-------
5.4.2 Subsurface Drains
5.4.2.1 General Description
Subsurface drains include any type of buried conduit
used to convey and collect aqueous discharges by
gravity flow. Subsurface drains essentially function as
an infinite line of extraction wells. They create a
continuous zone of influence in which ground water
flows toward the drain. Subsurface drainage com-
ponents are illustrated in Figure 5-20.
Figure 5-20.
Collector
Subsurface drainage system
components (EPA 1985a).
Discharge
Collection
Sump
The following are the major components of a
subsurface drainage system:
Drain pipe or gravel bed~for conveying flow to a
storage tank or wet well. Pipe drains are used
most frequently at hazardous waste sites. The
use of gravel beds or french drains and tile
drains is more limited.
Envelope-for conveying flow from the aquifer to
the drain pipe or bed
Filter-for preventing fine particles from clogging
the system, if necessary
Backfill-to bring the drain to grade and prevent
ponding
Manholes or wet wells~to collect flow and pump
the discharge to a treatment plant.
5.4.2.2 Application/Availability
Because drains essentially function like an infinite line
of extraction wells, they can perform many of the same
functions as wells. For example, they can be used to
contain or remove a plume or to lower the ground-
water table to prevent water from contacting the
released substance. Should a release occur, they also
may be installed as a control measure. The decision to
use drains or pumping is generally based on a cost-
effectiveness analysis. The application of subsurface
drains is often greatly restricted for response to
gasoline releases at a corner station because of the
numerous utilities in the immediate; area. Further-
more, the use of an open trench when lighter fraction
petroleums and ignitable fluids are present as free
product can be dangerous and should be undertaken
only with appropriate safety precautions. A variation of
subsurface drains (an open trench with a free-product
recovery system) is often used, however.
For shallow contamination problems, drains can be
more cost-effective than pumping, particularly in strata
with low or variable hydraulic conductivity. Under
these conditions, a pumping system would be difficult
to design and be cost-prohibitive to operate for the
maintenance of a continuous hydraulic boundary.
Subsurface drains may also be preferred over pump-
ing where ground-water removal is required over a
period of several years, as the operation and main-
tenance costs associated with pumping are sub-
stantially higher.
One of the biggest drawbacks to the use of
subsurface drains is that they are generally limited to
shallow depths. Although it is technically feasible to
excavate a trench to almost any depth, the costs of
shoring, dewatering, and hard rock excavation can
make drains cost-prohibitive at depths greater than 40
feet. In stable, low-permeability soils, however, where
little or no rock excavation is required, drains can be
cost-effective to depths of 100 feet.
The most widespread use of subsurface drains at LIST
sites is for the purpose of intercepting a plume
hydraulically downgradient from its source (Figure 5-
21 a). These interceptor drains, as they are commonly
called, are frequently used together with a barrier wall
(Figure 5-21 b) for two reasons. First, when a
subsurface drain is to be placed just upgradient of a
stream, the drainage system would, reverse the flow
direction of the stream and cause a prohibitively large
volume of clean water to be collected. The addition of
a barrier wall would prevent infiltration of clean water
from the stream and thus reduce treatment costs.
Second, subsurface drains are used in conjunction
with ground-water cutoff barriers: to prevent the
buildup of ground water upgradient of the barrier.
Evaluation of the suitability of subsurface drains as a
corrective technology is generally made by comparing
the cost-effectiveness of this alternative versus pump-
ing. Subsurface drains can be more difficult and costly
to install, particularly if extensive hard rock excavation,
subsurface utilities, and dewatering are required.
Safety of field workers is also more pf a concern with
subsurface drains because of the need for extensive
trench excavation, potential trench collapse, and gas
buildup.
5-34
-------
Figure 5-21. Use of a one-sided subsurface drain
for reducing flow from uncontaminated
sources (JRB Associates 1985).
Underground
Tank
Conventional
Subsurface
Drain
Original
Water Table
Clean Water
Recharging
from Stream
Low
Permeability
(a) Conventional Subsurface Drain
Underground
Tank
Subsurface Drain
. With Clay or
'Plastic Barrier
Low
Permeability
(b) One-Sided Subsurface Drain.
Drains have several advantages over pumping,
however. They are generally more cost-effective in
areas with low hydraulic conductivity, particularly when
pumping would be required for an extended period of
time. They are also easier to operate because water is
collected by gravity flow, and they are more reliable
because there are no electrical components to fail.
When drains fail as a result of clogging, breaks in the
pipes, or sinkhole formation, however, rehabilitation
can be costly and time-consuming.
5.4.2.3 Design and Construction
Considerations
The major elements to consider in designing a
subsurface drainage system include:
Location and spacing of drains to achieve
desired head levels.
Hydraulic design of the conduit, including pipe
diameter and gradient.
Properties and design of the envelope and filter
materials.
Design of a pumping station.
Each of these design elements is discussed in the
following subsections.
Location and Spacing of Drains
For design purposes, subsurface drains have been
divided into two categories based on their functions:
interceptor and relief drains. Interceptor drains are
installed perpendicular to ground-water flow and are
used to intercept ground water from an upgradient
source. These drains are the more commonly applied
to abate UST releases. Relief drains are installed
parallel to the direction of flow or around the perimeter
of a site where the water table is relatively flat. Relief
drains are used primarily to lower the water table
beneath a site. Figure 5-22 shows how interceptor
and relief drains alter the configuration of the water
table.
Determining the required location for an interceptor
drain is more often based exclusively on the use of
field data than on theoretical design. Site inves-
tigation data are used to develop potentiometric
surface maps, hydraulic conductivity data, plume
boundary limits, and geologic cross sections. With
these data in hand, the design engineer can pinpoint
and stake the design drain line.
To function properly, an interceptor drain should be
installed perpendicular to ground-water flow direction.
In stratified soils having greatly different hydraulic
conductivities, the drain should rest on a layer of low
hydraulic conductivity. If the trench is cut through an
impervious stratum, a significant percentage of the
product moving laterally could bridge over the drain
and continue downgradient. Similarly, if soil layers or
pockets with high hydraulic conductivity underly the
drain, the substance could flow beneath the drain.
Underflow can be minimized by placing impermeable
liner material at the base of the trench before laying a
thick (1- to 3-foot) gravel bedding. Where pockets of
highly permeable soils are found (e.g., in the scour
channel of an alluvial area), a manhole could be
constructed at the lowest point of the permeable soil
and a small lift station and force main could be installed
to carry the ground water from this low area back up to
the adjacent gravity flow section of the drainage
system (see Figure 5-23). A third solution is to install a
barrier wall downgradient of the drain and to key it in to
a low-permeability layer.
Filters and Envelopes
Filters and envelopes are often used in the
construction of drain systems. The primary function of
a filter is to prevent soil particles from entering and
clogging the drain. Filters should always be used
where soils have a high percentage of fines. Filter
fabrics may be clogged by highly viscous fluids.
5-l35
-------
Figure 5-22. The effect of relief and Interceptor drains In altering the configuration
of the water table (U.S. EPA 1985d).
Relief Drains
Water,
Table
Ground-Water Flow f. f. f
Water
Table
I = Hydraulic gradient
de = Effective depth of the drain
Le = Downslope influence of the drain
L *t Drain spacing
H ~ Hydraulic head
D = Depth bebw drains to the
impermeable barrier
Figure 5-23. Subsurface drain with a lift station (Giddlngs 1982).
Main Lift Station Manhole
And Pumi
Supplementary Manhole
And Lift Pump
Scour Channel
(View Looking Toward Site)
5-36
-------
The function of an envelope is to improve water flow
and to reduce flow velocity into the drains by providing
a material that is more permeable than the surrounding
soil. Envelopes also may be used to provide suitable
bedding for a drain and to stabilize the soil material on
which the drain is being placed. Envelopes are
required for most applications.
Although filters and envelopes have distinctly
different functions, well-graded sands and gravels can
be used to meet the requirements of both a filter and
an envelope. Geotextiles are also widely used as
filters. These are generally made of polypropylene,
polyethylene, polyester, or polyvinyl chloride. Selec-
tion of filter fabric should be based on its compatibility
with the contaminants.
Manholes are used in subsurface drainage systems to
serve as junction boxes between drains, silt and sand
traps, observation wells, and access points for pipe
location, inspection, and maintenance. Manholes
should be located at junction points, changes in
alignment or grade, and other designated points. No
criteria have been established for manhole spacing.
Contaminated ground water or free product is
collected by gravity flow in a drainage sump, from
which it is pumped to treatment or recovery systems
(Figure 5-24). The major steps in designing the sump
and pumping system include (Bureau of Reclamation
1978):
Determining the maximum inflow to the surnp.
Determining the amount of storage required.
Determining the pumping rate.
Qm = (Sv + QPtp)/tp
where
Qm = pumping rate, ftVmin
Sv = storage volume, ft3
Qp = maximum inflow, ftVmin
tp = running time of the pump, min
Determining the start, stop, and discharge
levels.
Determining the size of the sump.
Selecting the pump.
Construction Considerations
The major activities associated with construction of
subsurface drains are trench excavation, trench
stabilization, and installation of the drains, the filter,
and the envelope materials.
Trench excavation is critical in the determination of the
cost-effectiveness of drains. The need for extensive
rock fragmentation and relocation of subsurface
utilities may preclude the use of drains as a cost-
Figure 5-24. Typical design of an automatic drainage pumping plant (Bureau of Reclamation 1978).
, Meter Shelter^
Door
Stop Collar
Pipe Collector
El. 1296.0
Stop Level
Concrete Base
Pipe Collector
Plug
Stilling
Chamber
5-37
-------
effective correction action. Trenches are usually
excavated by trenching machines or backhoes.
Cranes, clamshells, and draglines are also used for
deep excavation. The factors that influence the rate
of trenching include 1) soil moisture; 2) soil charac-
teristics such as hardness, stickiness, and stones; and
3) the depth and width of the trench.
Trenchers used for continuous trenching can be
equipped with back-end modifications to provide
shoring, to install a geotextile envelope, to lay tile or
flexible piping, to blind the piping, and to backfill with
gravel or excavated soil. Backhoes can excavate earth
and fragmented rock for a trench the width of up to
one-half of the bucket diameter and to depths of up to
70 to 90 feet.
The crane and clamshell can be used for deeper
excavations or when access excludes the use of a
backhoe. The use of draglines is generally limited to
removal of loose rock and earth.
Excavation of a trench through material containing
numerous large boulders or hard rock layers results in
considerable construction delays and substantially
increases the cost of construction. Typically, these
materials must be fractured to facilitate their removal.
Proper grade control in a subsurface drain ensures
against ponding of water and provides for a nonsilting
velocity in the drainage pipe. Proper grade control
can be accomplished either by automatic laser or by
visual grade-control systems. Laser systems are
adaptable to a wide range of earth-moving equipment,
including trenchers and backhoes.
Proper installation of drains (maintenance of grades,
placement and alignment of pipes) generally requires
dewatering to produce a dry environment. Three basic
options are available for dewatering: open pumping,
predrainage with wellpoints or well systems, and
ground-water cutoff. These techniques may be used
separately or in combination.
Once trench excavation is completed, the com-
ponents of the subsurface drain can be installed. This
process includes laying the pipes, filter, and envelope
material; backfilling; and installing auxiliary com-
ponents.
All subsurface drains must be laid on a stable bed with
the desired grade. Trenches that have inadvertently
been overexcavated should be refilled with dry soil
and brought to grade with envelope material. Well-
graded gravel is then laid in an even layer several
inches thick to provide bedding for the pipes.
Gravel envelopes are installed around the pipe drain
to Increase flow into the drain and to reduce the
buildup of sediments in the drain line. They may be
placed by hand, backhoe, or hopper cart or truck. In
continuous trencher drain installation machines,
gravel filling may be ongoing along with other
operations. Filter fabrics are sometimes installed
around the gravel envelope to prevent fines from
clogging the envelope and drain pipe.
After the gravel envelope has been installed, the
trench must be backfilled to the original grade. Prior to
backfilling, the drain should be inspected for proper
elevation below ground surface, proper grade and
alignment, broken pipe, and thickness of the gravel
envelope. The inspector should ensure that pipe
drains and manholes are free of deposits of mud,
sand, and gravel, or other foreign matter, and that they
are in good working condition. Unstable soils may
preclude all but spot checks before backfilling.
When installation of the subsurface drain is complete,
the drain should be tested for obstructions. For a
small drainage system, this can be done visually by
shining a high-powered flashlight through a drain from
one manhole and observing the beam in another.
Television camera inspections can be used for large-
diameter drains. Mechanical methods can be used to
remove obstructions and to test for obstructions.
Flexible polyurethane foam plugs are available that
expand to wipe or scrape the pipe when water or air
pressure is applied. They are also available with a rope
through the center so they can be pulled through the
drain (Knapp, Inc. 1982).
Manholes and silt traps should be checked frequently
for the first year or two of operation for sediment
buildup. Less frequent inspection is required as the
system ages.
Piezometers may be installed in the various parts of
the drainage system to identify operational problems
with the filter, envelope, pipe, or other components of
the system. Piezometers can measure the loss of
head through a medium and ithereby identify
obstructions to flow, such as a clogged envelope or
filter.
Malfunction of subsurface drains can be attributed to
chemical clogging, clogging due to biological slimes,
or a variety of physical mechanisms. Problems caused
by these conditions are usually apparent at the
surface above the drain. Inspection of the area will
reveal soft or ponded surface conditions, areas of
subsidence, and areas of accelerated vegetative
growth. Chemical clogging of pipes and envelope
materials can occur by a number of mechanisms.
Calcium carbonate precipitates and iron and man-
ganese deposits can build up around collector pipes
or can cause cementation of the envelope material.
5.4.2.4 Costs
Costs for installation and operation of subsurface
drains can be divided into four categories: installation
costs, materials costs, engineering: supervision, and
operation and maintenance.
5-38
-------
Installation costs depend primarily on the depth of
excavation, stability of soils, extent of rock
fragmentation required, and ground-water flow rates.
The principal materials costs are for pipes, gravel,
manholes, pumps, and other accessories for the
drainage sump. Combined materials and installation
unit costs are summarized in Tables 5-14 through 5-
16.
Engineering and supervision involve such activities as
staking the drain line, checking for grade control and
alignment, and checking pipe specification and pipe
quality. For the installation of subsurface drains in
conventional agricultural and water conservation ap-
plications, engineering and supervision costs usually
run about 5 to 10 percent of the total cost. These
costs can be expected to be substantially higher for
UST site applications, however, and will vary con-
siderably depending on the geologic and hydro-
geologic conditions.
Capital costs associated with the installation of
subsurface drains are typically much higher than those
associated with pumping systems. This is particularly
true where substantial rock excavation is required and
where deep drains requiring extensive shoring are
needed. These factors may preclude consideration of
drains as a cost-effective remedial action; however,
operation and maintenance costs associated with
drains are generally lower than those associated with
pumping if the system is properly designed and
maintained. Lower operation and maintenance costs
become significant, particularly when plume removal
or containment is needed over a long period of time.
As with other remedial technologies, total capital costs
for drainage systems can vary widely with site
conditions. Two scenarios are briefly described here
to illustrate how widely capital costs may range.
At one particular hazardous waste site (Site A), a 261-
foot-long interceptor drain was installed to a depth of
12 to 17 feet. The leachate discharged into a 4-foot-
wide, 20-foot-deep sump that pumped the leachate at
a rate of 18 to 20 gal/min to a treatment system (EPA
1984; JRB Associates 1985). Construction of the
drainage system involved excavation of a 4- to 6-foot-
wide trench, which was supported with steel sheet
piling during construction. The trench was lined with
filter fabric, 6 inches of gravel, and a 12-inch,
perforated, concrete asbestos drain pipe. Additional
filter fabric supported by screening was then wrapped
around the pipe prior to backfilling. The total cost of
the drainage system was $269,721 or $67 to $88/ft2
(adjusted to 1986 dollars) (EPA 1984; JRB Associates
1985). Table 5-17 shows how the total capital costs
were distributed.
A second case history involved installation of a shallow
(3-foot-deep) interceptor trench at the A. W. Mauthe
site in Appleton, Wisconsin (EPA 1984). The drain-
age system was approximately 750 feet long and
consisted of 4-inch PVC drainage pipe laid in a gravel-
filled trench. Four-foot-diameter concrete sumps
were installed at two collection points and connected
by about 25 feet of PVC pipe so that the water
collected in one sump could be pumped to the other
sump. The total capital cost of the drainage system
was about $15,800 (updated to 1986 costs using
ENR Construction Cost Indices). This cost estimate,
however, also includes the cost of a 300-foot-long
surface-water-diversion drainpipe. Therefore, the unit
cost of the subsurface drain was less than $7.05/ft2
(EPA 1984).
These two case histories show an order of magnitude
difference in the unit cost for subsurface drainage
systems. Probably the most significant factor
contributing to these cost differences is the depth of
the drain. In the case of Site A, the trench was
excavated to a depth of 12 to 17 feet and shoring was
required to support the trench wall. In the case of the
Mauthe site, the subsurface drainage system was only
3 feet deep and trench excavation was greatly
simplified.
5.5 Subsurface Barriers
The term "subsurface barriers" refers to low-
permeability cutoff walls or diversions installed below-
ground to contain, capture, or redirect ground-water
flow in the vicinity of a site. The most commonly used
subsurface barriers are slurry walls, particularly soil-
bentonite slurry walls. Less common are cement-
bentonite or concrete (diaphragm) slurry walls,
grouted barriers, and sheet piling cutoffs. For shorter
periods of time (6 months or less), barriers can also be
created by artificial freezing techniques. Another
containment system involves the use of injection wells
to form hydraulic barriers that both contain a plume
and facilitate product recovery. These types of
subsurface barriers are discussed in the following
subsections.
5.5.1 Slurry Walts (EPA I985a)
5.5.1.1 General Description
Slurry walls are the most common subsurface barriers
because they provide a relatively inexpensive means
of greatly reducing ground-water flow in
Linconsolidated earth materials. All slurry walls have
one thing in common: they are all constructed in a
vertical trench excavated under a slurry (Figure 5-25).
The slurry, usually a mixture of bentonite and water,
acts essentially as a drilling fluid. It hydraulically shores
the trench to prevent collapse and simultaneously
forms a filter cake on the trench walls to prevent heavy
fluid losses into the surrounding ground. The
different types of slurry walls are differentiated by the
materials used to backfill the trench. Usually, an
engineered soil mixture is blended with the bentonite
5-39
-------
Table 5-14. 1986 Unit Costs for Trench Excavation and Associated Activities
Item Assumptions
Unit Cost ($)
Source*
Trench excavation
Trencher, ladder type
Backhoe, hydraulic
Dragline
Clamshell
Rock fragmentation
Jackhammer
Track-mounted air drill
Wall stabilization
Sheet piling
Wooden shoring
H-piles
Dowatering
Sump hole
Opening pumping
Gas, 5 ft deep, 8 in. wide
Diesel, 8 ft deep, 16 in. wide
4-ft-wide trench, damp sandy loam
1 yd3 capacity, 12 ft deep, 90 LFt/day
1.25 yd3 capacity. 14 ft deep, 90 LF/day
2.5 yd3 capacity, 18 ft deep, 1 15 LF/day
3.5 yd3 capacity, 20 ft deep, 136 LF/day
1.75yd3 capacity, 35 yd3 /h
1 .5 yd3 capacity, 65 yd3 /h
0.5 yd3 capacity, 20 yd3 /h
1.0 yd3 capacity, 35 yd3 /h
—
Includes pull and salvage:
15 ft excavation, 22 Ib/ft 2
25 ft excavation, 38 Ib/ft 2
40 ft excavation, 38 Ib/ft 2
Includes wales, braces, and spacers;
pull and salvage:
14 ft excavation
20 ft excavation
H-piles with 3-in. wood sheeting,
horizontal between piles; includes
removal of wales and braces:
15 to 22 feet
23 to 35 feet
36 to 45 feet
46 to 57 feet
Includes excavation and gravel:
with 12-in. corrugated pipe
with 15-in. corrugated pipe
with 18-in. corrugated pipe
with 24-in. corrugated pipe
Pumping 8 h, attended 8 h; includes
20 feet of suction hose and 100 feet
of discharge hose:
2-in. diaphragm pump
4-in. diaphragm pump
3-in. centrifugal pump
6-in. centrifugal pump
500/day
670/day
2.26/yd3
2.17/yd3
1.74/yd3
1.44/yd3
3.08/yd3
1.98/yd3
4.96/yd3
3.32/yd3
40/h
76/h
7.34/ft2
8.52/ft2
7.91/ft2
5.96/ft2
6.94/ft2
20 to 23/ft3
22 to 25/ft3
25 to 28/ft3
30to32ffi3
20.23/ft3
26/ft3
29/ft3
39/ft3
365/day
401/day
366/day
442/day
(1)
(D
(2)
(2)
(2)
(2)
d)
(D
(D
(1)
i
(2)
(2)
d)
(1)
d)
(1)
(D
0)
(1)
CD
(3)
i
(D
(D
(D
(D
(D
(D
(D
(D
5-40
-------
Table 5-14. (Continued)
Item
Assumptions
Unit Cost ($)
Source*
Submersible centrifugal Bronze, without installati
sump pump
Diaphragm pump
>n:
1/4 hp, 22 gpm, 1 0 ft head 205/each
1/2 hp, 68 gpm, 1 0 ft head . 334/each
1/2 hp, 94 gpm, 1 0 ft head 442/each
Cast iron, without installs tion:
1/4 hp, 23 gpm, 1 0 ft head 97/each
1/3 hp, 35 gpm, 10 ft head 108/each
1/2 hp, 68 gpm, 10 ft head 226/each
Cast iron starter and leve
witout installation; 2-in. d
10 gpm, 20 ft head
60 gpm, 20 ft head
120 gpm. 20 ft head
160 gpm, 20 ft head
control,
scharge:
313/each
411/each
770/each
1207/each
(1)
\ /
" (1)
d)
(1)
(1)
\ /
(D
(1)
\ /
(1)
\ /
(1)
.- \ •/
(1)
Wellpoint dewatering
Ground-water cutoffs
Grade control
Automatic laser control
See Section 5.1 for costs
See Sheet Piling, above, for costs
144/day
(1)
* Data from (1) Godfrey 1984a; (2) Godfrey 1984b; (3) McMahon 1984, adjusted to mid-1986 dollars
TLF = linear feet
5-41
-------
Table 5-15. 1986 Unit Costs for Pipe Installation
Item
Assumptions
Unit Cost ($)
Source
Drain pipe
PVC perforated underdrain
Corrugated steel or
aluminum, perforated,
asphalt-coated
Porous wall concrete
underdrain, extra
strength
Vitrified clay, extra-
heavy-duty strength,
premium joints
Rlter and envelope
Filter fabric
Gravel envelope
Backfill
Dozer backfill, no
compaction
Dozer backfill, air-
tamped
Compacted backfill,
vibrating roller
Compacted backfill,
sheepsfoot roller
10-foot length,
S.D.R. 35:
4-in.
6-in.
8-in.
10-in.
12-in.
6-in., 18-ga
8-in., 16-ga
10-in., 16-ga
6-in.
8-in.
10-in.
4-in.
5-in.
6-in.
8-in.
Polypropylene
laid in trench
Crushed bank run,
screened
0.75- to 0.50-in.,
in trench
Up to 300-foot haul,
900yd3/day
Up to 300-foot haul,
235yd3/day
6- to 12-in. lifts,
700yd3/day
6- to 12-in. lifts,
650 yd3/day
2.22/ft
3.74/ft
4.68/ft
6.98/ft
8.63/ft
4.76/ft
6.37/ft
8.22/ft
4.25/ft
5.96/ft
8.99/ft
4.58/ft
5.49/ft
6.52/ft
8.73/ft
1.17to1.53/yd2
9.45to10.83/yd3
1.14/yd3
5.60/yd3
1.58/yd3
1.72/yd3
(2) •
(2) •
(2)
(2)
(2)
(1)
(1)
(D
(1)
(1)
(1)
(2)
(2) !
(2)
(2)
(2)
(1) !
(1) '
(D
(D
(1)
' Data from (1) Godfrey 1984a; (2) Godfrey 1984b.
5-42
-------
Table 5-16. 1986 Installed Costs for Manholes*
Item
Concrete slab, cast in place, 8 in. thick
Precast concrete riser pipe,
4-ft inside diameter
6-ft inside diameter
Slab tops, precast, 8 in. thick
Frames and covers, watertight
Table 5-17.
Data from Godfrey 1984a.
Capital Costs for Interceptor Drain Installation*
Item
Materials
550 feet of 12-in.,_
(only about 261 feet were actually used)
perforated asbestos cem 3nt drainage pipe
147 feet of 2-in. carbon steel pipe for carry ng leachate
to treatment system
2 submersible pumps and accessories
2,700 ft2 vinyl-coated wire screen
338 yd2 filter fabric
Other materials costs not given
Subtotal
Labor/Equipment
Labor, equipment rental including excavation equipment
and sheet piling and gravel fill
Company in-house labor
Total
Data from EPA 1984.
M 986 dollars.
5-43
Assumptions
6 ft deep
8 ft deep
12 ft deep
16 ft deep
20 ft deep
6 ft deep
8 ft deep
12 ft deep
16 ft deep
20 ft deep
6 ft deep
8 ft deep
12 ft deep
16 ft deep
20 ft deep
4-ft diameter
5-ft diameter
6-ft diameter
24-in. diameter
32-in. diameter
Cost ($)
915
1310
1965
2625
3280
585
795
1205
1620
2030
1285
1720
2600
3485
4370
80
200
275
355
440
Costt ($)
3,909
1,591
1,069
721
284
459
12,033
64,290
669
76,992
-------
Figure 5-25. Slurry trench construction (Spooner ot al. 1984).
slurry and placed in the trench to form a soil-bentonite
(SB) slurry wall. In some cases, the trench is ex-
cavated under a slurry of portland cement, bentonite,
and water, and this mixture is left in the trench to
harden into a cement-bentonite (CB) slurry wall. In
those rare cases when great strength is required of a
subsurface barrier, precast or case-in-place concrete
panels are constructed in the trench to form a
diaphragm wall.
5.5.1.2 Application/Availability
As shown in Figure 5-26, slurry walls can be placed
(relative to the direction of ground-water flow)
upgradient, downgradient, or completely surrounding
the site of contamination. Circumferential installations,
by far the most common, offer several advantages.
This placement vastly reduces the amount of
uncontaminated ground water entering the site on the
upgradient side. Also, if no compatibility problems
exist between the contamination and the wall
materials, it will reduce the amount of contamination
leaving the downgradient side of the site. Moreover,
the use of this configuration in conjunction with an
infiltration barrier and a collection .system (or other
means of reducing the hydraulic head on the interior
of the wall) can maintain the hydraulic gradient in an
inward direction, which prevents escape of the
contamination.
5.5.1.3 Design and Construction
Considerations
Many factors must be considered in the design of a
slurry wall. First, a detailed, design-phase investi-
gation must be made to characterize subsurface
conditions and materials as well as to address the
disposition and nature of the contamination. The
issue of wall compatibility also should be addressed
early in the design stage by permeability testing of the
proposed backfill mixture with actual site materials.
The design-phase investigation results can be used
to decide on the optimum configuration and to select
any ancillary measures needed to enhance the
5-44
-------
Figure 5-26. Slurry wall placement (Spooner et al. 1984a).
Drain
Ground-
Water
Flow
Drain
Slurry Wall
Upgradient
Extraction Wells
Slurry Wall
Extraction Wells
Circumferential
' Slurry Wall
5-45
-------
performance of the wall. These and other design
considerations are covered in great detail in a report
by the EPA entitled Slurry Trench Construction for
Pollution Migration Control (Spooner et al. 1984).
Construction of a slurry wall is relatively straight-
forward. The required equipment depends on the
depth and length of the wall. For walls up to 80 feet
deep, a backhoe or modified backhoe is used for
excavation. Deeper installations require the use of a
mechanical or hydraulic clamshell or, in rare cases, a
dragline. In small-volume wall installations batch slurry
and backfill-mixing systems may be used, whereas
large jobs require flash slurry mixers and a large backfill
mixing area.
Regardless of the equipment used, the slurry is
introduced just after the trench is opened and before
the water table is reached. The primary function of the
slurry is to act as hydraulic shoring to prevent trench
collapse. There is also evidence to indicate that the
filter cake formed on the trench walls by the slurry
contributes to the low permeability of the completed
wall.
Proper quality control during wall installation is
essential. It is most important that checks be made of
trench continuity and backfill mixing and placement.
For backhoe-excavated trenches, the continuity of
the trench is relatively easy to verify. Inspection of the
excavated material indicates when and where the
confining layer is encountered, and observing the
motion of the backhoe arm will confirm lateral
continuity. With clamshell excavators, confirmation of
lateral continuity may be more complicated.
5.5.1.4 Costs
Costs of slurry walls are highly site-specific. A typical
installation based on a trench 4 ft wide x 40 ft deep x
100 ft long filled with bentonite/water slurry would cost
approximately $310 per linear foot of trench (updated
to 1986 dollars).
5.5.2 Grouting (EPA 1985a)
5.5.2.1 General Description
Grouting refers to a process whereby one of a variety
of fluids is injected into a rock or soil mass, where it is
set in place to reduce water flow and strengthen the
formation. Because they are costly, grouted barriers
are seldom used to contain ground-water flow in
unconsolidated materials around hazardous waste
sites. Slurry walls are less costly and have lower
permeability than grouted barriers. Consequently, in
remediation efforts at contaminated sites, grouting is
best suited for sealing voids in rock. Even when rock
voids are transmitting large water volumes, a grout can
be formulated to set before it is washed out of the
formation.
5.5.2.2 Application/Availability
Cement has probably been used longer than any
other type of material for grouting applications. For
cement grouting, hydraulic cement that sets, hardens,
and does not disintegrate in water is used, Because
of the large particles they contain, cement grouts are
more suitable for rock than for soil applications. The
addition of clay or chemical polymers to the grout,
however, can improve its range of usage. Cement
grouts have been used in both soil consolidation and
water cutoff applications, but their use is normally
restricted to more open soils.
Clays have been widely used as grouts, either alone
or in formulations, because they are inexpensive.
Only certain types of clay minerals; will swell in the
presence of water and form a gel structure at low-
solution concentrations.
Bentonite grouts can be used alone as void sealers in
coarse sands. Bentonite-chemicaJ grouts can be
used on medium to fine sands. Both of these grout
types can also be used to seal small rock fissures;
because of their low gel strengths, however, they
cannot support structures.
Alkali silicates receive the largest and widest use of
the chemical grouts. Sodium, potassium, and lithium
silicates are available, but sodium silicates are used
most often. Silicate grouts are used for both soil
consolidation and void sealing. Unless preceded by
cement grouting, however, they are not suitable for
open fissures or highly permeable materials because
of syneresis (water expulsion).
Organic polymer grouts represent only a small fraction
of the grouts in use. These grouts consist of organic
materials (monomers) that polymerize and cross-link to
form an insoluble gel. The organic polymer grouts
include acrylamide, phenolic, urethane, urea-
formaldehyde, epoxy, and polyester grouts.
The compatibility of these grouts with petroleum and
other chemical substances has not been studied in
great detail; only general incompatibilities are known.
Whenever grouting is considered as a remedial
option, thorough compatibility testing must be
performed.
5.5.2.3 Design and Construction
Considerations
The design of a grout system must be based on a
thorough site characterization. Analysis of site
characterization data, including boring logs, pump or
injection test results, and other data, is used to
determine if a site lends itself to the application of a
grout barrier and which grout is most suitable based
on viscosity, compatibility, and ultimate permeability.
This is a very involved process and should be
conducted by an experienced engineer.
5-46
-------
Construction of a grout barrier is accomplished by
pressure-injecting the grouting material through a
pipe into the strata to be waterproofed. The injection
points are usually arranged in a triple line of primary
and secondary grout holes (Figure 5-27). A
predetermined quantity of grout is pumped into the
primary holes and after it has had time to gel, grout is
injected into the secondary holes. The secondary
grout holes are intended to fill in any gaps left by the
primary grout injection.
5.5.2.4 Costs
Table 5-18 presents approximate costs of some
common grouts.
Table 5-18. Costs of Common Grouts*
Grout Type
Portland cement
Bentonite
Silicate
20 percent
30 percent
40 percent
Epoxy
Acrylamide
Urea-formaldehyde
Approximate Cost
of Solution ($/ga|t)
1.37
1.81
1.81
3.03
3.96
43.29
9.59
8.22
*Data from Spooner et al. 1984b.
T1986 dollars.
Figure 5-27. Semicircular grout curtain (Spooner et al. 1984b)
5.5.3 Sheet Piles (EPA 19853)
5.5.3.1 General Description
Sheet piling can be used to form a ground-water
barrier. Sheet piles can be made of wood, precast
concrete, or steel. Wood is an ineffective water
barrier, however, and concrete is used primarily where
great strength is required. Steel is the most effective
in terms of ground-water cutoff and cost.
5.5.3.2 Application/Availability
Steel sheet piling is seldom used as a ground-water
barrier because costs are high and wall integrity is
unpredictable. It is used more frequently for
temporary dewatering in other construction or as
erosion protection where some other barrier, such as
a slurry wall, intersects flowing surface water.
One of the largest drawbacks of sheet piling, or any
other barrier technology requiring pile driving, is that
rocky soils tend to damage or deflect the piles and
may render this approach ineffective as a ground-
water barrier.
5.5.3.3 Design and Construction
Considerations
The primary design parameters for any barrier are its
dimensions and permeability. Dimensional require-
ments, which are based on site characteristics, are
Semicircular Grout Curtain
\
ry Grout Tubes
Primary Grout Tubes
-------
straightforward. Depth limitations are governed by the
soil material at the site. On the other hand, design
factors for ultimate permeability of the cutoff are more
complicated and must assume some factor to account
for leakage through the interlocking joints.
Typical steel piling shapes and interlocks are
illustrated in Figure 5-28. For construction of a sheet
piling cutoff, the pilings are assembled at their edge
interlocks before they are driven into the ground to
ensure that earth materials and added pressures will
not prevent a good lock between piles. The piles are
then driven a few feet at a time over the entire length
of the wall. This process is repeated until the piles are
all driven to the desired depth.
When first placed in the ground, sheet piling cutoffs
are quite permeable. The edge interlocks, which are
necessarily loose to facilitate placement, allow the
easy passage of water. With time, however, fine soil
particles are washed into the seams and water cutoff is
effected. The time required for this sealing to take
place depends on the rate of ground-water flow and
the texture of the soil involved. In very coarse, sandy
soils, the wall may never seal. In such cases, it is
possible to grout the piling seams, but this is a costly
procedure.
5.5.3.4 Costs
Costs of installed steel sheet piling will vary with
depth, total length, type of pile (coated or uncoated),
and relative ease of installation. Average costs range
from approximately $6.68 per square foot up to
approximately $16.43 per square foot.
5.5.4 Hydraulic Barriers (CONCAWE
1979)
5.5.4.1 General Description
A plume of contaminated ground water can be
contained or manipulated by pumping and injection
wells. Similar to water-table adjustment, cones of de-
pression or recharge in the water table are developed
to modify prevailing hydraulic gradients. The move-
ment and size of a plume can be manipulated by
various pumping recharge strategies. Recovered
water may be treated at the surface and reinjected as
part of the plume-containment program.
5.5.4.2 Application/Availability
Hydraulic barrier technology can be applied to most
contaminated sites, particularly those directly under-
lain by moderate to highly productive aquifers. The
use of low-flow interdiction wells to control hydro-
carbon plumes in ground water has been demon-
strated successfully (Sammons and Armstrong 1986).
The principle of plume control by hydraulic methods is
to effect a change in the ground-Water flow pattern
such that the contaminant can be drawn to a specific
control point or points. This can be achieved by
discharging or recharging the aquifer or a combination
of both. The success of the method depends on
maintaining an artificial gradient in the ground-water
surface.
5.5.4.3 Design and Construction
Considerations
Establishing a hydraulic barrier requires strategic
placement of a series of injection wells. The location
and depth of these wells should be determined by
detailed examination of the site. Low-permeability soil
(clay, shale) and fractured or consolidated rock
aquifers limit the effectiveness of pumping/recharge
systems. ;
Required equipment includes a drilling rig (usually
truck-mounted) for well construction, a backhoe or
bulldozer for digging a recharge trench (if ap-
propriate), pumps for water withdrawal, and electricity
to operate the pump motor. Drilling in contaminated
soil or water will require special protection of workers.
5.5.4.4 Costs
Estimated costs were computed for a series of 6-inch-
diameter pumping and recharge wells connected by 8-
inch transfer piping, 200 feet per pair of wells, with
submersible pumps in the pumping wells and gravity
discharge in the injection wells, an average well depth
of 50 feet, and a pumping rate of 50 gal/rnin per well
(500 gal/min for the site). These costs are as follows:
Pumping/recharge wells
Transfer piping (PVC material)
Total
$27.73 per linear foot
10.78 per linear foot
$38.51 per linear foot
Operation and maintenance is estimated to be $72.79
per thousand gallons pumped per day.
5-48
-------
Figure 5-28. Typical steel piling shapes and Interlocks (U.S. E
Straight Web Type
Arch Web Type
Deep Arch Web Type
Z- Type
Y-Fitting
'Al985a).
5-49
-------
5.6 In Situ Treatment
In situ treatment of contamination resulting from an
UST leak offers an alternative to excavation and
removal and conventional pumping and treating
methods. In situ treatment processes are generally
divided into three categories: biological, chemical,
and physical. In situ biodegradation, or biostimulation,
is based on the concept of stimulating microorganisms
to decompose the contaminants of concern. In situ
chemical treatment involves the injection of a specific
chemical or chemicals into the subsurface to degrade,
immobilize, or flush out the contaminants. Physical
methods involve physical manipulation of the soil by
the use of air, steam, heat, freezing, or other means.
In many instances, a combination of in situ and above-
ground treatment will achieve the most cost-effective
treatment. Also involved are methods for delivering
solutions to the subsurface and methods for con-
trolling the spread of contaminants and treatment
reagents beyond the treatment zone.
Although not as developed as other currently
available technologies for restoring contaminated soils
and aquifers, some in situ treatment technologies
have been demonstrated successfully in actual UST
site remediations. Applicability of in situ methods must
generally be determined on a site-specific basis after
laboratory- and pilot-scale testing.
5.6.1 Soil Flushing (EPA 1985a)
5.6.1.1 General Description
Organic and inorganic contaminants can be washed
from contaminated soils by an extraction process
called "soil flushing," "solvent flushing," "ground
leaching," or "solution mining." Water or an aqueous
solution is injected into or sprayed onto the area of
contamination, and the contaminated elutriate is col-
lected and pumped to the surface for removal, re-
circulation, or onsite treatment and reinjection (Figure
Figure 5-29. Soil flushing system (Ehrenfeld and Bass 1983).
Spray Application
Pump
5-29). During elutriation, the flushing solution mo-
bilizes the sorbed contaminants by dissolution or
emulsification.
5.6.1.2 Application/Availability
Soil flushing technology is currently in the laboratory
stage. Studies have been conducted to determine
the appropriate solvents for mobilizing various classes
and types of chemical constituents. The technology
may be easy or difficult to apply, depending on the
ability to flood the soil with the flushing solution and to
install collection wells or subsurface;drains to recover
all the applied liquids. Provisions also must be made
for ultimate disposal of the elutriate! The achievable
level of treatment varies and depends on the contact
of the flushing solution with the contaminants, the
appropriateness of solutions for the contaminants,
and the hydraulic conductivity of the. soil. The techno-
logy is more applicable to highly permeable soils.
Water can be used to flush water-soluble or water-
mobile organics and inorganics. Hydrpphilic organics
are readily solubilized in water. Organics amenable to
water flushing can be identified according to their
soil/water partition coefficients, or estimated based on
their octanol/water partition coefficients. Organics
considered soluble in the environmental sense are
those with a partition coefficient (K) of less than about
1000. High-solubility organics, such as lower-
molecular-weight alcohols, phenols, and carboxylic
acids, and other organics with a coefficient less than
10 may already have been flushed from the site by
natural flushing processes. Medium-solubility organics
(K = 10 to 1000) that can be effectively removed from
soils by water flushing include low- to medium-
molecular-weight ketones, aldehydes, and aromatics,
and lower-molecular-weight halogenated hydro-
carbons, such as trichloroethylene (TCE) and tetra-
chloroethylene (PCE). Inorganics that can be flushed
••••• '
• • •!•!•!•!•!• I '!• -i- ' ! • ! • i • ; • ! • i •!•:•!•! •,'
''•:•:•:•: • : • : 1 1 Leacnate ; : j : j ! j ! j ! j : j : • ! • : j ! j
5-50
-------
from soil with water are soluble salts such as the
carbonates of nickel, zinc, and copper. Adjusting the
pH with dilute solutions of acids or bases will enhance
inorganic solubilization and removal. Soil flushing ap-
plicability is summarized in Table 5-19.
5.6.1.3 Design and Construction
Considerations (Sims et al. 1985)
The following basic information is required for
implementation of a soil-flushing process:
Characterization and concentration of chemical
constituents
Depth, profile, and areal distribution of
contamination
Partitioning of chemical constituents between
solvent(s) and soil
Effects of flushing agent (solvent) on physical,
chemical, and biological soil properties
Suitability of site for flooding and installation of
wells and/or subsurface drains
Whether soil and site can tolerate traffic
Flushing solutions may include water, acidic aqueous
solutions (sulfuric, hydrochloric, nitric, phosphoric,
and carbonic acids), basic solutions (e.g., sodium
hydroxide), and surfactants (e.g., alkylbenzene
Table 5-19. Applicability of Soil Flushing Techniques*
Compounds
Amenable to
Treatment
Flushing
Solution
Hydrophilic compounds
(high solubility, low KOW)
Hydrophobia compounds
(low solubility, high KOW)
Water
Aqueous solutions
of surfactants
Heavy metals
Dilute solutions
of acids, bases,
or chelates
Data from Wagner and Kosin 1985.
sulfonate). Water can be used to extract water-soluble
or water-mobile constituents. Acidic solutions are
used for metals recovery and for basic organic con-
stituents, including amines, ethers, and anilines.
Basic solutions are used for flushing metals, including
zinc, tin, and lead, as well as for some phenols, com-
plexing and chelating agents, and surfactants.
The addition of any flushing solution to a soil system
requires careful management and knowledge of re-
actions that may adversely affect the system.
For example, a sodium addition (e.g., sodium hy-
droxide) to soil systems may adversely affect soil
permeability by altering the soil/sodium adsorption
ratio. The user must understand both the chemical
reaction(s) between solvent and solute, and those
between solvent and site/soil system.
At a site contaminated by organic constituents, re-
cycling the elutriate back through the soil for treatment
by biodegradation may be possible. Proper control of
the application rate would provide an effective in-place
treatment for soil concentrations that would allow con-
trolled biodegradation of the waste constituents. This
approach could eliminate the need for separate
processes for treatment and disposal of the collected
waste solution, or at least provide for a combination
pretreatment/land application, which could be consid-
erably more economical than unit operations for
treatment of elutriate.
Process
Limitations
Contaminated soils are
flooded with water or a
water/chemical
mixture and the
elutriated solution is
collected
Contaminants are
mobilized into solution
by solubility formation of
emulsion or reaction.
Mobilized contaminants
are collected
Involves solubilizing
the metals followed
by extraction of the
metal ions
Not suitable for
compounds that are
adsorbed to soils
Some surfactants
are easily degraded
within the soil
environment
Extensive laboratory
testing may be
required to
determine optimum
mix of surfactant
Some metals are
strongly sorbed and
require treatment
with strong acids,
which may be toxic
Some chelating
agents will be
sorbed by soils
5-51
-------
When soils are contaminated with inorganic and
organic constituents, a combination pretreatment/land
application, in which the metal constituent(s) are
reduced or eliminated in the elutriate by precipitation
and the elutriate is then applied to the land, may be a
feasible, cost-effective method of treatment.
Soil flushing and elutriate recovery may also be
appropriate in cases where the use of chemical
oxidizing or reducing agents for chemical degradation
of waste constituents results in the production of large
amounts of oxygenated, mobile, degradation prod-
ucts. The most conservative and safest approach may
be to flush the soil after treatment for recovery and
possible controlled reapplication of the elutriate to the
soil surface.
