United States
              Environmental Protection
              A.gervcy
            Hazardous Waste Engineering
            Research Laboratory
            Cincinnati OH 45268
             I   ' •
            Office of Solid Waste and
            Emergency Response
            Washington DC 20460
EPA/625/6-87-015
January 1987
              Technology Transfer
oEPA
Underground Storage
Tank Corrective
Action Technologies

-------

-------
                                            EPA  625/6-87/015
                                            January 1987
   Underground
                  Storage  Tank
Corrective  Action  Technologies
     Hazardous Waste Engineering Research Laboratory
          Office of Research! and Development
          U.S. Environmental; Protection Agency
               Cincinnati, Ohio 45268
    Office of Underground Storage Tanks
Office of Solid Waste and Emergency Response
   U.S. Envirnnmpmtal PmtanHnn A^n»~..
          U.S. Environmental
              Washington,
                 Protection Agency
                 D.C. 20460

-------
                                        Notice
This document has been reviewed in accordance with U.S. Environmental Protection Agency policy and
approved for publication.  Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.

-------
                                           i

                                           Abstract
 Several factors influence the decision maker's (selection of a path of corrective action for dealing with a
 leaking underground storage tank (UST).  This document provides information to assist in the scientific
 and technical  direction of the response efforts.   It probes general background information on UST
 construction techniques, leak detection methods,  and failure mechanisms.  It also addresses transport
 pathways of released substances, techniques for  evaluating the extent of a release, factors influencing
 risk to human  health and the environment, techniques for selecting initial corrective-action  response
 technologies, and detailed technical profiles of corrective-action technologies.  Emphasis is on corrective
 actions associated with releases from gasoline land petroleum USTs; however, profiles are also provided
_for technologies used in response to chemical releases.

 Responses to  UST releases generally involvej consideration of two  steps:  1) initial corrective actions
 (source control),  and 2) permanent corrective actions. The degree  of detail and amount of resources
 applied in these steps  depends on site-specifiq conditions, such as the volume of the release, the time
 frame within which the release occurred, hydrogeologic conditions, and the proximity to  environmentally
 sensitive communities and human receptors.

 The characterization and assessment of an UST release usually include the following:  ascertaining the
 release mechanism, determining the extent of release (volume and release rate), and evaluating transport
 pathways (air, surface water, and ground water).! Site characteristics are important in the assessment of the
 rate of transport to receptor communities, the  evaluation  of the  effects of  releases  on receptor
 f*r\rY\mt in it if\c> <*«*^ +h/-\ *•» v*/tit *«*:*«  ~.t *U.A.  «.xx ^ — *• -..c :_u.:_i	i  	_.-__..                     *  .
 communities, and the analysis of the effects
 populations.
of initial  and corrective-action responses on receptor
 The initial corrective actions to mitigate a leaking UST will almost certainly involve removal of the product
 from the tank and either tank repair or removal, lit also may be necessary to contain, treat, or remove any
 substance released to the environment.  The case studies contained in this document offer examples of
 the technologies often used to mitigate the effects of leaking USTs. The selection of more permanent
 corrective  measures requires a higher level of analysis and may  involve the use of more sophisticated
 technologies.  This document identifies and [describes technologies that have been  used during a
 permanent corrective action response. Technology selection requires a detailed analysis of site-specific
 conditions  and the establishment of cleanup goals and objectives.

 This report was submitted in partial fulfillment of Contract No. 68-02-3995 by PEI Associates, Inc., under
 the sponsorship of the U.S. Environmental  Protection Agency.  This report covers the period from
 January 1986 to December 1986, and work was completed as of December 1986.

-------
;V

-------
                                          Contents
Abstract

Figures

Tables

Acknowledgments

1.   Introduction

    1.1   Background
    1.2  Objective

2.   Underground Storage Tank Design

    2.1   Types of Underground  Storage Tanks
    2.2  Failure  Modes

3.   Leak Detection and Environmental Assessment

    3.1   Tank Monitoring
         3.1.1   Leak test technologies
         3.1.2   Leak detection variables
    3.2
Contaminant Migration (Transport)  Pathways
3.2.1   Unsaturated zone (Vadose Zone)
3.2.2  Capillary zone
3.2.3  Saturated zone
3.2.4  Vapor phase transport
    3.3  Initial Assessment of Extent of Release
         3.3.1   Existing sampling points
         3.3.2   New sampling points

4.  Corrective Action Response Process

    4.1  Initial Corrective-Action Options
         4.1.1   Evaluation of release
         4.1.2   Initial corrective actions
    4.2  Permanent Corrective-Action Options
         4.2.1  Assessment and investigation
                use and potential impact on public health

         4.2.2   Factors affecting selection of corrective actions

    4.3  Risk Analysis

                                          I
                                  to determine surrounding land
Page

   iii

   viii

   xi

   xiv

   1-1

   1-1
   1-2

   2-1

   2-1
   2-5

   3-1

   3-1
   3-1
   3-7

   3-8
   3-8
   3-9
   3-11
   3-11

   3-12
   3-12
   3-13

   4-1

   4-1
   4-1
   4-6

   4-6

   4-6

   4-6

   4-9

-------
5.  Technology Profiles

    5.1  Tank Removal, Abandonment, and Rehabilitation Removal/Excavation
          of Soil and Sediments
         5.1.1   Tank removal, abandonment, and rehabilitation
         5.1.2  Soil excavation
         5.1.3  Sediment removal

    5.2  Onsite and Offsite Treatment and Disposal of Contaminants
         5.2.1  Solidification/stabilization
         5.2.2  Landfilling
         5.2.3  Landfarming
         5.2.4  Soil washing
         5.2.5  Thermal destruction
         5.2.6  Aqueous waste treatment
         5.2.7  Deep-well injection

    5.3  Free Product Recovery
         5.3.1  Dual-pump systems
         5.3.2  Floating-filter pumps
         5.3.3  Surface oil/water separators

    5.4  Ground-Water Recovery Systems
         5.4.1  Ground-water pumping
         5.4.2  Subsurface drains

    5.5  Subsurface Barriers
         5.5.1  Slurry walls
         5.5.2  Grouting
         5.5.3  Sheet piles
         5.5.4   Hydraulic barriers

    5.6   In Situ Treatment
         5.6.1   Soil flushing
         5.6.2   Biostimulation
         5.6.3   Chemical treatment
         5.6.4   Physical treatment

     5.7  Ground-Water Treatment
          5.7.1   Air stripping
          5.7.2   Carbon adsorption
          5.7.3   Biological treatment
          5.7.4  Precipitation/flocculation/sedimentation
          5.7.5  Dissolved air flotation
          5.7.6  Granular media filtration
          5.7.7  Ion exchange/resin adsorption
          5.7.8  Oxidation/reduction
          5.7.9  Neutralization
          5.7.10 Steam stripping
          5.7.11 Reverse osmosis
          5.7.12 Sludge dewatering

     5.8 Vapor Migration Control, Collection, and Treatment
          5.8.1  Passive collection systems
          5.8.2  Active vapor control
          5.8.3  Ventilation of structures
          5.8.4  Adsorption
          5.8.5  Flaring
                                            vi
                                                                                5-1
5-4
5-4
5-6
5-7

