost Manual (Fourth Edition). EPA 450/3-90-006 (NTIS PB90-169954). January 1990.
U.S.EPA. Handbook: Control Technologies for Hazardous Air Pollutants. EPA/625/6-91/014. Cincinnati, OH. June 1991.
Vatavuk, W.M., Estimating Costs of Air Pollution Control. Lewis Publishers. Chelsea, MI. 1 990
Selected Key References for Various Remediation Technologies
Excavation
' Hendlen Estimation of Air Impacts For the Excavation of Contaminated Soil. EPA 450/1-92-
Eklund B., D. Ranum, and A. Hendler. Field Measurements of VOC Emissions from Soils Handling Operations at Super-
fund Sites. EPA Contract No. 68-02-4392, WA64. Report to James Durham, U.S. EPA, Research Trfangle Park NC
oeptember 14, 1990.
Thermal Desorption
Troxlcr, W.L., J.J. Cudahy, R.P. Zink, and S.I. Rosenthal. Thermal Desorption Guidance Document for Treating Petroleum
Contaminated Soils. EPA ContractNo. 68-C9-0033. Report to James Yezzi, U.S. EPA, Edison, NJ. January 1992.
de Percin PR. Thermal Desorption Attainable Remediation Levels. In: Proceedings of the 17th Annual Hazardous Waste
Research Symposium. EPA/600/9-91/002, pp51 1-520. April. 1991.
142
-------
U.S. EPA. Engineering Bulletin- Low-Temperature Thermal Desorption Treatment. EPA/540/0-00/000. December 1990.
Incineration
Oppelt, E.T. Incineration of Hazardous Waste - A Critical Review. JAPCA, Vol. 37, No. 5, pp558-586. May 1987.
U.S. EPA. Engineering Bulletin - Mobile/Transportable Incineration Treatment. EPA/540/2-90/014. September 1990.
IT Corp. Screening Procedures For Estimating the Air Impacts of Incineration at Superfurid Sites: EPA-450/1 -92-003 (NTIS
PB92-171917). February 1992.
U.S. EPA. Guidance on Metals and Hydrogen Chloride Controls for Hazardous Waste Incinerators. Vol. IV of the Hazard-
ous Waste Incineration Guidance Series. August 1989.
Brna, T.G. and C.B. Sedman. Waste Incineration and Emission Control Technologies. EPA/600/D-87/147. May 1987.
••'•', '(''•"'•.
U.S..EPA. Performance Evaluation of Full-Scale Hazardous Waste Incineration. Five Volumes, (NTIS PB85-129500).
November 1984.
Soil Vapor Extraction
U1S. EPA. Handbook of Soil Vapor Extraction (SVE). EPA/540/2-91/003. 1991.
Hutzler, N.J., B.E. Murphy, and John S. Gierke. State of Technology Review ~ Soil Vapor Extraction Systems. EPA-600/2-
89/024 (NTIS PB89-195184). June 1989.
Johnson, P.C., M.W. Kemblowski, J.D. Colthart, D.L. Byers, and C.C. Stanley. 'A Practical Approach to the Design, Opera-
tion, and Monitoring of In-Situ Soil-Venting Systems. Ground Water Monitoring Review, Spring 1990.
U.S. EPA. Demonstration Bulletin for In-Situ Steam/Hot-Air Soil Stripping. EPA/540/M5-90/003. February 1990.
Biodegradation
U.S. EPA. Engineering Bulletin-Sluny Biodegradation. EPA/540/2-90/016. September 1990.
Nelson, M.J., J.V. Kinsella, and T. Montoya. In-Situ Biodegradation of TCE-Contaminated Groundwater. Env. Progress,
Vol. 9, No. 3, pp!90-196. August 1990.
Air Stripping
Vancit, M.A., et al. Air Stripping of Contaminated Water Sources - Air Emissions and Controls. EPA-450/3-87-017 (NTIS
PB88-106166). August 1987.
U.S. EPA. Air/Superfund National Technical Guidance Study Series: Air Stripper Design Manual. EPA-450/1-90-003.
May 1990.
U.S. EPA. Air/Superfund National Technical Guidance Study Series: Comparisons of Air Stripper Simulations and Field
Data. EPA-450/1-90-002. March 1990. ,
U.S. EPA. Selection Guide For Volatilization Technologies for Water Treatment. EPA/600/2-88/014 (NTIS PB88-165683).