Equipment used for soil flushing includes drains and
an elutriate collection and distribution system. Sol-
vents may also be required. Reapplication of col-
lected elutriate may require construction of a holding
tank for the elutriate.
5.6.1.4 Costs
Because the soil flushing technology is not widely
used at this time, and because this approach is
particularly site- and contaminant-related, cost infor-
mation is scarce.
5.6.2 Biostimulation (EPA 1985a)
5.6.2.1 General Description
A site contaminated by a leaking UST might be
restored through biostimulation, a technology
whereby naturally occurring soil microorganisms are
stimulated to biodegrade the waste. The basic con-
cept involves altering environmental conditions to
enhance microbial catabolism or cometabolism of
organic contaminants, which results in the breakdown
and detoxification of those contaminants. This tech-
nology has developed rapidly over recent years and
appears to be one of the most promising of the in situ
treatment techniques.
The biostimulation method that has been most
developed and is most feasible for in situ treatment is
one that relies on aerobic (oxygen-requiring) microbial
processes. This method involves optimizing environ-
mental conditions by providing an oxygen source and
nutrients, which are delivered to the subsurface
through an injection well or infiltration system for the
enhancement of microbial activity (Figure 5-30).
Indigenous microorganisms can generally be relied
upon to degrade a wide range of compounds given
proper nutrients and sufficient oxygen.
Figure 5-30. Biostimulation of soil and ground water (EPA 1985a).
Nutrients
In-line
Direction of Ground Water Row
Extraction Well
5-52
-------
In situ biodegradation technology also encompasses
ground-water seeding, which refers to the addition of
specially adapted or genetically manipulated micro-
organisms below the water table. (Some phases of
ground-water seeding are covered by patents.)
5.6.2.2 Application/Availability
Considerable research conducted over the past
several decades has confirmed that microorganisms
can break down many of the organic compounds
currently stored in underground tanks. Laboratory, pi-
lot, and field studies have demonstrated that
microorganisms can be used in situ to reclaim contam-
inated soils and ground water.
The feasibility of biostimulation as an in situ treatment
technique is dictated by waste and site characteristics.
The following factors determine the applicability of a
biostimulation approach:
Biodegradability of the organic contaminants
Enyironmental factors that affect microbial
activity
• Site hydrogeology
The most rapid and complete degradation of most
compounds occurs aerobically. Some compounds,
most notably the lower-molecular-weight halogenated
hydrocarbons, will only degrade anaerobically. In
general, aerobic techniques are most suitable for the
degradation of petroleum hydrocarbons, aromatics,
halogenated aromatics, polyaromatic hydrocarbons,
phenols, halophenols, biphenyls, organophos-
phates, and most pesticides and herbicides. An-
aerobic techniques under very reducing conditions
appear to be most feasible for the degradation of
lower-molecular-weight halogenated hydrocarbons
such as unsaturated alkyl halides (e.g., PCE and TCE)
and saturated alkyl halides (e.g., 1,1,1-trichloroethane
and trihalomethane). Aerobic degradation in the pres-
ence of methane gas, however, appears to be
promising for some low-molecular-weight halogenated
hydrocarbons.
The availability of the compound to the organism also
dictates its biodegradability. Compounds with greater
aqueous solubilities are generally more available to
degrading enzymes. For example, cis-1,2-dichIoro-
ethylene is preferentially degraded relative to trans-
1,2-dichloroethylene. The most likely explanation for
this is that "cis" is more polar than "trans" and is there-
fore more water-soluble. The use of surfactants can
increase the solubility and therefore the degradability
of compounds.
5.6.2.3 Design and Construction
Considerations
The feasibility and effectiveness of biostimulation as
an in situ treatment method is determined by the
5-53
microbial population, the biodegradability of the
organic contaminants, and a host of environmental
factors that affect microbial activity (Wagner and Kosin
1985). High concentrations of leaked materials from
an LIST and deficient soil conditions (such as low
moisture content) will adversely affect biodegradation,
as will extremes of pH, temperature, and nutrient
levels. In general, optimum environmental conditions
are 1) pH of 7.0 to 8.5, 2) temperature of 15° to 35°C,
3) nutrient levels of nitrogen and phosphorus, and
4) 40 percent by weight moisture in soil. Adequate
mixing (aeration or cultivation) is also needed.
Leaking substances from an LIST can destroy the
natural microbial population. For in situ biological
treatment to remain a viable option, the factors that
caused the sterilization must be corrected (e.g.,
neutralization with acid or base, or dispersing and
diluting to a certain extent). Even though the native
microbes have been destroyed, biological treatment is
still possible by the deployment of specialized mutant
strains of microbes; however, the value of mutant
organisms (super bugs) is still being debated. These
commercially produced strains are available in a fresh
liquid state, a powdered state, or freshly recon-
stituted. The potentially harmful secondary effect of
the addition of a foreign microorganism to the
environment will generally be minor because once the
hazardous material has been digested, the foreign-
added microorganisms will probably die and become a
source of nourishment for the naturally occurring
microorganisms.
One problem associated with the addition of micro-
organisms to contaminated water is the significant
increase in the consumption of dissolved oxygen.
Low dissolved oxygen levels could prove to be
detrimental to existing aquatic organisms. This
problem can be minimized, however, by providing
adequate aeration or adding only small amounts of
bacteria so that excessive oxygen consumption does
not occur.
Even if the active microbial population is substantial,
the wastes are biodegradable. Parameters can be
altered to optimize biodegradation in situ.
Biostimulation is not feasible, however, if the
hydrogeology of the site is not suitable. The hydraulic
conductivity must be great enough and the residence
time short enough so that added substances (oxygen
and nutrients) are not "used up" before reaching the
distal portions of the treatment zone. Sandy and
other highly permeable sites are far easier to treat than
sites containing clayey soils.
Added substances also can react with the soil
components. Oxidizing the subsurface could result in
the precipitation of iron and manganese oxides and
hydroxides. If precipitation is extensive, the delivery
system and possibly even the aquifer could become
-------
clogged. The addition of phosphates could result in
the precipitation of calcium and iron phosphates. If
calcium concentrations are high, the added
phosphate can be tied up by the calcium and would
therefore not be available to the microorganisms.
Oxygen can be provided to the subsurface through
the use of air, pure oxygen, hydrogen peroxide, or
possibly ozone. Air can be added to extracted ground
water before reinjection, or it can be injected directly
into the aquifer. The first method, known as in-line
aeration, involves adding air into the pipeline and
mixing it, with a static mixer for example. A pressurized
line can increase oxygen concentrations, as can the
use of pure oxygen.
The use of in situ aeration wells is a more suitable
method for injecting air into contaminated plumes. A
bank of aeration wells can be installed to provide a
zone of continuous aeration through which the
contaminated ground water would flow. Oxygen
saturation conditions can be maintained for degrading
organics during the residence time of ground-water
flow through the aerated zone. The required time for
aeration can be derived from bench-scale studies.
Microdispersions of air in water by the use of colloidal
gas aprons (CGA) create bubbles 25 to 50 micro-
meters in diameter. This newly developed method
holds promise as a means of introducing oxygen to
the subsurface (Michelsen, Wailis, and Lavinder
1985). With selected surfactants, dispersions of
CGA's can be generated that contain 65 percent air by
volume.
A blower also can be used to provide the flow rate and
pressure for aeration. At a ground-water biore-
clamation project in Waldwick, New Jersey, 5 psi
pressure was maintained in nine 10-foot aeration
wells, each with an airflow of 5 ft3/min (Groundwater
Decontamination Systems, Inc. 1983).
Oxygenation systems, either in-line or in situ, can also
be installed to supply oxygen to the biostimulation
process. Their advantage over conventional aeration
is that higher oxygen solubilities and hence more
efficient oxygen transfer to the microorganisms can be
attained. Solubilities of oxygen in various liquids are
four to five times higher under pure oxygen systems
than with conventional aeration. Therefore, in-line
injection of pure oxygen will provide sufficient dis-
solved oxygen to degrade 20 to 30 mg/liter of organic
material, assuming 50 percent cell conversion. The
higher oxygen solubilities also may provide some
flexibility in the design of cell banks, especially at
greater pressures, because the oxygen may not be
used up immediately as it is with aeration.
Hydrogen peroxide (H2O2) as an oxygen source has
been used successfully at the cleanup of several spill
sites. Hydrogen peroxide is cytotoxic at high con-
centrations, but research has demonstrated that it can
be added to acclimated cultures at;up to 1000 ppm
without toxic effects. Hydrogen peroxide offers the
following advantages:
Greater oxygen concentrations can be delivered
to the subsurface; 100 mg/liter H2O2 provides 50
mg/liter O2.
Less equipment is required to oxygenate the
subsurface. Hydrogen peroxide can be added
in-line along with the nutrient solution. Aeration
wells are not necessary.
Hydrogen peroxide keeps the well free of heavy
biogrowth. Microbial growth and subsequent
clogging is sometimes a problem in air injection
systems.
Bioreclamation also requires the presence of nutrients
in the soil to effect bipdegradation. Nitrogen and
phosphate are the nutrients most frequently present
in limited concentrations in soils. Other nutrients
required for microbial metabolism include potassium,
magnesium, calcium, sulfur, sodium, manganese,
iron, and trace metals. Many of these nutrients may
already be present in the aquifer in sufficient
quantities and need not be supplemented.
The optimum nutrient mix can be determined by
laboratory growth studies and geochemical
evaluations of the site. Caution must be exercised in
evaluating microbial needs based on soil and ground-
water chemical analyses. A chemical analysis does not
necessarily indicate what is available to the
microorganisms. In some cases, generalizations can
be made; e.g., if calcium is present at 200 mg/liter (a
very high concentration), calcium supplementation is
likely to be unnecessary.
One of the major factors in determining the success of
an in situ treatment system is to ensure that the
injection and recovery systems are designed to
accomplish the following:
Provide adequate contact between treatment
agents and contaminated soil or ground water.
Provide hydrplogic control of treatment agents
and contaminants to prevent their migration
beyond the treatment area.
Provide for complete recovery of spent
treatment solutions and/or contaminants where
necessary.
The several design alternatives available for delivering
5-54
-------
nutrients and oxygen to the subsurface and for
collecting and containing the ground water can
generally be categorized as gravity-flow or forced
methods. Most of the systems that have been used in
biostimulation projects involve subsurface drains
(gravity system), injection wells, and extraction wells.
The following criteria are important in the design of an
injection/recovery system:
The ground-water injection rate must be the
same as the rate determined during the field
testing program.
All injected ground water and associated
elements must be kept within the site boundary
to prevent the transport of contaminants to
adjacent areas. (This implies that some net
ground-water pumpage will take place at the
site.)
The distance between the injection and
pumping wells should be such that approx-
imately six injection-pumping cycles can be com-
pleted within a 6-month period.
Implementation of a remedial action involving bio-
stimulation will take longer than excavation and re-
moval of contaminated soils. Depending on the spe-
cific site, it could also take longer than a conventional
pumping and treating approach. The advantage of in
situ biodegradatipn over the latter approach is that in
situ biodegradation treats contaminated subsurface
soils and thus removes the source of ground-water
contamination.
5.6.2.4 Costs
Costs for biological in situ treatment are determined by
the site's geology and geohydrology, the extent of
contamination, the kinds and concentrations of con-
taminants, and the amount of ground water and soil
requiring treatment. There is no easy formula for
predicting costs. Costs provided for actual site clean-
ups indicate that biological treatment can be far more
economical as an alternative to, or in conjunction with,
excavation and removal or conventional pumping and
treating methods.
Total capital and research and development costs for
cleanup of the Biocraft site (Sims et al. 1985) were
$950,137 including $458,330 for process develop-
ment (R&D). Project costs also included a hydro-
geological study as well as the design and operation
of a ground-water injection and collection system and
a biostimulation plant. Total operating costs, based on
treating 13,680 gallons/day, were approximately
$232/day or $0.0170/gal. The total cost, including
amortization based on projected costs, was
$0.0367/gal over a 3-year period. Prior to initiation of
the biological treatment program, contaminated water
had been removed at a rate of 10,000 gal/month, and
the average disposal cost had been $0.36/gal. The
cost of biological treatment of an equal number of
gallons is an order of magnitude less than that for
disposal. The Biocraft site used surface biological
reactors and enhanced in situ treatment by rein-
filtrating oxygen and nutrient-treated ground water.
Costs of in situ treatment alone would have been less
because they would not include the process plant
design and equipment.
Table 5-20 presents the estimated site cleanup costs
for hypothetical sites at which hydrogen peroxide is
used as an oxygen source for the enhancement of in
situ biodegradation. The cleanup of 300 gallons of
gasoline from a sand/gravel aquifer (Site A) over a
period of 6 to 9 months is estimated to cost $72,000
to $123,000. Cleanup of 2,000 gallons of diesel fuel
from a fractured bedrock formation (Site B) is
estimated to require 9 to 12 months and cost
$164,000 to $257,000. The cost estimate for de-
grading 10,000 gallons of jet fuel from a fine gravel
formation is estimated to cost $411,000 to $61*6,000
and take 14 to 18 months.
Table 5-20.
Estimated Costs for Hypothetical Bio-
reclamations With Hydrogen Peroxide as
an Oxygen Source*
Site A
SiteB
SiteC
Contaminant
Formation
Row rate
Project time
Estimated cost
300 gallons
gasoline
Sand/gravel
50 gal/min
6 to 9 months
$72,00010
$123,000
2,000 gallons
diesel fuel
Fractured bedrock
1 0 gal/min
9 to 12 months
$164,000 to $257,000
10,000
gallons
jet fuel
Fine gravel
100 gal/min
14 to 18
months
$411,000 to
$616,00
Data fromFMC, 1985.
5.6.3 Chemical Treatment
5.6.3.1 General Description
The term "in situ chemical treatment" covers a variety
of technologies whereby organic and inorganic
contaminants can be immobilized, mobilized for
extraction, or detoxified. Technologies that fall into
the "immobilization" category include precipitation,
chelation, and polymerization. The technologies
used to mobilize contaminants for extraction fall into
the category of "soil flushing," which was covered in
Subsection 5.6.1. "Detoxification" techniques in-
clude oxidation, reduction, neutralization, and
hydrolysis.
These categories do not define the limits of each
technology. For example, a treatment method that
immobilizes a contaminant may also detoxify it; a
flushing solution that mobilizes one contaminant may
precipitate, detoxify, or increase the toxicity of
another.
5-55
-------
5.6.3.2 Application/Availability
The feasibility of an in situ treatment approach is
dictated by site geology and hydrology, soil
characteristics, and waste characteristics. The ap-
plication of many chemical in situ treatment
techniques to UST site reclamation is currently
conceptual in nature or in the developmental stage;
therefore, little hard data are available on the specific
site characteristics that may limit the applicability of
each method.
Table 5-21. Summary of In Situ Treatment Methods*
Table 5-21 summarizes the in situ chemical treatment
methods for organics and inorganics, respectively,
that are most promising or have been discussed most
widely in the literature. The summary includes the
compounds amenable to treatment, the treatment
reagents, and the processes.
5.6.3.3 Design and Construction
Considerations
Most in situ chemical treatments involve the delivery
of a fluid to the subsurface. Therefore, the same
Method
Precipitation
Sulfida
Carbonate/
hydroxide
Phosphate
Compounds
Amenable to
Treatment
Heavy metals
Heavy metals
Heavy metals
Treatment Reagents Process
Sodium or calcium sulfide Forms insoluble
metal precipitates.
Lime, calcium carbonate
Superphosphate fertilizer
Limitations
Carbonate and hydroxide
precipitation is effective
only over a narrow pH
range
Soil cations are likely
to compete for phosphate
ErWM,teie§3ft^
Oxidation
Reduction
Sorpfion by
natural and
synthetic
materials
Oxidation
(organics)
Reduction
(organics)
Trivalent arsenic
Hexavalent chromium
Heavy metals
Benzene and
substituted benzenes
Phenols
Halogenated phenols
Nitro aromatics
PAHs
Heterocyclic nitrogen
and oxygen
compounds
Aldehydes and
ketones
Sulfides, disulfides
Nitro aromatics
Chlorinated aromatics
Chlorinated aliphatics
Potassium permanganate
followed by precipitation
with ferrous sulfate, lime,
etc.
Ferrous sulfate
Organic matter, clay,
ion exchange materials
Ozone, hypochlorite, or
hydrogen peroxide
Catalyzed metal
Oxidizes As (III)
toAs(V).
Reduces Cr (VI) to
Cr (III).
Reduces Se (VI) to
Se(IV)
Exchanges metals
for soil cations or
innocuous cations
on an ion exchange
resin
Increases oxidation
state of compounds
by loss of electrons;
detoxifies
compounds or
renders them more
amenable to
biological degrada-
tion
Decreases oxida-
, tion state of
compounds by
addition of elec-
trons; detoxifies
compounds by
removal of halo-
gen or nitro group
or by saturation
of aromatic structure
01 n2o gas from sume
precipitation; precipitation
may reduce soil permeability
Treatment compounds are
nonspecific; volatile arsenic
may be formed
Potential for reoxidation of
chromium and selenium
under certain conditions
Potential for release of
sorbed metals; ion
exchange resins are
very costly
Potential for formation
of more toxic or soluble
degradation products;
process is nonspecific,
and compounds other
than the targeted com-
pounds may be oxidized;
use of hypochlorites may
result in formation of
chlorinated organics;
hydrogen peroxide and
ozone may decompose
rapidly
Treatment reagents may
be costly; very limited
research has been done
on chemical reduction
5-56
-------
Table 5-21. (continued)
Method
Hydrolysis
(base-
catalyzed)
Polymerization
Compounds
Amenable to
Treatment Treatment Reagents
Esters Water with lime or
Amides NaOH
Carbamates
Organophosphorus
carbamates .
Certain pesticides
(e.g., parathion,
malathion, 2-4D
esters, DDT)
Aliphatic, aromatic, Catalyst activation
and oxygenated
monomers (vinyl
chloride, isoprene,
acrylonitrile)
Process
Nucleophile (e.g.,
water or hydroxyl
ion) attacks an
electrophile (e.g..
carbon or phosphor-
ous), resulting in
bond cleavage and
displacement of the
leaving group
Converts a
compound to a larger
chemical multiple
of itself
Limitations
Sorbed organics may
be difficult to hydrolize;
very little research has
been conducted on the
feasibility of in situ
hydrolysis
Polymerization results
in a decrease in soil
permeability; therefore.
complex treatment would
be difficult to achieve and
would require dose
spacing of injection
points; potential for
reversal of the polymeri-
zation reaction
Sorption by
natural and
synthetic
materials
Hydrophobia organics,
organic cations
Sewage, sludge, activated
carbon, agricultural I
products and byproducts,
chelates, day, and ion
exchange resins
Complexes, chelates,
or sorbs hydrophobia
organics by chemical
bonding; sorbs
cationic organics
by ion exchange
Potential for release
of sorbed compounds;
feasibility and effective-
ness is highly site-
specific
Data from Wagner and Kosin 1985.
factors that limit the use of injection/extraction wells,
drains, or surface gravity application systems (such as
flooding and spray irrigation) will limit the applicability of
most in situ chemical treatments. Minimal permeability
requirements must be met for the treatment solution
to be delivered successfully to the contaminated
zone. Sandy soils are far more amenable to in situ
treatment than dayey soils. In addition, the
contaminated ground water must be contained within
the treatment zone. Measures must be taken to
ensure that treatment reagents do not migrate and
become contaminants themselves. Care must be
taken during the extraction process not to increase
the burden of contaminated water by drawing uncon-
taminated water into the treatment zone from an
aquifer far from hydraulically connected surface
waters.
Potential chemical reactions of the treatment reagents
with the soils and contaminants must be considered.
This form of treatment can reduce the permeability of
soils. In soils high in iron and manganese, for ex-
ample, oxidizing the subsurface could result in pre-
cipitation of iron and manganese oxides and hydrox-
ides, which could clog the delivery system and the
aquifer.
Mention of potential drawbacks should not preclude
consideration of chemical in situ methods; however,
laboratory (and possibly pilot-scale) testing probably
will be required in each case and delay implementation
of the remedial action. As with in situ biological treat-
ment methods, methods used to deliver and recover
treatment reagents also affect the reliability of
chemical in situ methods.
5.6.3.4 Costs
Costs for chemical in situ treatment approaches are
difficult to estimate because, for the most part, these
methods have not been demonstrated and no actual
cost data are available. In situ treatment costs will be
variable, but they could be less than those for ex-
cavation and removal methods and/or pumping and
treating methods. As with removal, in situ treatment is
a one-time effort, so as a rule no long-term operation
and maintenance costs would be involved.
5-57
-------
Costs for chemical treatments involving the delivery of
a reagent to the subsurface will depend on the
amount of material to be treated; the amount of
chemical reagent required; the costs for the delivery
system and the chemical feed system; and fees for
probing, excavation, and drilling. Costs for laboratory-
and pilot-scale studies also should be considered
when evaluating this approach to remediation.
5.6.4 Physical Treatment
5.6.4.1 General Description
Several methods currently under development
involve physical manipulation of the subsurface to
immobilize or detoxify contaminants. These tech-
nologies, which include in situ heating, vitrification,
and ground freezing, are in the early stages of de-
velopment and detailed information is not available.
Their application to contamination caused by UST
releases will probably be limited.
5.6.4.2 Application/Availability
In situ heating has been proposed as a method for de-
stroying or removing organic contaminants in the
subsurface through thermal decomposition, vapo-
rization, and distillation. Steam injection and radio
frequency (RF) heating are the recommended heating
methods. The radio frequency heating process has
been under development since the 1970's, and field
experiments have been conducted for the recovery of
hydrocarbons. This method involves laying a row of
horizontal conductors on the contaminated surface
and exciting them with an RF generator through a
matching network. The decontamination is accom-
plished in a temperature range of 300° to 400°C
(assisted with steam) and requires a residence time of
about 2 weeks. A gas- or vapor-recovery system is
required on the surface. No excavation, mining,
drilling, or boring is required. Preliminary design and
cost estimates for a mobile RF in situ decontamination
process indicate that this method is two to four times
cheaper than excavation and incineration. The
method appears to be very promising for certain
situations involving contamination with organics,
although more research is necessary to verify its
effectiveness.
In situ vitrification, a technology under development
for the stabilization of transuranic contaminated
wastes, is conceivably applicable to other hazardous
wastes. Several laboratory-scale and pilot-scale tests
have been conducted, and a full-scale testing system
is currently being fabricated. The technology is based
on electric melter technology, and the principle of
operation is joule heating, which occurs when an
electrical current is passed through a molten mass.
Contaminated soil is converted into durable glass, and
wastes are pyrolyzed or crystallized. Off-gases re-
leased during the melting process are trapped in an
off-gas hood. The depth of the contamination is a
significant limiting factor in the application of this
technology; 1 to 1.5 meters of uncontaminated over-
burden lowers release fractions considerably.
Artificial ground freezing involves the installation of
freezing loops in the ground and a self-contained
refrigeration system that pumps coolant around the
freezing loop. Although this technology has never
been used in an actual spill or leak containment
operation, its use as a construction method in civil
engineering projects is increasing. Artificial ground
freezing is not applied to the contaminant itself, which
may have a freezing point much lower than that of the
soil systems, but to the soil surrounding it. Freezing
renders the soil practically impermeable; however, its
value is temporary because of the thermal
maintenance expense.
5.6.4.3 Design and Construction
Consideration
The techniques discussed in this subsection (in situ
heating, vitrification, and ground freezing) are in the
early stages of development. Therefore, detailed
design and construction considerations are not avail-
able.
5.6.4.4 Costs
No actual cost figures are available for these
technologies because they are still in the develop-
mental stage. Some estimated costs are presented in
Table 5-22.
Table 5-22.
Technology
Estimated Costs for in Situ Physical
Treatment Methods
Cost Basis
Cost Range
Radio frequency Estimate for hypothetical $5 to $6 million
heating 1-acre area contaminated (operating costs
to a depth of 20 ft only)
Vitrification Estimate for a site con- ' $200 to $280/yd3
laminated to a depth of 13 ft
Ground freezing Estimate for hypothetical $154,000 to
1000 ft x 3 ft x 40 ft $310,000
deep frozen wall of soil
5.7 Ground-Water Treatment
Many of the concepts of ground-water treatment, an
old science that modern technology has improved,
have been developed through years of experience in
treating industrial wastewaters. Today's emphasis on
correcting contamination problems in diverse lo-
cations has prompted new applications of these con-
cepts to remedy extreme situations of ground-water
contamination, e.g., a leaking underground petroleum
product storage system.
Selection of treatment depends on the contaminants
to be removed. A system may consist of a combination
5-58
-------
1 of several technologies to effect a solution. Such a
• combination will result from treatability studies that
• should be conducted with representative samples.
• Some of the treatment technologies for ground water
•: that are more applicable to underground storage
• system leaks are described in this section. These des-
• criptions are intended as background for assembling a
• system to remedy a particular leak situation based on
• the treatability study. Table 5-23 lists the treatment
• processes discussed herein and the contaminants to
• which they can be applied.
I Table 5-23. Applicability of Ground-Water Treatment
1 Processes
• ' Gasoline Non-
• and Volatile Volatile
• Treatment Process Organics Organics Inorganics
1 • Air stripping x
1 ' Carbon adsorption x x
• , Biological treatment x x
I , Precipitation/flocculation/ x
1 sedimentation
1 : Dissolved air flotation x
1 ; Granular media filtration x* x* x
1 Ion exchange/resin x
1 adsorption
1 Oxidation/reduction x
1 Neutralization x
1 Steam stripping x
1 Reverse osmosis x
.Sludge dewatering x
: Pretreatment.
5.7.1 Air Stripping
15.7.1.1 General Description
In an air stripping process, a contaminated water
stream is mixed with a clean air stream, and the
intimate contact causes the air to remove the
dissolved organic substances from the water. The
different kinds of equipment used to carry out this
; process are classified primarily as towers and basins.
Basins are usually large installations with floating
aerators such as those found in municipal sewage
and/or water treatment plants. Towers are more
applicable for an LIST release because they are readily
available. Mobile units can be obtained from vendors
', and contractors. Only towers are discussed here.
Figure 5-31 shows typical air-stripping equipment
configurations.
5-
I
5.7.1.2 Application/Availability
A typical air-stripping tower is similar in construction to
a water cooling tower. Contaminated water is fed into
the top and flows down over an internal system of
baffles or packing designed to spread it out over a
large surface. As the water flows down through the
tower, a fan moves air up through the tower, where
the baffling or packing causes it to come into intimate
contact with the water. During the contact, the air
picks up dissolved organic substances.
The countercurrent packed tower appears to be the
most appropriate equipment configuration for treating
contaminated ground waters for the following reasons
(Canter and Knox 1986):
1 ) It provides the most liquid interfacial area.
2) High air-to-water volume ratios are possible
because of the low air pressure drop through the
tower.
3) Emission of stripped organics to the atmosphere
may be environmentally unacceptable; however,
a countercurrent tower is relatively small and can
be readily connected to vapor-recovery
equipment.
Air stripping has been successful in removing volatile
organics from contaminated ground waters. Table 5-
24 summarizes removals that have been achieved for
various organic contaminants at various air-to-water
ratios.
Table 5-24. Packed-Column Air Stripping of
Volatile Organics*
Air-to-
Water Influent Effluent
Organic Contaminant Ratio (ng/liter) (jig/liter)
1 ,1 ,2-Trichloroethylene 9.3 80 16
96.3 80 3
27.0 75 16
156.0 813 52
44.0 218 40
75.0 204 36
125.0 204 27
1,1,1-Trichloroethane 9.3 1200 460
96.3 1200 49
27.0 90 31
156.0 1332 143
1,1-Dichloroethane 9.3 35 9
96.3 35 1
1 ,2-Dichloropropane 27.0 50 <5
146.0 70 5
156.0 377 52
Chloroform 27.0 50 <2
146.0 57 2
Diisopropyl ether 44.0 15 7
75.0 14 6
125.0 4 4
Data from Canter and Knox 1986.
59
-------
Figure 5-31. Air stripping equipment configurations (Knox et al. 1984).
Exit Air
Influent <
Tnnruuu
*• Distributor
• Packing Material
• Support Plate
• Incoming Air
Effluent
Packed Column
Air Supply
Influent
Diffuser J
Grid
Effluent
Diffused-Air Basin
Air Outlet
Raw
Water-
Inlet
innnnnn
innnnnn
Distributing
Tray
I Splash
Aprons
'Aerated Water Basin
-Outlet
Coke-Tray Aerator
Water Inlet
Air Inlet
Water Outlet
Water Inlet
Air Inlet
Collection
Basin
Cross-Flow Tower
5.7.1.3 Design and Construction
Considerations
The design of a process for air stripping volatile
organics from contaminated ground water is
accomplished in two steps. The first step involves
determining the cross-sectional area of the column by
using the physical properties of the air flowing through
the column, the characteristics of the packing, and the
air-to-water flow ratio. A key factor is the establishment
of an acceptable air velocity. A general rule of thumb
used for establishing the air velocity is that an
acceptable air velocity is 60 percent of the air velocity
at flooding. Flooding refers to the condition in which
the air velocity is so high that it holds up the water in
the column to the point where the water becomes the
continuous phase rather than the air. If the air-to-water
ratio is held constant, the air velocity determines the
flooding condition. For a selected air-to-water ratio,
the cross-sectional area is determined by dividing the
air flow rate by the air velocity. The selection of the
design air-to-water ratio must be based on experience
or pilot-scale treatabiiity studies. Treatability studies
are particularly important for developing design infor-
mation for contaminated ground water (Canter and
Knox 1986). In the second step, the column height is
determined mathematically from the physical prop-
erties of the contaminant and the stripping air.
Installation of a unit usually requires field assembly of
equipment or placement of shop-fabricated/packaged
units. Installation of the complex tower internals is the
most labor-intensive task. Overall, however, instal-
lation of an air stripper is relatively simple and can be
done by most mechanical contractors.
The exiting air stream must be examined to determine
if it will be a source of air pollution. If it:is, cleanup tech-
nologies for vapor recovery can be added to remedy
that situation.
Because an air-stripping system is simple to operate
and occupies minimal space, it is a prime candidate for
treating contaminated ground water at service station
sites.
5.7.1.4 Costs
Although variations in the design of packed-column air
stripping systems result in varying costs, the major
5-60
-------
components of an air-stripping system for removing
organic contaminants from ground water include the
packed column, the air supply equipment, and
repumping equipment. Annual costs per 1000 gal of
water treated are shown in Table 5-25. Figure 5-32
presents the cost data in a graphic format. The costs
are from 1982 and have been updated to 1986 using
the Engineering News Record Construction Index.
They are based upon preliminary designs for
achieving 90 percent removal of trichloroethyiene
(TCE).
Table 5-25.
Costs for Packed-Column Aeration
Water rate (106 gal/day) Cost per 1000 gal water treated (0)
0.1
1.0
30 to 50
15 to 20
Figure 5-32. Annual costs for air stripping system.
80
§ 40
1
I
Includes amortized
capital costs.
0.1
0.2 0.3 0.4 0.5 0.6 0.8
Capacity, 106 gal/day
1.0
5.7.2 Carbon Adsorption
5.7.2.1 General Description (EPA 1985a)
The process of adsorption onto activated carbon
involves contacting a waste stream with the carbon,
usually by allowing itto flowthrough a series of packed-
bed reactors. The activated carbon selectively ad-
sorbs organic constituents by a surface attraction phe-
nomenon in which organic molecules are attracted to
the internal pores of the carbon granules. These
systems are portable skid-mounted devices that can
be rapidly deployed, which makes them especially
attractive for use in gasoline releases from under-
ground storage tanks.
Adsorption depends on the strength of the molecular
attraction between adsorbent and adsorbate, mole-
cular weight, type and characteristic of adsorbent,
electrokinetic charge, pH, and surface area.
When the micropore surfaces become saturated with
organics, the carbon is "spent" and must either be
replaced with virgin carbon or be thermally regen-
erated and returned to service. The time it takes to
reach "breakthrough" or exhaustion is the single most
critical operating parameter. Carbon longevity bal-
anced against influent concentration governs oper-
ating economics.
Most ground-water treatment applications involve the
use of adsorption units that contain granular activated
carbon (GAG) and operate in a downflow series mode
such as that shown in Figure 5-33.
' In general, the downflow fixed-bed series mode has
proved to be the most cost-effective and to produce
the lowest effluent concentrations relative to other
carbon adsorber configurations (e.g., downflow in
parallel, moving bed, upflow-expanded). The units
may be connected in parallel to increase the hydraulic
capacity.
5.7.2.2 Application/Availability (EPA
1985a)
Activated carbon is a well-developed technology that
is widely used in the treatment of contaminated
ground water. It is especially well suited for removal of
mixed organics from contaminated ground water.
Carbon adsorption is essentially an electrical
interaction phenomenon; therefore, the polarity of the
contaminant compounds will determine its effec-
tiveness to a great extent. Highly polar molecules
cannot be removed effectively by carbon adsorption.
Another factor that is likely to affect the effectiveness
of carbon adsorption is aqueous solubility. The more
hydrophobic (insoluble) a molecule is, the more
readily the compound is adsorbed. Low-solubility
humic and fulvic acids that may be present in the
ground water can sorb to the activated carbon more
readily than most waste contaminants and result in
rapid carbon exhaustion.
Activated carbon is an effective and reliable means of
removing low-solubility organics, and it is suitable for
treating a wide range of organics of widely varying
concentrations. Some metals and inorganic species
also have shown excellent to good adsorption po-
tential, including antimony, arsenic, bismuth, chro-
mium, tin, silver, mercury, cobalt, zirconium, chlorine,
bromine, and iodine. The highest concentration of
solute in the influent stream that has been treated on
a continuous basis is 10,000 ppm total organic carbon
(TOC), and a 1 percent solution is currently con-
sidered as the upper limit (De Renzo 1978).
Adsorption is not particularly sensitive to changes in
concentrations or flow rate and, unlike biological
treatment, it is not adversely affected by toxics. It is,
however, quite sensitive to suspended solids and oil-
and-grease concentrations. Thus, pretreatment is re-
quired for oil and grease and suspended solids.
5-61
-------
Figure 5-33. Two-vessel granular carbon adsorption system (EPA 1979).
Feed Water
Regenerated/Makeup
Activated Carbon
Backwash Effluent
Backwash Feed
Adsorber 1
Adsorber 2
Regenerated/Makeup
Activated Carbon
• Backwash Effluent
Spent Carbon
Concentrations of oil and grease in the influent should
be limited to 10 ppm. Suspended solids should be
less than 50 pprn for upflow systems; downflow
systems can handle much higher solids loadings.
Activated carbon is easily implemented into more
complex treatment systems. The process is well
suited to mobile treatment systems as well as to onsite
construction. Space requirements are small, startup
and shutdown are rapid, and numerous contractors
are experienced in the operation of mobile units.
5.7.2.3 Design and Construction
Considerations (EPA I985a)
Carbon adsorption is frequently used following
biological treatment and/or granular media filtration.
These treatments reduce the organic and suspended
solids load on the carbon columns. It is also used to
remove refractory organics that cannot be bio-
degraded. Air stripping also may be applied prior to
carbon adsorption to remove a portion of the volatile
contaminants and thereby reduce the organic load to
the column. These pretreatment steps all minimize car-
bon regeneration costs.
Like air stripping, carbon adsorption systems are
relatively compact and easy to operate. On the other
hand, the phenomenon of adsorption is extremely
complex and not mathematically predictable. Field
pilot-plant studies are necessary for the accurate pre-
diction of the performance, longevity, and operating
economics of carbon adsorption. The following data
Backwash Feed
Valve Closed
Valve Open
5-62
need to be established during pilot-plant testing for an
initial estimate of carbon column sizing:
Hydraulic retention time (hours)
Flow (gallons/minute)
Hydraulic capacity of the carbon (gallons
waste/pound carbon)
Collected volume of treated ground water at
breakthrough (gallons)
Carbon density (pounds carbon/cubic foot)
The use of several carbon adsorption columns at a site
can provide considerable flexibility. The various col-
umns can be arranged in series to increase service life
between regeneration of the lead bed or in parallel for
maximum hydraulic capacity. The piping arrangement
should allow for one or more beds to be regenerated
while the other columns remain in service.
The most obvious maintenance consideration
associated with activated carbon treatment is the
regeneration of spent carbon for reuse. Regeneration
must be performed for each column at the conclusion
of its bed-life so the spent carbon can be restored as
close as possible to its original condition for reuse;
otherwise, the carbon must be disposed of. Other
operation and maintenance requirements for the ac-
tivated carbon technology are minimal if appropriate
automatic controls have been installed.
The thermal destruction properties of the contaminant
should be determined prior to selection of an ac-
-------
tivated carbon treatment technology, as any chemicals
sorbed to activated carbon must eventually be de-
stroyed in a carbon regeneration furnace. Therefore,
of crucial importance in the selection of activated
carbon treatment is whether the sorbed contaminant
can be effectively destroyed in a regeneration fur-
nace; if not, they will become air pollutants when
introduced into the furnace.
The biggest limitation to the use of the activated
carbon process is the high capital and operating cost.
The operating costs can be substantially reduced by
pretreatment of the waste.
5.7.2.4 Costs (EPA 19853)
The cost of activated carbon units depends on the
size of the contact unit, which is influenced by the
concentrations of the target and nontarget organic
compounds in the ground water and the desired level
of target compounds in the effluent. Table 5-26
presents construction, operation, and maintenance
costs for cylindrical, pressurized, downflow steel
contactors based on a nominal detection time of 17.5
minutes and a carbon loading rate of 5 gal/min per ftA
The construction costs include housing, concrete
foundation, and all the necessary pipes, valves, and
nozzles for operating the unit plus the initial change of
carbon. The operation and maintenance costs
include the electricity and assume carbon
replacement once a year; however, systems for
unloading spent carbon and loading fresh carbon are
not included. Figure 5-34 presents these costs in a
graphic format.
Figure 5-34. Estimated costs of various sizes of
Activated Carbon Adsorption Units.
Capacity
(gal/min)
1.7
17
70
175
350
Column
Diameter (It)
0.67
2
4
6.5
9
Column
Length (ft)
5
5
5
5
5
Housing
Area (ft2)
60
150
300
375
450
Construction
Costs* ($)
12,500
24,200
43,300
65,200
95,600
O&M
Costst
($/yr)
1,720
2,360
4,890
8,270
12,70
100
80
60
40
Construction
O&M
6810 20 40 60 100 200 40C
Capcity, gal/min
Table 5-26. General Cost Data for Various Sizes of
Activated-Carbon Adsorption Units*
* Data adapted from Hansen, Gumerman, and Gulp 1979.
t Costs ware updated from 1979 to 1986 dollars by using the second-quarter
Marshall and Swift Equipment Index.
Several manufacturers market mobile activated-
carbon-adsorption systems. For example, Calgon
Carbon Corporation has a trailer-mounted carbon
adsorption treatment unit that can be shipped to a
treatment site within 24 to 48 hours. The system can
be configured with either single or multiple prepiped
adsorber vessels and it can handle a flow of up to 200
gal/min. The following costs are associated with a
mobile system consisting of two 10-ft-diameter, 10-ft-
high, skid-mounted vessels capable of handling up to
200 gal/min (Calgon Corp., undated):
Delivery, supervision of installation and startup, $25,200
tests to conduct reactivation of carbon, disman-
tling and removal of system (including freight to
and from the site)
Delivery and removal of one truckload of carbon $15,300
(2000 Ib). (Two truckloads of carbon required
for a two-vessel system)
Rental fee
$5100/rrionth
Calgon Carbon Corporation will take spent carbon
back for reactivation. Otherwise, disposal costs for
spent carbon must be added.
5.7.3 Biological Treatment
5.7.3.1 General Description
The function of biological treatment is to remove
organic matter from the ground water through
microbial degradation. The most prevalent form of
biological treatment is aerobic, i.e., in the presence of
oxygen. Several existing biological treatment pro-
cesses may be applicable for the treatment of ground
water from an UST, including the conventional
activated sludge process. Modifications of the
activated sludge process include the use of pure
oxygen-activated sludge, extended aeration, and
contact stabilization; and fixed-film systems, which
include rotating biological discs and trickling filters.
5-63
-------
In a conventional activated-sludge process, ground
water flows into an aeration basin, where it is aerated
for several hours. During this time, a suspended ac-
tive microbial population (maintained by recycling
sludge) aerobically degrades organic matter in the
stream and produces new cells. A simplified equation
for this process is:
Organics + O2 -» CO2 + H2O + new cells
The new cells produced during aeration form a sludge
that is settled out in a clarifier. A portion of the settled
sludge is recycled to the aeration basin to maintain the
microbial population, and the remaining sludge is
wasted (i.e., it undergoes volume reduction and
disposal). Clarified water is disposed of or receives
further processing.
In the pure oxygen-activated sludge process, oxygen
and oxygen-enriched air is used instead of normal air
to increase the transfer of oxygen. Extended aeration
Involves longer detention times than conventional
activated sludge and relies on a higher population of
microorganisms to degrade contaminants. Contact
stabilization involves only short contact of the
contaminants and suspended microbial solids; the
sludge is then allowed to settle before it is treated to
remove sorbed organics. Fixed-film systems involve
contact of the contaminants with microorganisms
attached to some inert medium such as rock or some
specially designed plastic material. The original
trickling filter consists of a bed of rocks over which the
contaminated water is sprayed. The microbes that
form a slime layer on the rocks metabolize the or-
ganics, while oxygen (air) moves countercurrently to
the water flow (Canter and Knox 1985).
Biological towers are a modification of the trickling
filter. The medium [e.g., polyvinyl chloride (PVC), poly-
ethylene, polystyrene, or redwood] is stacked into
towers that typically reach 16 to 20 ft. The con-
taminated water is sprayed across the top, and air is
pulled upward through the tower as the water moves
downward. A slime layer of microorganisms forms on
the media and removes the organic contaminants as
the water flows over the slime layer.
A rotating biological contactor (RBC) consists of a
series of rotating discs (connected by a shaft) set in a
basin or trough. The contaminated water passes
through the basin, where the microorganisms
attached to the discs metabolize the organics present
in the water. Approximately 40 percent of the disc's
surface area is submerged. This allows the slime layer
to come in contact with the contaminated water and
the air alternately, which provides oxygen to the
microorganisms (Canter and Knox 1985).
5.7.3.2 Application/Availability
c
Biological treatment offers considerable flexibility k
because of the variety of available processes and the a
5-64
adaptability of the microorganisms themselves. Al-
though many organic chemicals are considered
biodegradable, the relative ease of biodegradation
varies widely. Several generalizations can be made
with regard to the ease of treatability of organics by
aerobic biological treatment:
Unsubstituted nonaromatics or cyclic hydro-
carbons are preferred over unsubstituted aro-
matics.
Materials with unsaturated bonds, such as
alkenes, are preferred over those with saturated
bonds.
Soluble organics are usually more readily
degraded than insoluble materials. Biological
treatment is generally efficient in removing
dissolved or colloidal materials, which are more
readily attacked by enzymes.. This is not the
case, however, for fixed-film treatment systems,
which preferentially treat suspended matter.
The presence of functional groups affects
biodegradability. Alcohols, aldehydes, acids,
esters, amides, and amino acids are more
degradable than corresponding alkanes, olefins,
ketones, dicarboxylic acids, nitriles, and
chloroalkanes.
Halogen-substituted compounds are the most
refractory to biodegradation;: chlorinated ali-
phatics are generally more refractory than the
corresponding aromatics. The number of halo-
gens and their position are also significant in
determining degradation.
Nitro-substituted compounds are also difficult to
degrade; however, they are generally less re-
fractory than the halogen-substituted com-
pounds.
Several compounds are considered relatively resistant
to biological treatment; however, the treatabiiity of
waste should be determined through laboratory
biological oxygen demand (BOD) tests on a case-by-
case basis.
Despite the fact that industrial-type wastes may be
refractory to biological treatment, microorganisms can
be acclimated to degrade many compounds that are
initially refractory. Similarly, whereas heavy metals are
inhibitory to biological treatment, the biomass also can
be acclimated, within limits, to tolerate elevated
concentrations of metals.
Table 5-27 presents the applications and limitations of
the available biological treatment processes. The
completely mixed activated-sludge process is the
most widely used process for treatment of
contaminated ground water with relatively high organic
loads; however, the high-purity oxygen system has
advantages for UST site corrective actions.