5-8
I5-8
5-9
5-11
5-11
5-12
5-14
5-14

5-15
5-16
5-18
5-18

5-19
 5-20
 5-34

 5-39
 5-39
 5-46
 5-47
 5-48

 5-50
 5-50
, 5-52
 5-55
 5-58

 5-58
 5-59
 5-61
 5-63
 5-68
 5-72
 5-73
 5-75
 5-77
 5-78
 5-79
 5-81
 5-84

  5-90
  5-90
  5-92
  5-95
  5-95
  5-96

-------
    5.9  Surface Water/Drainage Controls
         5.9.1   Diversion/collection systems
         5.9.2   Grading
         5.9.3   Capping
         5.9.4   Revegetation

    5.10 Restoration of Contaminated Water Supplies and Utility/Sewer Lines
         5.10.1  Alternative central water supplies
         5.10.2  Alternative point-of-use water supplies
         5.10.3  Treatment of central water supplies
         5.10.4  Treatment of point-of-use water supplies
         5.10.5  Replacement of water and sewer lines
         5.10.6  Cleaning/restoration of waterlines and sewerlines

6.  Reference Matrix for Case Histories

    6.1   Purpose of Case Histories
    6.2  Case History Matrix

References

Appendix Case Histories

Index
5-97
5-98
5-101
5-104
5-105

5-110
5-110
5-111
5-112
5-112
5-113
5-114

6-1

6-1
6-2

R-1

A-1

1-1
                                          vil

-------
Figures
Number
2-1
2-2
2-3
2-4
2-5
3-1
3-2
3-3
3-4
3-5
4-1
4-2
4-3
4-4
4-5
5-1
5-2
5-3
5-4


Isometric of three tanks connected to one or more pumps
Typical anchoring systems where subsurface water
conditions exist
Components for tank installations with suction pumping
Components for tank installations with submerged pumping
Piping arrangement of external pump
Diagram of typical subsurface environment of USTs
Shape of hydrocarbon plume in vadose zone under different
leakage rates and physical conditions
Typical flow pattern for hydrocarbons in the capillary zone
Typical behavior of hydrocarbons in soil following a sudden
high-volume release
Behavior of hydrocarbons after release has stabilized
Typical initial corrective-action process
Site information needs for evaluation of corrective-action alternatives
Potential initial response situations and associated corrective actions
Permanent corrective-action process
Matrix of permanent corrective action categories for specific
site problems
Flow chart of activities involved in the initial response to an UST
release
Flow chart of permanent corrective action in response to an UST
release
Potentially applicable corrective action technologies relative to
release volume and chemical characteristics
Flow diagram for EPA mobile soil washer
viii
Page
2-1
2-3
2-4
2-4
2-5
3-8
3-9
3-10
3-10
3-1 1
4-1
4-4 ;
4-7
4-7
4-8
5-1
5-2
5-3
5-12


-------
1
1 5-5
1
1 5-6
1 : 5-7
1 5-8
1: 5-9
1 5-10
1 5-11
1 : 5-12
1 ! 5-13
1 ; 5-14
• |
1 ; 5-15
1 : 5-16
1 ; 5-17
; 5-18
5-19
5-20
5-21
5-22
: 5-23
5-24
i 5-25
5-26
; 5-27
; 5-28
5-29
5-30
5-31
!
.
Diagram of a conventional industr
sandstone
Dual-pump recovery system
Floating-filter recovery system
Oil/water separator
Containment by the use of extraci

al injection well completed in
5-15
5-16
5-18
5-19
ion wells 5-20
Extraction and injection well patterns for plume removal 5-21
Plume diversion using injection wells 5-22
Suction wells
5-24
Multiple suction lift wells (Well point system) 5-25
Submersible pump
Downhole turbine
, Deep well jet pump
Variations of ejector wells
Single-pipe ejector
Types of wells
Subsurface drainage system corr
Use of a one-sided subsurface dr;
uncontaminated sources
The effect of relief and interceptoi
configuration of the water table
Subsurface drain with a lift station
5-25
5-26
5-26
5-27
5-27
5-28
ponents 5-34
lin for reducing flow from
5-35
drains in altering the
5-36
5-36
Typical design of an automatic drainage pumping plant 5-37
Slurry trench construction
Slurry wall placement
Semicircular grout curtain
Typical steel piling shapes and int
5-44
5-45
5-47
srlocks 5-49
Soil flushing system 5-50
| - . -
Biostimulation of soil and ground v
Air-stripping equipment configura
.
/ater 5-52
ions 5-60
x

-------
5-32
5-33
5-34
5-35
5-36
5-37
5-38
5-39
5-40
5-41
5-42
5-43
5-44
5-45
5-46
5-47
5-48
5-49
5-50
5-51
5-52
5-53
5-54
5-55
5-56
5-57
5-58
6-1

Annual cost for air stripping
Two-vessel granular carbon adsorption system
Estimated costs of various sizes of activated carbon adsorption units
Representative configuration employing precipitation, flocculation,
and sedimentation
Solubility of metal hydroxides and sulf ides
Representative types of sedimentation
Capital and annual O&M costs for flocculation
Dissolved air flotation system
Capital and annual O&M costs for dissolved air flotation
Example granular media filtration bed ,
Pertinent features of ion exchange systems
Construction and annual O&M costs for ion exchange
Construction and annual O&M costs for chemical oxidation and
reduction
Construction and annual O&M costs for neutralization
Steam stripping system
Capital and annual O&M costs for steam stripping
Reverse osmosis membrane module configurations
Concentration and annual O&M costs for reverse osmosis
Drum-type rotary vacuum filter
Filter press plate
Belt filter press
Types of centrifuges
Capital and annual O&M costs for vacuum filters and filter presses
Capital and annual O&M costs for belt presses and centrifuges
High-permeability passive vapor control system
Low permeability passive vapor control system
Active vapor control system
Case history matrix
X
5-61
5-62
5-63
5-68
5-68
5-70
5-71
5-72
5-73
5-74
5-76
5-77
5-78
5-79 !
5-80 :
5-81
5-83
5-84
5-84
5-85
5-86
5-87
5-89
5-89 :
5-91
5-92
5-93 ;
6-:2


-------
Tables
Number
2-1
2-2
3-1
3-2
3-3
3-4
3-5
3-6
4-1
4-2
5-1
5-2
5-3
5-4
5-5
5-6
5-7
5-8
5-9
5-10
5-11
5-12
5-13

-
Underground tank installation pr<

ictices
Comparison of containment systems for UST releases
Summary of volumetric leak detection methods
Summary of nonvolumetric leak defection methods
Summary of other leak detection methods
Variables affecting accuracy of leak detection
Typical values for Rv
Water monitoring techniques
Petroleum products stored in un
Selected chemicals (462) stored
Soil excavation costs
Solidification/stabilization procei
Solidification costs
Summary comparison of relative
solidification alternatives
Summary of commonly used inc
Approximate costs for hazardous
Product recovery costs
Criteria for well selection


derground storage tanks
in underground storage tanks

ses

1986 costs of stabilization/
ineration technologies
> waste incineration
.