January 1988.
U.S. EPA. Performance of Air Stripping and GAG for SOC and VOC Removal From Groundwater. EPA/600/14 (NTIS
PB89 110274). September 1988. .
143
-------
McKinnon, RJ. and J.E. Dyksen. Removing Organics From Groundwater Through Aeration Plus GAC J of AWWA
pp42-47. May 1984.
Stabilization/SolidiGcation
Ponder, T.C. and D. Schmitt. Field Assessment of Air Emissions From Hazardous Waste Stabilization Operations. In-
Proceedings of the 17th Annual Hazardous Waste Research Symposium. EPA/600/9-91/002. April. 1991.
U.S. EPA. Handbook for Stabilization/Solidification of Hazardous Wastes. EPA-540/2-86/001. U.S. EPA, Cincinnati, OH.
June 1986.
Wcitzman, L., L. Hamel, P. dePercin, B. Blaney. Volatile Emissions from Stabilized Waste. In: Proceedings of the Fifteenth
Annual Research Symposium. EPA-600/9-90/006. February 1990.
Solvent Extraction/Soil Washing
Hall, D.W., J.A. Sandrin, and R.E. McBride. An Overview of Solvent Extraction Treatment Technologies. Env Progress
Vol. 9, No. 2, pp98-105. May 1990. ' • '
U.S.EPA. Engineering Bulletin - Soil Washing Treatment. EPA/540/2-90/017. September 1990.
Miscellaneous
U.S.EPA. AP-42: Compilation of Air Pollutant Emission Factors, Fourth Edition. U.S. EPA, Office of Air Quality Plan-
ning and Standards, Research Triangle Park, NC. September 1985.
144
-------
Appendix D
Categorization of Commonly Encountered Compounds
Class of Compounds
Example Compounds
Aromatic Hydrocarbons
Aliphatic Hydrocarbons
Halogenated Hydrocarbons
Ketones/Aldehydes
Other Oxygenated Hydrocarbons
Inorganic Gases
Benzene
Toluene
Xylenes
Ethylbenzene
Hexane
Heptane
Methylene chloride
Chloroform
Carbon tetrachloride
1,1 -dichloroethane
Trichloroethylene
1,1,1 -trichloroethane
Tetrachloroethylene
Chlorobenzene
Acetone
Methyl ethyl ketone
Methyl isobutyl ketone
Cyclohexanone
Formaldehyde
Acetaldehyde
Methanol
Ethylene glycol
Cellulose
Ethers
Phenols
Epoxides
Hydrogen sulfide
Hydrogen chloride
Sulfur dioxide
Nitrogen oxide
Nitrogen dioxide
145
-------
Class of Compounds
Example Compounds
Metals
Polynuclear Aromatics
Pesticides/Herbicides
Miscellaneous
Mercury
Lead
Chromium
Arsenic
Cadmium
Zinc
Beryllium
Copper
Polychlorinated biphenyls (PCBs)
Benzo(a)pyrene
Naphthalene
Anthracene
Chrysene
Chlordane
Lindane
Parathion
Asbestos
Cyanides
Radionuclides
Potential Air Contaminants by Generic Type of Contaminant
Volatiles (>1 mm Mercury vapor pressure at 25°C)
• AH monochlorinated solvents; also trichloroethylene, trichloroethane, tetrachloroethane.
• Most simple aromatic solvents: e.g., benzene, xylene, toluene, and ethylbenzene.
• Most alkanes up to decane (C10).
• Inorganic gases: e.g., hydrogen sulfide, chlorine, and sulfur dioxide.
Semi-Vofatiles (l-10'7mm Mercury vapor pressure at25°C)
• Most polychlorinated biphenyls, dichlorobenzenes, aniline, nitroaniline, and phthalates.
• Most pesticides: e.g., dieldrin, toxaphene, and parathion.
• Most complex alkanes: dodecane and octadecane.
• Most polynuclear aromatic's: e.g., naphthalene, phenanthrene, and benz(a)anthracene.
• Mercury
Non-Volatttes or Particulate Matter (<10'7 mm Mercury vapor pressure at25°C)
• Larger polynuclear aromatics: e.g., chrysene.
• Metals: e.g., lead and chromium.
• Other inorganics: e.g., asbestos, arsenic, and cyanides.
4U.S. GOVERNMENT PRINTING OFFICE: 1993 -yso-ooa' 60188 •
146
-------