-------
Table 5-27. Summary of Applications/Limitations of
Biological Treatment Process*
Process Applications/Limitations
Conventional
Completely
mixed
conventional
Extended
aeration
Applicable to low-strength wastes; subject to
shock loads
Resistant to shock loads
Requires low organic load and long detention
times; low sludge volume; available as
packaged plant
Contact
stabilization
Pure oxygen
Trickling filters
Rotating
biological
disc
Not suitable for soluble BOD
Suitable for high-strength wastes; low sludge
volume; reduced aeration tank volume
Most effective for removal of colloidal and
suspended BOD; used primarily as a
roughing filter
Can handle large flow variations and high
organic shock loads; modular construction
provides flexibility to meet increased or
decreased treatment needs
* Data from EPA 1985a.
Other parameters that may influence the performance
of a biological treatment system include concentration
of suspended solids, oil and grease; organic load
variations; and temperature. Table 5-28 lists the
parameters that may affect system performance, their
limiting concentrations, and the kind of pretreatment
required prior to biological treatment.
Although biological treatment can effectively treat a
wide range of organics, it has several drawbacks in
LIST site applications. The reliability of the process
can be adversely affected by shock loads of toxics.
Startup time can be slow if the organisms need to be
acclimated to the contaminant, and the detention time
can be long for complex contaminants. The existence
of cultures that have been previously adapted to
ground-water contaminants, however, can dramatically
decrease startup and detention time.
Several cleanup contractors have used biological
treatment as part of a mobile treatment system. The
high-purity oxygen treatment process is well suited for
mobile treatment applications because the high
oxygen efficiency enables use of smaller reactors, a
shorter detention time, and less power consumption
relative to other activated sludge processes. A hazard
associated with the high-purity oxygen process is that
the presence of low-flash-point compounds can
5-
present a potential fire hazard; however, the system is
equipped with hydrocarbon analyzers and control
systems that deactivate the system when dangerously
high concentrations of volatiles are detected
(Ghassemi, Yu, and Quinlivan 1981). Loss of volatile
organics from other biological treatment processes
also can pose some localized air pollution and a health
hazard to field personnel. These systems may be
restricted when gasoline is the contaminant of
concern.
Rotating biological contactors also have advantages
for LIST site treatment. These compact units can
handle large flow variations and high organic shock
loads, and they do not require use of aeration
equipment.
Sludge produced in biological ground-water treatment
may be a hazardous waste itself because of the
sorption and concentration of toxic and hazardous
compounds contained in the wastewater. If the
sludge is hazardous, it must be disposed of in a RCRA-
approved manner. Nonhazardous sludge should be
disposed of in accordance with State sludge disposal
guidelines.
5.7.3.3 Design and Construction
Considerations
Design of the activated sludge or fixed-film systems
for a particular application are best achieved by first
representing the system as a mathematical model, and
then determining the necessary coefficients through
laboratory or pilot tests.
The following models have been found to be reliable
for use in the design of biological treatment systems
for ground water containing priority pollutants (Canter
and Knox 1986).
Activated sludge:
FSj/X
U
max
-K,
B
Sj-Se
Biological tower and rotating biological contactor:
FSi
U
max
-K,
where
B
7
V = volume of aeration tank, ft3
F = flow rate, ftj/day
X = mixed liquor volatile solids, mg/liter
S; = influent BOD, COD, TOC, or specific
organics, mg/liter
65
-------
Table 5-28. Concentration of Pollutants That Make Prebiological
Primary Treatment Advisable
or
Pollutant or
System Condition
Suspended solids
Oil or grease
Toxic ions
Pb
Cu±Ni+CN
Cr43+
pH
Alkalinity
Limiting Concentration
>50to 125 mg/liter
>35 to 50 mg/liter
£0.1 mg/liter
S1 mg/liter
<3 mg/liter
S10 mg/liter
<6, >9
0.5 Ib alkalinity as
Kind of Treatment ;
Sedimentation, flotation,
or lagooning
Skimming tank or separator '
Precipitation or ion exchange
Neutralization
Neutralization for excessive
Acidity
Organic load variation
Sulfides
Phenols
Ammonia
Dissolved salts
Temperature
CaCO3/lb BOD removed
Free mineral acidity
>2:1 to 4:1
>100 mg/liter
>70 to 300 mg/liter
>1.6g/liter
>10to 1 eg/liter
13° to 38°C in reactor
alkalinity
Neutralization
Equalization
Precipitation or stripping
with recovery
Extraction, adsorption, or
internal dilution
Dilution, ion exchange, pH
adjustment, or stripping
Dilution or ion exchange
Cooling or steam addition
Data from Conway and Ross 1980.
5-66
-------
Se = effluent BOD, COD, TOC, or specific
organics, mg/liter
Umax and KB = biokinetic constants, day-'
A = surface area of biological tower or
rotating biological contactor, ft2
After the biokinetic constants are determined by. con-
ducting laboratory or pilot-plant studies, the required
volume of the aeration tank or the required surface
area for a biological tower or rotating biological con-
tactor can be determined for any flow rate; the influent
concentration of BOD, COD, TOC, or specific organic;
and the required.effluent concentration of BOD, COD,
TOC, or specific organic.
The design approaches of conventional, completely
mixed, activated sludge systems are well established.
The considerations are rate-limiting and establish the
size of the reaction basin and sludge settling zone
(i.e., the solids settling rate and the BOD and organic
removal rates).
Biologically degradable organics can be in solution or
in the form of a paniculate solid. Only soluble organics
can be absorbed and metabolized by activated sludge
microorganisms. The rate of metabolism is a function
of time, the concentration gradient, and toxicity. Solid
degradable organics are removed by adsorption on
the activated sludge microorganisms. A relationship
between the sludge retention time and the
biodegradable fraction remaining can be established
for any given waste. To reproduce and function
properly, the microorganisms .must have a source of
energy, carbon, and nutrients. The addition of
phosphoric acid to the wastewater is usually
necessary to maintain a proper carbon/nutrient ratio
for biooxidation. The effluent values of biodegradable
pollutants may be calculated by the following formula:
Se - So/11 + (Kp)(M.LVSS)(HRT)]
where Se = effluent pollutant, mg/liter
S0 = influent pollutant, mg/liter
Kp = average pollutant removal rate
coefficient for specified
temperature, hour -i
MLVSS = mixed liquor volatile suspended
solids, mg/liter
HRT = hydraulic retention time, hours
As the equation indicates, the longer it takes the
microorganisms to metabolize a given pollutant, the
higher the MLVSS required or the longer the
hydraulic retention time needed to reach the effluent
limitations.
Solids (sludge) removal and disposal depend largely
on the efficiency of the solid/liquid separation phase
70
140
350
694
79,300
86,500
108,000
162,000
4,400
6,600
10,200
16,000
of the treatment system. Solids will accumulate in the
activated sludge system unless a portion is wasted.
This accumulation results from 1) the removal of inert
materials, and 2) production of cellular material
through microorganism synthesis. Because clarifier
(sedimentation) efficiency is related both to its over-
flow rate and to the settling velocity of the sludge
entering it, the separation system is usually designed
for peak flow conditions.
5.7.3.4 Costs
Costs of various sizes of activated sludge units are
presented in Table 5-29. The costs for these units
assume a detention time of 3 hours and the use of
aeration basins, air-supply equipment, piping, and a
blower building. Clarifier and recycle pumps are not
included. The basins are sized to 50 percent recycle
flow. The influent BOD is assumed to be no greater
than 130 ppm, and the effluent BOD is assumed to be
40ppm.
Table 5-29. General Cost Data for Various Sizes of
Activated SludgeTreatment Units*
Capacity (gal/min) Construction Costst($) O&M Costst($/yr)
* Data adapted from EPA 1980.
t Costs were updated from 1978 to 1986 dollars by using the
second-quarter Marshall and Swift Equipment Index.
The operation and maintenance costs assume that
the hydraulic head loss through the aeration tank is
negligible. Sludge wasting and pumping energy are
not included.
Union Carbide manufactures a high-purity-oxygen-
activated sludge system (UNOX) suitable for mobile
system applications. The mobile UNOX systems have
a hydraulic capacity of 5 to 40 gal/min, are contained
within 40-ft van trailers, and include an external
clarifier. The oxygen required is also supplied by
Union Carbide. The customer is expected to provide
installation labor, operating manpower, analytical sup-
port, and utilities. A typical installation requires 3 or 4
days (Ghassemi, Yu, and Quinlivan 1981).
The mobile UNOX system can be rented or purchased
from the Union Carbide Corporation. Estimated rental
costs are as follows:
$6600 for the checkout and refurbishment of
equipment to make it operational
$560/day for onsite service, including
engineering consultation on program planning
and execution
$10/day for rental of equipment
5-67
-------
Transportation charges to get the equipment
from the manufacturer to the site of operation
and back again.
The purchase price of a UNOX mobile unit is between
$262,000 and $333,000 (Ghassemi, Yu, and
Quinlivan 1981, updated using 1986 second-quarter
Marshall Swift Index).
5.7.4 Precipltatlon/Flocculatton/Sedimentation
5.7.4.1 General Description
Precipitation is a physiochemical process whereby
some or all of a substance in solution is transformed
into a solid phase. The process is based on alteration
of the chemical equilibrium relationships affecting the
solubility of inorganic species. Removal of metals as
hydroxides or sulfides is the most common precip-
itation application in ground-water treatment. Gen-
erally, lime or sodium sulfide is added to the con-
taminated ground water in a rapid-mix tank along with
flocculating agents. The contaminated ground water
flows to a flocculation chamber in which adequate
mixing and retention time is provided for ag-
glomeration of precipitate particles. Agglomerated par-
ticles are separated from the liquid phase by settling in
a sedimentation chamber and/or by other physical
processes, such as filtration. Figure 5-35 illustrates a
typical configuration for the precipitation, flocculation,
and sedimentation processes.
Although precipitation of metals is governed by the
solubility product of ionic species, in practice, effluent
concentrations equal to the solubility product are
rarely achieved. Because of the common ion effect,
the amount of lime added is usually about three times
the stoichiometric amount that would be added to
reduce solubility. Figure 5-36 gives solubilities of
various metal hydroxides and sulfides at various pH
levels. The metal sulfides have significantly lower
solubility than their hydroxide counterparts, and more
complete precipitation is achieved. Metal sulfides are
also stable over a broad pH range. Many metal
hydroxides, on the other hand, are stable only over a
narrow pH range; metals reach a minimum solubility at
a specific pH, but further addition of lime causes the
metal to become soluble again. Therefore, accurate
control of lime dosages is needed.
Figure 5-36. Solubility of metal hydroxides and
sulfides (Ghassemi, Yu, and Quinlivan).
102
10°
I 10-2
E
i io~4
5 10-6
"5
1 10-8
o
10-
10'
,-10
Pb(OH)2 |
'CR(OH)3
Zn(OH)2
Ag(OH)
Cu(OH)2|
Ni(OH)2|
Cd(OH)2
PbS
Ag2S
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
PH
Flocculation refers to the process by which small,
unsettleable particles suspended in a liquid medium
are made to agglomerate into larger, more settleable
Figure 5-35. Representative configuration employing precipitation, flocccuiation,
and sedimentation (DeRenzo 1978).
Precipitation
Flocculation
Sedimentation
Precipitation
Chemicals
Flocculating
Agents
Inlet
Liquid
Stream
Rapid Mix
Tank
Outlet I
Liquid
Stream
Flocculation
Chamber
Sedimentation
Basin
5-68
-------
particles. The mechanisms by which flocculation oc-
curs involve surface chemistry and particle change
phenomena. In simple terms, these various phe-
nomena can be grouped into two sequential mech-
anisms (Kiang and Metry 1982):
Chemically induced destabilization of the req-
uisite surface-related forces, which allows par-
ticles to stick together when they touch
Chemical bridging and physical enmeshment
between the now nonrepelling particles, which
allows for the formation of large particles
Flocculation involves three basic steps: 1) addition of
the flocculation agent to the waste stream, 2) rapid
mixing to disperse the flocculating agent, and 3) slow
and gentle mixing to allow for contact between small
particles. Typically, chemicals used to promote floc-
culation include alum, lime, various iron salts (ferric
chloride, ferrous sulfate), and organic flocculating
agents, often referred to as "polyelectrolytes." These
materials generally consist of long-chain, water-
soluble polymers such as pplyacrylamides. They are
used either in conjunction with the inorganic
flocculants (such as alum) or as the primary flocculating
agent. A polyelectrolyte may be termed cationic,
anipnic, or ampholytic (depending on the type of
ionizable groups), or nonionic if it contains no
ionizable groups. The range of physical/chemical char-
acteristics (e.g., density, viscosity, toxicity, and
molecular weight) of the several thousand available
polymers is extremely broad.
The inorganic flocculants (such as alum, lime, or iron
salts) make use of precipitation reactions. Alum
(hydrated aluminum sulfate) is typically added to
aqueous waste streams as a solution. Upon mixing,
the slightly higher pH of the water causes the alum to
hydrolyze and form fluffy, gelatinous precipitates of
aluminum hydroxide. These precipitates, partially be-
cause of their large surface area, enmesh small
particles and thereby create larger particles. Lime and
iron salts also have a tendency to form large fluffy
precipitates or "floe" particles. Many precipitation
reactions, such as the precipitation of metals from
solution by the addition of sulfide ions, do not readily
form floe particles; rather, they precipitate as very fine
and relatively stable colloidal particles. In such cases,
flocculating agents such as alum and/or poly-
electrolytes must be added to cause flocculation of
the metal sulfide precipitates (Canter and Knox 1985).
After suspended particles have been flocculated into
larger particles, they usually can be removed from the
liquid by sedimentation, if a sufficient density
difference exists between the suspended matter and
the liquid.
Sedimentation relies on gravity to remove suspended
solids. The fundamentals of a sedimentation process
include (Kiang and Metry 1982):
5
•-'69
A basin or container of sufficient size to maintain
the liquid to be treated in a relatively quiescent
state for a specified period of time
A means of directing the liquid to be treated into
the basin mentioned above in a manner con-
ducive to settling
A means of physically removing the settled
particles from the liquid (or liquid from the settled
particles)
Sedimentation can be carried out either as a batch or
as a continuous process in lined impoundments,
conventional settling basins, clarifiers, and high-rate
gravity settlers. Modified above-ground swimming
pools often have been used for sedimentation in
temporary, short-term treatment systems at hazardous
waste sites. Figure 5-37 shows three different design
configurations for sedimentation. In sedimentation
ponds, the liquid is merely decanted as the particles
accumulate on the bottom of the pond. Backhoes,
draglines, or siphons are used periodically to remove
settled solids. Sedimentation basins and clarifiers
usually include a built-in solids collection and removal
device such as a sludge scraper and draw-off
mechanism. Sedimentation basins are generally
rectangular, normally use a belt-like collection
mechanism, and are used primarily for removal of truly
settleable particles from liquid.
Clarifiers are normally circular and are used in
applications involving precipitation and flocculation as
well as sedimentation. Many clarifiers are equipped
with separate zones for chemical mixing and
precipitation, flocculation, and sedimentation (Kiang
and Metry 1982).
5.7.4.2 Application/Availability
Precipitation is applicable for removal of most metals
from ground water, including zinc, cadmium, chro-
mium, copper, fluoride, lead, manganese, and mer-
cury. Certain anionic species (such as phosphate,
sulfate, and fluoride) also can be removed by pre-
cipitation. Although precipitation is useful for most
contaminated ground-water streams, limitations may
be imposed by certain physical or chemical char-
acteristics. |n some cases, organic compounds may
form organometallic complexes with metals that could
inhibit precipitation. Cyanide and other ions in the
ground water also may complex with metals, and make
treatment by precipitation less efficient.
Flocculation is applicable where particles must be
agglomerated into larger more settleable particles prior
to sedimentation or other types of treatment. There is
no concentration limit for precipitation or flocculation.
Sedimentation is commonly applied whenever high
suspended solid loadings are encountered. Sed-
imentation is also required as a pretreatment step for
many chemical processes, including carbon ad-
-------
Figure 5-37. Representative types of sedimentation (DeRenzo 1978).
Inlet Liquid
Settling Pond
Overflow Discharge Weir
Accumulated Settled
Particles Periodically
Removed by
Mechanical Shovel
Inlet Zone
Inlet Liquid
N
Settled Particles Collected
and Periodically Removed
Baffles to Maintain
Quiescent Conditions
Settling Particles Trajectory
Outlet Zone
|»>Outlet Liquid
Belt-Type Solids Collection Mechanism
Settling
Zone -
Inlet Liquid •
Sedimentation Basin
Circular Baffle
Revolving Collection
Mechanism
Annular Overflow Weir
Inlet Zone
* \
TFF^
-**
T
/ I
Liquid ,,.'"1'
Flow r
Sludge Drawoff
i
•^"Outlet
Liquid
Settling
"Particles
Settled Particles Collected
and Periodically Removed
Circular Clarifier
5r70
-------
sorption, ion exchange, air stripping, reverse osmosis,
and filtration. This technology is applicable to the re-
moval of suspended solids heavier than water. Sus-
pended oil droplets or oil-soaked particles may not
settle out and have to be removed by some other
means. Some sedimentation units are fitted with skim-
mers to remove oil and grease that float to the water
surface; however, these would not be effective in re-
moving emulsified oils.
5.7.4.3 Design and Construction
Considerations
Selection of the most suitable precipitate or flocculant
and their optimum dosages are determined through
laboratory-jar test studies. In addition to determining
the appropriate chemicals and optimum chemical dos-
ages, the following important parameters need to be
determined as part of the overall design (Canter and
Knox1985):
Most suitable chemical addition system
Optimum pH requirement
Rapid mix requirements
• Sludge production
Sludge flocculation, settling, and dewatering
characteristics
The system is relatively simple. The process requires
only chemical pumps, metering devices, and mixing
and settling tanks, all of which are readily available and
easy to operate. Precipitation and flocculation can be
easily integrated into more complex treatment sys-
tems, and they pose minimal safety and health haz-
ards to field workers. The entire system is operated at
near ambient conditions, which eliminates the danger
of high-pressure/high-temperature operations. The
chemicals employed are often skin irritants, but they
can easily be handled in a safe manner.
Sedimentation is frequently considered in terms of
ideal settling. The ideal settling theory results in the
following equation for surface loading or overflow rate:
where V0 = settling velocity
Q = flow through the basin
A = surface area of the basin
Sedimentation basin loadings are often expressed in
terms of gallons per day per square foot. Thus, under
ideal settling conditions, sedimentation is in-
dependent of basin depth and detention time and
depends only on the flow rate, basin surface area, and
properties of the particle. In reality, however, sedi-
mentation does not perform according to ideal
settling conditions because settling is affected by
5-71
such conditions as turbulence and bottom scour.
Therefore, removal of particles actually does depend
on basin depth and detention time as well as flow rate,
surface area, and particle size. The performance of a
sedimentation basin on suspended discrete particles
can be calculated, but it is impossible to calculate
sedimentation basin performance on suspended floc-
culating particles (such as in wastewater) because
settling velocities change continually. Laboratory set-
tling tests, however, can be used to predict sedi-
mentation basin performance.
Sedimentation provides a reliable means of removing
suspended matter from ground water, if the sus-
pended matter is settleable and the treatment process
(including the use of flocculants/coagulants) has been
appropriately designed from laboratory settling tests.
Most clarifiers are capable of removing 90 to 99
percent of the suspended solids. The process is
somewhat space-intensive, however, and its use may
be limited by the available space at the site.
5.7.4.4 Costs
Figure 5-38 shows capital and operating costs for a
flocculation system, including chemical storage,
chemical feeding, and rapid mix. A polymer dosage of
1 mg/liter at 0.25 percent solution is assumed.
Figure 5-38. Capital and annual O&M costs for
flocculation (U.S. EPA 1982).
20
15
10
_I
Note: It is coincidental that
capital and O&M costs
are identical.
I
50
100
Capacity, gal/min
200
300
Construction costs include piping and a building to
house the feeding equipment and bag storage. For a
1000 gal/day or smaller plant, feeding is done
manually. Costs include two systems of tanks and
feeders. For a 10,000 gal/day plant, the cost includes
-------
two solution feeders and mix tanks. For a 100,000
gal/day plant, costs include four feeders and mixing
tanks, two holding tanks, and ten solution feeders.
The rapid-mix tank is concrete and is equipped with a
stainless steel mixer and handrails. No separate
building is required for a 100 gal/day plant. Operation
of the feeder, mix tank, solution, and holding tank is
manual.
5.7.5 Dissolved Air Flotation
5.7.5.1 General Description
In dissolved air flotation (DAF), highly-pressurized air
forms bubbles that remove suspended solids. A por-
tion or all of the ground-water feed or a portion of re-
cycled effluent is saturated with air at a pressure of 25
to 70 Ib/in.z (gauge). The ground water is held at this
pressure for 0.5 to 3.0 minutes in a retention tank and
then released to the flotation chamber. The sudden
reduction in pressure results in the release of
microscopic air bubbles in the flotation chamber,
which attach themselves to oil and suspended par-
ticles in the ground water. This results in agglom-
eration that, because of the entrained air, results in
greatly increased vertical rise rates of about 0.5 to 2.0
ft/min. The floated materials rise to the surface to form
a froth layer, which is continually removed by specially
designed flight scrapers or other skimming devices.
The retention time in the flotation chamber is usually
about 20 to 60 minutes. The effectiveness of dis-
solved air flotation depends on the attachment of the
bubbles to the suspended oil and other particles that
are to be removed from the contaminated ground
water. The attraction between the air bubble and
particle results primarily from the particle surface char-
Figure 5-39. Dissolved air flotation system (EPA 1980).
ges and bubble-size distribution. The more uniform
the distribution of water and microbubbles, the shal-
lower the flotation unit can be. Generally, the depth of
effective flotation units is between 4 and 9 feet.
In certain cases, the surface sludge layer can attain a
thickness of many inches and remain relatively stable
for a short period. The layer thickens with time, but un-
due delays in removal will cause a release of par-
ticulates back into the liquid.
Equipment normally associated with an air gas flotation
system includes dissolved air flotation units, air com-
pressors, and skimmers. A flow diagram of an air
flotation system is shown in Figure 5-39.
5.7.5.2 Application/Availability
A DAF system is used to remove lighter suspended
materials whose specific gravity is only slightly in
excess of 1.0. They normally are used to remove oil
and grease, but they are sometimes used when
existing clarifiers are overloaded hydraulically because
converting to DAF requires less surface area.
Used for many years to treat industrial wastewaters,
DAF also is commonly used to thicken sludges,
including those generated by municipal wastewater
treatment. It is not widely used to treat municipal
wastewaters, however. Significantly modified systems
use pressurized raw waste and pressurized recycle. In
addition, gases other than air can be used. For ex-
ample, the petroleum industry has used nitrogen, with
closed vessels, to reduce the possibilities of fire.
Although proven to be reliable, DAF units are subject
to variable influent conditions that result in widely
Sludge Removal Mechanism
T
a-... ,5,w.x^x::.x.:,:;ss*,.v,rfssg;i;wxv.. Sludge Blanket
«VVX**V*»:«+:-*XXWfXXXXX:&,. ....°.
Flow Zone
Sludge Discharge
Recirculation /f\ Recycle Flow
Pump
Air Feed
f
Reaeration Pump
5-72
Retention Tank
Air Dissolved
-------
varying performance. Very little use of land is re-
quired, and air released from the unit is unlikely to con-
tain volatile organic material.
Chemicals are normally added to aid in the coagulation
of colloidal solids and to break emulsions. Such chem-
ical additions include alum, ferric chloride (FeCh),
lime, ancl polymer, which can be added prior to the
actual flotation step.
5.7.5.3 Design and Construction
Considerations
Typical design criteria for gas flotation systems are as
follows:
Design criterion
Pressure
Air-to-solids ratio
Float detention
Surface hydraulic loading
Recycle (where employed)
Solids loading
Value
25 to 70 psig
0.01 to 0.1 Ib/lb
20 to 60 min
500 to 8000 gal/day
per fta
5 to 120 percent
0.5 to 5 Ib/ft2 per hour
Typical design criteria for dissolved air flotation with
chemical addition of alum, FeCI3, lime, and polymers
are as follows:
1) Alum addition is determined by jar testing and
generally is 5 to 20 mg/liter as aluminum.
2) Ferric chloride addition is determined by jar
testing and 20 to 100 mg/liter is common.
3) Typical lime additions are shown below:
Approximate
lime addition
pH fmg/litert fas CaCK
9.5 185
10.5 270
9.5 230
Feed water alkalinity
(mg/liter) (as CaCCU
300
300
400
400
10.5 380
4) Polymer addition is determined by jar testing.
The materials contacting polymer solutions
should be Type 316 stainless steel, FRP, or
plastic. The solutions should be stored in a cool
and dry place, and storage time should be
minimal. Viscosity must be considered in the
feeding system design.
The materials of construction of the chemical feed
equipment, chemical storage area, and mixers require
special attention because of the corrosive nature of
the materials handled. Stainless steel or other cor-
rosion-resistant materials should be selected, de-
pending on each application.
5.7.5.4 Costs
Figure 5-40 indicates the total capital and annual
operating costs for dissolved air flotation systems.
5-73
Figure 5-40. Capital and annual O&M costs for
dissolved air flotation.
2000
1500
§"1000
I
I 300
o>
o 400
1 30°
1
O 200
100
40
O&M
10 20 30 40 60 100
Capacity, gal/min
200 30C-
The cost estimates assume the following design
characteristics:
Operating characteristic Assumed value
Air injection
Recycle
Float detention time
Surface hydraulic loading
1.25ftV1000gal
33 percent
25 min
500 to 8000
gal/day per \\?
Capital costs and annual operating costs will be
greater for air flotation with chemical addition than for
air flotation alone because of the required additional
equipment, installation, and chemical costs. The ad-
ditional equipment includes chemical feed equip-
ment, a rapid-mix tank, a chemical storage facility, and
a stainless steel mixer. Specific chemical addition
capital and operating cost data can be found in the
Treatability Manual, Volume IV. Cost Estimating (FPA
600/8-80-042d, July 1980).
5.7.6 Granular Media Filtration
5.7.6.1 General Description
Granular media filtration of contaminated ground water
is a physical process whereby suspended solids and
colloidal impurities are removed from solution by
forcing the liquid through a porous granular medium.
The filter media consists of a fixed bed of granular
particles (typically sand or sand with finely ground
anthracite) (Figure 5-41). The bed is contained within
a basin and is supported by an underlain system
(typically perforated pipes) that allows the filtered
liquid to be drawn off while the filter media is retained
in place. The bed may be operated under a typical
hydraulic head of 2 to 4 ft of water, or the entire basin
may be enclosed and operated under higher
-------
Figure 5-41. Example granular media filtration bed.
Backwash.
Drain
Air -
Diflusor
High Head
Backwash
-Trough
I Raw Feed
Under Drain
Air
, Backwash
Effluent
pressures. As water laden with suspended solids
passes through the media bed, the particles become
trapped on top of and within the bed. As the bed be-
comes loaded with solids, the filtration rate decreases
as a result of increased pressure drop through the
bed. Plugging is prevented by intermittently back-
flushing the filter with water at high velocity to dislodge
the particles. The backwash water must be treated fur-
ther because it contains high concentrations of solids
(De Renzo 1978). The backwash system may be sup-
plemented with an air-diffusion system.
5.7.6.2 Application/Availability
Granular media filtration can be used to handle ground
water containing less than about 200 rng/liter sus-
pended solids, depending on the required effluent
level. Greater suspended solids loading will reduce
run lengths and require excessively frequent back-
washing (De Renzo 1978). The suspended solids
concentration of the effluent depends largely on
particle size distribution, but granular media filters
usually are capable of producing a filtered liquid with a
suspended solids concentration as low as 1 to 10
mg/liter. Large flow variations have a deleterious effect
on effluent quality.
Granular media filters are often preceded by sedi-
mentation to reduce the suspended solids load on
the filter (De Renzo 1978). Granular media filtration
also is frequently installed ahead of biological or
activated carbon treatment units to reduce the sus-
pended solids load and, in the case of activated
carbon, to minimize plugging of the carbon columns
(De Renzo 1978).
The granular media filtration process is only marginally
effective in treating colloidal-size particles. These
particles often can be made larger by flocculation, but
this generally reduces run lengths. When such
particles cannot be flocculated (as in the case of many
oil/water emulsions), more advanced techniques
(e.g., ultrafiltration) may be appropriate (De Renzo
1978).
5-74
5.7.6.3 Design and Construction
Considerations
The composition and sizing of the filtration bed is an
important design consideration. Beds frequently used
are as deep as 4 ft and composed of sand with a size
distribution of about 10 percent less than 0.5 mm and
90 percent less than 1 mm (U.S. EPA 19i35a). Sand
may be mixed with slightly larger sized anthracite, and
the entire bed may be supported on 3/4- to 1/2-in.
gravel on the underdrain system; however, deep-bed
filters are also available. It is recommended that pilot-
plant studies be conducted to determine the optimum
size and combination of filter material, filtering rates,
and filtering schedule.
A filter bed can function properly only if the back-
washing system is effective in cleaning the material
from the filter. Methods that can be used for back-
washing include water backwash alone, water back-
wash with auxiliary surface-waterwash, water wash pre-
ceded by air scour, and simultaneous air and water
wash.
The backwash cycle is usually automated, and the
duration of the backwash is about 20 minutes per
cycle. Backwash water, which amounts to 1 to 5
percent of the total flow, can then be routed to a
clarifier via a storage vessel to allow flow equalization.
Several filters may be used in parallel to allow con-
tinuous processing during backwashing.
Filtration is a reliable and effective means of removing
low levels of solids from wastes if, the solids content
does not vary greatly ancl the filter is backwashed at
appropriate intervals. Filtration equipment is relatively
simple, readily available in a wide range of sizes, and
easy to operate and control. Filtration is also easily
integrated with other treatment steps.
Because of its small space requirements and relatively
simple operation, filtration is well suited for mobile
treatment systems as well as onsite construction,
Granular media filters have been used extensively at
hazardous waste sites.
The EPA physical/chemical treatment system, which
has been in operation for more than 9 years,
incorporates three "dual" media (sand-anthracite) fil-
ters connected in parallel in its treatment train. The
filters are designed for a maximum hydraulic loading of
7 gal/min per ft2 or 67 gal/min (Ghassemi, Yu, and
Quinlivan 1981). Also,"several manufacturers pro-
duce packaged plant systems suitable for trailer moun-
ting.
The most obvious maintenance consideration with
granular media filtration is the handling of the
backwash, which generally will contain high con-
centrations of contaminants and require subsequent
treatment.
-------
5.7.6.4 Costs
Capital costs for relatively small granular media filters
with capacities of about 300 gal/min are not readily
available. These costs would vary with the construc-
tion materials, depth of media, and filtration rate. The
approximate capital cost for small, open, coarse-media
filters is $400/ft,2 of surface area.* Operation costs in-
clude electricity to pump the feed and effluent
streams, treatment chemical, if any, and filter media
replacement.
5.7.7 Ion Exchange/Resin Adsorption
5.7.7.1 General Description
Ion exchange is a process whereby the toxic ions are
removed from the aqueous phase by being
exchanged with relatively harmless ions held by the
ion exchange material. Modern ion exchange resins
are primarily synthetic organic materials that contain
ionic functional groups to which exchangeable ions
are attached.
These synthetic resins are structurally stable (i.e., they
can tolerate a range of temperature and pH
conditions), exhibit a high exchange capacity, and can
be tailored to show selectivity toward specific ions.
Exchangers with negatively charged sites are cation
exchangers because they take up positively charged
ions. Anion exchangers have positively charged sites
and therefore take up negative ions. The exchange
reaction is reversible and concentration-dependent,
and it is possible to regenerate the exchange resins
for reuse. Sorptive (macroporous) resins are also avail-
able for removal of organics, and the removal mech-
anism is one of sorption rather than ion exchange
(Ghassemi, Yu, and Quinlivan 1981).
5.7.7.2 Application/Availability
Ion exchange is used to remove a broad range of ionic
species from water, including:
All metallic elements when present as soluble
species, either anionic or cationic.
Inorganic anions such as halides, sulfates,
nitrates, and cyanides.
Organic acids such as carboxylics, suifonics, and
some phenols at a pH sufficiently alkaline to form
the ions.
Organic amines when the solution acidity is
sufficiently acid to form the corresponding acid
salt (De Benzol 978).
Sorptive resins can remove a wide range of polar and
nonpolar organics.
* Extrapolated from data on large-sized filters in Gulp,
Wesner, and Gulp 1978.
5-75
A practical upper concentration limit for ion exchange
is about 2500 to 4000 mg/Iiter. A higher concen-
tration results in rapid exhaustion of the resin and
inordinately high regeneration costs. Suspended
solids in the feed stream should be less than 50
mg/Iiter to prevent plugging the resins, and waste
streams must be free of oxidants (De Renzo 1978).
5.7.7.3 Design and Construction
Considerations
Specific ion exchange and sorptive resin systems
must be designed on a case-by-case basis. Of the
three major operating models (fixed-bed cocurrent,
fixed-bed countercurrent, and continuous counter-
current), fixed-bed countercurrent systems are the
most widely used. Figure 5-42 illustrates the fixed-
bed countercurrent and continuous countercurrent
systems. The continuous countercurrent system is
suitable for high flows. Complete removal of cations
and anions (demineralizationj can be accomplished by
using the hydrogen form of a cation exchange resin
and the hydroxide form of an anion exchange resin.
For removal of organics as well as inorganics, a
combination adsorptive/demineralization system can
be used. In this system, lead beds carry sorptive
resins that act as organic scavengers, and the end
beds contain anion and cation exchange resins. The
use of different types of adsorptive resins (e.g., polar
and nonpolar) permits removal of a broad spectrum of
organics (Ghassemi, Yu, and Quinlivan 1981).
Ion exchange is a well-established technology for
removal of heavy metals and hazardous anions from
dilute solutions. Ion exchange can be expected to
perform well for these applications when fed wastes of
variable composition, provided the system's effluent is
continually monitored to determine when resin bed
exhaustion has occurred. As mentioned previously,
however, the reliability of ion exchange is markedly
affected by the presence of suspended solids. The
use of sorptive resins is relatively new, and reliability
under various conditions is not as well known.
Ion exchange systems are commercially available from
several vendors. The units are relatively compact and
are not energy-intensive. Startup or shutdown can be
accomplished easily and quickly (Ghassemi, Yu, and
Quinlivan 1981). These features allow for convenient
use of ion exchange and sorptive resin systems in
mobile treatment systems.
Although exchange columns can be operated
manually or automatically, manual operation is better
suited for application at hazardous waste sites
because of the diversity of wastes encountered. With
manual operation, the operator can decide when to
-------
Figure 5-42. Pertinent features of Ion exchange systems (Chemical Separations Corporation, no date).
Types
Countercurrent Fixed Bed
Service Regeneration
Continuous Countercurrent
Service
Regeneration
Resin
Flow
Description of Regeneration flows opposite in direction
Process to influent. Backwash (in regeneration)
does not occur on every cycle to pre-
serve resin stage heights. Resin bed is
locked in place during regeneration.
Indications Handles high loads at moderate through-
for Use put or low loads at high throughput (gpm
x TDS or gpm x ppm removal = 40,000 or
more). Where effluent quality must be
relatively constant, regeneration cost is
critical. Disposal of single batch waste
volume is no problem.
Advantages Moderate capital cost. Can be operated
with periodic attention. Moderate re-
generation cost. Lesser volume of waste
due to less frequent backwash. Consistent
effluent quality.
Disadvantages More controls and instrumentation and
higher cost. Requires mechanism to lock
resin bed. Large single batches of waste
disposal. Moderate water consumption
through dilution and waste. Requires sub-
stantial floor space.
Multistage Countercurrent movement of
resin in closed loop, providing simultaneous
treatment, regeneration, backwash, and
rinse. Operation is only interrupted for
momentary resin pulse.
High loads with high throughputs (gpm x
TDS or gpm x ppm removal = 40,000 or
more). Where constant effluent quality
is essential, regeneration costs are critical.
Total waste volume requires small concentrated
stream to be controllable. Where loss of product
through dilution and waste must be minimized.
Where available floor space is limited.
Lowest regeneration cost. Lowest resin
inventory. Consistent effluent quality.
Highest throughput to floor space. Large-
capacity units factory preassembled. Con-
centrated low-volume waste stream. Can
handle strong chemical solutions and slurry.
Fully automatic operation.
Requires automatic controls and instrumentation,
higher capital cost. More headroom required.
stop the service cycle and begin the backwash cycle;
however, this requires a skilled operator familiar with
the process (Ghassemi, Yu, and Quinlivan 1981).
The use of several exchange columns at a site can
provide considerable flexibility. As described pre-
viously, various resin types can be used to remove
anions, cations, and organics. Also, various columns
can be arranged in series to increase service life
between regeneration of the lead bed or in parallel for
maximum hydraulic capacity. The piping arrangement
allows for one or more beds to be taken out for re-
generation while the other columns remain in service
(Ghassemi, Yu, and Quinlivan 1981).
Consideration must be given to the disposal of
contaminated ion exchange regeneration solution.
Another important operational consideration involves
the selection of regeneration chemicals to ensure the
compatibility of the regenerating chemical with the
ground water being treated. For example, the use of
nitric acid to regenerate an ion exchange column
containing ammonium ions would result in the for-
mation of ammonium nitrate, a potentially explosive
compound.
5.7.7.4 COStS
Costs of various sizes of ion exchange units are
presented in Table 5-30 and Figure 5-43. The con-
struction costs assume fabricated-steel contact ves-l
sels with baked phenolic linings, a resin depth of 6 ft,
housing for the columns, and all piping and backwash
facilities.
5-76
-------
Table 5-30. General Cost Data for Various
Sizes of Ion Exchange Units*
Plant Capacity
(gal/min)
50
195
305
438
597
Construction
Costt ($)
85,600
118,400
137,300
156,800
183,500
O&M
Costs* ($/yr)
14,800
21,700
24,700
28,100
32,100
Adapted from Hansen, Gumerman, and Gulp 1979.
TCosts were updated from 1979 to 1986 dollars by using
the second-quarter Marshall and Swift Index.
Figure 5-43. Construction and annual O&M costs for
ion exchange (Hansen, Gumerman,
and Gulp 1979).
200
150
100
ra 80
!.
§. 50
1 40
5 30
8
o
20
15
10
Construction
O&M
SO 60 SO 100 150 200 300 400 500 600
Capacity, gal/min.
Operation and maintenance costs include electricity
for backwashing (after 150 bed volumes have been
treated) and periodic repair and replacement costs.
Costs for regenerant chemicals are not included be-
cause they will depend on the types and con-
centrations of target chemicals to be removed from
the wastewater.
5.7.0 Oxidation/Reduction
5.7.8.1 General Description
Reduction-oxidation (redox) reactions are those in
which the oxidation state of at least one reactant is
raised while that of another is lowered. In chemical
oxidation, the oxidation state of the treated com-
pound^) is raised. For example, in the conversion of
cyanide to cyanate under alkaline conditions and
using permanganate, the oxidation state of the cya-
nide ion is raised as it combines with an atom of oxy-
gen to form cyanate. This reaction can be expressed
as follows:
2NaCN + 2KMnO4 + KOH -*2K2MnO4 + NaCNO + H2O
5-77
Common commercially available oxidants include
potassium permanganate, hydrogen peroxide, chlo-
rine gas, and calcium and sodium hypochlorite.
5.7.8.2 Application/Availability
Chemical oxidation is used primarily for detoxification
of cyanide and treatment of dilute contaminated
ground water containing oxidizing organics. Among
the organics for which oxidative treatment has been
reported are aldehydes, mercaptans, phenols, ben-
zidine, unsaturated acids, and certain pesticides
(Kiang and Metry 1982). Chemical oxidation also can
be an effective way of pretreating wastes prior to
biological treatment; compounds that are refractory to
biological treatment can be partially oxidized, which
makes them more amenable to biological oxidation.
Equipment requirements for chemical oxidation are
simple and include readily available equipment such
as contact vessels with agitators to provide suitable
contact of the oxidant with the waste, storage vessels,
and chemical metering equipment. Some instru-
mentation is required to determine pH and the degree
of completion of the oxidation reaction. Because
some oxidizing reagents react violently in the
presence of significant quantities of readily oxidizable
materials, reagents must be added in small quantities
to avoid momentary excesses.
One of the major limitations of chemical oxidation is
that the oxidation reactions frequently are incomplete
(reactions dp not proceed to CO2 and H2O). Incom-
plete oxidation may be due to oxidant concentration,
pH, oxidation potential of the oxidant, or formation of a
stable intermediate (Kiang and Metry 1982). The
danger of incomplete oxidation is that more toxic
oxidation products could be formed. Chemical oxi-
dation is not well suited to high-strength, complex,
waste streams. The most powerful oxidants are rela-
tively nonselective, and any oxidizable organics in the
ground water will be treated. If the ground water is
highly contaminated, large concentrations of oxidizing
agents will have to be added to treat target com-
pounds. Some oxidants (e.g., permanganate) can be
decomposed in the presence of high concentrations
of alcohols and organic solvents (Kiang and Metry
1982).
Chemical reduction involves the addition of a reducing
agent, which lowers the oxidation of a substance as a
means of reducing toxicity or solubility or transforming
it to a form that can be more easily handled. For ex-
ample, when sulfur dioxide is used in the reduction of
hexavalent chromium [Cr(VI)] to trivalent chromium
[Cr(lll)], the oxidation state of Cr changes from 6+to 3+
(Cr is reduced) and the oxidization state of sulfur
increases from 4+ to 6+ (sulfur is oxidized). The de-
crease in the positive valence or the increase in the
negative valence with reduction takes place simul-
taneously with oxidation in chemically equivalent ratios
(Kiang and Metry 1982).
-------
2H2CrO4 + 3SO2 + 3H2O -* Cr2(SO4)3+5H2O
Commonly used reducing agents include sulfite salts
(e.g., sodium bisulfite, sodium metabisulfite, sodium
hydrosulfite), sulfur dioxide, and the base metals (iron,
aluminum, and zinc).
Chemical reduction is well demonstrated for the
treatment of lead, mercury, and chromium. For com- i
plex waste streams containing other potentially re-
ducible compounds, however, laboratory- and pilot-
scale tests will be required to determine appropriate,
chemical feed rates and reactor retention times.
Chemical reduction can be carried out using simple,,
readily available equipment and reagents. Capital and
operating costs are low and the process is easy to
implement.
5.7.8.3 Design and Construction
Considerations
Implementation is complicated because every oxi-
dation/reduction reaction system must be designed
for the specific application. Laboratory- and/or pilot-
scale testing is essential to determine the appropriate
chemical feed rates and reactor retention times in
accordance with reaction kinetics. Oxidation/reduc-
tion has not been widely used to treat ground water.
A major consideration in electing to use oxidation is
that the treatment chemicals are invariably hazardous
and require great care in handling. In particular, many
oxidizing agents are potentially hazardous to handle
and suppliers' instructions should be carefully
followed.
Oxidation can produce some undesirable byproducts.
For example, the addition of chlorine can result in the
formation of bioresistant end products that can be
odorous and more toxic than the original compound.
The possibility of this undesirable side reaction needs
to be considered when chlorine is used for oxidation
of contaminated ground water (Conway and Ross
1980).
The required equipment for chemical reduction is very
simple. It includes storage vessels for the reducing
agents and perhaps for the contaminated ground
water, metering equipment for both streams, and con-
tact vessels with agitators to provide suitable contact
of reducing agent and ground water. Some instru-
mentation is also required to determine the con-
centration and pH of the ground water and the degree
of completion of the reduction reaction. The reduc-
tion process may be monitored by an oxidation-
reduction potential (ORP) electrode (Kiang and Metry
1982).
5.7.8.4 Costs
Capital costs for both chemical oxidation and chemical
reduction include costs for chemical storage, chemical
feeding, and chemical mixing. Approximate costs are
shown in Figure 5-44.
Figure 5-44.
Construction and annual O&M costs
for chemical oxidation and reduction.
20
18
16
14
12
10
9
8
7
Note: It is coincidental that
construction and O&M
costs are identical.
Construction
O&M
50 60 70 80 90100 150
Capacity, gal/min.
200 250 300 400
5.7.9 Neutralization
5.7.9.1 General Description
Neutralization consists of adding acid or base to
contaminated ground water to adjust its pH. The most
common system for neutralizing acidic or basic ground-
water streams involves a multiple cornpartmental
basin, usually constructed of reinforced plastic or lined
concrete.