Common methods of well installation
1986 costs for selected pumps and accessories
^'
ellpoints
1 986 costs tor completion ot 2- to 4-inch diameter wells
Summary of seven recovery sys

em cost scenarios
xl
Page
2-2
2-6
3-2
3-4
3-5
3-7
3-9
3-14
4-2
4-3
5-7
5-9
5-9
5-10
5-13
5-14
5-17
5-22
5-29
5-31
5-32
5-32
5-33


-------
5-14
5-15
5-16
5-17
5-18
5-19
5-20
5-21
5-22
5-23
5-24
5-25
5-26
5-27
5-28
5-29
5-30
5-31
5-32
5-33
5-34
5-35
5-36
5-37
5-38
5-39
1986 unit costs for trench excavation and associated activities
1 986 unit costs for pipe installation
1986 installed costs for manholes
Capital costs for interceptor drain installation
Costs of common grouts
Applicability of soil flushing techniques
Estimated costs for hypothetical bioreclamations with hydrogen
peroxide as an oxygen source
Summary of in situ treatment methods
Estimated costs for in situ physical treatment methods
Applicability of ground-water treatment processes
Packed-column air stripping of volatile organics
Costs for packed-column aeration
General cost data for various sizes of activated-carbon
adsorption units
Summary of applications/limitations of biological treatment process
Concentration of pollutants that make prebiological or primary
treatment advisable
General cost data for various sizes of activated sludge treatment units
General cost data for various sizes of ion exchange units
Results of pilot scale testing of a reverse osmosis unit
1986 unit costs for components of passive vapor-control systems
Materials and equipment for active vapor-control systems
1986 unit costs for components of active vapor-control systems
1986 capital costs of elevated flares (for vapor with heat
content of 60 Btu/ft3)
Supplemental fuel requirements for elevated flares
Summary of surface water/drainage controls
1986 unit costs associated with surface water diversion and
collection structures
1986 unit costs associated with grading
5-40
5-42
5-43
5-43
5-47
5-51
5-55
5-56
5-58
5-59
5-59
5-61
5-63
5-65
5-66
5-67
5-77
5-81
5-92
5-94
5-94
5-97
5-97
5-98
5-102
5-104
XII

-------
5-40         1986 unit costs associated with capping
5-41         Important characteristics of grasses and legumes
5-42         1986 unit costs assorted with revegetation
5-43         1986 unit costs for replacement of water and sewer lines
5-106
5-107
5-109
5-113
                                        xfii

-------
                                   Acknowledgment
This document was  prepared for EPA's  Office of  Research and  Development, Hazardous Waste
Engineering Research Laboratory, in partial fulfillment of contract No. 68-02-3995.  Mr. Douglas Ammon
served as  the EPA Technical Monitor and directed the technical efforts of the project. Technical
assistance was supplied by Mr. John Hefflefinger, Office of Underground Storage Tanks, Mr. Jack Farlow
and David Berg, Office of Research and Development, and Mr. William Foskett, consultant.  Clarence
demons, Dale Dietrich,  and Norm Kulujian, Center for Environmental  Research Information, are
acknowledged for their technical and editorial assistance. Mr. Jack Greber served as PEI Associates
Project Director and Mr. Bob Cochran served as PEI Associates Project Manager. Other major contributors
include Ms. Judy McArdle, Mr. Fred Hall, Mr. Bill Trapp, Ms. Barb Locke, Mr. Doug Bailey, Ms. Catherine d.
C. Hartman, Mr. Don Henz, Mr. Bill Thompson, and Ms. Jan Zieleniewski. We are particularly grateful for the
substantial contributions of the Technical Editor, Ms. Marty Phillips, and of the Copy Editor, Mr. Jerry Day.
Great appreciation is extended to the following peer reviewers and contributors:
     Dr. David Chin
     Dr. Rudy White
     Mr. Jeffrey Reame
     Ms. Anne Mason
     Dr. Jay Lehr
     Mr. Hari Bindal
     Mr. Jeff van Ee
     Mr. Jerry Jones
     Mr. Kevin Smolenyak
                           American Petroleum Institute
                           American Petroleum Institute
                           Chemical Manufacturers Association
                           Chemical Manufacturers Association
                           National Water Well Association
                           Department of the Air Force
                           U.S. Environmental Protection Agency
                           U.S. Environmental Protection Agency
                           Clark Engineers and Scientists
We would also like to acknowledge the contributions of the PEI graphics and typing staff, Ms. Marty
Gillaspy, Ms. Stephanie Gresham, and Ms. Connie Hutchinson, for their contribution in producing the
document.
                                         XIV

-------
                                             Section 1
                                           Introduction
1.1  Background
Somewhere between three  and five  million under-
ground tanks in the United States are used to store
liquid petroleum and chemical substances.  Estimates
indicate that 100,000 to 400,000 of these tanks and
their associated piping systems may be leaking.  The
large percentage of these underground storage tanks
that contain petroleum products are of particular con-
cern. Recent studies also indicate that, if these leaks
are left unattended, their number will likely increase,
and  releases from these sources can result in the
contamination  of  subsurface  soils, the migration  of
toxic  or explosive vapors,  and the contamination  of
ground- and  surface-water systems. Environmental
and tank monitoring, tank replacement, and corrective
action technologies can reduce the potential of future
releases or remedy the effects of prior releases from
underground storage tanks.
Industry sources  estimate that the average cost  of
tank cleanup at a site is about $70,000; however, if
tank removal  and treatment of surrounding soils are
required, these costs can total more than $1 million.
Should  underground  tank  leakage impact  ground
water, surface water, drinking water supplies, reser-
voirs, sewers, or utility  trenches, the costs of cor-
rective action can quickly reach well beyond the $1-
million range.  The extent of resources required for
corrective action depends on the magnitude of re-
lease, the toxicity of the substance, and the agreed-
upon level or objective of the proposed cleanup.
Concern for the status of underground storage tanks
has prompted  the development of legislation regard-
ing the registration and monitoring of existing tanks,
design and installation of new tanks, and  corrective
actions for releases.  The 1984 Hazardous and Solid
Waste Amendments (HSWA) to the Resource Conser-
vation and Recovery Act (RCRA) added a new Subtitle
I, "Regulation of Underground Storage Tanks."  Sub-
title  I requires the  U.S.  Environmental  Protection
Agency (EPA) to develop a comprehensive program
for regulating certain underground storage tanks that
contain  regulated substances. Section 9001(1) de-
fines an "underground storage tank" as:

      "...any one  or combination of tanks  (including
      underground pipes connected thereto) which is
      used to  contain an accumulation of regulated
                                                1-1
      substances, and the volume of which (including
      the  volume  of  the underground  pipes con-
      nected thereto) is ten percent or more beneath
      the surface of the ground."
The definition does not include noncommercial farm
or residential motor fuel tanks with capacities of less
than  1100 gallons, heating  oil tanks, septic  tanks,
flpwthrough process tanks, and gas and oil production
lines.
Section 9001 (2) defines "regulated substances" as:

      "(A) any substance defined in Section 101(4) of
      the Comprehensive Environmental Response,
      Compensation, and Liability Act of  1980 (but
      not including any substances regulated as a
      hazardous waste under Subtitle C), and

      (B) petroleum, including crude oil or any fraction
      thereof which is liquid at  standard conditions of
      temperature  and  pressure  (60  degrees
      Fahrenheit and 14.7 pounds  per  square inch
      absolute)."
In Subtitle I, Congress requires  the EPA Administrator
to promulgate underground storage  tank regulations
"as may be necessary to protect human health and the
environment"  [Section 9003(a)]. , The  EPA  is  re-
sponsible  for  developing "requirements for  taking
corrective  action  response  to a release from  an
underground storage tank" [Section 9003(c)(4)] and
"requirements for reporting of releases and corrective
action taken  in  response to  a release from  an
underground  storage  tank"   [Section  9003(c)(3)].
Section 9001(5) defines  "release"  as "any spilling,
leaking, emitting, discharging, escaping,  leaching, or
disposing  from an underground storage  tank into
ground water, surface water,  or subsurface  soils."
"Corrective action" is not defined  in Subtitle  I, al-
though the term is used  to describe actions carried
out in response to releases from underground stor-
age tanks.