To reduce the volume of the neutralization basin to
the required level, mixers are installed in each
compartment to provide more intimate contact be-
tween the contaminated ground water and neu-
tralizing reagents. This speeds up reaction time.
Stainless steel plates mounted on the floor of the pit
and directly below the mixers reduce corrosion
damage to the structure. Basin inlets are baffled to
provide flow distribution; effluent baffles can help to
prevent foam from being carried over into the
receiving stream (Conway and Ross 1980).
5.7.9.2 Application/Availability
Neutralization can be applied to any ground-water
stream requiring pH control. It commonly precedes
biological treatment because bacteria are sensitive to
rapid pH changes and values outside a pH range of 6
to 9. Aquatic ecosystems are similarly pH-sensitive;
therefore, ground water must be neutralized prior to
discharge to a receiving water body. When con-
taminated ground water is hazardous because of its
corrosivity, neutralization may be required before its
acceptance for disposal. It is also used as a pre-
treatment for several chemical treatment tech-
nologies, including carbon adsorption, ion exchange,
air stripping, wet-air oxidation, and chemical oxi-
dation/reduction processes. A pH adjustment is also
5-78
-------
dictated in several other situations; e.g., for protection
of construction materials, breaking of emulsions,
insolubilization of certain organic materials, and control
of chemical reaction rates (Conway and Ross 1980).
Neutralization is a relatively simple treatment process
that can be performed with readily available equip-
ment. Only storage and reaction tanks with accessory
agitators and delivery systems are required. Because
of the corrosivity of the wastes and treatment re-
agents, appropriate materials of construction are
needed to provide a reasonable service life for equip-
ment. The process is reliable provided pH monitoring
units are used. The feed of the neutralization agent
may be regulated automatically by the pH monitoring
unit, which ensures effective neutralization and
minimizes worker contact with corrosive neutralizing
agents.
5.7.9.3 Design and Construction
Considerations
The amount of neutralizing reagent is readily com-
puted from the stoichiometry of the acid-base reaction
and the concentration of the reagent and the waste
stream. The choice of an acidic reagent for neu-
tralization of an alkaline ground water is generally
between sulfuric acid and hydrochloric acid. Sulfuric
acid is normally selected because of its lower cost.
Hydrochloric acid generally forms soluble reaction end
products that may be advantageous.
The choice of a caustic reagent is generally between
sodium hydroxide and various limes; magnesium
hydroxide and ammonium hydroxide are also used.
The factors to be considered in choosing the most
suitable reagent include purchase cost, neutralization
capacity, reaction rate, storage and feeding re-
quirements, and neutralization products. Although
sodium hydroxide costs much more than the other
materials, it is frequently used because of its
uniformity, ease of storage and feeding, rapid reaction
rate, and soluble end products. The lime materials
have the advantage of relatively low cost; this low cost
is at least partially offset by increased capital and
operating costs for the more complex feeding and
reaction system that is required (Conway and Ross
1980).
Although the rate of reaction between the completely
ionized sodium hydroxide and a strong acid-
contaminated ground water is virtually instantaneous,
the reactions of lime bases require considerable time
for completion. Reaction time can be minimized by
several approaches: a relatively high end-point pH
level (approximately 9), efficient mixing, and slurry
feeding as opposed to dry feeding (Conway and Ross
1980).
Neutralization of ground water can produce air
emissions. Acidification of streams containing certain
salts (e.g., sulfide) will produce toxic gases. Feed
tanks should be totally enclosed to prevent the
escape of fumes. Adequate mixing should be pro-
vided to disperse the heat of reaction if concentrated
wastes are being treated. The process should be
controlled from a remote location if possible. A sturdy,
shielded electrode that can be routinely cleaned or
replaced should be mounted near the discharge point
and connected to a pH meter.
5.7.9.4 Costs
Capital costs for a neutralization system include costs
for chemical storage, chemical feeding, and mixing.
Reagent cost is the primary operating cost.
Approximate costs are shown in Figure 5-45.
Figure 5-45. Construction and annual O&M costs for
neutralization.
20
18
16
5 14
I
| 12
o
g 10
of 9
8
7
6
Note: It is coincidental that
construction and O&M
costs are Identical.
Construction
*. '
O&M
50 60 708090100 150 200 250 300 400
Capacity, gal/mln.
5.7.10 Steam Stripping
5.7.10.1 General Description
Steam stripping is used to remove gases or volatile
organics from dilute contaminated ground-water
streams: This process is essentially a fractional
distillation of volatile compounds from a ground-water
stream. The volatile component may be a gas or
volatile organic compound with solubility in the
wastewater stream. In most instances, the volatile
component (e.g., methanol or ammonia) is quite water
soluble.
Steam stripping is usually conducted as a continuous
operation in a packed tower or conventional
fractionating distillation column (bubble cap or sieve
tray) with more than one stage of vapor/liquid contact.
The preheated contaminated ground water from the
heat exchanger enters near the top of the distillation
column and then flows by gravity countercurrently to
the steam and organic vapors (or gas) rising up from
5-79
-------
the bottom of the column. As the contaminated
ground water passes down through the column, it
contacts the vapors rising from the bottom, which
contain progressively less volatile organic compound
or gas, until it reaches the bottom of the column,
where the ground water is finally heated by the in-
coming steam to reduce the concentration of volatile
component(s) to their final concentration. Much of the
heat in the water discharged from the bottom of the
column is recovered during preheating of the feed to
the column.
Reflux (condensing a portion of the vapors from the
top of the column and returning it to the column) may
or may not be practiced, depending on the desired
composition of the vapor stream. Although many of
the steam strippers in industrial use introduce the
wastewater at the top of the stripper, introducing the
feed to a tray below the top tray has its advantages
when reflux is used. Introducing the feed at a lower
tray (while still using the same number of trays in the
stripper) will either reduce steam requirements (be-
cause less reflux is needed) or will yield a vapor stream
richer in volatile component. Combining reflux with
the introduction of the feed at a lower tray will increase
the concentration of the volatile organic component
beyond that obtainable by reflux alone. Figure 5-46 is
a flow diagram of a steam stripping system.
5.7.10.2 Application/Availability
Steam stripping has been used for many years for the
recovery of ammonia from coke oven gas. With the
recent advent of more stringent water effluent
regulations, contaminated ground-water streams are
being treated by steam stripping for removal of volatile
organic components (i.e., methanol from pulp mill
condensate),
Three common examples of product recovery by
steam stripping are ammonia recovery (for sale as
ammonia or ammonia sulfate) from coke oven gas
scrubber water, sulfur from refinery sour water, and
phenol from water solution in the production of
phenol. This technology has been recently applied to
ground-water treatment; even newer applications
include removal of phenols, mercaptans, and chlo-
rinated hydrocarbons from wastewater.
Equipment is nearly the same as that required for
conventional fractional distillation (i.e., packed column
or tray tower, reboiler, reflux condenser and feed
tanks, and pumps); however, the heat exchanger is
used for heating feed entering the column and
cooling the stripped contaminated ground water
leaving the column. The reboiler is often an integral
part of the tower body rather than a separate vessel.
5.7.10.3 Design and Construction
Considerations
Typical design criteria are shown below.
Column height, 20 to 60 ft
Column diameter, 3 to 6 ft
Steam requirements, 0.6 to 2.9 Ib/gal
Typical wastewater flow, 200 gal/min
The selection of materials of construction depends on
the operating pH and presence (or absence) of cor-
rosive ions (sulfides, chlorides).
Steam-stripped volatiles are usually processed further
for recovery or incinerated. If stripped volatiles contain
sulfur and are incinerated, the impact of SO2 emis-
sions must be considered. The impact of the stripped
ground water depends on the quantity and type of
residual volatile organics remaining in the stripped
Figure 5-46. Steam stripping system.
Treated Wastewater
1^. Concentrated
Vapors
• Steam
5-80
-------
ground water. Land requirements are small, and the
only discharge generally is the treated ground water.
Use of steam stripping requires a source of steam.
Steam stripping is thus not suitable for emergency
field use unless the contaminated ground water can
be transported from the site to a steam stripping facility
(EPA 1980).
5.7.10.4 Costs
The total capital costs and annual operating costs are
shown in Figure 5-47 (in mid-1986 dollars).
Figure 5-47.
Capital and annual O&M costs for steam
stripping.
1000
700
-3> 500
| 400
300
§
<»
200
1S°
100
50
Capital
50 60 70
100
150 200 300
Capacity, gal/min.
400 500600700
Table 5-31. Results of Pilot Scale Testing of a Reverse Osmosis
5.7.11 Reverse Osmosis
5.7.11.1 General Description
In reverse osmosis (RO), high pressure is used to
force a solvent (e.g., water) through a membrane that
is permeable to the solvent molecules but not to the
solute molecules. In industrial applications, it is used
primarily to demineralize brackish waters and to treat a
variety of industrial wastewaters.
The basic components of an RO unit are the
membrane, a membrane support structure, a
containing vessel, and a high-pressure pump. The
membrane and membrane support structure are the
most critical elements.
5.7.11.2 Application/Availability
Reverse osmosis is used to reduce the
concentrations of dissolved solids, both organic and
inorganic. In the treatment of contaminated ground
water, use of RO would be limited primarily to polishing
low-flow streams containing highly toxic contaminants.
In general, good removal can be expected for high-
molecular-weight organics and charged anions and
cations. Multivalent ions are treated more effectively
than are univalent ions. Recent advances in
membrane technology have made it possible to
remove such low-molecular-weight organics as
alcohols, ketones, amines, and aldehydes (Gooding
1985). Table 5-31 shows removal results obtained
during testing of a mobile RO unit using two favorable
membrane materials (Whittaker 1984).
Reverse osmosis units are subject to chemical attack,
fouling, and plugging. Pretreatment requirements
Unit*
Percent Removed in Permeate
Chemical
Dichloromethane
Acetone
1,1-DichIoroethene
Tetrahydrofuran
Diethyl ether
Chloroform
1 ,2-Dichloroethane
1 ,1 ,1-Trichloroethane
Trichloroethene
Benzene
Bromoform
Hexane
Feed
Concen-
tration
(PPb)
406
110
34
17,890
210
270*
99
659
24
539
12t
10t
Concentra-
tion in I Polyether-
Concentrate polysulphone
(ppb) I Membrane
I
203
355
795
467
439
567
415
651
346
491
633
704
58
84
99
98
97
98
92
99.8
99
99
99.1
99.8
Polyester/
Amide
Polysulphone
Membrane
52
76
95
89
89
92
85
97
92
99
98
97
Data from Whittaker 1984.
No standard available; concentration estimated
5-81
-------
can be extensive. Contaminated ground water must
be pretreated to remove oxidizing materials such as
iron and manganese salts; to filter out particulates; to
adjust pH to a range of 4.0 to 7.5; and to remove oil,
grease, and other film forms (De Renzo 1978). The
growth of slimy biomass on the membrane surface or
the presence of organic macromolecules may also foul
the membrane. This organic fouling can be minimized
by perchlorination, the addition of biocides, and/or
pretreatment with activated carbon (Ghassemi, Yu,
andQuinlivan 1981).
Compact RO units are commercially available, can be
started easily and shut down relatively quickly, can be
serviced conveniently, produce only a small volume of
residue (10 to 25 percent of the feed), do not require
skilled labor, and can be operated with electric power
produced on site. Thus, RO meets many of the re-
quirements of a mobile system; however, a significant
amount of time may be required to obtain and as-
semble the necessary components. Another major
shortcoming is membrane susceptibility to fouling or
degradation caused by the presence of suspended
solids or strong oxidizers in the contaminated ground
water or ground water with a very low pH. For this
reason, pretreatment of the ground water (e.g., by
coagulation/flocculation) is necessary before treat-
ment by RO. Depending on the specific contaminants
to be removed and the type of membrane used, RO is
generally used as a final polishing step. Another
consideration is that a certain amount of concentrated
waste material will require further treatment or dis-
posal. Also, if the system is to be located on a remote
site, power consumption can be a problem. A 10-
gal/min system will require about 5 kW of power. This
is generally not a problem at service station sites,
however.
Finally, RO will not reliably treat wastes with a high
organic content because the membrane may dissolve
in the waste. Lower levels of organic compounds may
also be detrimental to the unit's reliability because
biological growth may form on a membrane fed an
influent containing biodegradable organics.
5.7.11.3 Design and Construction
Considerations
The most critical design consideration applicable to
RO technology is the design of the semipermeable
membrane. In addition to achieving the required
degree of separation at an economic flux level under
ideal conditions, the design of the membrane must be
such that it can be incorporated in an operating
system that satisfies the following practical
requirements (Conway and Ross 1980):
• Minimum concentration polarization, i.e., ratio of
impurity concentration at the membrane surface
to that in the bulk stream.
• High packing density, i.e., membrane surface
area per unit volume of the pressure module.
5-82
Ability to handle any particulate impurities (by
proliferation if necessary).
• Adequate support for the membrane and other
physical features such as effectiveness of seals,
ease of membrane replacement, and ease of
cleaning.
Membranes are usually fabricated in flat sheets or
tabular forms and then assembled into modules. The
most common materials of construction are cellulose I
acetate and other polymers, such as polyamides and
polyether-polysulphone. There are three basic
module designs: tubular, hollow fiber, and spiral-
wound (Figure 5-48). Each has its own advantages
and limitations. The tubular module provides the
largest flow channel and allows for turbulent fluid flow;
thus, it is least susceptible to plugging caused by
suspended solids and has the highest flux. Because
of its small area/volume ratio, however, the total
product recovered per module is small. The cost of a
tubular module is approximately five times greater than
that for the other modules for an equivalent rate of
water recovery, and the total space requirement is
about three to five times greater than that for the spiral-
wound system (Ghassemi, Yu, and Quinlivan 1981).
A hollow-fiber membrane is constructed of polyamide
polymers (duPont) and cellulose triacetate (Dow). The
polyamide membrane permits a wider operating pH
range than does cellulose acetate, which is commonly
used for the construction of spiral-wound and tubular
membranes. The flow channel and the flux are about
an order of magnitude lower than on the other
configurations. This small flux, however, is com-
pensated for by the large surface area/volume ratio;
therefore, the total product water per module is close
to that obtainable with spiral-wound modules. Be-
cause of the small size of the channels (about 0.004
in.) and the laminar fluid flow within the channels,
however, this module is susceptible to plugging, and
extensive pretreatment may be required to protect the
membrane (Ghassemi, Yu, and Quinlivan 1981).
The spiral-wound module consists of an envelope of
flat-sheet membranes rolled around a permeate
collector tube. This configuration provides for a higher
flux and greater resistance to fouling than does the
configuration of the hollow-fiber modules; it is also
less expensive and occupies less space than a tubular
module (Ghassemi, Yu, and Quinlivan 1981).
5.7.11.4 Costs
Figure 5-49 presents the costs of various sizes of RO
units. Construction costs include housing, tanks,
piping, membranes, flow meters, cartridge filters, acid
and polyphosphate feed equipment, and cleanup
equipment. These costs are based on influent total
dissolved solids concentrations of less than 10,000
ppm.
-------
CO
O)
c
0)
o
•o
o
O
O
a.
(O
O)
•o
o
E
"o
5-83
-------
Figure 5-49. Concentration and annual O&M costs for
reverse osmosis.
1000
£00
•400
zoo
« 60
20
10
4 6
10 20 40 60100 200 400 700
Capacity, galftnln.
The operation and maintenance costs include
electricity for the high-pressure feed pumps (450 psi
operating pressure), building utilities, routine periodic
repair, routine cleaning, and membrane replacement
every 3 years. Operation and maintenance costs do
not include costs for pretreatment chemicals because
usage rates vary greatly among plants.
5.7.12 Sludge Dewatering
5.7.12.1 General Description
Sludge dewatering can be achieved by several
different types of filtration mechanisms. The four
major dewatering methods are described briefly in the
following subsections:
Vacuum Filtration .
A rotary vacuum filter consists of a cylindrical drum
rotating partially submerged in a vat or pan of con-
ditioned sludge. The drum is divided radially into
several sections that are connected through internal
piping to ports in a valve body (plate) at the hub. This
plate rotates in contact with a fixed valve plate with
similar parts that are connected to the appropriate
service. Various operating zones are encountered
during a complete revolution of the drum. In the
pickup or form section, vacuum is applied to draw
liquid through the filter covering (media), which forms
a cake of partially dewatered sludge. As the drum
rotates, the cake emerges from the liquid sludge pool
while suction is maintained to promote further de-
watering. A lower level of vacuum often exists in the
cake-drying zone. If the cake tends to adhere to the
media, a scraper blade may be provided to assist in its
removal.
The three principal types of rotary vacuum filters are
the drum type, coil type, and belt type. The filters
differ primarily in the type of covering used and the
cake discharge mechanism employed. A cloth
medium is used on drum and belt types; stainless
steel springs are used on the coil type. Occasionally,
a metal medium is used on belt types. The drum filter
also differs from the other two in that the cloth
covering does not leave the drum, but is washed in
place when necessary. The design of the drum filter
provides considerable latitude in the amount of cycle
time devoted to cake formation, washing, and
dewatering; the design also minimizes inactive time.
A schematic diagram of a drum-type rotary vacuum
filter is shown in Figure 5-50.
Figure 5-50. Drum-type rotary vacuum filter.
Drum
Cake Scraper
Filtrate
Line
-Vat
Sludge
Filter Press Dewatering
The recessed plate press, which is the conventional
filter press used for dewatering sewage sludges,
consists of vertical recessed plates up to 5 ft in di-
ameter (or 5 ft on a side, if square) that are held rigidly
in a frame and pressed together between a fixed and
moving end. A filter cloth is mounted on the face of
each plate. The sludge is fed into the press at pres-
sures up to 225 psig and passes through feed holes
in the trays along the length of the press. The water
passes through the cloth; the solids are retained and
form a cake on the surface of the cloth. Sludge
feeding is stopped when the cavities or chambers
between the plates are completely filled. Drainage
ports are provided at the bottom of each press
chamber. The filtrate is collected in these ports, taken
to the end of the press, and discharged to a common
drain. At the beginning of a processing cycle, the
drainage from a large press can be on the order of
5-84
-------
2000 to 3000 gal/h. This rate falls rapidly to about 500
gal/h as the cake begins to form; when the cake
completely fills the chamber, the rate is virtually zero.
The dewatering step is complete when the filtrate is
near zero. At this point, the pump feeding sludge to
the press is stopped, and any back pressure in the
piping is released through a bypass valve. The elec-
trical closing gear is then operated to open the press.
The individual plates are moved, in turn, over the gap
between the plates and the moving end; this allows
the filter cakes to fall out. The plate-moving step can
be either manual or automatic. When all of the plates
have been moved and the cakes released, the
complete pack of plates is pushed back by the moving
end and closed by the electrical closing gear. The
valve to the press is then opened, the sludge feed
pump is started, and the next dewatering cycle
commences. Thus, a cycle includes the time required
Figure 5-51. Filter press plate (DeRenzo 1978).
Perforated
Fabric
Filter
Medium
for filling, pressing, cake removal, media washing, and
press closing. Figure 5-51 is a schematic diagram of a
filter press plate.
Belt Filter Dewatering
A belt filter consists of an endless filter belt that runs
over a drive and guide roller at each end, much as a
conveyor belt does. The upper side of the filter belt is
supported by several rollers. A press belt above the
filter belt runs in the same direction and at the same
speed; its drive roller is coupled with the drive roller of
the filter belt by means of a pressure-roller system
whose individual rollers can be adjusted horizontally
and vertically. The sludge to be dewatered is fed on
the upper face of the filter belt and is continuously
dewatered between the filter and press belts. After
the sludge passes through the pressure zone, further
dewatering in a reasonable time cannot be achieved
Rectangular Frame
Inlet Liquid to
be Filtered
Fabric
Filter
Medium
Entrapped
Solids
Plates and Frames are
Pressed during Filtration
Cycle
Rectangular
Metal Plate
-------
by the application of static pressures alone; however,
a superimposition of shear forces can effect this fur-
ther dewatering. The supporting rollers of the filter
belt and the pressure rollers of the pressure belt are
adjusted in such a way that the belts and the sludge
between them form an S-shaped curve. This causes a
parallel displacement of the belts relative to each other
due to the differences in the radii. After further
dewatering in the shear zone, the sludge is removed
by a scraper.
Some units operate in two stages; the initial draining
zone is on the top level, and pressing and shearing
take place in an additional lower section. A significant
feature of the belt filter press is its use of a coarse-
mesh, relatively open-weave, metal-medium fabric.
The use of such fabric is possible because of the rapid
and complete cake formation obtainable when proper
tlocculation is achieved. Belt filters do not need
vacuum systems, and they do not have the sludge
pickup problem occasionally encountered with rotary
vacuum filters. The belt filter press system includes;
auxiliaries such as equipment for preparing polymer
solution and automatic process controls. A schematic
diagram of a belt filter press system is shown in
Figure 5-52.
Figure 5-52. Belt filter press.
Conditioned
Sludge
Wash Water
Sludge
Cake
Centrifugal Dewatering ',
The solid-bowl continuous centrifuge assembly con-
sists of a bowl and conveyor joined through a plane-
tary gear system designed to rotate the bowl and the
conveyor at slightly different speeds. The solid cylin-
drical bowl, or shell, is supported between two sets of
bearings. A conical section at one end of the bowl
forms the dewatering beach over which the helical
conveyor screw pushes the sludge solids to outlet
ports and then to a sludge cake discharge hopper.
The opposite end of the bowl is fitted with an
adjustable outlet weir plate for regulating the level of
the sludge pool in the bowl. The centrate flows
through outlet ports either by gravity or by a centrate
pump attached to the shaft at one end of the bowl.
Sludge slurry enters the unit through a stationary feed
pipe that extends into the hollow shaft of the rotating
bowl and passes to a baffled, abrasion-protected
chamber for acceleration before it is discharged
through the feed ports in the rotating conveyor hub
into the sludge pool. The centrifugal forces cause the
sludge pool to take the form of a concentric annular
ring on the inside of the bowl. Solids settle through
this ring to the wall of the bowl, where they are picked
up by the conveyor scroll. Separate motor sheaves or
a variable-speed drive can be used to adjust the bowl
speed for optimum performance.
Bowls and conveyors can be constructed from a large
variety of metals and alloys to suit special applications.
For dewatering of ground-water sludges, nnild steel or
stainless steel is normally used. Because of the
abrasive nature of many sludges, hardfacing materials
are applied to the leading edges and tips of the
conveyor blades, the discharge ports, and other
wearing surfaces. New wearing surfaces may be
welded on when required.
In the continuous concurrent solid-bowl centrifuge,
incoming sludge is carried by the feed pipe to the end
of the bowl opposite the discharge. Centrate is
skimmed off and the cake proceeds up the beach for
removal. As a result, settled solids are not disturbed
by incoming feed.
In the disc-type centrifuge, the incoming stream is
distributed between a multitude of narrow channels
formed by stacked conical discs. Suspended particles
have only a short distance to settle, so that small and
low-density particles are readily collected and dis-
charged continuously through fairly small orifices in
the bowl wall. The clarification capability and through-
put rate are high, but sludge concentration is limited
by the necessity of discharging through orifices 0.050
to 0.100 in. in diameter. Therefore, the disc-type
centrifuge is generally considered a thickener rather
than a dewatering device.
In the basket-type centrifuge, flow enters the machine
at the bottom and is directed toward the outer wall of
the basket. Cake continually builds up within the
basket until the centrate, which overflows a weir at the
top of the unit, begins to increase in solids. At that
point, feed to the unit is shut off, the machine de-
celerates, and a skimmer enters the bowl to remove
the liquid layer remaining in the unit. A knife is then
moved into the bowl to cut out the cake, which falls
out of the open bottom of the machine. This is a batch-
drive unit with alternate charging of feed sludge and
discharging of dewatered cake.
Schematic diagrams of the various types of cen-
trifuges are shown in Figure 5-53.
5.7.12.2 Application/Availability
Vacuum filtration is the most common method of
mechanical sludge dewatering in the United States.
Vacuum filtration units are available in various sizes,
5-86
-------
Figure 5-53. Types of centrifuges (EPA 1979).
Solid-bowl
Feed
Feed
Sludge discharge
•Rotor bowl
Rotor r ozzles
Sludge discharge
Recycle flow
Disc-type centrifuge
Dewatered
Solids
centrifuge
Polymer
Skimmings
Feed
Knife
Cake Cake
Basket-type centrifuge
5-87
-------
and they are generally used in larger facilities where
space is limited or when incineration is necessary for
maximum volume reduction. The operation is sen-
sitive to the type of sludge and conditioning pro-
cedures. Solids capture ranges from 85 to 99.5 per-
cent and cake moisture is usually 60 to 90 percent,
depending on feed type, solids concentration,
chemical conditioning, machine operation, and man-
agement. Dewatered cake is suitable for landfilling,
heat drying, incineration, or land spreading.
Filter-press dewatering is used for hard-to-handle
sludges and when filtration must be accomplished in a
small area. Batch discharge requires equalization of
pressed cake prior to incineration if adopted. For input
sludges of varying types with a total suspended solids
(TSS) content of 1 to 10 percent, typical filter press
production data show cake solids concentrations of
50 percent with fly-ash conditioning (100 to 250
percent on dry solids basis) and cycle times of 1.5 to
2.0 h. Cake solids concentrations of 45 percent have
been achieved with chemical conditioning (5 to 7.5
percent FeCI3 and 10 to 15 percent lime) and cycle
times of 1.0 to 2.0 h. In general, cakes of 25 to 50
percent solids concentrations are achieved.
Belt filter press use in the United States is on the
increase. Its major advantages are that it is the least
energy-intensive filtration-method and it can be used
to treat sludges that are difficult to dewater. This pro-
cess can be used where a filtration must be accom-
plished in a small area. Its disadvantage is that it is
sensitive to incoming feed characteristics and requires
sludge conditioning.
Centrifugal dewatering with solid-bowl and disc-type
centrifuges is in widespread use. Solid-bowl and disc-
type centrifuges are generally used for dewatering
sludge in larger facilities where space is limited or
where sludge incineration is required. Basket-type
units are used primarily for partial dewatering at small
plants. Disc-type centrifuges are more useful for
thickening and clarification than for dewatering.
5.7.12.3 Design and Construction
Considerations
Vacuum Filtration
The design load is a function of feed solids con-
centrations, subsequent processing requirements,
and chemical preconditioning. Typical loads for
vacuum filtration are shown below.
Filter Press Dewatering
Typical design criteria for filter presses are as follows:
Sludge Type
Raw primary
Digested primary
Mixed digested
Typical Loading (Ib dry
solids/ftz per hour)
7 to 15
4to7
3.5 to 5
Chamber volume
Filter areas
Number of chambers
Sludge cake thickness
Sludge feed rate
0.75 to 2.8 ftVchamber
14.5to45ft2/chamber
Up to 100
1 to 1.5 in.
Approximately 2 Ib/cycle
per ft" (dry solids basis)
The presses must normally be installed well above
floor level so that the filter cake can drop onto con-
veyors or trailers.
Belt Filter Dewatering
Typical design criteria for loadings, based on active
belt area, are as follows:
Sludae Type
Sludge Loading
• per hour)
Dry Solids
Loading
hour)
Raw primary 27 to 34 13.5 to 17
Digested primary 20 to 24 20.5 to 24
Digested mixed/ 13 to 17 6.7 to 8.4
secondary
Centrifugal Dewatering
Each installation is site-specific and depends on the
manufacturer's product line. Maximum capacities of
about 100 tons/h of dry solids are available in solid-
bowl units with diameters up to 54 in. and power re-
quirements up to 175 hp. Disc-type units are available
with capacities up to 400 gal/min of concentrate.
Centrifugation requires a sturdy foundation because
the operation generates vibration and noise. Ade-
quate electric power also must be provided for the
large motors required. The major .difficulty encoun-
tered in the operation of centrifuges has been the
disposal of the centrate, which is relatively high in
suspended nonsettling solids. With disc-type units,
the feed must be degritted and screened to prevent
pluggage of discharge orifices.
5.7.12.4 Costs
Figures 5-54 and 5-55 present the total capital and
annual operating costs (in 1986 dollars) for vacuum
filters and filter presses and for belt presses and
centrifuges, respectively.
5-88
-------
Figure 5-54. Capital and annual O&M costs for vacuum filters and filter presses (EPA 1982).
a>
(5
o
-o
o
o
o
a>
O
O
3600
3000
2000
1500
1000
800
600
500
400
300
200
100
TJ&M (Filter Press)
Capital (Filter Press)
O&M (Vacuum Filter)
50 60 70 80 100 150 200 300400500600700
Capacity, gal/min.
Figure 5-55. Capital and annual O&M costs for belt presses and centrifuges (EPA 1982)
10,000 p— T 1 | | | | II
1,000
•8
CO
o
te-
8 100 -
Capital (Centrifuge)
Capital (Belt Press)
O&M (Centrifuge)
O&M (Belt Press)
Sludge Load, tons of dry solids/day
5-89
-------
5.8 Vapor Migration Control,
Collection, and Treatment
Leaks from underground storage systems can
present problems due to vapors emanating from the
plume. Many things contribute to the path these
vapors take. Normally, they vent vertically through the
cover material; however, if this vertical path is sealed
by frost, rain-saturated cover soil, or pavement, they
tend toward lateral migration. A sand/gravel environ-
ment generally promotes greater lateral movement of
vapors than does a clay environment. Because vapor
migration and venting can create significant hazards,
special control systems have been developed to
alleviate these problems. This section presents infor-
mation on the following technologies for collecting,
controlling, and disposing of vapors from under-
ground spills: passive collection systems, active col-
lection systems, ventilation of structures, adsorption,
and flaring. ;
5.8.1 Passive Collection Systems (EPA
1985a)
5.8.1.1 General Description
Subsurface migration of vapors beyond the leak site
can be prevented by the use of passive vapor-control
systems, i.e., systems that control vapor movement by
altering the paths of flow without the use of me-
chanical components. Passive systems may be fur-
ther categorized as high-permeability or low-perme-
ability systems.
High-permeability systems entail the installation of
highly permeable (relative to the surrounding soil)
trenches or wells between the plume and the area to
be protected. High-permeability systems generally
take the form of trenches or wells excavated outside
of the plume limit and backfilled with a highly
permeable medium, such as a coarse crushed stone
(Figure 5-56).
Low-permeability systems effectively block vapor flow
into areas of concern by the use of barriers between
the plume and the area to be protected (Figure 5-57).
These two concepts (high- and low-permeability) of
passive vapor control are often combined in the same
system to provide controlled venting of vapors and
blockage of available paths for vapor migration.
5.8.1.2 Application/Availability
Passive vapor-control systems can be used at virtually
any site where an excavation can be trenched or
drilled to at least the same depth as the plume.
Limiting factors include the presence of a perched
water table or rock strata. Passive vents are generally
less effective in areas of high rainfall or prolonged
freezing temperatures.
5.8.1.3 Design and Construction
Considerations
Passive collection is usually controlled through the
use of trenches constructed around the plume.
Minimum trench widths of 3 feet are often specified to
ensure an open trench over the full depth. The depth
of the trench is dictated by local site conditions and in
some applications the trench need not be very deep
just as long as it is deep enough to intercept all
possible avenues of vapor migration.
Crushed stone or river gravel is normally used as the
permeable medium for trench backfill. Stone sizes
greater than 1/4-inch are recommended; fine material
should not be used. Horizontal perforated pipe and
vertical solid wall-riser pipes are often used to ensure
that paths of vapor flow to the atmosphere remain
open in the event that the top of the trench becomes
blocked by ice, snow, vegetation, etc. The ground
surface should be graded to drain away from the
trench to prevent soil from washing into the voids of
the stone. Any drainage swales that must cross the
trench should be installed by using enclosed conduits
or paved channels.
Trenches for low-permeability systems would normally
be situated and excavated in a manner similar to that
used to install a high-permeability system. Width and
depth requirements would be essentially the same. In
lieu of the highly permeable stone backfill, however,
low-permeability material would be placed in the
trench. Synthetic membranes are normally used as
barriers in this application.
Trenches excavated for passive vapor control systems
are normally cut with backhoes; however, other
conventional trenching equipment capable of
providing adequate depth and width could be used.
Rounded gravel or crushed stone (washed of fines)
should be used for venting trench backfill material,
and only rounded gravel should be used as a
permeable medium in conjunction with a synthetic
membrane liner to avert tearing or puncturing the
membrane. Although virtually any pipe material can be
used for perforated and riser pipe, the use of 3- or 4-
inch PVC pipe is customary. Polyvinyl chloride (PVC),
polyethylene (PE), chlorinated polyethylene (CPE),
Hypalon, and other materials have been used for
impermeable synthetic membranes. A minimum
thickness of 20 mils is recommended. Lap joints are
cemented or heat-welded and may be made at the
factory or in the field.
During trench excavation for passive systems, care
should be taken to ensure that workers are not
overcome by venting vapors or exposed to explosion
hazards. Open flames and smoking should be
prohibited in the work area. Regular monitoring of gas-
oline, methane, oxygen, and other vapors of concern
5-90
-------
Figure 5-56. High-permeability passive vapor control system
Permeable
Vent Trench
Area to be
Protected
Drainage Swale
fe
Monitoring Probe
4-in.PVC Vent Pipe.
4-in. PVC Perforated Collector.
(Continuous)
Natural
Ground
Drainage Monitoring
Swale Probe "
Low Ground-Water
Table, Bedrock
3ft •*•
Sect on View
Natural
Ground
Gravel or Stone
• (1/4-in. Win. Size)
5-91
-------
Figure 5-57. Low permeability passive vapor control
system
Area to be
Protected
Backfilled Trench
Synthetic Membrane
«
Monitoring Probe
Plan View
Synthetic Membrane
V
Monitoring
Probe
Backfilled Trench
Section View
should be conducted. Depending on soil char-
acteristics and trench depth, sloping of trench walls
may be required to avoid instability; alternatively,
shoring and bracing can be used to support trench
walls. This does not adversely affect the installation;
however, additional backfill material is required.
During the installation of synthetic membrane barriers,
extreme care must be taken to ensure that lap joints
are properly sealed and that tears and punctures are
averted in the process of placing the membrane and
backfilling the trench.
Monitoring the effectiveness of passive vapor-control
systems normally consists of periodic sampling of
subsurface vapors from probes installed in the area
being protected. Installation of a vapor-monitoring
probe requires the drilling of a hole in the soil to a
depth above which monitoring is desired. A probe
pipe, which is perforated except for the upper several
feet, is installed in the drilled hole, and the hole is
backfilled with permeable material (sand or pea gravel)
to a height above the perforations. The remainder of
the hole is backfilled with soil, which acts as a seal to
prevent the intrusion of air. (Seals are sometimes
used to keep this soil from entering the permeable
material.) A vapor sample can then be withdrawn from
the probe at the surface.
5.8.1.4 Costs
Because the lengths, depths, and widths of passive
system trenches vary, the unit cost per linear foot of
landfill border will be totally site-specific. The capital
costs in Table 5-32 are given in units that can be
readily determined for a given site, with some
judgment on the part of the estimator. Costs shown
are based on the following particulars:
Trench depth of 30 feet; length 2500 feet; width
3 feet.
• Collector piping installed over whole length of
trench with vent pipes 10 feet long and spaced
at 50 feet.
• Vapors vented to atmosphere.
Monitoring probes spaced at 100 feet, each 30
feet deep.
Table 5-32. 1986 Unit Costs for
Components of Passive
Vapor-Control Systems* '
Item Cost Range ($)
Excavation and disposal
of material
Gravel backfill
Bank sand backfill
Piping
Synthetic membrane
Monitoring probes
2 to 4/yd3
12to18/yd3
6 to 9/yd3
4 to 6/ft2
2 to 4/ft
10to15/ft
* Data from EPA 1985a.
5.8.2 Active Vapor Control (EPA 1985a)
5.8.2.1 General Description
Subsurface vapor migration also can be controlled by
active vapor-control systems, which consist of vapor
extraction, vapor collection headers, vacuum blowers
or compressors, and/or vapor treatment or utilization
systems (Figure 5-58).
Blowers or compressors establish a pressure gradient
through collection headers and wells to the area
surrounding the plume. The subsurface vapors then
flow through the collection system to the treatment or
recovery system.
The use of forced-air venting can also significantly
lower gasoline vapor concentrations in the soil above
5-92
-------
Figure 5-58. Active vapor control system
(EPA 1985a)
Blower/
Burner
Facility
Vapor 9
Collection Area to be
Header Protected
Plan View
Control
Valve
Gas V
Collection \
Header \4 /i
J^
(VUST *
\\Leak ^
KVOs
Ground water VsN
^
Natural
Ground
^*
Monitoring
Probe
Vapor
— Extraction
Well
Section View
an underground gasoline release and can also
remove some gasoline from the underground
environment. This relatively simple system consists of
an air intake and air exhaust system to force air
through the soil medium. Only experimental data are
currently available; however, this looks like a promising
means of reducing vapors resulting from medium to
large releases of petroleum or petroleum products.
5.8.2.2 Application/Availability
Active vapor-control systems can be used at any site
where the needed excavation can be made for their
installation. Systems of extraction wells, control
trenches, and header piping are normally required,
along with vapor pumping and disposal equipment.
Active vapor-control systems are not sensitive to the
freezing or saturation of surface or cover soils.
5.8.2.3 Design and Construction
Considerations
Active vapor-control systems consist of several
components, all of which require different materials of
construction and installation techniques. Table 5-33
summarizes the major requirements. Specific material
selection is at the discretion of the designer; however,
the materials listed are those that experience has
proven to perform satisfactorily. Corrosion resistance,
flexibility, and ease of installation are of particular
importance in the selection of materials and the
design of the components for these systems, which
generally will be somewhat temporary (i.e., expected
system life is but a few years).
Approximate values for design criteria should be
determined by vapor extraction tests on one or more
test wells, during which the change in pressure
gradient radially from the wells should be monitored.
Other parameters that should be monitored during the
tests are vapor extraction flow rates, subsurface
negative (vacuum) pressures at various depths and
distances from the well(s), and negative pressures
within the well.
In general, vapor extraction wells should be
constructed before a header pipe is installed because
wells are often relocated in the field during
construction for a variety of reasons and realignment
of header configurations may be desirable.
Blower/burner facility construction usually may begin
at any time because its location is dictated by factors of
accessibility. Associated header alignments may be
adjusted to accommodate the facility.
The first step in designing a vapor-collection header is
to estimate vapor flow rates from the individual
extraction wells. Preliminary flow rate estimates may
be inaccurate; therefore, a factor of safety should be
used to adjust the flow rate upward. Cumulative vapor
flow rates along the header line are estimated by
summing the individual well flow rates "upstream" from
the point under consideration.
Blower capacity or extraction wells can readily be
added to existing active systems to improve
performance. Shutdown and other nonperformance
alarms can be provided to identify the need for
emergency maintenance, which increases reliability.
Regular operation and maintenance of mechanical
systems (motors, bearings, belts, etc.) is required on
all active systems.
Active vapor-control systems can be implemented
with relatively conventional equipment, labor, and
materials. Some mechanical equipment may require
delivery lead times of several months. Well drilling is
accomplished with caisson, auger, and bucket rigs. A
few systems having high-torque capacity also are
needed to excavate through large obstacles that may
be present. Pipe laying is similar to utility pipeline
construction.
Implementation of active vapor control systems
requires little time. Several wells can be completed by
a single crew in a day (equivalent to several hundred
feet of perimeter per day), and collection piping and
mechanical components can be installed concur-
rently. Vapor control can be effected upon completion
and startup, and immediate results (as measured in
monitoring probes) are realized.
5-93
-------
Table 5-33. Materials and Equipment for Active Vapor-Control Systems*
Item Materials
Well drilling
Well piping 2- to 6-in. PVC, schedule 40 to
80, perforated and solid-wall
Well backfill 1 -in. washed crushed stone or
river gravel
Header piping 3-in. or greater (depending on flow/
pressure requirements) PVC,
polyethylene, or fiberglass (resistant
to chemical attack)
Valves Compatible with pipe size; gate, ball,
or butterfly type; PVC or other
chemical-resistant material
Vacuum blower Material or coating that resists
chemical attack; size varies with
flow/pressure requirements
Safety devices Specific items manufactured for
use at refineries, sewage digesters,
etc.
Vent stacks Any corrosion-resistant pipe of
adequate size and strength; may
require support
Installation
Auger, caisson, or bucket drill rig
Crane for deep wells, backhoe for
shallow wells
Placed slowly by hand
Conventional trench excavating
equipment, specialized jointing
equipment for some pipe materials
Jointing similar to piping materials
Foundation and installation per
manufacturer recommendations
Installed with piping
Same as header piping
'Datafrom EPA 1985a.
5.8.2.4 Costs
The capital costs of active vapor-control systems vary
greatly and depend on the size and depth of the
plume, the nature of the contaminant, and the
selected design criteria. Table 5-34 shows unit costs
for typical components of a vapor-control system. The
large range of units costs is due to the variable nature
of the system, which depends on the characteristics
of the plume in question. Unit costs for deep
extraction wells will be greater than those for shallower
wells because more specialized equipment is
needed. Likewise, large-diameter header pipe is
more costly than smaller pipe because material and
labor costs are higher. Blower/treatment facilities may
vary in scale from a small blower with a vent stack to
multiple, high-volume blowers with or without multiple
and/or high-volume burners, automatic timers, valves,
switches, and recorders.
Annual operating and maintenance costs also vary
with the size of the system. For example, a blower
driven by a 5-horsepower motor operating
continuously will consume about $2000 to $3000
worth of electricity at 5 cents/kilowatt-hour. Other
electrical costs for lighting or automatic controls are
nominal in comparison. The cost of replacement parts
also should be small because the system has few
mechanical components. Small material costs can be
expected for tools, lubrication, replacement of belts,
Table 5-34. 1986 Unit Costs for Components of
Active Vapor-Control Systems*
Item Range of Unit Costs ($)
Vapor-extraction well (drilling,
stone, piping, etc.), in place
Well connection lateral (10-ft
piping valve, excavation,
fittings, etc.), in place
Vapor-cpllectipn header (piping,
excavation, fittings, etc.), in place
Blower facility (blower(s), safety
devices, valves, foundation,
piping, fencing, electrical
components, and service
connection), in place
Monitoring probe (drilling, pipe,
fittings, backfill, etc.), in place
Operation and maintenance
Monitoring with portable meter
50 to 80/vertical foot
1,000 to 1,550 each
20 to 105/linearfoot
50,000 to 105,000 (total)
10 to 15/verticalfoot
5,000 to 20,500/year
10 to 15 each visit
Data from EPA 1985a.
fuses, etc. Manpower costs, assuming an average of
two or three person-days per month on a contract
basis, should be on the order of $5000 annually; the
5-94
-------
costs will vary with the scale and sophistication of the
system. Other annual costs that may be considered
are insurance, security, interest, and administration or
overhead.
5.8.3 Ventilation of Structures
5.8.3.1 General Description
Leakage from underground storage systems can
affect subsurface structures when the leaking
compound volatilizes and the vapors penetrate
basement walls, sewer systems, manholes, conduits,
drains, etc., that are not constructed of impervious
materials. The condition created by the existence of
vapors in underground structures can be grouped
into two classes: 1) flammable, and 2) injurious to life.
The latter condition results from the toxic or
suffocating properties of the vapors (NFPA 1982).
When these conditions arise, steps must be taken to
ventilate the affected structure to remove the
hazardous condition.
5.8.3.2 Application/Availability
The first symptoms of a hydrocarbon loss often are
manifested in the form of vapors, which result from the
volatilization of a portion of the fugitive organic
product. These vapors usually are more noticeable
during periods of the year when soil conditions are
wet from prolonged rains and when frost is in the
ground. Both of these conditions form an imperme-
able cap that causes the vapors to seek other
avenues for venting to the surface, such as utility
conduits or building basements. Because these
vapors often have the potential to reach combustible
levels, their presence usually elicits public demand for
immediate attention (Yaniga 1984).
Ventilation systems and equipment for remedying the
situation are readjly available for installation by
experienced heating/ventilating/air conditioning con-
tractors. Local fire departments are a source of
ventilating equipment for emergency situations and
could be a reference to other local industrial sources.
Tool rental agencies also carry ventilation equipment
that can be obtained quickly and will suffice until more
suitable models or remedies are available.