As of December 1986, EPA is evaluating proposed
regulations for  underground storage tanks.   These
proposed  regulations  will require that owners and
operators take corrective  actions to remedy releases
from underground storage tanks. They will most likely
require immediate corrective action to reduce fire and
explosion  hazards and to recover free product and

-------
remove contaminated soils. In addition, they probably
will require that these immediate responses be fol-
lowed  by longer-term corrective actions when  such
actions are necessary to address ground-water con-
tamination.

1.2 Objective
The EPA's newly formed  Office  of Underground
Storage Tanks (OUST), which is part of the Office of
Solid Waste  and  Emergency Response, is respon-
sible for establishing the Agency's regulatory program
for  controlling underground storage  tanks.    The
Hazardous Waste Engineering Research Laboratory
(HWERL) of EPA's Office of Research and Devel-
opment is responsible for providing engineering and
scientific support to OUST. The objective of this effort
is to establish a forum for technology transfer among
the various  regulatory  agencies  and  the  user
community.   To this  end,  this document contains
information for determination of the state  of the art,
applicability,  cost,  and  effectiveness  of  alternative
corrective actions  for leaking  underground storage
tanks (USTs).
This report presents  profiles on technologies cur-
rently being used to correct  leaking USTs. As part of
this study, technologies or practices used in  other
environmental remediation programs (e.g., Superfund
 Remedial Actions) were analyzed to  determine their
cross-over potential for UST corrective actions.
 Focus in this document is on the problems associated
with underground storage tanks containing gasoline
 and petroleum products because 1) most of the USTs
 in the United States are designed and used for storing
 gasoline and petroleum  products, and 2) technical
 information  on  releases  of  chemicals  listed  as
 Comprehensive Environmental  Response, Compen-
 sation,  and   Liability   Act  (CERCLA)  hazardous
 substances   is  available  through   the  Superfund
 program technical literature.
Section 2  describes  the design  and construction
materials of typical USTs. It contains a brief summary
of the types of  storage tanks  and piping  systems
currently in  use  and the most  commonly occurring
tank failure modes and release events.
Section 3 describes methods and  practices used to
assess the integrity of tank systems and to determine
the extent of any releases.  The  focus  is on tank
testing devices and methods for the speedy assess-
ment of the extent of release and the determination of
immediate environmental and public health impacts.
The contaminant's fate  and transport pathways are
also summarized.

Section 4 presents methodologies for possible use in
selecting  the appropriate corrective-action-response
alternative.  The brief  guidance on: selection tech-
niques  concerns  those  applicable  to initial  response
technologies (rapid response)  and long-term correc-
tive action remedies. The section also presents infor-
mation on  the types  of site characterization  data
necessary for the selection, design, and implemen-
tation of permanent remedies.  Data also include the
chemical and physical  properties of release  sub-
stances and their impact on the selection of corrective-
action technologies.

Section 5  contains  a  detailed discussion  of  the
technologies surveyed,  including descriptions of ap-
plication and availability, design and construction con-
siderations,  ranges of  unit costs,  and  additional
reference materials.

Section 6 summarizes  case history information.   It
includes a  matrix that cross-references case history
 information with  items  critical to  the  selection  and
 implementation of corrective-action technologies. Ac-
tual case histories are presented in the Appendix.
                                                  1-2

-------
                                             Section 2
                              Underground Storage Tank Design
The intent of this section is to familiarize the reader
with the design features of underground storage sys-
tems. This information will help to establish a base for
examining these systems in the field.

The configurations of the thousands of underground
storage tanks in this country vary to suit several con-
straints, including the geography of the site, the type
of material  stored, insurance  underwriter require-
ments, and the owner's operation.  Regardless of the
original design criteria,  each installation will probably
have these basic components:

   One or more tanks

   Antiflotation anchorage (in regions having high
   ground-water tables)

   Piping system

   Pumps

   Means for level gaging

   System for corrosion protection (if tanks are metal)

Tanks  are  usually placed  underground to  store
materials that are classed as hazardous because of
their flammability or combustibility.  Many tanks were
installed underground on the premise that they would
never  develop  leaks,  and little consideration  was
given to the consequences if  any leakage should
develop.

2.1  Types of Underground Storage
Tanks

Two broad  categories of underground tank appli-
cations are gas  station  and  industrial/commercial
installations.  Underground tanks in gasoline stations
are used for storage of the following products:

•   Gasoline-leaded, unleaded, and premium grades

    Diesel oil

    Waste oil (may contain some gasoline)

Most  gasoline  stations  have paved  service areas;
therefore,  many of these tanks are located under
pavement.  Depending on the station's business vol-
ume, multiple tank installations (as shown in Figure 2-
1) are often used.   In gasoline stations, the pump is
usually part of the product meter/dispenser located on
the service  island  (commonly  referred to  as  "the
pump"); however, submerged pumps are also used.
       Figure 2-1.     Isometric of three tanks connected to one or more pumps (API 1979).
                   Double Swing Joint
                                                                Union (Typical)
             Extractor or Angle
             Check Valve
                                                                                 Suction Line


                                                                       •<	1— Double Swing Joint

-------
Industrial/commercial underground  tank installations
store a wide  variety of materials, including solvents
and various hydrocarbons.  Both new materials and
waste products  are often stored  underground be-
cause their  contents  are hazardous or  explosive.
These installations  usually have individual tanks for
each stored material rather than multiple tanks con-
nected by common pump suction piping.  Piping for
industrial storage installations is not always located
below grade; instead, it is sometimes run on overhead
racks. In commercial installations such as dry cleaning
establishments,  system  piping  into  the  building  is
generally below grade.

Underground  tanks  vary in size,  shape, and materials
of construction.  Metal tanks are usually welded and
have some  kind of exterior coating for protection
against corrosion.   Tanks fabricated from fiberglass,
epoxy, or other nonmetallic material, which are com-
mon in   newer  installations, generally  require  no
coating for corrosion resistance. Newer installations
also might have tanks with double-wall construction.
The  annular  space between the  walls  is  used  in
   various ways to detect product leaks from the inner
   tank wall or ground-water leaks into the outer tank wall.

   Proper installation of underground storage tanks  is
   crucial to their remaining leak-free.  Most major tank
   manufacturers warrant their tanks against failure only if
   they  are  installed and used in accordance with the
   manufacturer's instructions.  Therefore, USTs should
   be installed in  strict conformance  with  the manu-
   facturer's recommendations.

   Fiberglass and plastic tanks are more fragile than steel
   tanks, and thus  require more careful handling during
   installation. Also, fiberglass and plastic tanks receive a
   substantial portion (as much as 90%) of their structural
   support from backfill and bedding materials; hence,
   proper installation is extremely important if tank failures
   are to be avoided.  Table 2-1 presents a summary of
   installation  practices for fiberglass-reinforced plastic
   (FRP) and steel tanks.
   Depth  of tank  burial varies,  but  the  tank top  is
   generally not  more than one tank diameter  below
   finished grade.  At some installations, tanks are placed
Table 2-1.   Underground Tank Installation Practices
            Item
                                                         Minimum Recommended Dimension, in.t
FRP (Capacity
<20,000 gal)
FRP (Capacity
>20,000 gal)
                                                                                      Steel
            Distance between adjacent tanks               18

             Distance between tank and excavation          24
             sidowall

             Thickness of compacted bedding               12

             Top slab extension beyond tank               18

             Maximum burial depth                       84
             Top slab thickness assuming 34,000 Ib
             per axle load

               Reinforced concrete plus compacted          24
               backfill

               Asphaltic concrete plus compacted           36
               backfill

             Top slab thickness assuming no traffic

               Reinforced concrete plus compacted          16
               backfill

             Anchor slab extension beyond tank (if           18
             required because of buoyancy)

             Anchor slab thickness (if required              8
             because of buoyancy)
                 24

                 24


                 12

                 24

                 84
                  42


                  48





                  30


                  24


                  8
                 12

                 24


                 6

                 12

                 1.5 tank
                 diameters
                  24


                  26




                  16


                  18


                  8
             ' Data from AP11979; NFPA 1981.
             t Except as noted.
                                                    2-2

-------
in mounds and only partially buried below grade;
however, they usually have the same components as
underground tanks.  The ground above underground
tanks is usually paved if any traffic will occur.  Man-
holes, caps, or other hardware are provided to cover
and  protect  tank  appurtenances  such as the fill
connections and gage pipes.  Some large  tanks are
equipped with a manhole to allow access into the tank.

Unless the water table  is well below the  tank ex-
cavation, underground tanks have an anchor system
of some sort to prevent them from floating upward and
out of the ground.   Figure 2-2 shows diagrams of a
screw or expandable type anchor, a deadman anchor,
and a concrete slab anchor.

Most  tanks are filled and  emptied through   con-
nections on the top side of the tank.  Because the
geometry of underground tank installations  precludes
liquid transfer by gravity, pumps are required. These
pumps will either be submerged in  the tank or be
located above ground at a spot remote to the tank to
supply material from the tank to the user (Figures 2-3
and  2-4).   In these arrangements, one submerged
pump is required for each  tank, whereas an external
pump can draw from more than one tank, as shown in
Figure 2-5. Installations rarely  have  supply pumps
connected  to the  tank  bottoms  because  such an
arrangement  has the potential for complete loss of
tank contents if the pump suction piping should fail.
Underground  storage  tanks  generally  include  a
method for tracking tank  inventory, such as the use of
a level gage.  The most common level gage for under-
ground tanks is a  metering  stick inserted in the fill
tube. More elaborate systems might include bubbler-
type pneumatic level sensors or a differential pressure
instrument, which requires  connections to the top and
bottom of the tank.
Figure 2-2.     Typical anchoring systems where subsirface water conditions exist (API 1979).
     For metal tanks, systems for protection against cor-
     rosion are necessary to prevent piping or tank failures
     (leaks).  Cathodic protection is commonly used. Other
     protection systems  might  be nothing more  than a
     coating  of mastic on the tank and the wrapping or
     taping of the underground piping.

     Containment systems for  releases from USTs  rep-
     resent the second line of defense against  the prop-
     agation  of soil  or ground-water contamination.  They
     also enhance the effectiveness of early-warning leak-
     monitoring systems  by confining the release prior to
     its detection. The control technology used to contain
     underground  releases consists  of  establishing  a
     barrier around the storage tank to prevent the escape
     of the leaked liquid from the immediate area..   The
     following barrier materials are used for containment:

         Liner with low soil permeability
         Synthetic membrane  liners
         Soil cement or bentonites
         Concrete vaults

         Double walled tanks

     The selection of the  proper containment material for a
     particular application depends  upon several factors,
     including the type  of material being stored,  local
     environmental  conditions, and  legislative require-
     ments.  The containment system should be sloped to
     a sump, from which a sample can be taken for analysis.
     It also  should  have a siphoning or pumping system
     capable of removing the liquids contained  within the
     system. Table 2-2 presents a comparison of contain-
     ment systems for UST releases.
                                                                                      Grade
    Sand or Gravel (Typical)

    Cable or Strap (Typical)
    Firm Soil •
                      a) Screw or Expandable
                        Type Anchor
b) (Concrete Bell (or
  Deadman) Anchor
                                                                                              12 in.
c) Concrete Slab •
  Anchor
                                                2-3

-------
Figure 2-3.     Component:
                       for tank Installations with suction pumping (AP11979).


             -Suction Lines to Pump Islands        .Manhole                                       Vent Line to Appropriate
              (Slope to Tanks)                  /  Rl| Cg     Clay Tile                           Location (Slope to Tank)

                      , Reinforced Concrete Slab » / |'|   *^/	3L         / Double Swing Joint


             *^^^^^^                          !   '
             Existing      />
             Soil         />
             (Undisturbed)  ^18 in.
                          M—
                            Min.
                      ///        Kft.
                     W/sss?-
                      " PEA Gravel for Nonmetallic
                       Tanks (Or Mfr.-Approved
                       Alternative)
                                                      Note: Eliminate concrete
                                                           slab in nontraffic areas
                                                           and construct concete
                                                           (only) around manholes.
                                                                                                     uion ^oiope u
                                                                                                     a



-J — ' V "il^.
Plug
/Suction

.^


Gasoline
Storage
Tank
"*"Tube
6 in. Max.

. Suction
PTube

                                                                                            Overfill Prevention
                                                                                            Float Fill Valve
                                                                                           S      Sand or Gravel*
                                                                                         18 in.
                                                                                                 '£
                                              Piping Details For Tanks Under Concrete
 Figure 2-4.    Components for tank Installations with submerged pumping (AP11979).


                                                        .Manhole
                 Reinforced
                 Concrete Slab
                                        Manhole
       Leak
       Detector
Supply Line to
Pump Islands
ll*/xg^^il
                                                                                                   Vent Line to Appropriate
                                                                                                   Location (Slope to Tank)
                                                                                                                     'T
                                                                                                                       6 in.
                                                                                                     Overfill Prevention
                                                                                                     Float Fill Valve
                                                                                                           Sand or Gravel*
                           ' PEA Gravel for Nonmetallic
                            Tanks (or Mfr.-Approved
                            Alternative)
                                                           Note: Eliminate concrete
                                                                slab in nontraffic areas
                                                                and construct concete
                                                                (only) around manholes.
                                                                The extr. angle check riser
                                                                is not required with earth cover.
                                             Piping Details For Tanks  Under Construction
                                                              2-4

-------
Figure 2-5.
  Pitch 1/8 in
   per foot
Swing
 Joint
 Piping arrangement of external pump
 (AP11979).
 Angle or
 Extractor
Check Valve,
Suction
 Line


Swing
 Joint
 Union
(Typical)

Keep Stubs
 at Same
 Elevation
2.2  Failure Modes
Leaks from underground storage tank systems  (i.e.,
the storage tank and associated piping and pumping)
can result from corrosion; from system rupture due to
overloading, external stresses, or puncture; and from
faulty construction and installation.  Overfilling is an-
other potential source of a product release from a tank
system.

The most common failure mode of underground  tank
systems is corrosion of the tank or piping. Corrosion
may be traced to failure of the corrosion protection
system  due  to pinholes  in the coating or  taping
system, depletion of the sacrificial anodes, corrosion
from the inside due to the stored product, and various
other reasons/Many corrosion-related leaks are found
in systems that have no corrosion protection at all;
these are systems that were installed in the  1960's
and 1970's and are  now approaching the age when
failure is immminent.  Tanks having corrosion protec-
tion  also  can corrode,  however,  if  the protective
coating  is damaged  during  installation,  sacrificial
anodes are not replaced when required, the current is
switched  off in impressed  current systems,  or the
protection system is not designed properly for the soil
conditions and stored liquids.

Fiberglass-reinforced plastic (FRP)  tanks and piping
are inherently corrosion-resistant, but this material of
construction has lower structural strength than steel
and requires extra care during installation and oper-
ation.  Reinforced ribs are usually incorporated into
their design to withstand both the internal  stresses
from the stored liquid and the external stresses from
backfill.  Some resins used for FRP tank construction
may lose structural strength when exposed to certain
chemicals; others may dissolve, soften, or become
brittle in acidic or saline soil environments.  Failure to
observe  the design  limitations and to  follow the
handling and installation requirements for FRP tanks
can lead to tank failure and the resultant leaks.