5.8.3.3 Design and Construction
Considerations
Whereas the best long-term method for vapor
abatement is to recover or neutralize the fugitive
contaminants, it is sometimes necessary to abate the
symptoms quickly so that imminent danger to the
public is minimized. Three general methods are
available to address symptomatic vapor problems:
1) The first and most common method is to
ventilate the affected structure(s) so that the
atmosphere is exchanged or cleansed, either
continuously or on command. This method is
effective, but it can raise the cost of heating or
cooling the structure beyond the point of
economic feasibility. This approach also has the
tendency to induce vapor movement into the
structure by creating a negative pressure to
which vapors will migrate.
2) Another approach that has been used
successfully involves the induction of a positive
pressure within the structure that inhibits the
inflow and accumulation of vapors. This is
accomplished by adding a volume of makeup air
to the structure that is greater than its collective
air loss. This makeup air is seasonally heated
and cooled as necessary.
3) The third approach involves the forced
ventilation of vapors from soils before they can
accumulate within the structure. This is
accomplished by installing horizontal or vertical
collection pipes below grade and connecting
them to a high vacuum pump that pulls the
vapors to a central area for safe venting and/or
destruction (see Subsection 5.8.2).
The preferred approach to addressing vapor
mitigation in buildings is a combination of methods 2
and 3 (Yaniga 1984b).
The objective in ventilating the vapor-contaminated
space is to dilute the atmosphere so the
concentration of the contaminant in the air is below its
lower explosive limit (LEL) or below its threshold limit
value (TLV). Instrumentation will be necessary to
detect these properties, and a system or routine
should be established for monitoring them.
Any ventilating equipment should be equipped with
explosion-proof motors and switch gear. Particular
caution should be exercised to deenergize or render
harmless any ignition sources that may be present in
vapor-contaminated space (e.g., vapor-fired hesiters,
light switches, nonexplosion-proof motors, and
electrical items).
5.8.3.4 Costs
Remedial action for the situations discussed will be
very site-specific; therefore, representative cost
figures cannot be presented here. Emergency
ventilating fans can be rented for between $10 and
$20 a day; more suitable ducted blowers with makeup
heaters and controls may cost $1000 to $5000,
installed.
5.8.4 Adsorption
5.8.4.1 General Description
Vapors collected at an LIST leak site can be treated by
adsorption, a process for transferring and
concentrating contaminants from one medium to
95
-------
another. The most commonly used adsorbent is
activated carbon; other adsorbents are specially
manufactured resins.
Adsorber systems consist of ducting or piping and a
blower for moving the contaminated air and a vessel(s)
that contains the adsorbent. In practice, the con-
taminated air is blown through the adsorber vessel,
where the contaminant is removed.
5.8.4.2 Application/Availability
(EPA 1985a)
Carbon adsorption can be used to treat vapors
containing volatile hydrocarbons and most halo-
genated organics. It can also control oxides of sulfur
and nitrogen and carbon monoxide. Resins are
capable of removing most organic contaminants from
gaseous streams; however, they are not widely used
in this application.
Vapor flow rate and influent and discharge adsorbate
concentrations must be monitored to determine
changeover/regeneration schedules. Automatic moni-
tors and microprocessors may be warranted for highly
complex and variable systems. Alarms and/or
shutdown controls may also be warranted for complex
systems or in sensitive or populated areas.
5.8.4.3 Design and Construction
Considerations (EPA I985a)
Adsorption techniques are well established for
removal of organic compounds and some inorganic
compounds from gaseous streams. Adsorption is
highly reliable if adsorbates and adsorbent are
properly matched, sufficient contact time is allowed,
and the adsorbent is regenerated or replaced before
saturation (and desorption) is reached. Many
adsorption systems are prepackaged and can be
quickly installed and put into operation by contractors,
suppliers, or manufacturers. Specially designed sys-
tems use off-the-shelf towers, blowers, and other
equipment and require additional installation time.
Operation of properly designed adsorption vapor-
treatment systems is essentially as automatic as the
vapor-delivery system; however, manual or special
automatic adjustments may be needed for highly
variable flows or adsorbate concentrations. Change-
over or regeneration of the adsorbent bed must be
conducted on a predetermined basis to ensure
continuous effective treatment.
Multiple-bed vessels are often required to allow
adequate contact time to optimize the frequency of
adsorbent changeover or regeneration. Partial or total
redundant capacity is often provided by extra bed
vessels to allow continuous operation during
changeover or regeneration.
Spent adsorbents can be disposed of in appropriate
landfills, incinerated, or can be regenerated.
Regeneration drives off the adsorbate and allows the
adsorbent to be reused for treatment. Granular
activated carbon is regenerated by heating it in a
reduced-pressure atmosphere. Resins are regen-
erated by washing them with appropriate solvents.
Adsorbed materials (e.g., solvents) can be recovered
from the regeneration process and reused.
5.8.4.4 Costs
Capital and operating costs for an adsorption system
are highly dependent on the specific system type,
vapor flow volumes, and the contaminant
concentrations to be treated.
Direct capital costs can be estimated by the equation:
DC = 8.3 (Q) + 34
where
DC = Direct capital costs, $1 OOO's
Q = Vapor flow rate, 1000 cfm
(300 cfm < Q < 10,000 cfm)
Operation/maintenance costs can be estimated by the
equation:
OM = 12.2(Q)
where
OM = Annual operating and maintenance costs,
$1000 per year
(These include maintenance, power, makeup
carbon, and spent carbon hauling/disposal.)
Q = Vapor flow rate, 1000 cfm
(300 cfm < Q :S 10,000 cfm)
5.8.5 Flaring
5.8.5.1 General Description
Flaring is a special category of combustion in which
vapors are exposed to an open flame; no special
features are used to control temperatures or time of
combustion. It is a means of disposing of vapors that
are easily burned and have no harmful products of
combustion. Supplementary fuels may be needed to
sustain continuous combustion.
5.8.5.2 Application/Availability
Flares are commonly used in the oil industry to
dispose of waste vapors and fumes at refineries; at
sewage treatment plants to dispose of digester
vapors; and at sanitary landfills to dispose of landfill
vapors. Flaring is applicable to gaseous waste streams
consisting of relatively simple hydrocarbons, such as
vapors from fuel tanks (EPA 1985a).
Flaring may be a suitable technology for disposing of
vapors ventilated from a subsurface structure if they
5-96
-------
are burnable and venting them directly to the
atmosphere is not feasible. Properly designed and
operated, flares pose no unusual safety impacts to
operators or others. The public sometimes considers
the presence of a visible flame to be a nuisance.
5.8.5.3 Design and Construction
Considerations (EPA 1985a)
Flaring systems, by virtue of their relative lack of
controllability, are generally considered to be incon-
sistent in performance. They are relatively simple to
fabricate and to install. Conventional steel plate, pipe,
and welding are used in fabrication.
Supplemental fuel is required to sustain a flame when
vapors have a very low heating value; however, vapors
with heating values as low as a few hundred Btu per
cubic foot can sustain a flame (natural gas has a
heating value of approximately 1000 Btu per cubic
foot).
Flame sensors, pilot flames, automatic sparkers, and
alarms are often used to detect loss of flame, to
attempt reignition, and to alert operators to system
performance problems. Shields can be strategically
placed as windbreaks to prevent the flame from being
"blown out."
The flow rate dictates the diameter and height of the
flare and the number of flares required. The flare must
be designed so that the flame is largely contained
within the body of the flare for safety reasons and to
allow adequate mixing of vapor and air. The oxygen
content of the vapor influences the air/gas ratio that is
sought in the combustion area of the flare.
A blower usually moves the waste vapor to the flare
through piping, a moisture knockout pot, and a stack.
The maximum allowable pressure drop is approx-
imately 60 in. H2O for the system.
5.8.5.4 Costs
Operating costs for flares are high because substantial
quantities of natural gas and steam (in the smokeless
type) are consumed. If the waste gas must be driven,
fan power costs for overcoming pressure drops also
may be high.
The capital costs of flare systems depend primarily on
the waste-gas flowrate, and secondarily on design and
elevation. The typical costs listed in Table 5-35
include ladders, platforms, knockout drums with seals,
and stacks high enough to ensure grade-level
radiation no greater than 1500 Btu/(h) (ft*). These
costs represent self-supporting type elevated flares
approximately 40 feet high. Costs of elevated flares
supported by guy wires (nominally 100 feet tall) range
from 30 percent higher (as opposed to the self-
supporting type) at flow rates of 250,000 Ib/h to 80
percent higher at flow rates of 2500 Ib/h.
Table 5-35.
Waste Vapor
Flow Rate (Ib/h)
1986 Capital Costs of
Elevated Flares (for vapor
with heat content of 60 Btu/ft3)
Cost ($)
3,000
10,000
50,000
100,000
7,200
9,800
10,700
11,000
* Data from Vatavuk and Neveril 1983, updated to 1986.
Elevated flares require supplemental fuel (in addition
to gas for pilots and purging) when a low-Btu vapor is
being burned. The required supplemental fuel
(natural gas), based on 880 h/yr operation, is listed in
Table 5-36.
Table 5-36.
Supplemental Fuel Requirements for
Elevated Flares*
Waste Vapor Flow
Rate (Ib/h)
Natural Gas
(106Btu/yr)
3,000
10,000
50,000
100,000
3,800
12,000
55,000
110,000
* Data from Vatavuk and Neveril 1983.
Steam consumption for smokeless flares (or others
requiring steam injection) is estimated to be 0.6 Ib of
steam per pound of vapor.
5.9 Surface Water/Drainage Controls
(EPA1985a)
Surface-water/drainage controls most applicable to
UST releases include diversion and collection
systems, grading, capping, and revegetation, which
are designed to minimize contamination of surface
waters, to prevent surface-water infiltration, and to
prevent offsite transport of surface waters that have
been contaminated. The following is a list of the
methods and technologies required for each control
activity.
Prevention of run-on/interception of runoff: The
following technologies are used to divert or
intercept surface water. Technologies thait are
designed to prevent or reduce run-on include
dikes, diversion channels, flood walls, terraces,
grading, and revegetation. Temporary diversion
dikes, diversion channels, and terraces are
constructed upslope of a site to direct run-on
from offsite to a collection system or away from
the site. Terraces are used in combination with
dikes or ditches to channel water stopped by the
terraces away from the site.
5-97
-------
Prevention of infiltration: The primary method for
preventing infiltration of onsite surface water is
capping. Grading also helps to minimize
infiltration by maximizing the amount of water
that will run off without causing significant
erosion. Revegetation can either promote or
minimize infiltration.
Collection and transfer of water: The following
technologies are used to collect diverted water
and discharge or transfer it to storage or
treatment. Chutes (or flumes) and downpipes
are designed to transfer water away from
diversion structures such as dikes or terraces to
stabilized channels or outlets. Waterways can
be used to intercept or divert water as well as to
collect and transfer water diverted elsewhere.
Storage and discharge of water: Technologies
for this purpose include seepage basins and
ditches, sedimentation basins, and storage
ponds. Their function depends on the level of
contamination of the water they receive.
Seepage basins and ditches are used to
discharge uncontaminated or treated water
down and away from the site. Sedimentation
basins are used to control suspended solid
particles in surface-water flow.
The most effective strategy for managing surface flow
is often a combination of several control technologies.
Tabla 5-37. Summary of Surface Water/Drainage Controls*
Table 5-37 summarizes the major surface water
controls and indicates their general function.
5.9.1 Diversion/Collection Systems
5.9.1.1 General Description
This section addresses the various surface water
diversion and collection methods. These include
dikes and berms, channels and waterways, terraces
and benches, chutes and downpipes, seepage
basins and ditches, and sedimentation basins and
ponds. Each is addressed separately in the following
two subsections.
5.9.1.2 Application/Availability
Dikes and Berms
Dikes and berms are temporary structures used to
prevent excessive erosion of newly constructed
slopes until more permanent drainage structures are
installed or until the slope is stabilized with vegetation.
These structures are frequently used to provide
temporary isolation of wastes prior to removal or
effective containment. Their use is especially wide-
spread during excavation and removal operations
where it is necessary to isolate contaminated soils that
have been temporarily staged on site. These tem-
porary structures are designed to handle relatively
small amounts of runoff; they are not recommended
for drainage areas larger than 5 acres. Diversion of
General Function
Technology
Capping
Grading
Revegetation
Dikes and berms
Prevent or
Intercept
Run-on/Runoff
X
X
X
Prevent or Collect
Minimize and
Infiltration Transfer
X
X
Store
and
Discharge
Channels and
waterways
Terraces and
benches
Chutes and
downpipes
Seepage basins
and ditches
Sedimentation
basins and ponds
Data from EPA 1985a.
5-98
-------
storm runoff will decrease the amount of water
available to infiltrate the soil cover.
Channels and Waterways
Channels and waterways are excavated ditches,
usually wide and shallow, with trapezoidal, triangular,
or parabolic cross sections. They are used primarily to
intercept runoff or slope length, and they may or may
not be stabilized. Channels stabilized with vegetation
or stone rip-rap are used to collect and transfer
diverted water offsite or to onsite storage or treatment.
Terraces and Benches
Terraces and benches are embankments constructed
along the contour of very long or very steep slopes to
intercept and divert flow and to control erosion by
reducing slope length. These structures are classified
as bench terraces or drainage benches. Bench
terraces are used primarily to reduce land slope,
whereas drainage benches on broad-based terraces
are used to remove or retain water on sloping land.
Chutes and Downpipes
Chutes and downpipes are structures used to carry
concentrated flows of surface runoff from one level to
a lower level without erosive damage. They generally
extend downslope from earthen embankments (dikes
or berms) and convey water to stabilized outlets
located at the base of terraced slopes.
Chutes and downpipes often represent key elements
in combination surface control systems. They are
especially effective in the temporary prevention of
erosion on long, steep slopes, and they can be used
to channel storm runoff to sediment traps, drainage
basins, or stabilized waterways for offsite transport.
Chutes are limited to heads of about 18 feet or less,
and downpipes are limited to drainage areas up to 5
acres in size. These structures provide a quick
solution for emergency situations on which down-
slope ditches or waterways overflow during severe
storms and threaten to erode the base of disposal fill
areas.
Seepage Basins and Ditches
Seepage or recharge basins and ditches are used to
discharge water collected from surface water
diversions and ground-water pumping or treatment.
They may also be used in in situ treatment to force
treatment reagents into the subsurface.
Seepage basins and ditches are most effective in
highly permeable soils so that recharge can be
performed. They are not applicable at sites where
collected runoff or ground water is contaminated.
Basins and ditches are normally used in areas with
shallow ground-water tables, as very deep basins or
trenches can be hazardous.
Sedimentation Basins and Ponds
Sedimentation basins are used to control suspended
solids entrained in surface flows. A sedimentation
basin is constructed by placing an earthen dam across
a waterway or natural depression, by excavation, or by
a combination of both. The purpose of a sedimen-
tation basin is to impede runoff containing solids, and
thus to allow sufficient time for the paniculate matter to
settle.
Sedimentation basins are usually the final step in the
control of diverted, uncontaminated runoff prior to its
discharge. They are especially useful in areas where
the runoff has a high silt or sand content. They are an
essential part of any good surface flow-control system.
5.9.1.3 Design and Construction
Considerations
Dikes and Berms
Dikes and berms ideally are constructed of erosion-
resistant, low-permeability, clayey soils. Compacted
sands and gravel, however, may be suitable for
interceptor dikes and berms. The general design life
of these structures is generally no more than one
year; however, seeding and mulching or chemical
stabilization of dikes and berms may extend their life
expectancy. Stabilization with gravel or stone rip rap
immediately upslope of diversion dikes will also
extend performance life.
The techniques for constructing dikes and berms are
well-established, and the necessary excavation and
grading equipment is frequently already available at
the disposal site. The required earthfill is often
available on site as well.
Channels and Waterways
As an alternative to excavated ditches, channels can
be constructed of half-round pipe. The pipe can be
constructed of cut corrugated metal pipe (CMF'), or
sectional slope drains made of asbestos-impregnated
asphalt can be purchased for this purpose. The
channels are formed by placing the half-round pipe
below ground. This type of channel is easier to install
than earthen channels, and maintenance costs are
lower. In addition, they decrease infiltration into the
site. Like earthen channels, half-round pipe channels
may be constructed on the perimeter of an UST site
and moved as needed to protect other portions of the
site. They also are effective for carrying storm water
runoff over a filled area when it is not practical to divert
the runoff around the fill.
Channels and waterways should be designed to
accommodate flows resulting from 10- or 25-year
rainfall events (storms). More importantly, they should
be designed and constructed to intercept and convey
99
-------
such flows at nonerosive velocities. The design of
channel and waterways is generally based on the
Manning formula for steady uniform flow in open
channel.
Terraces and Benches
Although benches and terraces are slope-reduction
devices, they are generally constructed with reverse
or natural fall. Diversions and ditches included in
bench/terrace construction may be seeded and
mulched, sodded, or stabilized with riprap or soil
additives. Stabilization may include any combination
of these methods. Lining the channels with concrete
or grouted riprap is a more costly alternative.
In the design of a terrace system, proper spacing and
location of terraces and providing a channel with
adequate capacity are major considerations. The
spacing between benches and terraces will depend
on the length and steepness of the slope and the
type of soil. In general, the longer and steeper the
slope and the more erodible the cover soil, the less
the distance should be between drainage benches to
maximize erosion reduction. For slopes with steep-
ness greater than 10 percent, the maximum distance
between drainage benches should be approximately
100 feet, i.e., one bench for every 10 feet of rise in
elevation. When the slope is greater than 20 percent,
one bench should be placed for every 20 feet of rise
in elevation.
Benches and terraces are constructed with a variety of
commonly used excavation equipment including
bulldozers, scrapers, and graders. The usual well-
established techniques are applied, and local fill
material is used. Benches and terraces must be
sufficiently compacted and stabilized with appropriate
vegetation to accommodate local topography.
Chutes and Downpipes
Chutes and downpipes can be constructed quickly
and inexpensively. No special materials or equipment
are required. Temporary downpipes may be con-
structed by joining half-round sections of bituminous
fiber or concrete pipe. These structures provide a
quick solution for emergency situations when down-
slope ditches or waterways overflow during severe
storms and threaten to erode the base of disposal fill
areas.
Seepage Basins and Ditches
Considerable flexibility is possible in the design of
seepage basins and ditches. A seepage basin
typically consists of the basin itself, a sediment trap, a
bypass for excess flow, and an emergency overflow.
Because a considerable amount of recharge occurs
through the sidewalls of a basin, the use of a previous
material of construction is advisable. Gabions are
frequently used to make sidewalls.
Well-established techniques and procedures are
used in the construction of seepage basins and
ditches. Much of the necessary equipment and
material will be found on site. Such things as piping
and gravel may have to be ordered, but they should
be readily available.
Dense turf on the side slopes of these basins will not
only prevent erosion and sloughing, but will also allow
a high infiltration rate. Prevention of scouring by the
inlet is an important consideration to reduce main-
tenance requirements. This can be accomplished by
use of a "hydraulic jump" or an impact stilling basin
before the water flows into the recharge basin.
Percolation can be improved by construction of gravel-
filled trenches along the basin floor. ;
Sedimentation Basins and Ponds
The removal of suspended solids from waterways is
based on the concept of gravitational settling of the
suspended material. A typical design for a sedi-
mentation basin embankment includes a principal and
emergency spillway, an antivortex device, and the
basin. The principal spillway consists of a vertical pipe
or riser jointed to a horizontal pipe (barrel) that extends
through the dike and outlets beyond the water
impoundment. The riser is topped by an antivortex
device and trash rack that improve the flow of water
into the spillway and prevent floating debris from
being carried out of the basin. The riser should be
watertight and, except for a dewatering opening at the
top, should be free of holes, leaks, or perforations.
The riser base should be attached to a watertight
connection, and it should weigh enough to prevent
the riser from floating. The water discharged from the
sediment basin through the principal spillway should
be conveyed in an erosion-free manner to an existing
stable stream.
Before construction begins, the areas under the
embankment and any structural works should be
cleared, grubbed, and stripped of topsoil to remove
trees, vegetation, roots, or other objectionable mate-
rial. Fill material for the embankment should be clean
mineral soil, free of roots, woody vegetation, over-
sized stones, rocks, or other objectionable material.
Areas should be scarified before they are filled. The
moisture content of the fill material should be high
enough to permit the material to be formed by hand
into a ball without crumbling. This will facilitate proper
compaction. Compaction is obtained by routing the
hauling equipment over the fill in such a manner that
the entire surface of the fill is traversed by at least one
wheel or tread track of the equipment, or by using a
compactor.
The riser of the principal spillway should be securely
attached to the barrel by a watertight connection, and
the barrel and riser should be placed on a firmly
compacted soil foundation. The base of the riser
5-100
-------
should be firmly anchored to prevent floating.
Pervious materials such as sand, gravel, or crushed
stone should not be used as backfill around the barrel.
At least 2 feet of fill material should be placed around
the pipe in thin layers and compacted by hand at least
to the same density as the embankment before it is
crossed with construction equipment.
5.9.1.4 Costs
Costs for surface water diversion/collection structures
include those for ditch linings, riprap slope protection,
soil testing, corrugated metal pipe, sheet piling,
backflow valves, and sumps. Table 5-38 summarizes
these unit costs and the structures to which they
apply.
All cost estimates should be determined on a site-
specific basis and be based on the specific structures
to be installed, all associated earthwork, and any
special appurtenances that may be required. A
general methodology for estimating costs for
construction of surface-water diversion/collection
structures should contain the following elements:
Source of required earth fill (onsite vs. offsite)
and hauling distances.
Amount of fill required (cubic yards).
Type and quantity of other materials required
(cubic yards of pipe, square feet of riprap, etc.).
Costs of installation or placement of these
materials (using unit costs).
Costs of required stabilization for earthen
structures (berms, etc.) based on the area (in
square yards) to be stabilized; revegetation,
riprap, or gravel stabilization.
Required maintenance or repair costs for a given
time period based on reasonable assumptions;
for example, assuming the diversion requires
rebuilding (new fill and compaction) twice a year,
after major storms, costs will be...
Summation of all calculated costs to arrive at the
total estimated construction and maintenance
expenditures.
Costs are derived by simply multiplying unit costs
shown in the table by the required quantities of the
material or service. These costs will give gross
estimates only; they are to be used as general
guidelines for the decision-maker in evaluating
alternative strategies.
5.9.2 Grading
5.9.2.1 General Description
Grading is the general term applied to techniques
used to reshape the land surface to manage surface-
water infiltration and runoff and to control erosion.
The required spreading and compaction steps are
techniques practiced routinely by earthwork con-
tractors. The equipment and methods used in grading
are essentially the same throughout the country, but
their application will vary by site. Grading is often
performed in conjunction with surface sealing and
revegetation as part of an integrated site-closure plan.
5.9.2.2 Application/Availability
The techniques and equipment used in grading
operations are well established and widely used in all
forms of land development. Contractors and equip-
ment are usually available locally, which expedites the
work and avoids extra expenses.
5.9.2.3 Design and Construction
Considerations
Grading is used to modify the natural topography and
runoff characteristics of a site and thereby to control
infiltration and erosion due to surface water. Con-
tinuous grades are established to ensure that runoff
water does not pond. The choice of specific grading
techniques for a given site will depend on the desired
site-specific functions of a graded surface. A graded
surface may reduce or enhance infiltration and detain
or promote runoff.
Erosion control may be considered a complicating
variable in the design and performance of a grading
scheme. The design of graded slopes at sites should
balance infiltration and runoff control against possible
decreases in slope stability and increases in erosion.
The design of specific slope configurations, the
choice of cover soil, the degree of compaction, and
the types of grading equipment used will all depend
on local topography, climate, future land use, and
drainage methods.
All States and localities publish guidelines and
regulations for performance of any grading work.
These publications should be consulted before a
grading system is designed.
The equipment used to construct graded slopes
consists of both standard construction and spe-
cialized earth-moving vehicles. Excavation, hauling,
spreading, and compaction of cover materials are the
major elements of a complete grading operation.
Grading vehicles include scrapers, crawler bulldozers
and loaders, rubber-tired bulldozers and loaders, and
compactors.
It is important that grades be constructed in ac-
cordance with design specifications. Particular atten-
tion should be given to the backfill and compaction
methods used by the contractor. Normally, compac-
tion is achieved through repeated passes of leveling
equipment over the area in several different direc-
tions. Any compaction tests specified in the design
should be performed to assure the permanence and
stability of the grading work.
101
-------
Table 5-38. 1986 Unit Costs Associated With Surface Water Diversion
Description
Excavation, hauling, grading
(spreading and compaction)
Trench excavation
Loam, sand, and loose gravel
1 to 6 ft deep. 1/2:1 sides
6 to 10 ft deep
Compacted gravel and till
1 to 6 ft deep. 1/2:1 sides
6 to 10 ft deep
Building embankments;
spreading, shaping,
compacting
Material delivered by
scraper
Material delivered by
back dump
Placement of ditch liner
pipe, 1/3 section
15-in. radius
18-in. radius
24-in. radius
Catch basin sump,
3ftx4ftx 1.5ft
Corrugated galvanized
steel underdrain pipe,
asphalt-coated, perforated
12-in. diameter, 16-gauge
18-in. diameter, 16-gauge
Corrugated galvanized metal
pipe, with paved invert
18-in. diameter, 14-gauge
36-in. diameter, 12-gauge
48-in. diameter, 12-gauge
20 ft deep. 27 Ib/ff
25ftdeep,38lb/fp
Backflow preventer; gate
valves, automatic operation,
flanged, 10-in. diameter
Floating baffles
Applicable
Structures
All
D/B, D/D/W,
BT. L, DT/B
All
D/D/W, DB, CD
i DT/B
'
DT/B
C/C, SB
L (seepage control)
L (drainage control)
SB
and Collection Structures
Range of Unit Costs ($)
See Table 5-1
0.86to1.03/yd3
0.86 to 1.02/yd3
0.86to1.21/yd3
0.86to1.09/ycP
0.43 to 0.85/yd3
0.91 to 1.35/yd3
15/ft
21/ft
31/ft
224 each
22/ft
31/ft
35.04/ft
89.17/ft
66.88/ft
10.94/ft2
12.68/ft2
16.12/ft2
9,500 each
15to51/ft
Source"!"
a)
(2)
(2)
(2)
(2)
(3)
5-102
-------
Table 5-38. (Continued)
Description
Applicable,
Structures
Range of Unit Costs ($)
Source"!"
Sump pumps, 10-fthead,
automatic
Bronze
Cast iron
Revegetation, mulching,
maintenance
Loose gravel, excavation,
loading, hauling 5 miles,
spreading, and compacting
Stone riprap; dumped from
trucks, machine-placed
Soil testing
Liquid and plastic limits
Hydrometer analysis,
specific gravity
Moisture content
Permeability
Proctor compaction
Shear tests
Triaxial shear
Direct shear
Temporary diversion dike
L (backwater drainage)
D/B, D/D/W, BT, L
All (slope protection;
drainage)
All (slope protection;
channel and outlet
stabilization)
All (preconstruction
evaluation)
D/B
Temporary sediment
construction, drainage area
1 to 25 acres
50 to 75 acres
75 to 100 acres
100 to 125 acres
Sediment-removal from SB
basins
Paved flume, installed C/D, SB
Level spreader D/B, D/D/W, BT, C/D
construction
405 to 2, 138
4,048 to 6,645
6,645 to 8,565
8,565 to 11,208
4.05 to 9.35/yo3
27to41/yd2
3.32 to 6.64/linear ft
(4)
(4)
(4)
(4)
W
(4)
(4)
25 to 520 each
150 to 300 each
See Table 5-40
8 to 8.75/yd3
22/yd3
31/test
51/test
7.35/test
67/test
110to125/test
150to290/test
100 to 270/test
1,35to2.69/linearft
(2)
(2)
(1)
(2)
(2)
(2)
(2)
(2)
(2)
(2)
(4)
Key: D/B = dikes and berms; D/D/W = ditches, diversions, ar d waterways; BT = bench terraces;
CD = chutes and downpipes; L = levees; DT/B = drainage trer cnes and basins; SB = sediment basins;
DB = drainage benches. |
T Data from (1) McMahon 1984; (2) Godfrey 1984a; (3) EPA 1985a; (4) Virginia SWCC 1980.
5-
103
-------
5.9.2.4 Costs
Table 5-39 presents unit costs associated with
grading equipment and methods. Costs associated
with heavy equipment maintenance (fuel, repairs, etc.)
are not addressed. Costs of excavation, hauling,
spreading, and compaction will vary with equipment
type and size, the cover material being graded, haul
distance, support labor required, and unforeseen
construction difficulties. These costs are represen-
tative of average contractor bid prices for performance
of the work and therefore include charges for
overhead and profit. All costs are reported in 1986
dollars. ;
5.9.3 Capping
5.9.3.1 General Description
Capping refers to the process used to cover con-
taminated materials in place to prevent their contact
with the land surface and ground water. The designs
of modem caps usually conform to the performance
standards in 40 CFR 264.310, which addresses
RCRA landfill closure requirements. These standards
include minimum liquid migration through the wastes,
low-cover maintenance requirements, efficient site
drainage, high resistance to damage by settling or
subsidence, and a permeability lower than or equal to
the underlying liner system or natural soils. These
performance standards may not always be ap-
propriate, particularly when the cap is intended to be
temporary, in places where precipitation is very low,:
and when the capped waste is not leached by
infiltrating rainwater.
A variety of cap designs and capping materials are
available. Most cap designs are multilayered to
conform with the aforementioned design standards;
however, single-layered designs are also used for
special purposes. The selection of capping materials
and cap design are influenced by specific factors such
as local availability and costs of cover materials,
desired functions of cover materials, the nature of the
wastes being covered, local climate and hydro-
geology, and projected future use of the site in
question.
5.9.3.2 Application/Availability
Capping is necessary whenever contaminated
materials are to be buried or left in place at an UST
release site. Capping also may be performed when
extensive subsurface contamination at a site pre-
cludes excavation and removal of the soil because of
potential hazards and/or unrealistic costs.
Capping is often performed together with ground-
water extraction or containment technologies to
prevent (or significantly reduce) further plume
development. This combined effort reduces the time
needed to complete ground-water cleanup opera-
tions. Ground-water monitoring wells also are often
used in conjunction with caps to detect any
unexpected migration of the capped material. A vapor-
collection system should always be incorporated into a
cap when the capped material is volatile. Surface water
control technologies such as ditches, dikes, and
berms are also associated with capping, as these
structures are often designed to control rainwater
drainage from the cap. Two other surface-water
Table 5-39. 1986 Unit Costs Associated With Grading
Description
Topsoil (sandy loam) from borrow pits, excavation
hauling, spreading, and grading (within 25 miles);
labor, materials, and equipment
Onsite excavation, hauling, spreading, and
compaction of earth (1000- to 5000-ft haul);
labor and equipment only
Sandy loam topsoil; material only
Excavating, hauling 2 miles, spreading and
compacting loam, sand, or loose gravel (with
front-end loader); labor, material, and equipment
Grading, site excavation, and fill (no compaction):
75-hp bulldozer, 300-ft haul
300-hp bulldozer, 300-ft haul
Field-density compaction testing of soils
Unit Cost ($)
Source
16/yd3
2 to 4/yd3
2.30/yd3
6.85 to 7.00/yd3
3.50/yd'
2.40/yd3
110/day
(1)
(1)
d)
(1)
(2)
(2)
(D
' Data from (1) McMahon 1984; (2) Means 1984a.
5-104
-------
control technologies-grading and revegetation-are
incorporated into multilayered caps.
Availability of capping material is somewhat site-
specific because the local soils are typically used with
admixtures to form parts of the cap. Synthetic mem-
branes, which are available in varying sizes, can be
overlain and spliced in the field. In general, capping
should be a readily available technology for any LIST
release site.
5.9.3.3 Design and Construction
Considerations
The primary purpose of a cap is to prevent rainwater
infiltration. The two basic designs are multilayered and
single-layered. Although multilayered caps are the
more common, a single-layered cap may be accept-
able when a site is being covered temporarily, in an
area where little or no ground water exists, or when
continual maintenance of the integrity of a cap cannot
be absolutely assured. A vapor collection system
should always be included in the design of a cover
when there is any indication that the underlying
contaminant is volatile.
The design of multilayered caps generally conforms to
EPA's guidance under RCRA, which recommends a
three-layered system consisting of an upper vege-
tative layer, a drainage layer, and an underlying low-
permeability layer. Single-layered caps can be con-
structed of various low-permeability materials; how-
ever, natural soil admixes are not recommended
because they are disrupted by freeze/thaw cycles and
exposure to drying causes them to shrink and crack.
The most effective single-layer caps are composed of
concrete or bituminous asphalt.
Construction considerations for single-layered caps
vary depending on the cap materials used (e.g.,
concrete, asphalt, clay); therefore, appropriate con-
struction guidance should be acquired according to
the cap material being considered. The EPA docu-
ment entitled "Lining of Waste Impoundment and
Disposal Facilities," SW-870, contains references for
guidance in the construction of caps made of several
different materials.
Construction of a multilayered cap typically includes a
vegetative layer based on topsoil, a drainage layer
composed of sand, and a low-permeability layer
formed by a combined synthetic and soil liner system.
This type of cap function diverts infiltrating liquids from
the vegetative layer through the drainage layer and
away from the underlying contaminated materials.
The low-permeability layer of a multilayered cap can be
composed of natural soils, admixed soils, a synthetic
liner, or any combination of these materials; however,
a synthetic liner overlying at least 2 feet of low-
permeability natural soil or soil admix is recommended
because the synthetic liner allows virtually no liquid
penetration for at least 20 years, whereas the soil layer
provides assurance of continued protection even if
the synthetic liner fails.
>
5.9.3.4 Costs
The cost of a cap depends on the type of materials
selected, the thickness of each layer, and the area to
be covered. General material and installation costs for
multilayered caps are presented in Table 5-40, which
indicates the complexity of this type of capping. In a
recent RCRA Part B Permit Application for a 4-acre
hazardous waste landfill, the estimated installed cost
of a multilayered cap was $5.45/ft2. The design for this
cap included 3 feet of topsoil overlying a 1 -foot sand
layer overlying 1 foot of compacted clay overlying a 30-
mil high density polyethylene (HOPE) liner overlying 2
feet of compacted clay. Filter fabric was specified
between the topsoil and sand drainage layer to
prevent clogging. Quality control testing of each layer
of the cap was included in the installation cost
estimates.
On the other hand, a single-layer cap comprised of a
sprayed asphalt membrane, as might be more typical
for an underground storage system leak situation,
might cost $2.33 to 3.96/sq. yd.
5.9.4 Revegetation
5.9.4.1 General Description
The establishment of a vegetative cover is a cost-
effective method to stabilize the surface of newly
graded and/or capped sites. Revegetation decreases
erosion by wind and water and contributes to the
development of a naturally fertile and stable surface
environment. Also, the technique can be used to
restore the appearance of sites following any cleanup
operations.
A systematic revegetation plan will include: 1) se-
lection of suitable plant species, 2) seed bed
preparation, 3) seeding/planting, 4) mulching and/or
chemical stabilization, and 5) fertilization and
maintenance.
5.9.4.2 Application/Availability
Grasses such as fescue and lovegrass provide a quick
and lasting ground cover, with dense root systems
that anchor soil and enhance infiltration. Legumes
(lespedeza, vetch, clover, etc.) store nitrogen in their
roots, enhancing soil fertility and assisting the growth
of grasses. They are also readily established on steep
slopes. Shrubs such as bristly locust and autumn
olive also provide a dense surface cover, and certain
species are quite tolerant of acidic soils and other
possible contaminated site stresses. Trees are
generally planted in the later stages of site restoration,
after grasses and legumes have established a stable
ground cover. They help provide long-term protective
cover and build up a stable, fertile layer of decaying
leaves and branches. A well-mixed cover of grasses,
5-i05
-------
Table 5-40.
1986 Unit Costs Associated With Capping*
Itemt
Unit Cost ($)
Clearing and grubbing
Excavation
Earthfill
Berms and levees
Soil liners
Backfill
Soil import
Drainage sand
Drainage rock (rounded)
Soil placement
Vegetation, mulch, and hydroseed
Geotextile fabrics ;
Bentonite admix (2 to 9 Ib/yd3)*
Membrane liners
Nonreinforced
30-mil PVC
30-mil CPE
30-mil Butyl/EDPM
30-mil Neoprene
100-mil HOPE
Reinforced
36-mil Hypalon (CSPER)
60-mil Hypalon (CSPER)
36-mil Hypalon
Installation, excluding earthwork
1,130/acre
1.65/yd3
2.15/yd3
3.20/yd3
3.20/yd3
10.80/yd3
10.80/yd3
1.05/yd3
1,130/acre
1.00 to 3.20/yd2
0.20to1.15/ft2
0.25to0.35/tt2
0.35 to 0.45/ft 2
0.45 to 0.60/ft2
0.70 to 0.80/ft 2
1.10 to 1.65/ft2
0.50 to 0.65/ft2
0.80to1.05/ft2
0.50 to 0.65/ft2
0.60to1.25/ft2
Based on costs for a 400,000 ft2 area presented in Cope, et al.,
1984, and updated to 1986 dollars by using the ENR Construction
Cost Index.
t Key: PVC = polyvinyl chloride; CPE = chlorinated polyethylene;
EDPM = ethylene-propylene-eiene-monomer; CSPER = chlorosulfbnated
polyethylene (reinforced); HOPE = high density polyethylene.
$ Includes mixing and placing.
shrubs, and trees will ultimately restore both economic
and aesthetic value to a site and will provide suitable
habitat for populations of both humans and wildlife.
Table 5-41 summarizes the suitability of various
grasses and legumes for revegetation purposes.
Native species of trees and shrubs, particularly those
with a shallow root system, should generally be
specified.
Local landscaping contractors should be hired to
perform revegetation work, preferably someone who
has experience in growing local varieties of vegetation
and who can make recommendations on the suitability
of the different varieties.
5.9.4.3 Design and Construction
Considerations
The selection of suitable plant species for a given
disposal site depends on several site-specific
variables. These variables include cover soil charac-
teristics (grain size, organic content, nutrient and pH
levels, and water content), local climate, and site
hydrology (slope steepness and drainage charac-
teristics). Individual species must be chosen on the
basis of their tolerance to such site-specific stresses
as soil acidity and erodibility. Other important
considerations include the compatibility of the species
with other plants selected to be grown on the site,
resistance to insect damage and diseases, and
suitability for future land use.
5-106
-------
Table 5-41. Important Characteristics of Grasses and
Characteristics
Texture
Legumes'
Common Examples
Fine
Coarse
Growth height
Short
Medium
Tall
Growth habit
Bunch
Sod former
Reproduction
Seed
Vegetative
Seed and vegetative
. Annual
Summer
Winter
Perennials
Short-lived
Long-lived
Maintenance
Difficult
Moderate
Easy
Shallow-rooted
Weak
Strong
Deep-rooted
Weak
Strong
Kentucky blue grass, bentgrass, red fescue
Smo< th brome grass, reed canary grass, timothy
I
Kentucky bluegrass, buffalo grass, red fescue
Redtop, perennial rye grass
Smooth brome grass, timothy, switch grass
Timothy, big bluestem, sand dropseed,
peren'nial rye grass
Quick grass, smooth brome grass, Kentucky
bluegrass, switch grass
Red and alsike clover, sand dropseed,
rye, perennial rye grass, field brome grass
Prairie cord grass, some bent grasses
White clover, crown vetch, quack grass,
Kentt cky bluegrass, smooth bromegrass
Rabbit clover, oats, soybeans, com, sorghum
Rye, hairy vetch, field brome grass
Timothy, perennial rye grass, red and white
clover
Bird'sjfoot trefoil, crown vetch, Kentucky
bluegrass, smooth brome grass
Tall fe'scue, reed canary grass, timothy, alfalfa
Kentucky bluegrass, smooth brome grass
Crown vetch, white clover, bird's-foot trefoil,
big bluestem
Sand dropseed, crab grass, foxtail, white clover
Timothy, Kentucky bluegrass
Many weeds
„ .. T. stem, switch grass, alfalfa, reed
canary grass
5-107
-------
Tablo 5-41. (Continued)
Characteristics
Moisture
Dry
Moderate
Wet
Temperature
Hot
Moderate
Cold
Common Examples
Sheep fescue, sand dropseed, smooth
brome grass
Crested wheat grass, red clover
Reed canary grass, bent grass
Lehman love grass, four-wing saltbush,
rye grass
Orchard grass, Kentucky bluegrass, white clover
Alfalfa, hairy vetch, smooth brome grass, slender
wheat grass
Data from Lutton 1982.
Long-term vegetative stabilization and site reclamation
require the proper planting of compatible mixes of
grasses, legumes, shrubs, and trees. Short-term
revegetation efforts generally require the use of low-
cost, quick-growing perennial and self-seeding annual
species, usually grasses. In areas where a quick
vegetative cover is essential for preventing erosion
and pollutant transport, the use of an approved sod
could prove beneficial.
The optimum time for seeding depends on local
climatic considerations and the individual species.
Early fall seeding is recommended for most perennial
species in most localities. Spring and early summer are
usually the best times for seeding annuals, but they
can be planted for quick vegetation whenever soil is
damp and warm. In mild climates (e.g., southeastern
United States), the growth of both summer and winter
grasses will extend the range of erosion resistance for
cover soils.
Mulches or chemical stabilizers may be applied to
seeded soils to aid in the establishment of vegetative
cover. Organic mulches such as straw, hay, wood
chips, sawdust, dry bark, bagasse (unprocessed
sugar cane fibers), excelsior (fine wood shavings), and
manure protect bare seedbed slopes from erosion
prior to germination. Also, thin blankets of burlap,
fiberglass, and excelsior can be stapled down or
applied with asphalt tacks to form protective mulch
mats for germinating seedbeds.
5.9.4.4 Costs
Total 1986 cost estimates for revegetation range from
$1290 to $8525/acre. Costs for revegetation vary:
widely depending upon the site conditions. The
lowest cost estimate represents a hypothetical site
that required hydroseeding (lime, fertilizer, field seed)
only. The highest cost estimate represents a pro-
posed restoration of a secondary growth, temperate,
deciduous forest that required heavy liming to
neutralize the highly acidic soils. Unit costs associated
with revegetation are reported in Table 5-42. All costs
are reported in 1986 dollars and include contractors'
overhead and profit.
5-108
-------
J
Table 5-42. 1986 Unit Costs Associated With Revegetation
Description
Hydraulic spreading (hydroseeding), lime,
fertilizer, and seed
Mulching, hay
Loam topsoil
Loam topsoil, removal and stockpiling
6 feet deep on site with a 200-hp bulldozer
200-ft haul
500-ft haul
Hauling loam 2 miles on site
Spreading loam 2 to 6 inches deep
Slopes
Level areas
Plant-bed preparation (unspecified), 18
inches deep, by machine
Hydraulic seeding and fertilization of large
areas with wood-fiber mulch
Handspreading of mulch (wood chips) 2
inches deep
Liming
Level areas
Slopes
Fertilizing (no insecticides)
Level areas
Slopes
Seeding
Level areas
Slopes
Jute mesh, stapled (erosion control)
Sodding 1 inch deep
Level areas
Slopes
Maintenance
Grass mowing
Slopes
Level areas
Refertilization
Weeding/pruning shrubs
Onsite planting
Trees
Evergreens
Black pines
Yews
Junipers
Shade trees (balled and
burlapped)
o™
Unit Cost ($)
873/acre
317/acre
6.70/yd3
1.20/yd3
4.56/yd3
3.94/yd3
1.07to3.20/yd3
0.79 to 2.33/yd 3
5.60/yd2
0.42/yd2
1 .26/yd2
90/acre
590/acre
460/acre
660/acre
675/acre
873/acre
0.87/yd2
2.93/yd2
3.37/yd2
93.20/acre
38.90/acre
249/acre
2000/acre
30 to 36 inches, 1 1 5 ea.
36 to 42 inches, 117ea.
42 to 48 inches, 165 ea.
4 to 5 ft, 208 ea.
5 to 6 ft, 265 ea.
2.5 to 3 ft, 28 ea.
2 to 2.5 ft, 38 ea.
4 to 5 ft, 50 ea.
109
Source*
(1)
(1)
(1)
(2)
(2)
(D
(1)
(1)
(2)
(2)
(2)
(D
(1)
(1)
(1)
(D
<1)
(2)
(2)
<2)
(1)
(D
(1)
(1)
(1)
(1)
(1)
(1)
(1)
(2)
(2)
(2)
-------
Table 5-42.