A  substantial percentage  of  leaks  in  underground
storage systems occur in piping systems, probably
because the threaded joints used in most of these
systems are vulnerable to corrosion from outside if
adequate corrosion protection is not provided. Piping
joints also may  leak  because  of improper  sealing
during installation or loosening due to vibration, tem-
perature cycling, and the like.  According to practices
recommended by the State of New York (no date), the
principal causes of pipe failure are as follows:

    Corrosion

    Settlement of backfill

    Frost heave

•   Traffic loadings on poorly protected piping

    Stresses resulting from changes in temperature

Overfilling is another source of releases  (spills) from
tanks. Spills of this type can be caused by human
error, failure of the delivery source's shutoff  valves,
and failure of the tank level indicator.  These spills are
generally small (less than a few gallons), but they have
the potential of becoming much larger  if the equip-
ment is left unattended during tank filling or if shutoff
valves fail.

The   Pennsylvania  Department  of  Environmental
Resources (PADER), through investigations of  leak-
ing underground storage systems, categorized leaks
as 49 percent from tanks, 39 percent from piping, and
12 percent  from both tanks and  piping  (Water In-
formation Center, Inc. 1986). They further report that
53 percent  of  the piping leaks occurred within 10
years of tank installation and 77 percent within the first
15 years.  In contrast, 71 percent of the tank leaks
occurred more than 10 years after installation and 46
percent after 15 years.   According  to  PADER, the
three major causes of leaks were  corrosion (62  per-
cent), loose fittings (19 percent), and improper  tank
installation (13  percent).  They also found a prepon-
derance of systems that had no cathodic protection
whatsoever.
                                                 2-5

-------
Table 2-2.   Comparison of Containment Systems for UST Releases
      Type of System
Advantages
Disadvantages
                                                                                                        Relative Cost
       Clay Liners
       Polymeric Liners
       Soil Cement
       Bentonite
       Concrete Vaults
        Double-Walled Tanks
If available close to the site,
day provides the least
expensive liner; use of clay
is a well-established practice
and standard testing
procedures are available
Provide well-established
solution to problem of
containing petroleum
products; particularly good
for temporary storage;
high resistance to bacterial
deterioration
 Is durable; resistant to
 aging and weathering
 Has low permeability;
 does not deteriorate with
 age; is self-sealing
 Has good strength and
 is durable
 Constructed of material
 (FRP or coated steel) that
 is resistant to the stored
 product and to external
 corrosion; tank design
 includes leak-detection
 system
Subject to drying and cracking;          Low
therefore, must be protected
with soil cover; subject to
leaching of components when              :
exposed to ground water or
other solutions; subject to ion
exchange when exosed to
water containing acids, alkalis,
or dissolved salts; subject to                ,
destabilization when exposed
to some organic solvents

Require subgrade preparation           Moderate to
and sterilization to reduce               high
risk of puncture; must be
protected for damage,
particularly that due to
vehicular traffic; must be
protected from sunlight and
ozone; may be attached by
hydrocarbon solvents, partic-
ularly those with high
aromatic content; good oil
resistance and good low-
temperature properties do not
normally go hand in hand

Subject to degradation due to           Moderate
frost heaving of subgrade; in-
place soil normally used;
permeability varies with the
type of soil

Untreated bentonite may                Moderate
deteriorate when exposed to
contaminant; requires
protective soil cover, typically
 18 inches; subject to
destabilization when exposed              ',
to some organic solvents

 Requires surface coating to             High
 ensure impermeability; subject
 to cracking when exposed to
 freeze/thaw cycles

 Some models only available in          High
 tank sizes up to 4,000 gal.
         Data from State of New York, 1983.
                                                              2-6

-------
                                             Sections
                       Leak Detection and Environmental Assessment
This section presents a brief review of techniques for
detecting  underground tank leaks and  methods for
making ah initial assessment of the extent of the
resulting product release to the environment.  These
techniques are discussed in greater detail in a recent
EPA report (Niaki and Broscious 1986).  Both early
identification  of leaks and a prompt assessment of
their impact are necessary to minimize  the  adverse
effects of releases from USTs. A continuing tank moni-
toring program and accurate leak detection systems
are essential.  In addition,  when  a release is  dis-
covered, accurate characterization of the extent of
release and the pathways of migration  is critical to
planning response actions.

3.1  Tank Monitoring
Methods for detecting leaks in underground storage
tanks fall into four general classes:

    1)  Volumetric (quantitative) leak testing and leak
       rate measurement

    2)  Nonvolumetric (qualitative) leak testing
    3)  Inventory control

    4)  Monitoring of leak effects
These  methods can be used  independently  or in
combination.  Efforts are  being made by EPA's Haz-
ardous  Waste Engineering Research Laboratory to
evaluate the  relative effectiveness  of select technol-
ogies  within  these   groups  at the  Edison,   New
Jersey .facility by using control tanks.

3.1.1  Leak Test Technologies
3.1.1.1  Volumetric (Quantitative)
Leak Testing

This type of testing is based on detecting a change in
tank volume by measuring parameters such as liquid
level, temperature, pressure,  and density-  Among the
techniques used  are an  air bubbling system that
monitors pressure changes resulting from changes in
product level in the tank; a "J" tube manometer that
measures pressure  changes;  and two  methods to
measure changes in liquid levels:  a laser beam and its
reflection,  and a "dip-stick" type device.  Another ap-
proach is  to measure any volume  change by main-
taining a constant level. Some volumetric leak testing

                                                3-
methods can reportedly detect  leak rates  as low as
0.02 gal/h (Niaki  and Broscious 1986).  Table  3-1
presents  a summary of information  on  available
volumetric leak detection methods. This listing is not
considered exhaustive.   Costs  listed under "cost of
testing" do not necessarily include  all testing costs
incurred during leak detection.


3.1.1.2  Nonvolumetric (Qualitative)
Leak Testing

Testing for leaks by qualitative  methods usually
involves the  use  of  another material  besides  the
product (tracer material).  When the tracer material,
typically helium, is used to pressurize a tank, a loss of
pressure or the detection of the tracer gas outside the
tank by a  sniffer mass spectrometer indicates a leak.
Helium's rapid diffusivity allows it to escape through a
tank leak as small as 0.005 inch (Niaki and Broscious
1986).   Other  nonvolumetric methods  of detecting
leaks include pressurizing a tank with gas or placing it
under vacuum and using  sound detection to listen for
a hiss of gas escaping or the bubbling sound of gas
entering the tank and rising through the liquid con-
tents. The primary concern with the use of  qualitative
testing methods is the potential for causing greater
leakage, their effect on product quality, the possibility
of an explosion hazard, the inability to measure  the
leak rate,  and the time required for testing (Niaki and
Broscious 1986).  This form of testing is best suited
for determining the presence of a leak, followed by
volumetric testing to  determine the  rate of leakage.
Table 3-2 presents  a summary  of information  on
available  no.nvolumetric  leak  detection  methods.
Again, this listing is not considered exhaustive.

3.1.1.3 Inventory Control or
Inventory Monitoring

This is perhaps the simplest and most economical leak-
detection method.  Table 3-3 presents a summary of
information on the  use of inventory control (and leak
effects monitoring). The information presented in this
table is not considered exhaustive.  Leaks can be
detected by keeping  records of tank inventories and
noting  any unexplained  change in  liquid  levels or
amounts.  Inventory monitoring can be performed by
gage stick, electronic level measurement, or by weight

-------
 o
•3

 8


I

JK


J
etrl




ipte
P

o
                       o
                                                                         a
                                                                         c
                                                                         o
                                                                                     I
                           o
                           o
          S|

          2.3-

          83-
                       s

                       I
           fl
                       l
                                                            I


                                                             a.


                                                            I
                                                             tn
                                                                  en
                                                                  g

                                                                  '•5.
                                                                  1
           |
           .2
MFP-414

detector
                                    •3
                                    o
                                    en
                                                             I
                                            o
                                            O
                                                                         1
                                                                                                       a.

                                                                                                       aj
                                                                                    2 —   S"
                                                                                   08   =.s
                                                3-2

-------
= $
cctuK
,f
ill
if
*-|
"i1
|—
B
3
o


L


•1
o






IX

f
3
.C
C
Q
O,

-..
5 -o
Cl) O
2 •=
(8 CD

"•
JC
•4* '5.o"Sa ^3






1 f |
I l|
O O) 9
t-etl'S
§ fe-S
' *O -W W O
' c oJ JS ©
L £ S^I




• .
(8 ®
S sis
1 £lf
I-2 F,E «"








1
••e
Ul
^e
I
c
T>
•D
§

CO
s
ji
•5 •
CD
Q] *Q.
.Q "O
ii
C <0 d
= £.i
186,
fi manufacturer or differs
abilization with testing w
ibe temperature equilibr
^Ito*'
§|-B§
0 S 
-------
ires
Empty/Full
Tank for Tes
Requ

    o
    o
C
£

o
Q)

ffl
Q
CO
O
>
C
o
ra

E
E

CO
     ex
     '3

     •c
                           I
                           LLI
                   8
                   fi-
                  •s-s-s
                  

                                                                       = 1
                                                                       oj
                                                                       tn
                                                                      S
                                                                      §
                                                                      .§•
                                  •D
         iisp.
ll-l
•§88
>8°
S co^
CL
   O>
  C> 3
  000)

  s^-g

   (3 co
   Q) ^o
                                                           i

                                                           'x
                               I
                     gffi-g-Sg"0  |.
                     Q 2 O .2 £ 5  D- .
                                                    c,
                                                    0]
                                                    Q

                                                    =8

                                                    •£
                                                                      &

                                                                      Jo

                                                                      1'
                                                                      co:
°-58l

•tills-
*OJ 4= ., P (0
.g to to g £ E g
                                                                      1
                                                                      CO
                                                                      S
                                                                      CO
                                                           ui
                                                           c
                                                           o

                                                           'to
                                                                              §.
                                                                              (0
                                                                              S
                                                                              co

                                                                              CO
                                                                              c
                                                                              'o
                                                                              2

                                                                              i
                                                                              •
                                                                              I
                                                                              CC

                   5 o
                   es

                            Si
                            CO
                                                            CO
                                                            £
                                                            o
                                                           ts
                                                           0)
                                                                       is
                                         3-4

-------
Acc
l/h)
Claimed
I

g
=
8
o
Q
1C

J
 0)
-f
ro



W
CO
co
0)
S
       0)
       .0.
       "8
           = •3
           n.g
           TJ CO
           -S? S

            '"5


           ll
                                      JZ
                                      CM Q.
                   CD
                   S
          fj
          §s
                   S
                          8
sitive to O.
level chang
in
than 0.0
                                                                   £
00
S
CO
                                                       I
                                                       §
                                                       o
                                                       i
                                                       £

Less than 0.05
          His
          o:S-8.E
Certi-Tec testing
k sen
Ethyl
                                        3-5
                                             CD
                                             g.
                                             'a.
Fluid-stati
testing
Helium differe
pressure test

-------
           ^>
      g

      "o
•g
       T3

       Jg
                         g
                         o
                         I
                         8
                         o
                         1
           lo

           Is
           o-S
           10 -5
           o >
             ^
                                   2
                                   CL
o
z
c
o
fe
•§
g
c
8


w


1
5

i.
o
c
tn
Si
Determining soil temperature
characteristic change due to
the presence of hydrocarbons


D)

to
C
CD
to
1
fC
S
o
i
o
•z.

0
§

1
c
8


w
1

1
5

1
o
c
8 o
Q IS
Hydrocarbon detection by
ground-penetrating radar,
electromagnetic induction,
or resistivity techniques






.8
t
O
QJ

O o
w E
o
z
g
o
b
•Q
2

8


t/>
1

1
•S

1
0
c
S m
Q 2
Product sensing in liquid;
collection sump for product
directed through a horizontal
pipe installed under a tank








en
CO

0
z

o
h
•£)
s
c
8


to
.03
CO

1
5

1
0
c
S CO
Q S
Monitoring of vapor through
monitoring well







JO
"33
1









to
c
o
s
tf=
'8
i
to
c
i
•a
TJ
to
J2
1
"o
c
us 1986.
d with manufacturer for different nun
O fl)
'i|
r~n 5"
m ffi


•-3 "^
|1
o W
11
QO
* 4—
                                                            3-6

-------
monitoring (i.e., using pressure and density measure-
ments to determine weight changes in tank contents).
Although  the  problem of  keeping  records  of  tank
inventory is complicated by the fact that  petroleum
products and other chemicals are volatile and  losses
due to  evaporation  are  unavoidable,  the inventory
method can be used as a first and convenient method
for gross leak monitoring. A long-term (greater than 4
months) tabulation  of  inventory  records,  properly
graphed with adjustments for temperature, aids  in leak
detection.

3.1.1.4  Monitoring of Leaks Effects

Leaks in  USTs can be detected by monitoring the
environmental effects of a leak inside or outside the
tank.  This type of monitoring typically  entails drilling
monitoring wells,  installing  monitoring  casings,  and
performing  chemical analyses.  With environmental-
effects  monitoring, it may  be  difficult  to determine

Table 3-4.  Variables Affecting Accuracy of Leak Detection*
    Variable                               Impact
                     which tank is leaking  when  there is more  than one
                     tank; however, leak detection may be more conclusive
                     than quantitative testing  methods  if  no  interfering
                     substances are  present.  These methods  do not pro-
                     vide information on leakage  rates or the  size of the
                     leak; however,  once  installed,  a  leak effects moni-
                     toring system enables more  frequent checkinq to  be
                     made for leaking tanks than do the other approaches.

                     3.1.2  Leak Detection Variables
                     How accurately  leak-detection methods pinpoint leaks
                     and measure rates of  leakage depends  on  several
                     variables as well as the method itself.  Table 3-4 lists
                     the major variables that affect the accuracy of currently
                     available leak-detection methods.

                     During tank testing, most of these variables  can  be
                     compensated for by such techniques  as measuring
                     temperatures  and pressures in the tank and con-
                     sidering  measured variations in  the pertinent calcu-
   Temperature change

   Water table


   Tank deformation


   Vapor pockets



   Product evaporation

   Piping leaks


   Tank geometry


   Wind


   Vibration

   Noise


   Equipment accuracy

   Operator error


   Type of liquid stored



   Power vibration

   Instrumentation limitation

   Atmospheric pressure


   Tank inclination
Expansion or contraction of a tank and its contents can mask leak and/or leak rate.

Hydrostatic head and s jrface tension forces caused by ground water may mask tank leaks
partially or completely.

Changes or distortions of the tank due to changes in pressure or temperature can cause an apparent
volume change when none exists.