(Continued)
Description
Unit Cost ($)
Source*
Shade trees (balled and
burlapped)
Birch
Oak
Shrubs (balled and burlapped)
Honeysuckle shrub
6 to 8 ft, 79 ea. (1)
8 to 10 ft, 111 ea. (1)
1.5- to 2.5-in. diameter, 384 ea. (1)
2.5- to 4.0-in. diameter, 462 ea. (1)
6 to 8 ft, 99 ea. (2)
8 to 10ft, 500 ea. (2)
2 to 3 ft, 60 ea. (1)
3 to 4 ft, 90 ea. (1)
4 to 5 ft, 110 ea. (1)
3 to 4 ft, 20 ea. (2)
* Data from (1) McMahon 1984; (2) Means 1984a.
5.10 Restoration of Contaminated
Water Supplies and Utility/Sewer
Lines (EPA 1985a)
Contaminants from leaking USTs can enter public;
water systems through a wide variety of pathways and
thus contaminate these systems as well as the water in
them. Once contaminated, a water system can
become a secondary source of contamination to
which system users can be exposed over long
periods of time.
Sanitary and storm sewers can become contaminated
by infiltration of the leaking fluid or contaminated
ground water through cracks, ruptures, or poorly
sealed joints in piping and by direct discharges into
the system. Potable water supply mains also can
become contaminated by contact with contaminated
water that may inadvertently flow through them, or by^
infiltration of the leaking fluid or contaminated ground
water. Water mains are less susceptible to the
infiltration of contaminants, however, because they
are generally full-flowing, pressurized systems. The
potential public health consequences of the con-
tamination of municipal mains carrying potable water
supplies to commercial and residential consumers are
obviously much greater than the consequences of
contaminated sewage flowing to a treatment plant or
runoff draining into surface waters.
5.10.1 Alternative Central Water Supplies
5.10.1.1 General Description
Unless the discovered leak in an underground
storage system is a catastrophic one that grossly
affects a municipality's central water supply system, an
alternative central water supply may not be needed.
Nevertheless, some general discussion of the subject
is in order. Providing alternative central water supplies
or sources of water that serve many users through
central distribution systems generally involves one or
more of the following approaches:
Purchase of water from another supply
Provision of a new surface water intake(s)
Provision of a new ground-water well(s)
5.10.1.2 Application/Availability
The contaminated water supply may be abandoned,
or it may be blended with the new supply to achieve
acceptable water quality by dilution. Combinations of
the possible approaches may be applied either
concurrently (multiple replacement supplies) or
consecutively (emergency water purchased from a
neighboring supply unit, followed by new wells or
intakes).
Purchase of treated water from another supply
requires a cross-connection(s) between the systems.
Many neighboring public water departments, author-
ities, and companies maintain networks of inter-
connections that allow ready flow between systems
for emergencies such as droughts, fires, line
breakage, or malfunction of treatment facilities. Where
cross-connections do not exist, water transmission
lines can be installed.
Provision of a new surface water intake may be
feasible where a groundwater source is to be replaced
or where a replacement surface water intake would
hydraulically isolate the water supply system from
contaminated surface water.
Provision of new ground-water wells is often feasible
where the extent of aquifer contamination is relatively
confined and would not be expected to be drawn to
the area of influence of the new weHs, or where other
(usually deeper) aquifers can be tapped as a
replacement water supply.
5-110
-------
5.10.1.3 Design and Construction
Considerations
Surface water is drawn from rivers, lakes, and reser-
voirs through submerged intake pipes, or through
fairly elaborate tower-like structures that rise above the
water surface. In the design and operation of intakes,
it is important that the water drawn be as clean,
palatable, and safe as the source of supply can
provide. River intakes are constructed well upstream
from points of discharge of sewage and industrial
wastes. The selected location should take advantage
of deep water, a stable bottom, and favorable water
quality and should be protected against floods,
debris, ice, and river traffic. Small streams may be
dammed up by diversion or intake dams to keep intake
pipes submerged and preclude hydraulically wasteful
air entrainment. In the siting of lake intakes, con-
sideration must be given to sources of pollution,
prevailing winds, surface and subsurface currents,
and shipping lanes. Shifting the depth of draft makes
it possible to collect clean bottom water when the wind
is offshore, and conversely, to collect clean surface
water when the wind is onshore. Reservoir intakes
resemble lake intakes, but they generally lie closer to
shore in the deepest part of the reservoir.
The feasibility of providing new surface-water intakes
depends on the following numerous case-specific
requirements and conditions:
Proximity of the point of intake to the water
supply system.
Peak demand flow versus historic and predicted
low flow in the water body.
Downstream environmental, recreational, and
commercial effects of reduced flow.
Quality of the surface water and corresponding
treatment requirements.
Section 5.4 addressed the design and construction
considerations for ground-water wells.
5.10.1.4 Costs
Costs for an alternate central water supply vary widely
depending on the population to be served, ground-
water or surface-water source, treatment require-
ments, and accessibility to an existing alternate water
supply. Ground-water well installation costs were
addressed in Section 5.4.
5.10.2 Alternative Point-of-Use Water
Supplies
5.10.2.1 General Description
Central water supplies that are contaminated at the
source or while flowing through pipelines can be
replaced permanently or temporarily with an inde-
pendent supply at each point of usage. Such sup-
plies could include one or a combination of the
following: bottled and bulk water, point-of-use wells,
and collection of rainwater.
Bottled water and bulk water are commonly used as
temporary water supplies on an emergency basis until
arrangements can be made for a more permanent
water supply. Bottled water is widely available in small
quantities from common retail outlets (grocery and
drug stores) and in large quantities from commercial
distributors. Larger bottles (e.g., five-gallon "water
cooler" bottles) require dispensers for convenience
purposes. Their full weight (approximately 50 pounds)
may present handling and changeover problems for
some users.
Point-of-use wells, or individual wells for each user
establishment, may offer a permanent alternative to a
contaminated central supply, provided the available
ground water is and can be expected to remain clean.
Rainwater is rarely an immediate source of municipal
water supplies, but it could be used to replace a
contaminated water supply. The use of rainwater is
generally confined to farms and towns in semiarid
regions where no satisfactory ground water or surface-
water supplies exist. For individual users, rainwater
running off the roof is led through gutters and
downspouts to a cistern situated on or below the
ground. For municipal service, roof runoff may be
combined with water collected from sheds or catches
on the surface of ground that is either naturally
impervious or rendered so by grouting, cementing,
paving, or similar means.
5.10.2.2 Application/Availability
Bulk water can be provided in portable tanks (trailers or
tank trucks) by commercial, clean-water contractors
and by public emergency service organizations (e.g.,
Army National Guard). Tanks that are normally usedfor
other purposes, such as milk tank trucks, also have
been used. The tanks are typically made available to
home owners at temporary, centrally located,
distribution points, where small containers can be
filled for home use. Whole tanks can be made avail-
able to commercial and institutional establishments.
New wells can be installed as long as they will be
pumping from an uncontaminated aquifer. The eco-
nomics of installing a well should be compared with
those for other possible solutions for providing
potable water.
The gross yield of rainwater supplies is proportional to
the receiving or drainage area and the amount of
precipitation. Because of the relatively small catch-
ment area available, roof drainage cannot be expected
to yield an abundant supply of water, and a close
analysis of storm rainfalls and seasonal variations in
precipitation must be made if catchment areas,
standby tanks, filters, and cisterns are to be
5-111
-------
proportioned and developed properly.
Bottled water is probably the most readily available
solution for temporary alternate water supply, as it can
be purchased from local distributors, groceries, con-
venient stores, and drug stores.
5.10.2.3 Design and Construction
Considerations
Schemes for providing bulk water, an alternate central
supply, a new surface-water source, new wells, or
rainwater collection will be highly site-specific. Local
expertise and contractors should be called upon to
help devise the total solution; obviously, the local
water company should be included in any planning
and implementation programs.
For bottled water supply, price, quantity, and delivery
are the main considerations. User requirements will
need to be ascertained, and a regular delivery sched-
ule should be established and maintained. Logistics
of delivery should be left to the supplier (e.g., where
to leave the boxes of bottles if there isn't anyone
home).
5.10.2.4 Costs !
Bottled water costs range from 500 to $1 per gallon
delivered in quantities. Bulk water costs will depend
on the mode of delivery, availability of supply, and
other local factors.
The cost of a new well will depend greatly on local
conditions, i.e., depth to ground water, availability of
ground water, etc. The cost of a 4-inch, 5-gal/min well,
200-feet deep with PVC casing, submersible pump,
tank, distribution piping, and installation would be
approximately $11,500; annual operating and
maintenance costs would run about $360.
Costs for a rainwater system are very site-specific.
They depend on the collection and storage system
required and piping/pumping required to supply the
user.
5.10.3 Treatment of Central Water
Supplies i
5.10.3.1 General Description
Central water supplies that are contaminated at the
source can be treated at central treatment systems to
upgrade them to an acceptable quality level. In small
communities that pump ground water directly to
distribution systems without treatment, providing
central treatment may necessitate the installation of
new facilities. In large communities that already treat
surface water before distribution, upgrading of
existing treatment with the installation of polishing
units may be necessary.
5.10.3.2 Application/Availability
Available water treatment methods include physical,
chemical, and biological technologies, and com-
binations of these methods may be required for the
removal of some contaminants.
Many of the technologies described in subsection 5.7
for ground-water treatment also apply to treatment of
contaminated water supplies. In general, however,
those technologies that are normally associated with
"polishing" (i.e., removal of low .concentrations of
contaminants), such as activated carbon, ion
exchange, and reverse osmosis, are most applicable
to treatment of public water supplies.
5.10.3.3 Design and Construction
Consideration
Design and construction considerations for the
treatment of ground-water supplies were addressed in
subsection 5.7.
5.10.3.4 Costs
Costs were provided in subsection 5.7.
5.10.4 Treatment of Polnt-of-Use Water
Supplies
5.10.4.1 General Description
Modern technology has produced a new point-of-use
treatment device for contaminated water systems, and
a variety of these devices is available on the market
today. The following configurations are typical:
Line-bypass device. The treatment unit is con-
nected to the water line by saddle valve, and
treated water is dispensed through a separate
faucet that is usually mounted on the kitchen
sink. Models are also available for residential
icemakers.
Faucet-mounted device. This unit is mounted
on the kitchen or lavatory faucet and treats all the
water that flows from that faucet.
Whole-house unit. This large unit is installed in
the main waterline to treat all the water used in a
household or business establishment.
These units should be considered as alternatives to
supplying bottled water to replace contaminated water
systems. Although these treatment devices may be
more costly initially, an economic breakeven or
savings could result, depending on the length of time
the alternative water source is required.
5.10.4.2 Application/Availability
The point-of-use water treatment devices available on
the market today, which use activated carbon as the
treatment media, are probably the most appropriate for
5-112
-------
converting water contaminated by a leaking UST into
potable water. These are readily available from well-
known suppliers (e.g., Sears, Culligan, Water Pic), as
are the needed replacement carbon cartridges.
5.10.4.3 Design and Construction
Considerations
Several items should be considered in the selection
of a point-of-use device. These include being sure
the unit is appropriate for the contaminant(s) of
concern, that it has the appropriate hydraulic capacity
for the application, and that criteria and schedules for
maintaining the units are supplied. This information is
generally available from the manufacturer or supplier.
The most preferred style of treatment cartridge is one
that uses activated carbon in solid block or granular
(not powder) form. Also, the cartridge configuration
should be such that water flows through the whole
cartridge and cannot bypass any section of it.
5.10.4.4 Costs
Bypass-type point-of-use treatment devices with
activated carbon elements can be purchased at prices
ranging from $70 to $400 each. Installation by a
plumber will cost,another $50 to $100 or more,
depending on the complexity of the installation.
Whole-house units will cost upwards from $1000
installed, depending on the size required and the
complexity of the installation.
j
5.10.5 Replacement of Water and Sewer
Lines
5.10.5.1 General Description
Replacement of water and sewer pipelines that are
contaminated by contact with hazardous substances
is seldom more cost-effective than rehabilitation, but it
often is the only practical alternative. Replacement
involves excavation of trenches, laying of new lines
with uncontaminated pipe materials, laying new
connections and/or tying in connections,,, and the
associated backfilling and surface restoration.
Contaminated pipelines may either be abandoned in-
place or be removed during trench excavation.
Construction of water and sewer lines is common in
land development projects, and the associated
methods, materials, and equipment are well es-
tablished.
5.10.5.2 Application/Availability
Pipeline replacement is applicable in virtually all cases
of pipeline contamination. The primary disadvantage
of pipeline replacement is its high cost. Analyses to
determine the cost-effectiveness of replacement
must include all costs associated with the
replacement; typically, these include pavement re-
moval and. replacement, excavation, possible sub-
stitution of select backfill to replace poor quality
misting or contaminated material, dewatering and
shoring, pipe materials, and traffic control. Higher
costs can result from interference with other under-
ground utilities. Narrow easements or limited space
for construction also must be considered, as well as
the need for temporary rerouting of the flow to
maintain service. Depending on the service life as-
sumed for other rehabilitation methods, these high
capital costs may be offset somewhat by the longer
service life a new line provides.
5.10.5.3 Design and Construction
Considerations
In general, new pipeline systems will be much like the
systems they replace (size, material, grade, location,
capacity, etc.). The considerations that govern the
design of new systems will apply, but not control.
Logistics and the presence of fixed constraints will
dictate how the replacement system is designed. The
need for replacement, however, may provide an
'opportunity" to upgrade the systems in terms of
capacity, improved materials and methods, location,
and/or direction of flow. Also, some consideration can
be given to the criteria for the design of new systems.
Various conventional and unconventional methods
are available for constructing water and sewer lines.
The most common method, open-trench excavation,
often requires lateral bracing of trench walls in deep
cuts or noncohesive soils. Other construction
methods include augering or boring, jacking, and
tunneling. Their use applies in limited situations only,
however.
.10.5.4 Costs
Representative costs for replacement of water and
sewer lines are shown in Table 5-43.
iTableS-43.
Item
1986 Unit Costs for Replacement of
Water and Sewer Lines*
Cost Per Unit ($)
Sewer pipe, material and installation,
n-place
8-inch diameter 6 to 10/lin. ft.
18-inch diameter 13 to 32/lin. ft.
36-inch diameter 34 to 125/lin. ft.
Water pipe, material .and instal-
lation, in-place
2-inch .diameter 3to7/lin. ft.
4-inch diameter 5 to 11/lin. ft.
12-inch diameter 16 to 28/lin. ft.
Pipe bedding material
14to26/yd3
Trench excavation, backfill,
and compaction
Water lines 1 to 3/lin. ft.
Sewer fines 6to10/yd3
5-1
* Data from Means 1984a.
13
-------
5.10.6 Cleaning/Restoration of Water and
Sewer Lines
5.10.6.1 General Description
Available techniques for inspecting and cleaning
sewerlines generally also apply to waterlines.
Water!ines are normally smaller in diameter than
sewerlines, however, and size is often a limiting factor
In the applicability of inspection and cleaning
technologies. Inspection techniques include smoke
testing, dye-water flooding, first-hand visual obser-
vation, and closed-circuit television visual observation.
Waterlines and sewerlines that have come in contact
with contaminated substances or have been infiltrated
by contaminated water can be lined or sealed in place
with chemically inert material. Available methods for
sealing out contaminants include:
• Insertion of new pipe inside of existing pipe
(sliplining)
In-place forming of new pipe inside of existing
pipe
Point repairs of leaks and other defects
Sliplining involves sliding a flexible liner pipe of slightly
smaller diameter into an existing circular pipeline and
then reconnecting the service connections to the
new liner. Polyethylene is the most common material
used for sliplining pipelines.
A patented system called "inversion lining" involves
the use of a flexible lining material that is thermally
hardened. Access to the pipeline can be made
through manholes or excavations. After the lining
system has been installed and cured, a special cutting
device is used to reopen service connections (with
the help of a closed-circuit television camera).
Because inversion lining can be accomplished
relatively quickly and without excavation, this method
is particularly well suited for repairing pipelines located
under existing structures or large trees. It also is
particularly useful for repairing pipelines located under
busy streets or highways, where traffic disruption must
be minimized. Inversion lining is relatively new in the
United States, and its cost-competitiveness has not
yet been fully established. It is available only through
a limited number of licensed contractors.
5.10.6.2 Application/Availability
Cleaning of waterlines and sewerlines removes
deposits and debris from the pipelines. Its purpose is
to improve flow conditions and capacity, to allow visual
inspections, and to provide clean surfaces for
placement of repair materials.
Available sewer-cleaning techniques include me-
chanical scouring, hydraulic scouring and flushing,
bucket dredging, suction cleaning with pumps or
vacuum, chemical absorption, or a combination of
these methods. Access to sewerlines for interior
cleaning and repair is usually through manholes.
Basin inlets and service connections provide ad-
ditional points of access. Service and fire hydrant
connections afford access to municipal waterlines.
Inversion lining, which involves the use of water to
cure the resins, is generally used in pipelines with
diameters less than 57 inches and manhole-to-
manhole segments less than 1000 feet long. Larger-
diameter pipelines (up to 108 inches) have been lined
by inversion techniques in which air is used.
Chemical grouting is commonly used to seal leaking
joints in structurally sound sewer pipes. Small holes
and radial cracks also can be sealed by chemical
grouting.
Sliplining is used to rehabilitate extensively cracked
pipelines, especially those lain in unstable soil con-
ditions. This technique is also used to rehabilitate
pipe installed in a corrosive environment and in areas
where sewer pipes have massive, destructive, root-
intrusion problems. The flexible liner pipes have the
advantage of being able to accommodate a normal
amount of future settlement or moderate horizontal or
vertical deflection.
5.10.6.3 Design and Construction
Considerations
Design of waterline and sewerline rehabilitation efforts
consists primarily of planning for the logistics of
implementation. Sections of pipeline to be rehabil-
itated are identified by television or other inspection
methods. Critical points of operation are selected,
such as access manholes, base of operation, and
material storage. Methods of managing disruption of
services (water or sewer) and of surface activities such
as traffic are also planned. Affected parties are notified
in advance of the planned work.
Before a liner pipe is installed, the existing pipeline
should be inspected by closed-circuit television to
identify all obstructions (e.g., displaced joints,
crushed pipe, and protruding service laterals) and to
locate service connections. The existing pipe is
thoroughly cleaned immediately, before sliplining
begins.
HOPE sliplining is pulled through existing pipelines by
a cable fed through the section to be lined. The cable
and pipe are advanced by a winch and pully assembly.
An approach trench is excavated at the insertion end
of the existing pipe section to allow a gradual transition
from the ground surface, where sections of HOPE
pipe are heat-fused to form a continuous pipe to an
opened section of pipe. Several thousand feet of
waterline or sewerline can be sliplined in a single
setup of such an operation. Fiberglass-reinforced
5-114
-------
pipe can be sliplined in a similar manner, but a
combination of pushing and pulling of the pipe may be
required.
In-place forming of new pipe inside of existing pipe is
accomplished with portland cement grout and mortar,
chemical grouts, and synthetic resins. Chemical
grouts can be used to seal fractures and leaking joints
to waterproof points of infiltration/exfiltration. Grout
materials used for this application include acrylamide,
acrylate, urethane, and polyurethane.
Grouting combined with sealing rings requires the use
of a small control panel, chemical and water pumps,
and various other accessories, depending on the type
of sealing grout being used. A worker must enter the
line, manually place the ring over the joint, and inflate
the ring to isolate the joint. Sealing grout is pumped
into the small void between the pipe wall and the face
of the ring with a hand-held probe. As the pressure in
the void increases, the grout solution is forced into
the joint and surrounding soil. A catalyst solution is
injected, and the grout cures, thereby sealing the joint
from infiltration.
5.10.6.4 Costs
Waterlines can be inspected and cleaned by a variety
of methods and at varying costs. Television inspec-
tion and light high-pressure water cleaning (the
minimum required in preparation for repairing or lining
pipelines) typically cost $100 to $155 per hour, or
$0.40 to $0.60 per linear foot to cover 2000 feet of
pipe per 8-hour day and $0.80 to $1.25 per linear foot
to cover 1000 feet of pipe per day. Costs of other
inspection and cleaning methods are highly variable
and depend on the type of pipeline and nature of the
material being removed.
Costs of sliplining waterlines and sewerlines vary v/ith
the diameter and depth of the pipeline. Costs for
relatively small-diameter (less than 15-inch) HOPE
sliplining projects range from $20 to $30 per linear
foot. Larger-diameter sliplining projects are seldom
undertaken and must be costed on a project-specific
basis.
Inversion lining costs are normally given on aper-linear-
foot basis for initial television inspection, cleaning,
bypass pumping, and post-construction television
inspection combined. The following are represen-
tative unit costs for typical inversion lining of
sewerlines:
Diameter (inches)
8
10
12
5-1
Cost f$/linear ft)
45 to 51
47 to 53
49 to 55
Reconnection of laterals typically costs $100 to $260
each, depending on logistics and the number of
laterals in a given project.
Grout repairs to sewer pipelines are generally
accomplished by pumping grout into a joint until soil
voids are filled, as determined by back pressure. A
wide range of grout volumes can be pumped into a
joint, and sewer grouting work is typically conducted
on a per-hour basis for manpower and equipment
($100 to $155 per hour) and on a per-gallon basis for
grout ($5 to $10 per gallon for chemical grout).
-------
-------
Section 6
Reference Matrix for Case Histories
6.1 Purpose of Case Histories
Real-world experience in the remediation of releases
from underground storage tanks is poorly document-
ed. An extensive review of the literature revealed
fewer than 40 reported cases of leaking underground
tanks/piping and attempts to implement corrective-
action programs. Given the broad population of under-
ground tanks and the high probability of tank failure,
this is a small number indeed. Nevertheless, the
cases reported in the literature provide a basis on
which to evaluate the applicability of more conven-
tional corrective-action technologies to subsurface
releases of petroleum and chemical products under a
variety of site conditions. The purpose of including
the case histories in this report is to provide the reader
with real-world applications of various technologies
along with information on performance and costs.
Ten case histories relating UST corrective action ex-
perience are presented in the appendix. Each of
these gase histories provides the following infor-
mation:
• Background - A brief description of the circum-
stances leading to the discovery of the leaking
underground tank/pipeline.
Site description - A detailed description of the
site's geologic/hydrogeologic characteristics, in-
cluding depth to water-table and proximity to any
drinking water supply wells (where available).
• Nature and extent of contamination - Estimates of
the volume of product (petroleum, chemical) re-
leased and the vertical and horizontal dimensions
of the plume.
6-
• Corrective actions - A summary of the initial and
remedial measures to control the release, contain
the plume, and clean up contaminated soils,
ground water, and surface water.
• Project evaluation - An assessment of the ef-
fectiveness of the corrective actions in meeting
the cleanup objectives in a timely manner.
• Costs - A summary of the total capital and annual
operating expenditures (where available) to im-
plement the corrective-action program.
It is hoped that this information will be of value to site
owners/operators, cleanup contractors, and regu-
latory personnel contemplating alternative corrective-
action responses at other UST locations.
6.2 Case History Matrix
The matrix presented in Figure 6-1 summarizes the
salient aspects of each of the case histories contained
in the appendix. As shown in the matrix, the case
histories encompass a broad range of facility types,
hydrogeologic settings, and types and quantities of
product(s) released. Sixteen of the corrective action
technologies profiled in Section 5 are represented in
the case histories.
The matrix leads to two conclusions: 1) a successful
corrective-action program requires implementation of
a combination of response technologies, and 2)
ground-water extraction will be an integral part of
nearly every corrective-action program. These points
should be considered in the response to releases
from underground tanks.
-------
Figure 6-1. Case history matrix.
Case History
A. Gasoline Pipeline,
Glondalo, CA
100.000 to 250,000 gal
gasoline released; ground-
water aquifer, alluvial deposit;
70 ft to wator table; 400 ft from
drinking water supply
B, Gasoline Pipeline, Ambler, PA
100,000 gal gasoline released;
highly fractured dolomite; 30 to
100 ft to water table; 300 ft from
drinking water supply
C. Retail Gasoline Station
Gonesee County, Ml
Undetermined amount of gasoline
released; ground-water aquifer, fine
glacial sand; 15 to 20 ft to water
table
D. Retail Gasoline Station
Montgomery County, PA
Undetermined amount of gasoline
released; ground-water aquifer,
fractured sha!e sillstone, 15 to
20 ft to wator table; 200 ft from
drinking water supply
E U.S. Coast Guard
Air Station Traverse City, Ml
200 + gal of jet fuel and aviation
fuel released (benzene and
toluene); 1 200 ft from drinking
wator supply
Environmental
%
0
Monitoring
1
E
6
•
0
9
0
1
M
cl
0
••§
0
">
I
O)
1
s
ii
0
0
0
Corrective Action Technologies
c
1
"J
c§
0
is"
>
§1
tJM
II
Q. Q.
IL O
0
>,
> E
§ !,
if
0 £
O- m
oi -a
£
U. CO
0)
m
0
I
J
1
1
0
9
m
0
1
c
1
£•
0
c
1
Q
i
3
•§
^
0
0
o
1
c
0
•Cl
s
1
TS
£.
co
5
I
01
1
0
c
o
'£3
1
m
O
O
E
'§;
3
i
S-
J
O
S1"""1
HO
1
'§>
•5
5
c
'S
IJL
i
•5
c
S5
_g
H
a
0)
CO
"S
I
fl>
6-2
-------
Figure 6-1. (Continued)
F. Bulk Fuel Storage and
Distribution Center
200,000 gal of fuel oil released;
ground-water aquifer, fractured
dolomite; 50 ft to water table
G. Midwestern Laboratory
Facility
1000 to 1500 gal of fuel oil and
solvents released; ground-water
aquifer, glacial till
H. Chemical Pipeline
Undetermined amount of methylene
chloride released; 100 ft to water
table
I. Biocraft Laboratories,
Waldwick, NJ
33,000 gal of methylene chloride,
acetone, m-butyl alcohol, and dimethyl
aniline released; ground-water aquifer,
glacial till and fractured shale; 0 to 9 ft to
water table; 1 000 ft from drinking water
supply
J. Fairchild Camera &
Instrument Corp.,
South San Jose, CA
43,000 gal of 1,1,1-trichloroethane,
xyleno, acetone, and isopropyl alcohol
released; ground-water aquifer, alluvial
deposits; 10 to 40 ft to water table; 2000
ft from drinking water supply
Environmental
Monitor! no.
H
«
e
5
5
3
3
«
•
•
•
«
>
i
:
I
5
CO
•
^
9
_
•
•
'
»
L
I
i
•
e
•
e
Corrective Action Technologies
E
&
Jr
Q.
E
'. a.
: Q
9
«
;" p
i s
5 5,
• <£
3 E
3 —j
3 Q_
-
-------
-------
References
American Petroleum Institute. 1979. Installation of
Underground Petroleum Storage Systems. Publi-
cation 1615. Washington, D.C.
American Petroleum Institute. 1980. Underground
Spill Cleanup Manual. Publication 1628.
Washington, D.C.
American Petroleum Institute. 1981. Recommended
Practice for Abandonment or Removal of Used
Underground Service Station Tanks. Publication
1604. Washington, D.C.
Drainage Manual.
Bureau of Reclamation. 1978.
Denver, Colorado.
Bureau of National Affairs. 1985. Cities Must Begin
Registering Tanks to Meet May Deadline, EPA
Official Says. Environmental Reporter, 16(35):
1664-1665.
Camp, Dresser, and McKee, Inc. 1986. Fate and
Transport of Substances Leaking From Under-
ground Storage Tanks. Volume 1, Technical
Report. Interim report prepared for the U.S.
Environmental Protection Agency under Contract
No. 68-01-6939.
Canter, L. W., and R. C. Knox. 1985. Ground Water
Pollution Control. Lewis Publishers, Inc., Chelsea,
Michigan.
Chemical Separations Corp. Undated. Ion Exchange
Systems (a sales brochure) Oak Ridge, TN.
Cheremisinoff, P. N., et al. 1986. Special Report:
Underground Storage Tank Control. Pollution
Engineering, 18(2):22-29.
CONGA WE. 1979. Protection of Groundwater From
Oil Pollution. Report No. 3/79. The Hague,
Netherlands.
Conway, R. A., and R. D. Ross. 1980. Handbook of
Industrial Waste Disposal. Van Nostrand Reinhold
Co., New York.
Cope, F. W., G. Karpinski, and R. L. Steiner. 1984.
Use of Liners for Containment at Hazardous Waste
Landfills. Pollution Engineering, 14(3):22-32.
Cullinane, M. J., and L. W. Jones. 1985. Technical
Handbook for Stabilization/Solidification of Haz-
ardous Waste. Prepared for U.S. Environmental
Protection Agency, Hazardous Waste Engineering
Research Laboratory, Cincinnati.
Culp, R. L., Wesmer, and G. L. Gulp. 1978.
Handbook of Advanced Wastewater Pretreatment.
Van Nostrand Reinhold Environmental Engi-
neering Series, New York.
De Renzo, D. (ed.) 1978. Unit Operation for
Treatment of Hazardous Wastes. Nqyes Data
Corporation, Park Ridge, New Jersey.
Driscoll, F. G. 1986. Groundwater and Wells. 2nd Ed.
Johnson Division, UOP, St. Paul, MN.
Ehrenfeld, J. R., and J. M. Bass. 1983. Handbook for
Evaluating Remedial Action Technology Plans.
EPA-600/2-83-076.
Environmental Law institute. 1985. Compendium of
Cost of Remedial Technologies at Hazardous
Waste Sites. A Report to the Office of Emergency
and Remedial Response, U.S. Environmental
Protection Agency.
FMC Corporation. 1985. An Introduction to FMC's
Enhanced Bioreclamation. FMC Aquifer Reme-
diation Systems, Princeton, NJ.
Ghassemi, M., K. Yu, and S. Quinlivan. 1981.
Feasibility of Commercialized Water Treatment
Techniques for Concentrated Waste Spills.
Prepared for the U.S. Environmental Protection
Agency, Municipal Research Laboratory,
Cincinnati, Ohio.
Giddings, T. 1982. The Utilization of a Groundwater
Dam for Leachate Containment at a Landfill Site. In:
Proceedings of Aquifer Restoration and Ground-
water Rehabilitation, 2nd National Symposium on
Aquifer Restoration and Groundwater Monitoring,
Columbus, Ohio, May 26-28,1982. pp. 23-29.
Gooding, C. H. 1985. Reverse Osmosis and Ultra-
filtration. Chemical Engineering, January 5, 1985,
pp. 56-62.
R-1
-------
Hansen, S. P., R. Gumerman, and R. Gulp. 1979
Estimating Water Treatment Costs. Volume 3-
Cost Curves Applicable to 2500 gpd to 1 mgd
Treatment Plants. EPA-600/2-79-162C. U.S. En-
vironmental Protection Agency, Municipal Environ-
mental Research Laboratory, Cincinnati, Ohio.
Johnson Division, UOP, Inc. 1975. Groundwater and
Wells. A Reference Book for the Water-Well
Industry. St. Paul, Minnesota.
JRB Associates. 1985. Leachate Plume Manage-
ment. EPA No. 540/285-004. Prepared for the
U.S. Environmental Protection Agency, Cincinnati,
Ohio.
Kiang, Y., and A. R. Metry. 1982. Hazardous Waste
Processing Technology. Ann Arbor Science
Publishers, Inc., Ann Arbor, Ml.
Knapp, Inc. 1982. Polly-Pig. Technical Bulletin
B5/82/5M. Knapp, Inc., Houston, TX.
Knox, R. C., et al. 1984. State-of-the-Art Aquifer
Restoration. Volumes 1 and 2. EPA/600-2-84-
182aandb.
Lundy, D. A., and Mahan, J. S. 1982. Conceptual
Designs and Cost Sensitivities of Fluid Recovery
Systems for Containment of Plumes of Contam-
inated Groundwater. In: Proceedings of National
Conference on Management of Uncontrolled
Hazardous Waste Sites, Washington, D.C.,
November 29-December 1,1982. pp. 136-140.
Lutton, R. 1982. Evaluating Cover Systems for Solid
and Hazardous Waste. SW-867. U.S. Army
Engineers Waterways Experiment Station. Pre-
pared for U.S. Environmental Protection Agency,
Municipal Environmental Research Laboratory,
Cincinnati, Ohio.
McMahon, L. A. 1984. 1985 Dodge Guide to Public
Works and Heavy Construction Costs. Annual
Edition No, 17. McGraw-Hill Information Systems
Company, Princeton, NJ.
Robert Snow Means Company, Inc. 1984a. Means
Building Construction Cost Data 1985. 39th ed. R
S. Godfrey, ed. Kingston, MA.
Robert Snow Means Company, Inc. 1984b. Means
Site Work Cost Data 1985. 4th ed. R. S. Godfrey
ed. Kingston, MA.
Michelsen, D. L., p. A. Wallis, and S. R. Lavinder.
1985. In Situ Biodegradation of Dispersed Organ-
ics Using a Microdispersion of Air in Water In-
Proceedings of the Sixth National Conference on
Management of Uncontrolled Hazardous Waste
Sites, Washington, D.C., November 4-6, 1985.
Hazardous Materials Control Research Institute,
Silver Spring, Maryland.
Montgomery, R. E., D. P. Remeta, and M. Gruenfeld.
1985. Rapid On-Site Methods of Chemical Anal-
ysis. In: Contaminated Land. Michael A. Smith,
ed. Plenum Publishing Corporation, New York.
National Fire Protection Association. 1981. Flam-
mable and Combustible Liquids Code. NFPA 30
Quincy, Massachusetts.
National Fire Protection Association. 1982. Recom-
mended Practice for the Control of Flammable and
Combustible Liquids and Gases in Manholes,
Sewers, and Similar Underground Structures
NFPA 328-1982. Quincy, Massachusetts.
New York State Department of Environmental
Conservation. Undated. Recommended Practices
for Underground Storage of Petroleum. Albany
New York.
Niaki, S., and J. A. Broscious. 1986. Underground
Tank Leak Detection Methods: A State-of-the-Art
Review. EPA-600/2-86-001.
Oil Recovery Systems, Inc. Undated. Floating-filter
Recovery system. Promotional Literature. Novi,
Ml.
Powers, J. P. 1981. Construction Dewatering. A
Guide to Theory and Practice. John Wiley & Sons
New York.
Sammons, J. H., and J. M. Armstrong. 1986. Use of
Low Flow Interdiction Wells to Control Hydrocarbon
Plumes in Groundwater. In: Proceedings of the
National Conference on Hazardous Wastes and
Hazardous Materials, Atlanta, Georgia, March 4-6,
1986. Hazardous Materials Control Research
Institute, Silver Spring, Maryland.
Schwab, G. O., et al. 1981. Soil and Water Conser-
vation Engineering. John Willey and Sons New
York.
Shepherd, W. D. No date. Practical Geohydrological
Aspects of Groundwater Contamination. Depart-
ment of Environmental Affairs, Shell Oil Company
Houston, Texas.
Sims, R. C., et al. 1985. Review of In-Place Treatment
Techniques for Contaminated Surface Soils
Volumes 1 and 2. EPA-540/2-84-003a and b.
Spooner, P. A., et al. 1984a. Slurry Trench
Construction for Pollution Migration Control. EPA-
540/2-84-001.
Spooner, P. A., et al. 1984b. Compatibility of Grouts
With Hazardous Wastes. EPA-600/2-84-015
Prepared by JRB Associates for the U.S.
Environmental Protection Agency.
Tusa, W. 1986. Underground Leaks: True Corporate
Costs. Pollution Engineering, 18(2):68-81.
R-2
-------
Ueguhardt, L. C., et al. 1962. Civil Engineering
Handbook. McGraw-Hill, New York.
U.S. Environmental Protection Agency. 1973. Pro-
cess Design Manual for Activated Carbon
Adsorption. EPA 625/1 -71 -002a.
U.S. Environmental Protection Agency. 1979.
Process Design Manual: Sludge Treatment and
Disposal. EPA 625/1-79-011. Municipal Environ-
mental Research Laboratory, Cincinnati, Ohio.
U.S. Environmental Protection Agency. 1980.
Innovative and Alternative Technology Assess-
ment Manual. EPA-430/9-78-009.
U.S. Environmental Protection Agency. 1982.
Design Manual: Dewatering Municipal Wastewater
Sludges. EPA-625/1-82-014.
U.S. Environmental Protection Agency. 1984. Case
Studies No. 1-23: Remedial Response at Haz-
ardous Waste Sites. EPA-540/2-84-002b.
U.S. Environmental Protection Agency. 1985a.
Handbook - Remedial Action at Waste Disposal
Sites (Revised). EPA-625/6-85-006.
U.S. Environmental Protection Agency. 1985b. In
Situ Flushing and Soils Washing Technologies for
Superfund Sites. A Collection of Papers Pre-
sented at the RCRA/Superfund Engineering
Technology Transfer Symposium by the Haz-
ardous Waste Engineering Research Laboratory,
Cincinnati, Ohio.
U.S. Environmental Protection Agency. 1985c. Fact
Sheet. Mobile System for Extracting Spilled
Hazardous Materials From Soil. Hazardous Waste
Engineering Research Laboratory, Releases
Control Branch, Edison, New Jersey.
U.S. Environmental Protection Agency. 1985d.
Leachate Plume Management. EPA/540/2-
85/004.
Vatavuk, W. M., and R. B. Neveril. 1983. Cost of
Flares. Chemical Engineering. Vol. 90, No. 4.
Virginia Soil and Water Commission. 1980. Virginia
Erosion and Sediment Control Handbook: Stan-
dards, Criteria and Guidelines. Richmond, VA.
Wagner, K., and Z. Kosin. 1985. In Situ Treatment.
In: Proceedings of the Sixth National Conference
on Management of Uncontrolled Hazardous Waste
Sites, Washington, D.C., November 4-6, 1985.
Hazardous Materials Control Research Institute,
Silver Spring, Maryland.
Water Information Center, Inc. 1986. Pennsylvania
Makes Assessment of Its LUST Investigations. The
Groundwater Newsletter, 14(9):2.
Whittaker, H. 1984. Development of a Mobile
Reverse Osmosis Unit for Spill Cleanup. In: Pro-
ceedings of Hazardous Material Spills Conference,
April 9-12. Government Institutes, Inc., Rockville,
Maryland.
Yaniga, P. M. 1984. Hydrocarbon Contamination of
Ground Water: Assessment and Abatement.
Testimony Presented by Paul M. Yaniga on March
1, 1984, to Hearings of the United States Senate
Committee on Environmental and Public Works.
-------
-------
Appendix
Case Histories
A.1 Case History A-Gasoline
Pipeline, Glendale, California
A.1.1 Background
On April 15, 1968, gasoline odors were detected at a
drinking fountain in Forest Lawn Memorial Park in
Glendale, California. By September 12, engineers
had traced the source of the odors to Well No. 4 (FL-
4). A buried 8-inch steel pipeline, which was used to
transport refined gasoline, traversed the area. Several
minor leaks (totalling less than 6000 gallons) in this
line had been reported (and repaired) since 1964;
howeve.r, preliminary estimates of the extent of
contamination indicated that between 100,000 and
250,000 gallons may have leaked (McFee, Laverty,
and Hertel 1972).
A.1.2 Site Description
Forest Lawn Memorial Park is located in a hilly area
bordering the somewhat narrow alluvial valley of the
Los Angeles River. The upper Los Angeles River
area is an important ground-water resource for
southern California. The pipeline release occurred on
the relatively flat ground-water divide between two
major well fields that supply water to the city of Los
Angeles (Figure A-1).
Geologic maps and numerous exploratory wells in the
Forest Lawn area indicate a clay layer at a depth of
about 100 feet below the ground surface. Depth to
ground water was approximately 70 feet in the fall of
1968, but had been recorded as shallow as 30 feet
and as deep as 100 feet during periods of floods and
droughts. Because the water table is nearly flat,
horizontal movement of the ground water in the
vicinity of the pipeline release was extremely slow
(McKee, Laverty, and Hertel 1972).
A.1.3 Nature and Extent of Contamination
Immediate action was taken to determine the extent of
contamination, and by August 1969, some 30
observation wells had been drilled. During the course
of the drilling, free gasoline was found floating on the
water table at thicknesses of 12 to 30 inches. Free
gasoline was estimated to cover an area of more than
160,000 square feet (Blevins and Williams 1985).
A-1
A.1.4 Corrective Action
When pumping of FL-4 was stopped in September
1968, a considerable depth of free gasoline
accumulated on top of the water table. In the first
month after pumping ceased, approximately 1000
gallons of gasoline was removed by bailing. Selected
observation wells were subsequently equipped with
skimmer pumps for recovery of free product.
Extracted gasoline/water was treated by gravity
separators at three small treatment plants constructed
in the Forest Lawn area. Floating gasoline was
removed and trucked away; the underlying water was
discharged to storm drains or to a lined flood-control
channel. About 20,000 gallons of gasoline was
recovered in this manner over the next year (McKee,
Laverty, and Hertel 1972).
In November 1969, however, free gasoline appeared
for the first time in a well 500 feet closer to one of the
major supply well fields (Pollock Field) than that which
had been detected initially. Pumping from the well
field was stopped immediately to minimize drawdown
toward the field. Further drilling and testing in the
Forest Lawn area helped to contain the plume by
creating localized cones of depression.
Extraction of ground water and separation of free
product continued through August 1971, when free
product recovery was essentially complete; however,
strong taste and odor problems persisted in many of
the supply wells in the Forest Lawn and Pollock Field
areas. Biodegradation of the pellicular gasoline by
naturally occurring bacteria (Pseudomonas and
Arthrobactei) over the next several years eventually
eliminated the bad taste and odors (Blevins and
Williams 1985).
A.1.5 Performance Evaluation
From September 1968 through August 1971, about
50,000 gallons of free product was recovered. By mid-
1971, the extent of gasoline contamination was
limited to a few wells in the immediate vicinity of Forest
Lawn (Blevins and Williams 1985).
Ground-water sampling through August 1971 showed
gasoline-degrading bacteria to be present at
concentrations of 50,000/ml or higher in wells that
contained traces of free product; bacteria were found
at concentrations of 5,000 to 50,000/ml in wells with
-------
Figure A-1. Site of gasoline pipeline break (Blevins and Williams 1985).
• Pumping Well
A Observation Well
taste and odor problems (Blevins and Williams 1985)^
This correlation between the amount of available
hydrocarbons and the concentration of bacteria
indicates that natural biodgradation of the pellicular
gasoline was occurring.
Since February 1976, measurements by sensitive
paste on a weighted tape have shown no trace of free
gasoline in any of the wells. No odors have been
detected by sniffing since April 1976. Infrared
analyses for hydrocarbons performed biweekly from
July through September 1976 produced essentially
negative results (Blevins and Williams 1985).
Because of the lack of any detectable odor or any
significant levels of hydrocarbons by analysis,
monitoring of the site was terminated on December
31,1976.
A. 1.6 Project Costs
Project costs have not been published.
A-2
-------
A.1.7 References
Blevins, M. L, and D. E.Williams. 1985. Management
of Gasoline Leaks - A Positive Outlook. In:
Innovative Means of Dealing with Potential
Sources of Ground Water Contamination;
Proceedings of the Seventh National Ground
Water Quality Symposium, Las Vegas, Nevada,
September 26-28, 1984. EPA-600/9-85-012.
McKee, J. E., F. B. Laverty, and R. M. Hertel. 1972.
Gasoline in Ground-water. J. of the Water
Pollution Control Federation , 44(2):293-302.
A.2 Case History B—Gasoline
Pipeline, Ambler, Pennsylvania
A.2.1 Background
In July 1971, a pipeline break near Ambler,
Pennsylvania, spilled an estimated 100,000 gallons of
high-octane gasoline into the subsurface, which
contaminated the Whitemarsh Township water supply.
A.2.2 Site Description
The spill occurred in a valley approximately 500 feet
east of a small creek and 300 feet north of a municipal
pumping station. The contaminated aquifer is
composed of a highly fractured dolomite. Depending
on its location relative to the creek, the depth to
ground water varies from 30 to 100 feet (FMC 1972).
Natural ground-water flow in the region is toward the
creek; however, the natural gradient is reversed when
the two municipal wells, which draw 1 million gallons
of water per day, are being pumped. Under these
conditions, the creek is a source of ground-water
recharge.
A.2.3 Nature and Extent of Contamination
Because the two municipal wells were being pumped
at the time the spill occurred, the gasoline was largely
confined to the area within the radius of influence of
the well system. Over the next 8 months, 46 wells
were drilled in the area to define the extent of
contamination and to act as recovery points (Figure A-
2). Details on the extent of contamination were not
available.
A.2.4 Corrective Action
Shortly after its discovery, the broken pipeline was
drained and replaced. Gasoline floating on the water
table was recovered through continued pumping of
the municipal wells. The extracted ground water was
discharged to a ditch to allow separation of any free
product before the water entered a nearby creek.
Physical methods were successful in recovering only
about two-thirds of the spilled product during the first
A-3
year (FMC 1972, Lee and Ward 1984). Much of the
remaining product was believed to be trapped in tiny
crevices and adsorbed to soil particles. Extraction of
the contaminated ground water, which contained 5
ppm dissolved hydrocarbons, and its treatment by
conventional means could take up to 100 years for
complete elimination of the contaminants and
restoration of the subsurface environment (FMC
1972). For this reason, an alternative cleanup
technique~biostimulation-was pursued.
The biostimulation program was initiated in 1972.
Nitrogen and phosphorus as 30 percent solutions of
ammonium sulfate [(NH4)2SO4)], disodium phosphate
(Na2HPO4), and monosodium phosphate (NaH2PO4)
were introduced through injection wells to provide the
natural gasoline-degrading bacteria with the nutrients
required for their growth. Dissolved oxygen was
supplied by sparging air into wells through diffusers
connected to paint-sprayer-type compressors. On
the average, 10 aeration systems supplying air at a
rate of 2.5 ftVmin were used. Ground-water flow was
controlled by a series of injection and production
wells.
A.2.5 Performance Evaluation
The introduction of nutrients led to an average 100-
fold increase in the number of gasoline-degrading
bacteria in wells within the spill area. The average
concentration of bacteria appeared to correspond to
the availability of gasoline and averaged 107/ml in the
center of the spill and 10Vml beyond its perimeter.
Thirty-two bacterial cultures, principally Norcardia,
Pseudomonas, and Arthrobacter, were isolated
during the project (FMC 1972).
Nutrient addition was stopped in February 1974. By
that time, the levels of gasoline in the production wells
had been reduced from 5 ppm to less than 2.5 ppm.
Within 6 months, gasoline in these wells was reduced
to less than detectable levels (0.5 ppm based on UV
analysis) (FMC 1972).
A.2.6 Project Costs
Project costs have not been published.
A.2.7 References
FMC Corporation. 1972. Case History - Gasoline Pipe-
line Leak Promotional Literature. Princeton, NJ.
Jamison, V. W., R. L. Raymond, and J. O. Hudson, Jr.
1975. Biodegradation of High-Octane Gasoline
in Groundwater. In: Developments in Industrial
Microbiology, Volume 16; Proceedings of the
31st General Meeting of the Society for
Industrial Microbiology, Memphis, Tennessee,
August 11-16, 1974. American Institute of
Biological Sciences, Washington, D.C.
-------
Figure A-2. Location of observation and recovery wells (Jamison, Raymond, and Hudson 1975).
Lee, M. D., and C. H. Ward. 1984. Reclamation of
Contaminated Aquifers: Biological Techniques;
In: Proceedings of 1984 Hazardous Material Spills
Conference, April 9-12, Nashville, Tennessee.
Government Institutes, Inc., Rockville, Maryland.
A.3 Case History C-Retail Gasoline
Station, Genesee County, Michigan
A.3.1 Background
In the spring of 1980, gasoline odors were reported in
the basement of a bank building in Genesee County,
Michigan. As part of a normal program of investigation,
the State regulatory agency requested an adjacent
service station to check its inventory records and
tanks for losses. No losses were documented, but
the odors persisted and no other source in the area
could be found. At the direction of the State agency,
an observation well was constructed on the station
property. The well revealed free product gasoline
atop the water table, which had evidently leaked into
the shallow ground water from one of the station's
underground tanks.
A-4
-------
A.3.2 Site Description
The subsurface strata in the Genesee County area is
characterized by fine glacial sands. Clay is present in
the general vicinity of the leak at depths below 30
feet. Observation wells on site showed a static
product/water level of 15 to 20 feet below grade
before retrieval operations were begun (Yaniga
1984a, 1985).
A.3.3 Nature and Extent of Contamination
Twelve observation wells were used to determine the
direction of ground-water flow (to the northwest) and
the areal extent of the product plume (Figure A-3). Six
wells (Nos. 1, 2, 3, 5, 6, and 7) showed product at
thicknesses from 0.5 to 3.8 feet (Yaniga 1984a,
1985).
A.3.4 Corrective Action
Because the oil company's previous experience in
dealing with hydrocarbon contamination of ground
water was somewhat limited, initial recovery attempts
entailed the use of a vacuum truck to skim the product
off the water surface in the well. This resulted in an
abundance of water contaminanted with small
amounts of product, and gasoline odors in the bank
persisted.
Subsequent attempts at product retrieval involved the
use of a dual-pump system. Well No. 2 served as the
product recovery well. A 3/4-hp explosion-proof water-
table depression pump was used to create a cone of
depression. Maintenance of a drawdown of
approximately 10 feet at the pumping well could
achieve a radius of influence of 200 feet. The product
(gasoline) flowed to the well and was collected by a
product-retrieval pump.
A.3.5 Performance Evaluation
Retrieval operations began in January 1981, and
product from observation well No. 7 adjacent to the
bank was removed by March 1981. In that time, odor
levels were greatly reduced within the structure itself.
Figure A-3. Configuration of the free product plume (Yaniga 1985).
Recovery Well
Observation Well
-------
By April 1982, product thicknesses had been
reduced to less than 1 foot over a small area, and more
than 7000 gallons of gasoline had been retrieved
(Yaniga 1984a, 1984b, 1985).
A.3.6 Project Costs
Project costs have not been published; however, cost
savings of more than $10,000 per day were reported
to result from keeping bank operations open (Yaniga
1984b).
A.3.7 References
Yaniga, P. M. 1984a. Ground-Water Abatement
Techniques for Removal of Refined Hydro-
carbons. In: Proceedings of the National Con-
ference on Hazardous Wastes and Environmental
Emergencies, March 12-14, 1984, Houston,
Texas. Hazardous Materials Control Research
Institute, Silver Spring, Maryland.
Yaniga, P. M. 1984b. Hydrocarbon Contamination of
Ground Water: Assessment and Abatement Testi-
mony presented by Paul M. Yaniga on March 1,
1984, to Hearings of the United States Senate
Committee on Environmental and Public Works.
Yaniga, P. M. 1985. Alternatives in Decontamination
for Hydrocarbon-Contaminated Aquifers. Ground-
Water Monitoring Review, 5(4):40-49.
A.4 Case History D-Retail Gasoline
Station, Montgomery County,
Pennsylvania
A.4.1 Background
In 1980, several inhabitants of a suburban residential
area of Montgomery County, Pennsylvania, noted
unusual tastes and odors in their well water. Sampling
and analysis of 10 domestic wells by the State
regulatory agency attributed the taste and odor
problems to dissolved gasoline-type hydrocarbons.
The contamination was traced to a low-level, long-term
loss of an undetermined amount of unleaded gasoline
from an underground storage tank at a nearby service
station. Soil and ground water in the area were
contaminated, but no free product was found.
A.4.2 Site Description ;
The geology of the impacted area consists of 6 to 7
feet of red-brown silty loam underlain by fractured
shale and siltstone. Ground water is encountered 20
to 25 feet below grade in the bedrock system.
Domestic water wells penetrate the subsurface 60 to
80 feet.
Ground-water movement through the area is
influenced by geologic structure. The direction of
ground-water flow is from the northeast to the
southwest (i.e., from the area of the service station
toward the residents whose water was impacted).
A-6
A.4.3 Nature and Extent of Contamination
Twelve monitoring wells were installed in the area as
part of a ground-water evaluation program. Samples
taken from these wells and existing domestic wells
showed the dissolved contaminant plume to extend
200 to 250 feet in a north-south direction and 300 to
350 feet east-west (FMC 1981). Isopach maps of the
dissolved product plume (Figure A-4) indicated certain
control on its spread by strike and dip of the geologic
units (strike: 30°N to 40°E; dip: 12°). Dissolved
hydrocarbon concentrations ranged from less than
the detection limit (10 ppb) on the periphery of the
plume to greater than 15 ppm near the center of the
plume. Dissolved oxygen levels within the impacted
area were reduced, and naturally occurring
hydrocarbon-utilizing bacteria were limited in numbers
(Yaniga 1984a, 1985).
Figure A-4. Configurations of the dissolved hydrocarbon
plume (Yaniga 1984a).
Scale
• Residents' Wells
O Observation Wells
A.4.4 Corrective Action
The leaking tank, which was found to be pitted with
small holes, was excavated from the tank pit area along
with the service station's other tanks. Contaminated
soil above the tanks was also removed.
The following remedial alternatives were evaluated for
abating the existing contamination and for preventing
additional contamination of downgradient wells:
• Deepening the wells to case off contamination.
Providing public water.
-------
Developing a ground-water decontamination pro-
gram for the removal of dissolved hydrocarbons.
Geologic conditions indicated some hydraulic
connection between the contaminated aquifer and
lower water-bearing horizons, such that recon-
struction of the wells, could not assure the abatement
of well contamination. In addition, any attempt to case
off the contamination would restrict further
contaminant movement. Public water supplies were
not immediately available, nor would their use facilitate
the containment or restriction of further contaminant
movement. Therefore, a program of hydrocarbon
containment and treatment was selected as the most
comprehensive alternative.
The ground-water decontamination program was
developed on the basis of hydrogeological and
geochemical data collected from field and secondary
sources. In addition, certain laboratory biocultural
studies were conducted to identify dominant
hydrocarbon-utilizing species, and bench-scale pilot
tests were conducted to determine the effectiveness
of air stripping. The resulting program, illustrated in
Figures A-5 and A-6, incorporates the following
elements:
Figure A-5. Cross section of the ground-water decontaminatic n program (Yaniga 1984a),
Centralized pumping well to contain contaminant
movement and induce flow to the recovery
location.
Piping system to convey the contaminated water
to an air-stripping tower.
Air-stripping tower for the removal of volatile
organics and the oxygenation of contaminated
ground water.
Infiltration gallery for recirculation of the treated
ground water and to facilitate flushing and
leaching of adsorbed hydrocarbons back to the
recovery well.
Air-sparging and nutrient-addition wells to facilitate
conveyance of necessary oxygen and nutrients
into the ground-water system to stimulate the
growth of hydrocarbon-utilizing bacteria.
Regular program of monitoring hydrogeologic,
geochemical, and microbiological conditions to
determine the success and prognosis of the
aquifer cleanup.
Infiltration Gallery
Air-Stripping
Tovyer
Control
Panel
•Nutrient
Addition
Concrete
Pad
Discharge Lin'e
Approximate Water
Table Level
Water-Table
Depression Pump
Central Recovery
Well
-------
Figure A-6. Plan view of the ground-water decontamination program (Yaniga 1984a).
N
Station / \
E3 Air-Sparging Wells
Air Compressor
W& Infiltration Gallery
_E|_ Air-Stripping Tower
•sll? Contaminant Plume
• Residents' Wells
O Observation Wells
O Recovery Well
The following mechanical components were used in
the treatment program:
• A 2-hp explosion-proof water-table depression
pump equipped with a water-level-control sensing
probe and pitless adapter.
• A 24-in.-diameter air-stripping tower (12 ft tall)
equipped with a 3-hp blower.
• Two 3-hp air compressors capable of delivering 20
ftVmin each.
Six 4 in. x 2 in. carborundum air diffusers.
250 ft of 2-in.-diameter ABS pipe for water
delivery to and from the air-stripping tower.
1200 ft of 1/2-in.-diameter flexible air-pressure
line.
A 20 ft x 30 ft x 10 ft infiltration gallery backfilled
with crushed stone ballast.
In implementing the abatement program, it was
determined that significant quantities of contaminated
soil (silty clay) existed in the former tank pit area. This
represented a potential long-term continuing source
of gasoline contamination to ground water via
leaching; therefore, it was removed for disposal at a
secure landfill. The resulting excavation was
subsequently converted to an infiltration gallery for
the water treated by the air stripper.
Because the impacted area is residential and subject
to high volumes of traffic, all electrical leads, water
lines, and air-transmission lines were trenched and
backfilled. This provided both for the security of these
services and for the overall aesthetic acceptability of
the system. To accommodate both noise reduction
and security, the air compressors were housed in
metal sheds, and the air-stripping tower was fenced.
A-8
-------
I
Before the abatement program was initiated, a pump
test was conducted to determine the optimum
pumping rate and the amount of drawdown required
to control ground-water movement within the
impacted area. A combined program including
analysis of time vs. drawdown and distance vs.
drawdown was established. The results indicated an
optimum pumping rate of 22 gal/min to maintain 10
feet of drawdown in the recovery well. Results also
indicated that the radius of influence could be
propagated 300 to 350 feet along the strike and down-
dip without dewatering adjacent domestic wells.
Pumping effects of the up-dip could only be
documented 100 to 150 feet from the pumping well
(Yaniga 1984a, 1985). This was far enough to contain
the dissolved product plume and to cause it to move
toward the pumping well.
A.4.5 Performance Evaluation
Results of the ground-water decontamination program
are explained briefly in the following subsections on
the performance of the individual components of the
system and the reduction of total gasoline-type
hydrocarbons achieved. Figure A-7 shows the
changes in dissolved hydrocarbon contamination over
time in the central, most concentrated area of the
dissolved plume. In December 1982, residents' wells
showed a 50 to 100 percent reduction in total
dissolved gasoline hydrocarbons. Overall, the
dissolved plume was reduced in magnitude by 60 to
70 percent within the first year (Yaniga 1984a).
A.4.5.1 Recovery Well
As predicted, the area of influence of the recovery
well was greatest along strike and down-dip. Although
influence up-dip was less, it was sufficient to contain
and reverse the direction of dissolved product
movement (Yaniga 1984a, 1985). This asymmetrical
configuration for pumping influence is related to the
aquifer's homoclinal structure.
A.4.5.2 Air-Stripping Tower
Based on laboratory bench-scale testing, the air
stripper was projected to achieve a removal efficiency
of 85 to 90 percent for dissolved hydrocarbons.
Performance of the air stripping tower, however,
exceeded expectations; removal efficiencies of nearly
100 percent were realized for influent concentrations
in the 4 to 6 ppm range [i.e., no gasoline
hydrocarbons were detected in the air stripper
discharge by infrared analysis procedures (0.1 ppm
detection limit)] (Yaniga 1984a, 1985).
Figure A-7. Changes in total hydrocarbon concentrations over time (Yaniga 1985).
6 -
5 -
CO
U)
c
o
I
1
i i"r"l"i"rTT-rvvT-r-i-v:"»
A-9
-------
A.4.5.3 Infiltration Gallery
The treated water had to be recirculated through an
infiltration gallery to accelerate cleanup of the aquifer.
Recirculation aided the physical desorption of
hydrocarbons bound to the silty soil and shaly
materials and increased the rate of movement of
contaminants to the recovery well. Because the
treated water was also aerated by the stripping
process, oxygen was supplied to the subsurface to
accelerate biologic decomposition of residual gasoline
hydrocarbons.
The infiltration gallery was excavated in the former tank!
pit area and backfilled with crushed stone ballast.
Monitoring of liquid levels within the gallery showed
that it could accommodate a flow of 32,000 gal/day of
treated water. Sampling and monitoring for nutrients
and tracers added to the gallery indicated water from
the gallery flowed initially down-dip and then along
strike back to the recovery well (Yaniga 1984a, 1985).
A.4.5.4 Air Sparging
The air-sparging system, which consists of two 3-hp air
compressors capable of delivering 5 ftf/min air to each
of six carborundum air diffusers, went on line in March
1982. The air-sparging wells were located on the
periphery of the plume; the diffusers were placed in
the wells 50 to 55 feet below grade.
For assessment of the effectiveness of the oxygen
exchange program, levels of dissolved oxygen (DO)
were measured in the observation wells. Before
sparging, the levels of DO in wells affected by
dissolved gasoline contamination were generally less
than 1 mg/liter. Several had DO levels of 0 mg/liter.
While the air-sparging system was in operation,
dissolved oxygen levels in the sparging wells rose to
near saturation levels of 9 and 10 mg/liters. At the
periphery of the plume, DO levels rose to 3 to 5
mg/liter. Dissolved oxygen levels within the core area
of the plume rose to 2.5 to 3 mg/liter in approximately
6 weeks (Yaniga 1984a, 1985).
Adequate oxygen supply proved to be a limiting factor
in the growth of hydrocarbon-degrading bacteria. The
air-sparging system was eventually replaced by the
continuous addition of an oxygen-enhancement
solution to the injection gallery. Dissolved oxygen
levels within the plume subsequently increased to 4
to 8 mg/liter (FMC 1981).
A.4.5.5 Nutrient Addition
In March 1982, nutrients began to be added to the
ground water to stimulate the growth of indigenous
bacteria. The nutrient broth contained a mixture of
ammonium chloride, sodium phosphates, and various
mineral salt tracers to track the spread and movement
of the nutrients through the aquifer system. Nutrients
were added to the treated ground water in batches
and injected through the gallery.
Monitoring of the tracers indicated slow ground-water
migration from the infiltration pit along strike to the
recovery well (Yaniga 1984a, 1985). Because nutrient
diffusion rates were slower than desired, the nutrient-
addition program was modified to incorporate the use
of additional observation wells and the more frequent
addition of nutrients. These modifications increased
the effectiveness of the bacteria in consuming the
hydrocarbons.
A.4.6 Project Costs
Project costs have not been published.
A.4.7 References
FMC Corporation. 1981. Case History—Leaking
Underground Gasoline Storage Tank. Promo-
tional Literature. Princeton, New Jersey.
Yaniga, P. M. 1984a. Groundwater Abatement
Techniques for Removal of Refined Hydro-
carbons. In: Proceedings of the National Con-
ference on Hazardous Wastes and Environmental
Emergencies, Houston, Texas, March 12-14,
1984. Hazardous Materials Control Research
Institute, Silver Spring, Maryland.
Yaniga, P. M. 1985. Alternatives in Decontamination
for Hydrocarbon-Contaminated Aquifers. Ground-
Water Monitoring Review, 5(4):40-49.
A.5 Case History E-U.S. Coast
Guard Air Station, Traverse City,
Michigan
A.5.1 Background
In July 1942, the United States Navy established an
Air Station at Traverse City, Michigan, a small isolated
community in the northwestern section of the Lower
Peninsula (Figure A-8). The purpose of the Air
Station was to conduct highly classified research and
development of pilotless drone aircraft. This research
effort was suspended in 1944. When the war ended,
the Air Station was turned over to the U.S. Coast
Guard (USCG) to serve as a major Search and Rescue
base for Lake Superior, Lake Huron, and the upper
portion of Lake Michigan.
In 1979, during the removal of two fuel farms
preparatory to the installation of a new system, soil
contamination was discovered in the jet fuel (JP-4)
storage area. This area was located some 1500 feet
upgradient and to the north of an area that was
ultimately implicated as the "geographical origin" of
the plume. The aviation fuel (115/145) farm
immediately adjacent to the "geographical origin" of
the plume was excavated at the same time, and little
A-10
-------
Figure A-8.
Location of U.S. Coast Guard Air Station,
(Sammons and Armstrong 1986).
Traverse City, Michigan
East Arm
Grand
Traverse
Bay
Principal
Study Area
West Arm
Grand
Traverse
Bay
U.S. Coast Guard
Air, Station
indication of any leakage was found. Some odor was
noted in the soil, but laboratory analyses did not
confirm gross contamination.
In 1979 and 1980, residents in the Avenue E area of
the Pine Grove Subdivision of East Bay Township
complained to the local health department that the
water from their wells was discolored, foamed, tasted
bad, and had a foul odor. The first residence reporting
the problem is located 1200 feet to the northeast of
the Coast Guard Air Station (CGAS). At that time there
A-11
Lake Erie
was no explanation for the contaminated wells, and
the health department did not test for any of the
possible organic contaminants. Later in 1980, the
Michigan Department of Natural Resources (MDNR)
did a limited hydrogeologic study in the area and
concluded that the source of the contamination was
from some unspecified site on the CGAS. In May
1982, the Coast Guard was notified of these findings.
Subsequent internal investigations by the USCG
revealed that an aviation fuel spill incident had
-------
occurred 11 years earlier (in November or December
1969) when a flange in an undergound pipeline
beneath a 115/145 high-octane aviation gasoline
fueling station failed. The failure resulted in a loss of
approximately 2000 gallons of product over a 12-hour
period.
A.5.2 Site Description
Descriptions of the site's stratigraphy and hydro-
geology were not available. .
A.5.3 Nature and Extent of Contamination
In June 1982, the Coast Guard retained the U.S.
Geological Survey (USGS) to conduct a thorough
hydrogeological study of the area to define the extent
of the contaminant plume and to determine its source.
By April 1983, the USGS had determined the
direction of ground-water flow through the area (to the
northeast) and had tentatively identified the
boundaries of the plume. Although the USGS was
unable to determine the source of the contaminants,
they did conclude that the majority of the con-
taminants identified were related to components in
fuels. They also indicated that some chlorinated
compounds were present.
In November 1983, the Coast Guard contracted the
University of Michigan (UM) to make a scientific study
of the site. The objectives of this study were to
analyze the temporal variation of the plume and to
determine positively the origin(s) of the con-
tamination. The UM study identified benzene and
toluene as the components in the plume presenting
the greatest health risk. The largest concentrations of
these compounds occurred in the vicinity of the
Hangar/Administration Building (HAB) at the geo-
graphical head of the plume (Figures A-9 and A-10),
although significant amounts of some compounds
(e.g., benzene) were found at some distance
downgradient. Other chemicals also were found in
the plume, but at smaller concentrations and reduced
distributions (Sammons and Armstrong 1986).
From the HAB, the plume followed ground-water flow
to the northeast and off the base, passed under an
industrial park, and turned slightly north. It narrowed
as it passed underneath Parsons Road and widened
out again under Avenue E (Figure A-11). The plume
was approximately 4300 feet long and varied from 180
to 400 feet in width. Its vertical dimension ranged from
25 to 80 feet (Sammons and Armstrong 1986).
Small concentrations of benzene and toluene were
detected in the water of East Bay. The USGS
reported maximum benzene values of 20 u.g/liter and
maximum toluene values of 3.1 u.g/liter approximately
330 feet from shore (Sammons and Armstrong 1986).
Both the UM study and the USGS study reported
numerous measurements of organics in the soils at
the Air Station. The UM study found the following
maximum concentrations: benzene, 25.4 u.g/liter;
toluene, 27.6 u,g/liter; and xylene, 299 u.g/liter.
Analyses for seven other hydrocarbons showed
negative results (Sammons and Armstrong 1986).
Analysis of soil borings indicated that much of the
organic material was retained in the soil in a 6- to 12-
inch thick layer in the capillary zone immediately above
the water table. This zone served as a continuing
source of ground-water contamination.
In February 1985, new hydrocarbon contamination
was discovered at the JP-4 fuel farm south of the
HAB. The four fiberglass underground storage tanks
at the station were tested, and three of them were
found to be leaking.
A.5.4 Corrective Action
Hydrocarbon contamination of the soil and ground
water at the Air Station initially occurred in 1969 and
remained undetected until 1979. Contaminated soil
was subsequently removed from the JP-4 storage
area under the direction of the MDNR. The three
leaking tanks discovered in 1985 were also removed.
After the various long-term treatment or cleanup
options available had been considered, it became
clear that the first step in any remedial program would
be to decrease or stop further movement of
contaminants off U.S. Coast Guard property. This
option was judged to have several advantages:
Reduction of any possible increase in risk to
human populations that may have been related to
fuel-based contaminants present in the ground
water.
Promotion of reductions in contaminant con-
centrations in the ground water either by dilution
or the possible biodegradation of fuels by
indigenous microbial populations present in the
subsurface soil-water system.
Provision of a better opportunity (e.g., more time)
to efficiently select and design appropriate
method(s) for dealing with the contaminants
present in the geographical origin of the plume.
A containment system consisting of seven extraction
wells spaced laterally across the plume was
constructed in the east-northeast area of the Air
Station to block further migration of the plume offsite.
Six-inch auger-drilled wells with full 10-slot stainless
steel screens running from the top of the aquifer to
the clay confining layer at the bottom were installed.
The full-screen configuration was necessary to
capture contaminants that were present throughout
the vertical extent of the aquifer.
A-12
-------
Figure A-9.
Area of benzene contamination near the Hangar/Administration
building (Sammons and Armstrong 19136).
524*
K9S.D
Rock Garden
S33
S31 /
K43S * /* S19' I flFl
S36
K44S
Hangar/Administration Building
S35
S25
Legend
K8D.S
Well location and number -
Letters D, S following well number
indicate Deep, Shallow
Ground-water flow -
Arrow indicates direction of flow
Area of benzene contamination
The water produced from the extraction well system
was piped to a carbon treatment system, which
consisted of four 20,000-lb carbon reactors. The
carbon reactors were specified to reduce the levels of
benzene and toluene in the water to less than 1
u.g/liter. The extracted water was discharged to the
Traverse City Publicly Owned Treatment Works.
Costs for processing the discharge were estimated at
between $4,000 and $10,000 per month (Sammons
and Armstrong 1986).
A-13
.o K35.D
S11
To monitor the effectiveness of the containment
system, a network of five monitoring wells was
constructed downgradient of and outside the zone of
influence of the extraction well system. Ground-water
samples were collected by standard field procedures
for the collection of volatiles and were analyzed within
a few hours of collection. Five sample pumps were
placed in each monitoring well: one each at the top
and bottom of the saturated zone and three at
equidistant imtermediate points.
-------
Figure A-10. Area of toluene contamination near thfe Hangar/Administration
building (Sammons and Armstrong).
! K9S.D
Rock Garden
Hangar/Administration Building
.» K35.D
S11'
S25
Legend
. Well location and number -
K8D *S Letters D, S following well number
indicate Deep, Shallow
Ground-water flow-
Arrow indicates direction of flow
Area of toluene contamination
A.5.5 Performance Evaluation
Toluene and benzene levels in the downgradient
monitoring wells were monitored in one of the five
wells (M2/TG12) on a daily basis for the first 8 weeks
and on a biweekly basis thereafter. The other four
wells are monitored once a week.
Figure A-12 shows the concentrations of benzene
and toluene in Well M-2 (immediately downgradient of
the extraction well system) from April to December
1985. Toluene levels decreased from a baseline level
of 10,329 ppb to less than 10 ppb in approximately
100 days (Sammons and Armstrong 1986). The
decrease in toluene levels together with the
accompanying rise in benzene between May and July
could be attributed to the demethoxylation of toluene
to benzene by microbial activity. The increase in
benzene from October to December is attributed to
the appearance of a red slime (oxidized iron
compounds and complexes, mineral deposits, and
A-14
-------
Figure A-11. Offsite migration of the contaminant plume (Sammons and Armstrong 1986)
\
Location of
Extraction Wells
603— Water Table Contour
^- Ground-Water Flow
8S888: Plume of Contamination
500 1000 ft
I I
i n
100 200m
biomass), which plugged pump screens, coated the
inside of the piping system, and reduced the hydraulic
capacity of the system.
A.5.6 Project Costs
Project costs have not been published.
A.5.7 References
Sammons, J. H., and J. M. Armstrong. 1986. Use of
Low Flow Interdiction Wells to Control Hydro-
carbon Plumes in Groundwater. In: Proceedings
of the National Conference on Hazardous Wastes
and Hazardous Materials, Atlanta, Georgia, March
4-6,1986. Hazardous Materials Control Research
Institute, Silver Spring, Maryland.
A-
A.6 Case History F-Bulk Fuel
Storage and Distribution Center
A.6.1 Background
The loss of more than 200,000 gallons of fuel oil from
a bulk storage and distribution center was first
detected via inventory reconciliation. The loss, which
initially went unnoticed because of the high volume of
fuels turnover at the facility, was originally believed to
be the result of a shortage or theft. Further investi-
gations by the owner, however, revealed a leak in one
of the facility's underground transmission lines that
served a seldom-used remote loading rack. A cursory
investigation concluded (incorrectly) that the product
had migrated miles from the area. Consequently,
retrieval of the lost product did not begin until 2 years
later, when fuel oil-type hydrocarbons were en-
countered during the construction of a downgradient
well.
15
-------
Figure A-12. Concentrations of benzene and toluene in Well M-2 (Sammons and Armstrong 1986).
Well M-2
1200
1000 -
800
600 —
400 -
200 -
Nov.
Dec
Time
A.6.2 Site Description
The fuel storage depot, which is situated in a light
industrial and commercial area, is immediately adjacent
to a local quarry operation (Figure A-13). The site is
underlain by a silty-clay overburden and limestone-
dolomite bedrock. Depth to bedrock varies from 10 to
50 feet over short lateral distances, which indicates
the presence of a pinnacle-and-trough rock profile.
The dolomite is prone to the evolution of solution
channels, and deep residual soil has developed along
joint sets, rock fractures, and certain bedding planes.
The dominant joint set is oriented in a NNW-SSE
direction.
A shallow aquifer is present at 50 to 70 feet below
grade, and a second, much deeper, aquifer is present
at a depth of 200 feet. Ground-water flow, which is
controlled by the dominant joint set, is west to
northwest toward the adjacent quarry. One local
industry relies solely on ground water (approximately
150,000 gal/day) for all manufacturing processes.
This causes some local impact on ground-water
movement (f.e, induced flow).
A.6.3 Nature and Extent of Contamination
To define the impact of the product loss on the local
ground-water regime, a detailed hydrogeologic
investigation was conducted during the summer of
1982. Six 4-inch observation wells were constructed
along the dominant joint set at intervals of 100 to 150
feet and depths of 70 to 110 feet. Two additional
observation wells were located off the dominant joint
set normal to the trend. The defined free-product
plume measured up to 60 feet thick, 900 feet long,
and 300 feet wide. True product thickness reached
its maximum in the vicinity of Well G-2 (Figure A-14).
The predominant direction of free-product migration
was along the dominant joint set to the northwest
(Yaniga and Demko 1983).
A.6.4 Corrective Action
The immediate response to the pipeline break
entailed repair and rerouting of the plumbing. Product
retrieval was not initiated until 2 years later.
Aquifer testing prior to product-recovery attempts
indicated wells constructed to the 70- to 100-foot
A-16
-------
Figure A-13. Location of bulk fuel storage and
distribution center (Yaniga and
Demko 1983).
• Recovery Well
Existing Site Wells
G-1 o Observation Wells
Figure A-14. Configuration of the free product
plume (Yaniga and Demko 1983).
depth could not control the total product plume from
any one location (Yaniga and Demko 1983).
Consequently, a 12-inch-diameter recovery well was
constructed in the vicinity of Well G-2 to a depth of
240 feet.
Water table control and free product collection were
achieved by creating a cone of depression with a 2-hp
explosion-proof, water-table depression pump. By
maintaining a minimum of 20 feet of drawdown at the
pumping well, a radius of influence along the
dominant joint set trend of approximately 450 feet was
achieved. Free product flowing into the well along the
created cone of depression was collected by a
product-retrieval pump, which operates by a sensing
probe that differentiates between water and product.
An in-line transfer pump was designed into the system
to provide additional delivery head and to transfer the
product to a remote storage area.
Water pumped from the recovery well contained
dissolved organic constituents at levels below the
discharge criteria of the local regulatory agency. An
infiltration gallery was constructed within the influence
of the recovery well to recirculate the water to the
ground-water system.
A.6.5 Performance Evaluation
The free-product recovery system constructed at this
site successfully reversed the migration of the fugitive
fuel. Reusable fuel-oil-type hydrocarbons were
recovered at an average rate of 500 to 700 gal/day
(Yaniga and Demko 1983).
A.6.6 Project Costs
Project costs have not been published.
A-
• Recovery Well
AD Existing Site Wells
o Observation Wells
A.6.7 References
Yaniga, P. M., and D. J. Demko. 1983. Hydrocarbon
Contamination of Carbonate Aquifers: Assess-
ment and Abatement. In: Proceedings of the
Third National Symposium on Aquifer Restoration
and Ground-Water Mentoring. National Water Well
Association, Worthington, Ohio.
A.7 Case History G-Midwestern
Laboratory Facility
A.7.1 Background
Vapors were discovered in a laboratory building at a
midwestern industrial facility. Investigation of this
problem traced the source of the vapors to a below-
ground tank vault adjacent to the building. The vault,
which contained seven 6000-galIon tanks, was used
for storing clean fuel oils and waste solvents from the
laboratory facility. Testing of the tanks in the vault
revealed that two of the seven were leaking.
A.7.2 Site Description
The vault, which measures 25 feet wide, 70 feet long,
and 12 feet deep, was constructed in a natural clay
strata (glacial till) and backfilled with medium pea
gravel. Descriptions of the site's stratigraphy and
hydrogeology are not available.
A.7.3 Nature and Extent of Contamination
A series of monitoring wells were drilled in the vault
and in the area adjacent to the vault. Free product was
found to be confined to the vault area. Ground-water
contamination was confined primarily to the vault, but
17
-------
some dissojved hydrocarbons were detected in the
clay strata immediately adjacent to the tank. Soils
throughout the vault were saturated with aromatic and
aliphatic hydrocarbons. Total contamination was
calculated to be 655 gallons of free product
(solvent/fuel mixture) and 300 to 900 gallons of
adsorbed hydrocarbons (Brenoel and Brown 1985:
FMC 1984).
A.7.4 Corrective Action
The first phase of remediation was removal of free
product. Water was pumped from the sump well at 15
to 25 gal/min with a surface-mounted, explosion-proof
pump to depress the water table and allow free
product to collect. Free product was separated/
recovered from the well with a dual-pump system.
Clean water from this operation was injected into the
vault to sweep the contaminated area. A total of 700
gallons of free product were recovered in this manner
(Brenoel and Brown 1985; FMC 1984).
The second phase of remediation involved treatment
of the contaminated fill and ground water. The total
hydrocarbon concentration in the ground water prior
to free-product recovery was 22,700 ppb; this
concentration increased to approximately 45,000 ppb
during free-product recovery because of the
increased circulation of the ground water. The vault
contained approximately 22,000 gallons of ground
water, which represented 4 to 10 Ib of dissolved
hydrocarbons (Brenoel and Brown 1985; FMC 1984).
This amount could not be significantly reduced,
however, without removing the source of the
dissolved fraction, i.e., the material adsorbed in the fill.
Two options were considered for the second phase of
remediation: 1) pumping and treating of ground water
with carbon adsorption or air stripping, and 2) bio-
stimulation. The first option was not chosen for the
following reasons:
1) The quantity of carbon needed to treat all of the
contaminated ground water would be too costly.
2) Air stripping would require special engineering
and permitting because of concerns with
atmospheric discharges.
3) Time was a problem; downtime had to be minimal
because this was a high-activity area.
4) Carbon adsorption and air stripping would not
effectively remove the adsorbed material.
Biostimulation was chosen for the second phase of
remediation because of its cost-effectiveness, timeli-
ness, efficiency, and minimal equipment require-
ments. The basic elements of the biostimulation
process installed at this site included 1) a ground-
water circulation system to sweep the contaminated
area (i.e., a system of injection and recovery wells),
A-18
and 2) nutrient and oxygen injection capabilities.
Bacterial counts were taken at the site before the
program was initiated to verify the presence of an
active bacterial community. Two types of counting
procedures were used, one for the total bacteria and
one for specific hydrocarbon degraders. Counts
taken on successive days were as follows (Brenoel
and Brown 1985):
Total bacteria
420,000
300,000
Hydrocarbon degraders
5,400
6,100
In the laboratory, the addition of nutrients to ground
water from the vault increased both total and
hydrocarbon-degrading bacteria by several orders of
magnitude. These data confirmed the feasibility of
implementing biostimulation at the site.
The ground-water recirculation system, which was
designed to sweep the entire vault, consisted of four
injection wells and a pumping well. The injection wells
were screened throughout their total depth. Ground-
water pumping rates averaged 15 to 25 gal/min for
most of the project.
The nutrients required for effective stimulation of
bacteria and controlled degradation of hydrocarbons
are nitrogen, phosphorus, trace minerals, and
oxygen. Two formulations were used at the site: a
blend of ammonium chloride and sodium phosphates
and a formulated oxygen-enhancement solution.
Because the available oxygen level is critical to
maintaining a rapid rate of degradation, the oxygen-
enhancement solution was added continually for the
entire biostimulation phase. The nitrogen/phos-
phorus (N/P) blend was added to the four injection
wells in batches to maintain N/P levels of 100 to 200
ppm.
The nutrient addition caused a 10-fold increase in total
bacteria and a 200-fold increase in hydrocarbon
degraders (Brenoel and Brown 1985; FMC 1984).
The hydrocarbon concentration dropped significantly
during the course of biostimulation. The greatest
drop occurred in the first 42 days of operation, which
paralleled the consumption of nutrients (Brenoel and
Brown 1985; FMC 1984).
At the end of the biostimulation phase, the injection
rate in the vault decreased from 15 to 25 gal/min to as
low as 1.5 gal/min, apparently the result of movement
of silts through the backfill and degradation of cement
in the vault causing a buildup of silicates and silts at
the well screens. Because of concern with the ability
to supply nutrients at the low injectivity and because
the hydrocarbon levels had been significantly
-------
I
reduced, the biostimulation phase of remediation was
terminated.
The final phase of remediation involved the use of an
activated carbon system for further reduction of
residual hydrocarbons in the ground water. The
system consisted of two activated carbon tanks
plumbed in series. Influent to the carbon columns was
via a surface-mounted pump at the vault sump. The
effluent was recirculated through injection well No. 3
for continued sweeping of the vault. This design
minimized site reconstruction and equipment.
During the operation of the carbon column, the
influent hydrocarbon concentration increased. This
increase was traced to a leaking line between the
laboratory and the vault that had not been detected
and repaired during the initial phase of remediation.
Consequently, a slow, low-volume leak had continued
throughout the remediation program. The leak
contaminated water entering the vault, but it did not
add significant quantities of hydrocarbons. Upon
repair of the leak, influent hydrocarbon levels
decreased because the carbon columns were able to
remove the residual dissolved hydrocarbons. The
rapid drop in influent hydrocarbon levels after the leak
was repaired demonstrates that it was no longer
contaminating the fill in the vault. After an additional 2
months of carbon column operation, dissolved
hydrocarbon levels were reduced to less than 10 ppb
(Brenoel and Brown 1985; FMC 1984).
A.7.5 Performance Evaluation
The remediation program eliminated the soil
contamination in the vault and significantly reduced
the ground-water contamination. This project illus-
trates how an integrated approach to site remediation
can solve severe contamination problems. Each
stage of remediation was chosen for maximum
efficiency in dealing with the contamination problem.
A well-designed free-product-recovery system elimin-
ated the severe, free-floating layer. Biostimulation
was effective in removing the adsorbed and most of
the dissolved hydrocarbons. Finally, carbon adsorp-
tion was also effective in removing low-level, residual,
dissolved hydrocarbons. This integrated approach
speeded up the recovery and reduced costs.
A.7.6 Project Costs
Project costs have not been published.
A.7.7 References
Brenoel, M., and R. A. Brown. 1985. Remediation of
a Leaking Underground Storage Tank With
Enhanced Bioreclamation. Presented at the
NWWA 5th National Symposium and Exposition,
Aquifer Restoration and Groundwater Monitoring,
Columbus, Ohio, May 21-24, 1985. National
Water Well Association, Inc., Worthington, Ohio.
A-
FMC Corporation. 1984. Case History - Midwestern
Industrial Facility. Promotional Literature.
Princeton, New Jersey.
A.8 Case History H-Chemical
Pipeline
A.8.1 Background
On August 16, 1983, investigation into a pressure
loss in the water main at an industrial facility led to the
discovery of a ruptured methylene chloride pipeline in
proximity to the water main. The buried pipeline had
leaked an undetermined amount of methylene
chloride into the soil and ground water before the
break was discovered.
stratigraphy and
A.8.2 Site Description
Descriptions of the site's
hydrogeology are not available.
A.8.3 Nature and Extent of Contamination
Monitoring wells installed in the vicinity of the pipeline
break indicated that contamination was confined to a
clay layer at the 20-foot level and had not reached an
aquifer at the 100-foot level. Coarse gravel lenses
and utility lines in the area, however, had permitted
methylene chloride to migrate laterally with little
resistance.
A.8.4 Corrective Action
Emergency response to the spill involved only
containment. Trenches were constructed in the
vicinity of the pipeline break to intercept the
contamination. Free product in the trenches was
collected with vacuum equipment and staged in
vessels for eventual onsite treatment.
The alternatives investigated for environmental
restoration of the site once the spill was contained
included excavation and disposal, physical contain-
ment, gravity collection, ground-water pumping, and
biological treatment. A combination of these alter-
natives was determined to be the most effective
means for site remediation.
The first stage of site remediation entailed excavation
of the highly contaminated soil. Approximately 160
yd" of soil (1.1 percent of the total contaminated soil)
was removed from the site to a Class A secure landfill
(Flathman and Caplan 1986; Flathman et al. 1985).
After the highly contaminated soil was removed,
pumping wells were installed, and contaminated
ground water was pumped to the surface by both
positive-placement and suction-lift techniques.
Packed-column air stripping was considered the
preferred physical treatment technology for removing
methylene chloride from recovered ground water
because of the compound's strippability and the
19
-------
relatively maintenance-free operation of the unit.
Also, the State regulatory agency did not require
vapor-phase scrubbing to control air emissions from
the stripping tower.
The recovered ground water was characterized and
bench-scale tested to determine necessary pre-
treatment requirements. A schematic diagram of the
physical treatment system, which was designed to^
operate at a flow rate of 10 to 15 gal/min, is presented
in Figure A-15. The recovered ground water was first
pumped through a downflow mixed-media filter
(anthracite, silica sand, and pea gravel) to remove
sand and other paniculate matter. It was then piped to
a separation tank with a residence time of 14 hours,
which allowed the denser methylene chloride to
separate from the water by gravity. Pure product was
recovered from the bottom of the tank. Supernatant
containing up to 150 ppm dissolved methylene
chloride was then pumped through a skid-mounted
shell-and-tube heat exchanger to raise the
temperature of the water from 10°C to more than 40°C.
The heated water was then pumped to the top of a 9-
foot air-stripping column packed with 2-inch ceramic
Raschig rings for removal of soluble methylene
chloride. The tower operating temperature was
maintained between 27°C and 60°C. Effluent
containing less than 20 ppm methylene chloride was
discharged to the facility's onsite wastewater
treatment plant for further processing. Contaminant-
laden air was vented to the atmosphere through a 10-
foot stack.
After 2 months of operation, air stripping reduced the
concentration of methylene chloride in the ground
water by an estimated 97 percent (Flathman and
Caplan 1986; Flathman et al. 1985). It became
increasingly more difficult to remove the residual
contamination by physical means, however, and
biological techniques were considered.
Bench-scale biodegradation studies indicated the;
presence of a naturally occurring microbial population
capable of degrading methylene chloride; however,
inorganic nitrogen and phosphorus nutrient additions
would be necessary. The biological treatment system,
illustrated in Figure A-16, consisted of a recovery/
injection system and a modified activated-sludge
system. The recovery system was used to withdraw
contaminated ground water for above-ground
treatment. Supernatant containing adapted micro-
organisms from the activated sludge process was
reinjected into the subsurface environment, which
created a closed-loop system. Biodegradation of
methylene chloride occurred in situ as well as above
ground. Biological treatment commenced on the
82nd day of field operations and was suspended after
41 days because of the onset of winter conditions.
A.8.5 Performance Evaluation
Figure A-17 presents the reduction in methylene
chloride concentration obtained in monitoring well B-5
(located 20 feet downgradient of the pipeline break)
during the 3-month operation of the air stripper. After
2 months, methylene chloride had been reduced 97
percent (from 9300 ppm to 300 ppm) (Flathman and
Caplan 1986, Flathman et al. 1985).
Biodegradation of methylene chloride in the ground
water was rapid. A 50 percent reduction in methylene
chloride concentration was observed in monitoring
well B-5 within 8 days (Figure A-18). This reduction in
methylene chloride was accompanied by a
corresponding increase in chloride concentration.
When biological treatment was suspended at the end
of the fourth month, 97 percent of the residual
contamination had been degraded for an overall
reduction in methylene chloride of greater than 99.9
percent (Flathman and Caplan 1986; Flathman et al.
1985).
A.8.6 Project Costs
Project costs have not been published.
Figure A-15. Schematic diagram of the physical treatment system (Flatham and Caplan 1986).
Foad from
Vacuum Recovery
Unit
Air Stripper
To Plant Wastewater
Treatment Facility
Solvent
Recovery
Air in
A-20
-------
Figure A-16. Schematic diagram of the biological treatment system (Flatham and Caplan 1986)
Nutrients
pH Adjustment
Feed from
Vacuum
Recovery"
Unit
To Underground
Injection System
MakeUp
Water
from Plant
Figure A-17. Reduction in methylene chloride
concentration in well B-5 achieved during
air stripping (Flathman and Caplan 1986).
Well B-5
Figure A-18.
Reduction in methylene chloride
concentration in well B-5 achieved during
biodegradation (Flathman and Caplan 1986).
Well B-5
10000
8000
o.
o.
cf
•1 6000
I
I
.2
O
_
I
4000
2000
\ I I I I I I I I I
A /
/ V
•-• Methylene Chloride
»<• Chloride
_L
_L
400
300
200
CL
Q.
1
100
90 100
110 120 130 140 150
Days
120
A-21
-------
A.8.7 References
Flathman, P. E., and J. A. Caplan. 1986. Cleanup of
Contaminated Soil and Ground Water Using
Biological Techniques. In: Proceedings of the
National Conference on Hazardous Wastes and
Hazardous Materials, March 4-6,1986. Hazardous
Materials Control Research Institute, Silver Spring,
Maryland.
Flathman, P. E., et al. 1985. In Situ Physical/
Biological Treatment of Methylene Chloride
(Dichloromethane) Contaminated Ground Water.
In: Proceedings of the Fifth National Symposium
on Aquifer Restoration and Ground Water
Monitoring, Columbus, Ohio, May 1985.
A.9 Case History l-Biocraft
Laboratories, Waldwick, New Jersey
A.9.1 Background
Biocraft Laboratories is a small synthetic-penicillin
manufacturing plant located on a 4.3-acre site in an
industrial park in Waldwick, New Jersey. In August
1975, environmental degradation was observed in
Allendale Brook and its tributary, a small creek 350
feet east of the site. The source of contamination in
these streams was traced to a storm-sewer line that
runs across the Biocraft plant site and discharges into
the creek. Subsequent investigations by Biocraft
revealed that two pipes leading from the plant to
underground waste-solvent storage tanks were
leaking and that a mixture of solvents had infiltrated
the storm-sewer line.
A.9.2 Site Description
The Biocraft Laboratories plant site is in the northern
section of Bergen County, New Jersey. The property
is relatively flat (0 to 3 percent slope). Approximately
30 percent of the area is paved or covered with
buildings, 10 percent is grassed, and the remaining
60 percent is lightly forested (Figure A-19).
Glacial till, composed of a poorly sorted mixture of
boulders, cobbles, pebbles, sand, silt, and clay,
underlies the surface at a thickness of 8 to 15 feet.
Permeability varies throughout the till layer and ranges
from 0.02 to 36 gal/day per ft2 (EPA 1984; Jhaveri and
Mazzacca 1983).
Figure A-19. Biocraft Laboratories site, Waldwick, New Jersey (Jhaveri and Mazzacca 1985).
Municipal Deep Well
Approx. 1000 ft from
Area of Contamination
A-22
-------
Approximately 40 feet of semiconsolidated silt and
fine sand underlie the till layer. Although no testing
has been conducted, visual inspection of this material
suggests it has a very low permeability and functions
as an aquiclude (EPA 1984; Jhaveri and Mazzacca
1983).
Brunswick shale underlies the semiconsolidated layer
at a depth of 50 to 60 feet and a thickness of several
hundred feet. The Brunswick formation is the primary
water-supply aquifer for the area. Primary ground-
water flow occurs in the interconnecting fractures,
vertical joints, and faults in the shale, whereas little or
no flow occurs in the rock. A municipal deep well is
located in the Brunswick formation approximately
1000 feet east of the spill site. Biocraft Laboratories'
Figure A-20.
Ground-water surface contours (Jhaveri
own deep well, which supplies water for their chemical
manufacturing operations, is located directly under
the contaminant plume.
As shown by the ground-water surface contours in
Figure A-20, ground-water flow at the Biocraft site is
somewhat irregular because of the heterogeneous
geology, diverse surface cover, and other factors. A
noticeable ground-water mound, corresponding to
the south and east edges of the blacktopped area, is
present; this represents an area of ground-water
recharge. Ground-water flow from the mound is
omnidirectional, but the major flow regimes move
toward the northwest, northeast, and south.
The average depth to ground water ranges from 0 to 9
feet, depending on seasonal fluctuations. Ground-
and Mazzacca 1983).
Legend
Ground-Water Flow
243.0
A-23
-------
water flow in the shallow aquifer is fairly rapid (average
rate is 0.4 ft/day) (EPA 1984; Jhaveri and Mazzacca
1983).
A.9.3 Nature and Extent of Contamination
Sometime between June 1972, when the plant
opened, and August 1975, when the contamination
was discovered, two pipes feeding underground
waste-solvent storage tanks began leaking a mixture
of methylene chloride, acetone, n-butyl alcohol, and
dimethyl aniline. Based on daily tank inventory
readings, Biocraft estimated that as much as 33,000
gallons (285,000 pounds) of waste solvents, as
identified in Table A-1, may have leaked into the
subsurface (EPA 1984; Jhaveri and Mazzacca 1983).
Table A-1. Estimated Quantities of Solvents
Released*
Substance Estimated Quantity (Ib)
Methylene Chloride
n-Butyl Alcohol
Dimethyl Aniline
Acetone
181,500
66,825
26,300
10,890
Total Solvents
285,515
* Data are from EPA 1984 and Jhaveri and Mazzacca 1983.
Contamination was confined to the shallow aquifer.
The contaminant plume flowed predominantly north
and northeast (toward the northern edge of the
property and a storm sewer) and south (toward the
southern property boundary). The contaminated area
covered approximately 1.75 acres and encompassed
12,000 yd3 of soil (Amdurer, Fellman, and Abdelhamid
1985; Jhaveri and Mazzacca 1983).
A.9.4 Corrective Action
The leaking underground feed lines to the storage
tanks were sealed in the winter of 1975, and above-
ground feed lines were installed. During January
1976, six ground-water monitoring wells (2-inch well
points with depths ranging from 10 to 15 feet) were
constructed on site. From June 1976 to early 1979,
16 additional wells were installed for monitoring and
pumping contaminated ground water. Figure A-21
shows the locations of these wells.
From January 1977 through 1978, five wells (Nos. 2,
3, 8, 10, and 13) were selectively pumped at a
combined rate of 10,000 gal/min, and the con-
taminated ground water was disposed of off site (at an
industrial wastewater plant, an incinerator, or a pre-
treatment facility) at an average cost of $0.35/gal
(Amdurer, Fellman, and Abdelhamid 1985; Jhaveri
and Mazzacca 1983).
In December 1978, dissatisfied with the progress of
these initial measures to clean up the contaminated
ground water, the State ordered Biocraft to accelerate
the decontamination process. Several alternative
response technologies were considered, including
the following:
Collecting and treating all discharge from the
storm sewer.
Isolating the storm sewer from the contaminated
flow by resleeving the existing pipe, grouting the
pipe joints, or replacing the sewer with a
noninfiltrating pipe.
Surrounding the contaminated area with grout or a
slurry cutoff wall and pumping and treating or
disposing of the ground water from within the
confined area.
Excavating the entire contaminated soil column
under the site.
Each of these alternatives was rejected, in turn, as
ineffective, impractical, or too costly.
In May 1979, Biocraft Laboratories and the State
settled on a biodegradation-biostimulation process
designed to provide both contaminant plume con-
tainment and removal of the source of contamination
in a cost-effective manner. The system, which has
been patented by Groundwater Decontamination
Systems, a subsidiary of Biocraft Laboratories, entails
the following:
• Collecting the contaminated plume downgradient
of the source area in a slotted-pipe collection
trench and two interceptor wells.
• Treating the collected ground water in a surface
aerobic biological treatment system.
Injecting the treated "bioactive" water upgradient
of the source area in two slotted-pipe recharge
trenches to flush the soil of contaminants.
Stimulating in situ biodegradatipn of contaminants
in the subsurface by injecting air through a series
of aeration wells along the path of ground-water
flow.
It was estimated that this system would require 5 years
for complete restoration of the Biocraft site compared
with an estimated 15 to 20 years for the initial ground-
water withdraw! and offsite disposal alternative
(Amdurer, Fellman, and Abdelhamid 1985).
The research and development phase of the project,
which spanned 2-1/2 years, included a hydrogeolic
investigation, bench- and pilot-scale studies, and
design and construction of system components.
Installation of the system was completed in June
1981. Major system components include a
A-24
-------
Figure A-21.
Location of ground-water monitoring and pumping wells (Jhaveri and Mazzacca 1985).
Legend
• 10—Ground-Water Monitoring Well
T34—Trench Monitoring Well
P13—Pumping Well
subsurface collection drain (Trench A), two interceptor
wells, a four-tank dual biological treatment system, two
effluent injection trenches (Trenches B and C), and a
series of nine in situ aeration wells (see Figures A-22
andA-23).
The primary ground-water collection system consists
of a subsurface drain about 80 feet long, 4 feet wide,
and 10 feet deep. Two 16-inch slotted steel
collection pipes were laid on a bed of gravel at the
bottom of the trench, sloped toward the center, and
connected to a central collection pumping well. This
well (No. 13) has a 12-inch-diameter steel casing with a
2.5-foot slotted screen and a 10 gal/min stainless
steel submersible pump. The trench has two 2-inch-
diameter PVC monitoring wells installed on each side
of the central collection well. The trench was
backfilled with washed stone and then covered with
15-mil plastic sheet, backfilled with earth, and finished
to grade.
Ground-water pumping is also being carried out in two
interceptor wells (Nos. 30 and 32A) on the southern
edge of the property to collect the southern
component of the contaminant plume. These wells
consist of trenches about 16 feet long, 4 feet wide,
and 10 feet deep. An 8-inch and a 12-inch PVC fully
slotted casing and 10 gal/min submersible pump were
installed in well Nos. 30 and 32A, respectively. The
trenches were then backfilled with washed stone and
earth and finished to grade.
An average of 13,680 gal/day of water is pumped from
the collection trench and the interceptor wells to the
biological treatment system. This system consists of
four tanks (two aeration tanks and two sludge settling
tanks), each with a capacity of about 5400 gallons.
Influent water from the collection trench and two
interceptor wells is pumped first to the aeration tanks,
where most of the biodegradation occurs. The
A-25
-------
Figure A-22.
Cross section of biodegradation-biostimulation system (EPA 1984).
300ft
10 gal/min >{)
Pumo V-X
15-mil
Plastic
Sheet
Washed
Stone
Gravel
Slotted
Pipe
Collection
Trench
Process Influent
Air (4 psig)
Bioactive
Water
TTT
I
I I
I I
I I
I I
I I
I I
I I
I I
I I 1 1 1 1 1 1 1
Nine Equally
Spaced Aeration
Wells
Aeration
Tanks
Settling
Tanks
15-mil
Plastic
Sheeting
Sand
Base
Recharge
Trench
microorganisms isolated from the aeration tanks
consist of Pseudomonas (40 percent), Agrobacterium
(40 percent), and Arthrobacter (20 percent), which are
naturally occurring soil bacteria (Jhaveri and Mazzacca
1983). Air is added to each tank through a series of
porous ceramic tube diffusers at a rate of 20 ftVmin.
Temperature is kept constant at 68°F by single-pass
steam coils installed in the tanks. The tanks have 2
inches of insulation to help buffer the effects of
ambient temperature. A nutrient solution is metered
in from mixing tanks in the pump house to obtain the
following concentrations in the aeration tanks:
Nutrient salt
NH3CI2
KH2PO4
K2HPO4
MgSO4
Na2SO4
Concentration,
nig/liter
500
270
410
14
9
Nutrient salt
CaCI2
MnSO4
FeSO4
Concentration,
ma/liter
0.9
1.8
0.45
The system now operates at an average flow rate of
9.5 gal/min or 13,500 gal/day; retention time in the
aeration tanks is 17.5 hours. The system can handle a,
flow of up to 14 gal/min or 20,000 gal/day with a
retention time of 12 hours, but the pumping wells are
already at capacity flow.
Effluent air from the aeration tanks passes through
vapor-phase carbon adsorbers to remove any
volatilized organics. Pilot-plant studies and the
infrequent need for replacement carbon indicate that
the amount of volatilization is not substantial.
The effluent streams from the aeration tanks are
combined and pumped to two sludge settling tanks,
where some biomass solids are settled out and
recycled to the aeration tanks (approximately 200
A-26
-------
Figure A-23. Plan view of biodegradation-biostimulatio'n system (EPA 1984),
Recirculating
Ground-Water
Row
Collection
Trench
P13
80ft
gal/day). Much of the biomass, however, is allowed to
pass with the supernatant into the recharge trenches
to provide continuous inoculation of the subsurface
with microorganisms. Waste sludge production is
minimal (approximately 11 gal/min) because sludge is
recycled to the aeration tanks and reinjection trenches
and cell reproduction rates associated with the
biodegradation of relatively refractory organics are low.
Effluent from the biological treatment plant is
reinjected through two recharge trenches located at
the ground-water mound to flush the soil and
subsurface with treated water to remove residual
contaminants. The dimensions of each trench are
approximately 100 feet long, 4 feet wide, and 10 feet
deep. The trenches are lined on the bottom, ends,
back, and top with a 15-mil plastic liner so that injected
water can exit only from the front side of the trench.
The bottom section of the liner was laid on 1 foot of
sand and then covered with 0.5 foot of sand. The
trenches were filled with 2-inch washed stone to a
thickness of 5 feet. A 2-inch vertical inlet pipe ending
in a "Y" connection was installed in the center of the
A-27
Extraction
:,.....,. ' wells
trench. Connected to the "Y" were two 20-foot
sections of 2-inch slotted pipe. The trench was then
backfilled to the surface with 2-inch washed stone. A
4-foot-high manhole was installed over the recharge
pipe for access. A 4-foot-high soil mound was then
placed over the top liner to insulate the trench from
freezing. Each trench has two monitoring wells, one
at either end of the trench. These wells can also be
used for flushing the system of sludge accumulation if
required. Average flow of effluent to the two trenches
is about 13,680 gal/day.
As effluent flows from the treatment plant to the
trenches, air is injected into the recharge line with a jet
ejector or air compressor when the flow rate is low.
Aeration of the reinjected effluent creates a biological
trickling filter in the trenches, which further increases
biodegradation of organics. The water level in the
trenches is kept at surface elevation to flush
contaminants in the shallow soil layers.
A series of nine continuous aeration wells were
installed in the subsurface along the major path of
contaminant plume movement. Air is injected into
-------
each well at a pressure of 4 to 9 lb/in.2 Adding air to
these wells creates a zone of subsurface aeration,
where contaminants in the groundwater passing near
the wells are aerobically biodegraded. The nine wells
are spaced on 30-foot centers and arranged in a
rectangular matrix about 30 feet wide and 100 feet
long. The arrangement of the wells was based on an
assumed 15-foot radius of influence. Residence time
through the aeration zone, assuming an average
ground-water velocity of 0.4 ft/day, ranges from 65 to
300 days, depending on the direction of ground-
water flow. Ground-water temperature averages 54°F,
which is adequate for biodegradation.
The nutrient tanks, pumps, flow meters, temperature
recorders, etc., are housed in a small control room.
Monitoring and rate adjustments are performed as
required.
A.9.5 Performance Evaluation
After 3 years of operation, the contaminant plume was
reportedly reduced by 90 percent (Figure A-24)
(Jhaveri and Mazzacca 1985). The reduction of COD
in monitoring wells No. 3 and No. 10, which were
highly contaminated prior to the biostimulation
process, is shown graphically in Figures A-25 and A-
26. Continuous core samples taken in the vicinity of
these wells show no detectable contamination from 0
to 12 feet at a detection limit of 0.8 mg/liter (Jhaveri
and Mazzacca 1985).
Figure A-24.
Reduction of contaminant plume (Jhaveri and Mazzacca 1985).
Legend
Original Plume
Present Plume
10—Ground-Water Monitoring Well
T34—Trench Monitoring Well
P13—Pumping Well
A-28
-------
Figure A-25.
500
Reduction of COD in well No. 3
(Jhaveri and Mazzacca 1985).
Well No. 3
l l i l i i i i i i i i i i
o
123412341
1982 1983
Year, quarters
2341234
1984 1985
The area near the pumping wells on the southern
edge of the plume, which shows the highest level of
residual contamination, is still biologically active. This
is evident from increased concentrations of dissolved
CO2 (a byproduct of aerobic respiration) in monitoring
wells in that zone (Jhaveri and Mazzacca 1985).
A.9.6 Project Costs
The total cost of research and development (R&D)
and capital design and construction of the
biostimulation operation at Biocraft was about
$926,000 (EPA 1984). These costs are reported in
Table A-2. About half of the total capital cost
($446,280) was for in-house process development,
including construction of a pilot plant. Virtually all of
this process development cost was a one-time-only
expense.
Table A-2.
Activity
Capital Costs, Biocraft Laboratories
Cost ($)
Hydrogeological study
Research and development
Ground-water collection/recharge
system design and installation
Biological treatment plant design
and construction
73,948
446,280
184,243
221,207
Total capital costs
925,678
Data from EPA 1984.
Figure A-26.
350
Reduction of COD In well No. 10
(Jhaveri and Mazzacca 1985).
Well No. 10
234123412341
1981 1982 1983
Year, quarters
2341
1984
234
1985
The operation and maintenance (O&M) costs
(reported in Table A-3) include utilities, maintenance
labor and overhead, and chemicals. Total O&M costs
are approximately $226.53/day (EPA 1984). At an
average daily treatment rate of 13,680 gal/day, unit
O&M costs are about $0.0165/gal.
Table A-3.
Component
Operation and Maintenance Costs,
Biocraft Laboratories
Cost Per Day ($)
Utilities (steam, electricity)
Maintenance
Nutrients
47.40
159.93
19.20
Total O&M costs
226.53
Data from EPA 1984.
Total capital and O&M costs of the biodegradation-
biostimulation process now in operation at the Biocraft
site are estimated to be a quarter of the total cost that
would have been incurred with the initial remedial
measure (i.e., pumping and offsite disposal)
(Amdurer, Fellman, and Abdelhamid 1985).
A.9.7 References
Amdurer, M., R. Fellman, and S. Abdelhamid. 1985.
In Situ Treatment Technologies and Superfund.
In: Proceedings of International Conference on
New Frontiers for Hazardous Waste Management.
EPA-600/9-85-025.
A-29
-------
Jhaveri, V., and A. J. Mazzacca. 1983. Bioreclama-
tion of Ground and Ground-water, Case
History. In: National Conference on Manage-
ment of Uncontrolled Hazardous Waste Sites,
Washington, D.C., October 31-November 2,
1983. Hazardous Materials Control Research
Institute, Silver Spring, Maryland.
Jhaveri, V., and A. J. Mazzacca. 1985. Bioreclamation
of Ground and Ground-water by In Situ
Biodegradation: Case History. In: The Sixth
Annual National Conference on Manage-
ment of Uncontrolled Hazardous Waste Sites,
Washington, D.C., November 4-6, 1985. Haz-
ardous Materials Control Research Institute, Silver
Spring, Maryland.
U.S. Environmental Protection Agency. 1984. Case
Studies No. 1-23: Remedial Response at Haz-
ardous Waste Sites. EPA-540/2-84-002b.
A.10 Case History J-Fairchild
Camera and instrument Corp.,
South San Jose, California
A.10.1 Background
On December 4, 1981, Fairchild Camera and
Instrument Corp. of South San Jose, California, a
manufacturer of semiconductor devices, reported a
leak in excess of 20,000 gallons from an underground
waste solvent storage tank to the California Regional
Water Quality Control Board (San Francisco Bay
Region). Fairchild's subsequent analysis of inventory
records revealed an estimated chemical loss of;
43,000 gallons over an 18-month period (June 1980
through November 1981). The leak was apparently
caused by chemical degradation of the fiberglass tank,
which contained a mixture of 1,1,1-trichIoroethane
(TCA), xylene, acetone, and isopropyl alcohol (IPA).
According to company records, the mixture contained
approximately 15 percent TCA.
A.10.2 Site Description
The Fairchild plant site is located on the Santa Teresa
Plain in the southern Santa Clara Valley ("Silicon
Valley") of California. The main plant is approximately a
half-mile west of the Santa Clara Valley Water District's
Coyote Percolation Pond, which is used to recharge
the area's ground-water aquifers. The ground water
flows from the percolation pond toward the Fairchild
facility and is extracted by the Great Oaks Water
Company at numerous wells downgradient.
The Fairchild site is underlain by two distinct alluvial
deposits separated by a buried, clayey-silt-filled valley
known as Edenvale Gap, which generally trends east
to west. The alluvial deposits on either side of the gap
consist of interlayed strata of sands, gravels, silts, and
clays. Many strata are discontinuous and may be
locally interrupted by buried stream channels. The
coarser-grained strata (sands and gravels) compose
an aquifer system under the plant. The less
permeable strata (silts and clays) tend to retard, but
not prevent, movement of water and contaminants
between aquifers.
Four aquifers, designated A, B, C, and D, have been
identified beneath the Fairchild site. The A aquifer,
which is transected by Edenvale Gap, generally
occurs in the upper 50 feet of the alluvium. It is first
encountered 10 to 20 feet below the ground surface
north of the gap (east of the plant) and 35 to 40 feet
below the ground surface south of the gap (west of
the plant). These two shallow aquifers are
hydraulically connected in the vicinity of the leaking
tank.
The B aquifer occurs at a depth of 60 to 100 feet and
is laterally continuous across the plant site. The A-B
aquitard, which ranges in thickness from 5 to 30 feet,
separates the shallow A aquifers from the deeper B
aquifer. The A and B aquifers appear to merge and
form a single A/B aquifer at a point a few hundred feet
west of the main plant.
The C aquifer occurs from 150 to 190 feet below the
ground surface and the D aquifer occurs from 220 to
270 feet below grade. (The D aquifer is poorly
defined and nonexistent in some areas.)
Ground-water flow in the shallow aquifers has both a
horizontal and a vertical component. The horizontal
component is generally northwest toward Edenvale
Gap and through the narrow hydraulic connection
between the A aquifers. The vertical component of
ground-water flow has resulted in downward migration
of contaminated ground water into deeper aquifers.
Pathways for this vertical migration are believed to be:
Through sand interbeds in the A-B aquitard.
By slow seepage through fine-grained strata
separating aquifers.
Through improperly cased or abandoned wells
that penetrate multiple aquifers.
A.10.3 Nature and Extent of Contamination
At the direction of the California Regional Water
Quality Control Board (RWQCB), the lead agency
involved in this incident, Fairchild conducted detailed
subsurface geologic and hydrogeologic investi-
gations to define the extent of soil and ground-water
contamination. Numerous soil borings and ground-
water monitoring wells were drilled both on site and
offsite over a period of several months following
discovery of the waste solvent leak.
Results of the soil exploration program showed
concentrations of solvents in the soil on site
A-30
-------
extending to a depth of 50 feet (i.e., to the A-B
aquitard). Solvents were concentrated in the area
immediately adjacent to the waste solvent tank
(personal communication regarding proposed
Fairchild's conceptual onsite remedial plan from P. E.
Antommaria, Canonie Environmental, Chesterton,
Indiana, to L. Amon, Manager, Real Estate and
Facilities Planning, Fairchild Camera and Instrument
Corp., Mountain View, California, July 26,1982).
Results of the ground-water monitoring program
revealed a plume of contamination roughly 2000 feet
long and 1000 feet wide that extended west from the
Fairchild facility to the Great Oaks Water Company well
No. 13 [GO-13(M)] (internal memo regarding under-
ground waste solvent leak at Fairchild Camera and
Instrument Corp. from P. W. Johnson and H. J.
Singer, Toxics Division, to F. H. Dierker, Executive
Officer, California Regional Water Quality Control
Board, San Francisco Bay Region, March 8, 1982).
The most extensive TCA contamination was found in
the B aquifer. Initially, TCA was detected in GO-13 at
levels up to 5.7 ppm; however, no other municipal
wells in the vicinity appeared to be affected (Levine
1981).
A.10.4 Corrective Action
Upon discovery and reporting of the waste solvent
leak, Fairchild immediately emptied the underground
tank and replaced it with a temporary aboveground
tank. The underground tank and the highly
contaminated soils adjacent to it were subsequently
excavated.
On January 19, 1982, Fairchild began extracting
ground water from GO-13 to create a zone of influence
that would draw the contaminant plume into the well
and prevent its migration further downgradient. The
extracted ground water was treated by three-stage
carbon adsorption before being discharged via a
storm sewer to Canoas Creek, a tributary of the
Guadalupe River. This discharge was permitted under
NPDES. Initially, GO-13 was pumped at a rate of 500
gal/min; production was more than tripled over the
next 6 months as additional treatment capacity was
added.
A.10.4.1 Onsite Remedial Activities
A plan outlining Fairchild's proposed onsite remedial
activities was submitted to the RWQCB on July 26,
1982. This plan, which was envisioned to minimize
the potential for migration of materials from the site
and to restore ground water beneath the site, called
for:
Excavation of solvent-contaminated soils to the
extent practicable in the area adjacent to the
former waste solvent tank.
A-31
Installation of a purge well in the vicinity of WCC-
14(B), where the A and B aquifers merge.
Treatment of the purged ground water.
Continued monitoring of the site.
Data from the soil exploration program showed
concentrations of solvents in the soil to a depth of 50
feet in the area immediately adjacent to the tank.
Because of the proximity of this area to the plant and
the depth of solvent concentration, it was originally
believed that soil removal from the site would be
technically and economically infeasible. Further study
of the problem, however, yielded a method for soil
removal based on a modified caisson installation
technique.
Excavation of highly contaminated soils began in
October 1982. As soil was removed from the
caissons, they were filled with concrete. Ground water
encountered during the excavation was pumped
through the onsite carbon treatment system and
discharged to Canoas Creek. Approximately 3000 yd=>
of soil were removed from the site in covered,
watertight trucks and disposed of at sites approved by
the California Department of Health Services (internal
memo of October 12, 1983, from P. W. Johnson,
Toxics Division, to F. H. Dierker, Executive Officer,
California Regional Water Quality Control Board, San
Francisco Bay Region, in response to Dr. D. Todd's
memo of September 19, 1983, to B. Roeder). Soil
removal was important to the reduction of the quantity
of chemicals that were subsequently removed by
ground-water purging.
On November 5, 1982, following completion of the
soil excavation phase of remediation, Fairchild
obtained written approval from the RWQCB to
implement a ground-water extraction/treatment pro-
gram designed to intercept the contaminant plume
closer to the source and thereby minimize the offsite
migration of solvents. A newly installed purge well,
RW-1(A,B), and an existing 8-inch observation well,
WCC-20(B), which were located in the main pollutant
plume near the plant, were pumped at a combined
flow rate of 1200 gal/min. The extracted ground water
from RW-1 and WCC-20 was treated by carbon
adsorption* prior to its discharge to a storm drain
tributary of Canoas Creek. Effluent limits for the
combined discharge from RW-1, WCC-20, and GO-13
as set forth in RWQCB Order No. 82-61 are presented
in Table A-4. In addition, the Bay Area Air Quality
Management District limited air emissions at the point
of discharge to 150 Ib/day of smog precursors (all
constituents except TCA) and 150 Ib/day of nonsmog
* Other treatment options investigated included air stripping and oxidation by
ozone or peroxide; however, these technologies failed to achieve the
specified effluent and air emission limitations.
-------
precursors (TCA) (personal communication regarding
Falrchild's extraction, treatment, and discharge of
contaminated ground water from F. H. Dierker,
Executive Officer, California Regional Water Quality
Control Board, San Francisco Bay Region, to L.
Amon, Fairchild Camera and Instrument Corp.,
Mountain View, California, November 5,1982). ;
Table A-4. Extracted Ground-Water Effluent
Limitations*
Effluent Limit
Constituent (daily maximum) (mg/liter)
Acetone
1,1-DichIoroethylene
Isopropyl alcohol
Sum of monocyclic compounds
(includes 1,2-dichtorobenzene,
ethylbenzene, toluene, xylene)
1,1,1-Trichloroethane
Tetrachloroethylene
50
0.30
50
0.1 Ot
5.00
0.10
Data from CRWQCB (undated).
T The discharge may contain monocyclic compounds in excess of
the 0.1 mg/liter daily maximum, but not to exceedl.O mg/liter, up
to 4 days per month.
Throughout the period of onsite remedial activities,
work areas were monitored for organic vapors with
organic vapor analyzers (OVA's), water levels in onsite
observation wells were recorded to determine field
drawdowns, and ground-water samples were
collected for routine chemical analysis. •
A.10.4.2 Off site Remedial Activities
In March 1982, contamination appeared in well
17N1(M), which indicated that pumping of GO-13 was
not effectively containing the solvent plume (personal
communication from B. B. Roeder, President, Great
Oaks Water Company, to J. S. Vigil, Redevelopment
Agency, City of San Jose, California, July 16, 1982).
Concern that further spread of the plume could
contaminate downgradient municipal water supply
wells was addressed by the development of an offsite
groundwater extraction system. The system con-
sisted of redundant purge capabilities at four locations
between the Fairchild facility and GO-4(M).
The first line of extraction wells was in the vicinity of
GO-13 and included existing wells 17N1 and
17N11(M) plus newly constructed wells in the B and C
aquifers [RW-2(B) and RW-3(C)]. Because the TCA
concentrations measured at these wells were well
below the established discharge limits, treatment of
the purged water was not required.
The second line of extraction wells was located
approximately 500 feet downgradient of GO-13 along
San Ignacio Avenue between Via del Oro and Santa
Teresa Boulevard. Pumping wells in this location
included WCC-18(C) and RW-12(B).
The third line of extraction wells, which included WCC-
32(C) and RW-14(B), was located approximately 2200
feet downgradient of GO-13. The fourth and final
extraction capability was a single well, RW-13(B),
located approximately 500 feet upgradient of GO-4.
Pumping capacity of the offsite purge system totaled
5200 gal/min. Individual wells were selectively
pumped in an effort to control the migration of the
contaminant plume.
A.10.4.3 Shallow Aquifer Flushing
To accelerate the removal of solvents from the shallow
A aquifer and to reduce the need for offsite ground-
water recovery, Fairchild implemented a shallow
aquifer flushing program at the San Jose plant site.
The program, which ran from March 8, 1984, to
December 31, 1984, involved the injection of clean
water into three shallow A aquifer wells near the main
plant building and recovery of the recharge water at
several downgradient A and B aquifer extraction wells.
Flushing of solvents from the soil was achieved by the
movement of the water through the A and B aquifers.
Treated ground water from extraction wells RW-1, RW-
25(B), WCC-1(B), and WCC-2(B) was injected into
Wells WCC-41(A), RW-15(A), and RW-16(A) to
recharge the shallow aquifer. The injected water was
treated either by carbon adsorption or by air stripping.
The recharged ground water created an artificial
hydraulic mound around the recharge wells. As the
recharge water flowed away from the mound, it
transported solvents through the soil to the recovery
wells. Most of the recharge water was recovered from
the B aquifer.
The shallow aquifer flushing program operated for 10
months, during which time solvent concentrations in
both the A and B aquifers were significantly reduced
(Canonie Engineers 1985). The flow path of the
recharging water, however, proved to be difficult to
control because of the irregular bottom topography of
the A aquifer, and the operation was suspended.
A.10.4.4 Slurry Wall
Despite continued purge pumping of the A and B
aquifers and attempts to flush contaminants from the
soil, solvent concentrations in the ground water on
site remained high. Purging operations had de-
watered the A aquifer and had significantly lowered
the water table level in the B aquifer. Residual
contaminants trapped in the unsaturated soil were
gradually leaching into the ground water.
To address this continuing source of ground-water
contamination, Fairchild constructed a 3500-foot-long
perimeter slurry wall in the fall/winter of 1985 for
A-32
-------
complete enclosure and isolation of the 15-acre site.
The wall, which was constructed of a 3 percent soil-
bentonite slurry, extends completely below the A and
B aquifers and ranges in depth from 70 to 140 feet.
The 3-fpot-wide barrier has a very low permeability
(approximately 10-? cm/s) and is keyed into the clay
aquitard beneath the B aquifer. Ground-water
pumping is maintained at a rate sufficient to establish
an inward pressure gradient.
Offsite plume management is continuing via pumping
of the B and C aquifers. Several options are being
considered as final remedial measures within the
walled area. These options include resaturation/
ground-water purging, air purging, in situ bio-
degradation, and no action. In the interim, however,
the slurry wall will function as a hydraulic barrier to
prevent the continued migration of pollutants offsite.
A 70.5 Performance Evaluation
The status of onsite and offsite remedial efforts is
documented each month in progress reports to
Fairchild prepared by their technical consultants. The
soil excavation operation in October 1982 sub-
stantially reduced the amount of TCA in the onsite
soils; approximately 3000 yd" of soil containing TCA in
excess of 100 ppm was removed. By October 1983,
TCA concentrations in the A and B aquifers near the
source area had been reduced an average of 88
percent and 95 percent, respectively. Solvent
concentrations at offsite locations also have been
substantially reduced (internal memo of October 12,
1983, from P. W. Johnson, Toxics Division, to F. H.
Dierker, Executive Officer, California Regional Water
Quality Control Board, San Francisco Bay Region, in
response to Dr. D. Todd's September 19, 1983,
memo to B. Roeder). Although decontamination of
the Fairchild site is proceeding effectively, complete
onsite and offsite remediation may be several years
away.
A.10.6 Project Costs
Costs to date for remedial activities at the site,
including removal of the tank and the soil beneath it,
installation of onsite and offsite extraction wells, and
construction of the slurry wall, exceed $23 million
(CRWQCB 1985; internal memo to file regarding
Fairchild Camera and Instrument Corp.'s proposed
slurry wall from M. Kurtovich, Associate Engineer,
i California Regional Water Quality Control Board, San
Francisco Bay Region, July 17, 1985). Because final
cleanup levels have not yet been established, total
project costs cannot be estimated.
A.10.7 References
California Regional Water Quality Control Board
(CRWQCB), San Francisco Bay Region. 1985.
Executive Officer Summary Report, Item 21:
Fairchild, San Jose-Report on Proposal for In-
stallation of Slurry Wall for Pollutant Containment,
July 17.
California Regional Water Quality Control Board
(CRWQCB), San Francisco Bay Region. No date.
Order No. 82-61, NPDES No. CA0028185, Waste
Discharge Requirements for Fairchild Camera and
Instrument Corp., Bernal Road, San Jose, Santa
Clara County.
Canonie Engineers. 1985. Progress Report: Aqui-
fer "A" Flushing Program, March 8 through
December 31, 1984, Fairchild Plant, San Jose,
California.
Levine, J. D. 1981. Fact Sheet: Fairchild Semi-
conductor Waste Solvent Leakage From Under-
ground Storage Tank. California Regional Water
Quality Control Board, San Francisco Bay Region
December 16.
A-33
-------
-------
Index
Active vapor control, 5-92 to 5-94
Adsorption, 5-95, 5-96
Air stripping, 5-5 to 5-61
Aqueous waste treatment, 5-14
Barrier materials, 2-3
Biological treatment, 5-63 to 5-68
Biostimulation, 5-50, 5-52 to 5-55
Capillary zone, 3-8 to 3-10
Carbon absorption, 5-61 to 5-63
Case studies
Biocraft Laboratories, Waldwick, NJ, A-22
Bulk Fuel Storage and Distribution
Center, A-15
Chemical Pipeline, A-19
Fairchild Camera and Instrument Corp., South
San Jose, California, A-30
Gasoline Pipeline, Amber, Pennsylvania, A-3
Gasoline Pipeline, Glendale, California, A-1
Midwestern Laboratory Facility, A-17
Retail Gasoline Station, Genesee County,
Michigan, A-4
Retail Gasoline Station, Montgomery County,
Pennsylvania, 4-6
U.S. Coast Guard Air Station, Traverse City,
Michigan, A-10
Central water supplies
alternatives to, 5-11
treatment of, 5-112
Chemical treatment, 5-55 to 5-58
Comprehensive Environmental Response,
Compensation, and Liability Act (CERCLA), 1-1
Containment migration
through unsaturated (vadose) zone, 3-8,3-9
through saturated zone, 3-11
through capillary zone, 3-9, 3-10
transport in vapor phase, 3-11
Containment systems
comparison of, 2-6
Control water supplies
treatment of, 5-112
alternative supplies, 5-110, 5-111
Corrective actions (initial)
evaluation of release, 4-1
determination of substance, 4-2,4-3
site characterization 4-2,4-4 to 4-6
1-1
Corrective actions (permanent)
assessment of land use and potential health impact,4-6
human health and environmental factors affecting, 4-9
institutional (regulatory) factors affecting, 4-9
technical factors affecting, 4-9
Costs
Active vapor-control systems, 5-94
Air stripping, 5-60,5-61
Alternative central water supplies, 5-111
Alternative point-of-use water supplies, 5-113
Biological treatment, 5-67, 5-68
Biostimulation, 5-55
Capping, 5-105, 5-106
Carbon adsorption, 5-63
Chemical treatment, 5-57, 5-58
Completion of 2- to 4-inch diameter wells, 5-32
Deep-well injection, 5-15
Dissolved air flotation, 5-73
Diversion/collection systems, 5-101,5-102, 5-103
Dual-pump systems, 5-17
Flaring, 5-97
Floating-filter pumps, 5-18
Grading, 5-104
Granular media filtration, 5-75
Ground-water pumping, 5-30
Grouting, 5-47
Hydaulic barriers, 5-49
Interceptor drains, 5-43
Ion exchange/resin adsorption, 5-76,5-77
Landf arming, 5-11
Landfilling, 5-11
Manholes, 5-43
Neutralization, 5-79
Oxidation/reduction, 5-78
Packed-column aeration, 5-61
Passive collection systems, 5-92
Permanent corrective actions, 4-9
Physical treatment, 5-58
Pipe installation, 5-42
Precipitation/flocculation/sedimentation, 5-71, 5-72
Product recovery costs, 5-17
Revegetation, 5-108, 5-109, 5-110
Reverse osmosis, 5-82, 5-83, 5-84
. Selected pumps and accessories, 5-31
Seven recovery system cost scenarios, 5-33
Sheet piles, 5-48
Sludge dewatering, 5-88,5-89
Slurry walls, 5-46
-------
Costs (continued)
Soil flushing, 5-52
Soil washing, 5-12
Solidification/stabilization, 5-9,5-10
Steam stripping, 5-81
Subsurface drains, 5-38,5-39
Surface oil/water separators, 5-19
Thermal destruction, 5-14
Treatment of central water supplies, 5-112
Treatment of point-of-use water supplies, 5-113
Trench Excavation, 5-40,5-41
Ventilation structures, 5-96
Water and sewer lines
replacement, 5-113
cleaning/restoration, 5-115
Well screens and well points, 5-32
Deep well injection, 5-14,5-15
Dissolved air flotation, 5-72,5-73
Diversion/collection systems
dikes and berms, 5-98,5-99
channels and waterways, 5-99,5-100
chutes and downpipes, 5-99,5-100
sedimentation basins and ponds, 5-99,5-101
seepage basins and ditches, 5-99,5-100
terraces and benches, 5-99,5-100
Downhole turbine, 5-27
Drainage controls (see surface watef)
Dredging
mechanical, 5-7
hydraulic, 5-7,5-8
Dual-pump systems 5-16,5-17
Fiberglass-reinforced plastic (FRP) tanks
description of 2-2,2-5,2-6
Floating-filter pumps, 5-18
Flaring, 5-96, 5-97
Grading, 5-101,5-104
Granular media filtration, 5-73 to 5-75
Ground-water pumping, 5-20 to 5-22
Ground-water recovery
ground-water pumping (see above)
subsurface drains, 5-34
Grouting, 5-46,5-47
Hydraulic barriers, 5-48,5-49
Incineration technologies, 5-13, 5-14
In situ treatment (see biostimulation, chemical
treatment, physical treatment, soil flushing)
Interceptor drains, 5-36,5-43
Inventory control, 3-1
Ion exchange/resin adsorption, 5-75 to 5-77
Landf arming, 5-11
Landfilling, 5-9 to 5-11
Leaks
causes of, 2-5
Leak detection methods
inventory control, 3-1,3-2
monitoring, 3-7
nonvolumetric, 3-1, 3-2, 3-4
other, 3-5, 3-6
variables affecting accuracy, 3-7
volumetric, 3-1,3-2,3-3
Onsite and offsite treatment and disposal of
contaminants, 4-4, 5-8
Oxidation/reduction, 5-77, 5-78
Passive collection systems, 5-90 to 5-92
Physical treatment, 5-58
Piping systems
causes of pipe failure, 2-5
leaks, 2-5
Point-of-use water supplies
alternatives to, 5-111,5-112
treatment of, 5-112,5-113
Precipitation/flocculation/desimentation, 5-68 to 5-72
Pumping (see ground-water pumping)
Redox (see oxidation/reduction)
Rehabilitation of tanks, 5-4
Released substances
chemical and physical characteristics of, 4-2
Removal of tanks, 5-5
soil excavation, 5-6
sediment removal, 5-7
Resource Conservation and Recovery Act (RCRA)
Subtitle 1
definition of, 1-1
Reverse osmosis, 5-81 to 5-84
Risk evaluation, 4-9
analysis, 4-9
exposure assessments, 4-6
hazard evaluations, 4-9
risk assessments, 4-7
Sampling points (existing) 3-12, 3-13
inventory records, 3-12
surface water bodies, 3-13
underground structure, 3-12
water supply wells, 3-12
Sampling points (new), 3-13
accumulator devices, 3-13
downhole flux chambers, 3-13
ground probes, 3-13
soil borings, 3-15
surface flux chambers, 3-15
Saturated zone, 3-11
Sediment removal, 5-7
Sewer lines (see water lines)
Sheet piles, 5-47, 5-48
Site characterization, 4-2
1-2
-------
Site characterization (continued)
geography/topography, 4-4, 4-6
hydrogeology, 4-2,4-4
water and land use patterns, 4-4,4-6
Sludge dewatering
filter press dewatering, 5-84,5-85, 5-88,5-89
belt filter dewatering, 5-85, 5-86, 5-88, 5-89
centrifugal dewatering, 5-86,5-88
vacuum filtration, 5-84, 5-88, 5-89
Slurry trench, 5-44
Slurry walls, 5-39 to 5-46
Soil flushing, 5-50 to 5-52
Soil washing, 5-11,5-12
Solidification/stabilization, 5-8, 5-9, 5-10
Steam stripping, 5-79 to 5-81
Submersible pump, 5-25
Subsurface barriers (see grouting, hydraulic barriers,
sheet piles, slurry walls)
Subsurface drains, 5-34
location and spacing of, 5-35
filters and envelopes, 5-35
Surface oil/water separators, 5-18, 5-19
Surface water/drainage controls
capping, 5-104
diversion/collection systems, 5-98
grading, 5-101
revegetation, 5-105
Thermal destruction, 5-12 to 5-14
Underground storage tanks (USTs)
types, 2-1
substances stored, 4-2, 4-3
Unsaturated zone (see vadose zone)
Vadose zone, 3-8, 3-9
Vapor mitration control, collection, and treatment
active collection systems, 5-90
adsorption, 5-95
flaring 5-96
passive collection systems, 5-90
ventilation of structures, 5-94
Vapor phase, 3-11
Volumetric leak detection, 3-1, 3-2, 3-3
Waterlines/sewerlines
cleaning of, 5-114, 5-115
replacement of, 5-113
restoration after contamination, 5-114, 5-115
Well point systems, 5-22, 5-24, 5-25
Wells
drilled, 5-28 to 5-30
driven, 5-28,5-29
ejector, 5-27
extraction, 5-21
injection, 5-15, 5-21, 5-22
jetted, 5-28, 5-29
suction lift, 5-24
Well screens, 5-32
1-3
•ft U . S . GOVERNMENT PRINTING OFFICE] 1987-748-121/40695
-------
-------