Vapor pockets formed \jvhen the tank must be overfilled for testing can be released during a test or
expand or contract from temperature and pressure changes and cause an apparent change in
 volume.
Product evaporation can cause a decrease in volume that must be accounted for during a test.

Leaks in piping can catlse misleading results during a tank test because many test methods cannot
differentiate between piping leaks and tank leaks.

Differences between the actual tank specifications and nominal manufacturer's specification can affect
the accuracy of change in liquid volume calculations.

When fill pipes or vents, are left open, wind can cause an irregular fluctuation of pressure on
the surface of the liquid and/or a wave on the liquid-free surface that may affect test results.

Vibration can cause waves on the free surface of the liquid that can cause inaccurate test results.
Some nonvolumetric test methods are sound-sensitive, and sound vibrations can cause
waves to affect volumetric test results.

Equipment accuracy can change with the environment (e.g., temperature and pressure).

The more complicated a test method, the greater the chance for operator error, such as not
adequately sealing the tanks if required by the test method in use.

The physical properties of the liquid (including effects of possible contaminants) can affect the
applicability or repeatability of a detection method (e.g., viscosity can affect the sound
characteristics of leaks; in acoustical leak detection methods).

Power vibration can affect instrument readings.
Instruments must be
                       within their design range or accuracy will decrease.
A change in this parameter has the greatest effect when vapor pockets are in the tank,
particularly for leak-rate determination.

The volume change per unit of level change is different in an inclined tank than in a level one.
    Data from Niaki and Broscious 1986.

-------
lations,  thorough training  of  operators to  minimize
operator errors,  inspection for any noticeable vapor
pockets, care in sealing vents and fill pipes where ap-
plicable, and thorough calibration of instruments and
other equipment. (More detailed discussions of these
techniques are  presented in Subsections  3.3  and
4.2.)

3.2  Contaminant Migration
(Transport) Pathways
An understanding of how contaminants released from
underground tanks migrate in the subsurface environ-
ment is critical to evaluating the impact of any release.
The  migration, or transport,  of these  releases de-
pends primarily on the quantity released, the physical
properties of the substance, and the structure of the
soils and rock through which the contaminant moves.
Figure 3-1  shows the typical subsurface environment
of USTs.

3.2.1 Unsaturated Zone  (Vadose Zone)
When a liquid  contaminant  is  released below the
ground surface, it  percolates downward to the un-
confined ground-water surface.  In the case of a small-
quantity release, the contaminant may be held in the
interstices of the soil particles in this zone and may not
reach the  saturated zone  except as it is slowly dis-
placed and carried downward by percolating rainwater.
In the case of large releases, however, movement
through this zone may be quite rapid.

If surface soils were homogenous and isotropic, the
only lateral movement of percolating product would be
from capillarity effects, and the  resultant contaminant-
soaked soil would be an inverted narrow cone.  Be-
cause typical  spill  sites are  nonhomogenous, the
resultant  product spread through the  unsaturated
zone is usually irregularly shaped (Figure 3-2).

If the soil and  product properties are known, whether
the product will reach the ground water can be esti-
mated.  The  following  formula provides a  useful
approximation for hydrocarbons:
where

      D = maximum depth of penetration, m

      V = volume of infiltrating hydrocarbon, m3

      A = area of spill, m2
      Rv = a constant reflecting the retention capacity
           of the soil and the viscosity of the product
           (see Table 3-5).
 Figure 3-1.   Diagram of typical subsurface environment of USTs (Camp, Dresser, and McKee 1986a).

                                                                Ground Surface •
                                                                      Saturated Zone
                                                                      (Ground Water)
                                                 \x	
                                                 Bedrock ^
                                                 3-8

-------
 Figure 3-2.   Shape of hydrocarbon plume in vadose zone under different leakage rates and physical
            conditions (AP11985).                     |
                 Slow Seepage
                     into
                  Permeable
                     Soil
High Volume
 Seepage
   into
 Permeable
   Soil
Table 3-5.  Typical Values for Rv*
                                 Rvt
Soil Gasoline
Coarse Gravel 400
Gravel to Coarse Sand 250
Coarse to Medium Sand 130
Medium to Fine Sand 80
Fine Sand to Silt 50
Kerosine
200
125
66
40
25
Light
Fuel Oil
100
62
33
20
12
* Data from Shepherd, no date.
t A constant value representing capacity of soil and viscosity
 of product.

The usefulness of  the  calculations  is limited by the
accuracy of the Rv values, which are extremely site-
specific. The Rv values given in Table 3-5 are esti-
mates for dry soil conditions.  Obviously, values for
wet or partially wet soils would be less  because pore
voids are already  partially filled.  Retention (atten-
uative) capacities for hydrocarbons vary from about 5
liters/ms in coarse gravel to more than  40 liters/m" in
silts.  A leaked substance such as gasoline can travel
25  feet through  unsaturated, permeable,  alluvial, or
glacial sediments in a few hours (or at most, a few
days) depending on specific conditions  at the release
site.

3.2.2  Capillary Zone

The capillary  zone has a significant impact  on the
movement of nonaqueous-phase liquids that are less
dense than water (gasoline and other petroleum prod-
ucts). Figure 3-3 presents a typical flow pattern  for
hydrocarbons in the capillary zone.
Seepage into
Stratified Soil
with Varying
Permeability
             The primary movement in the capillary zone is lateral.
             As more free oil reaches the capillary fringe, a layer of
             increasing thickness (a lens) begins to form  on the
             capillary fringe under the  influence of the infiltrating
             oil. The weight of the oil may result in the depression
             of the capillary fringe and possibly the water table
             (Figure 3-4).

             Buoyant forces act to restore the water to its initial
             level, and if a critical minimum thickness (which varies
             with soil and contaminant properties) is exceeded,
             lateral spreading begins.  Lateral spreading will occur
             in all directions, but the predominant movement will be
             with the slope of the water table (Figure 3-5).

             The characteristic  shape of the flow is called an "oil
             pancake." The shape of the pancake depends on the
             permeability of the soil, the percolation rate, and the
             local water-table configuration.  In general, the more
             permeable  the  soil, the  more the contaminant will
             spread and  the less thick the recharge lens will be.
             The steeper the  hydraulic gradient, the narrower the
             plume will be and it will be elongated in the direction of
             ground-water flow.

             As the water  table fluctuates over time, the  non-
             aqueous-phase liquid contaminant may spread out
             somewhat in the vertical plane.  In porous media, this
             can lead to an apparent decrease in the volume of free
             product, as  previously uncontaminated  soil will now
             retain (adsorb) some of the contaminant.  In fractured
             rock media,  on the other hand, an apparent increase
             in the  volume of free  product may occur as contam-
             inants trapped in dead-end fractures not previously
             unconnected with the water table may be washed out.
                                                  3-9

-------
Figure 3-3.   Typical flow pattern for hydrocarbons in the capillary zone (Shepherd, undated).
          Capillary
            Zone
                                                            Ground-water Flow
 Figure 3-4.    Typical behavior of hydrocarbons in soil following a sudden high-volume release (AP11980).


                                                    Spill Sjte  -                Ground Surface
                                                                     Soil Contaminated
                                                                     by Residual Product

                                                                     Product Migrating Downward
                                                                     iand Accumulating on Water Table
                                                                              Capillary Zone
                                                           3-10

-------
 Figure 3-5.   Behavior of hydrocarbons after release has stabilized (AP11980).
                                                               Soil Contaminated
                                                               by Residual Product
 3.2.3  Saturated Zone
 The transport of  contaminants  in the saturated zone
 can be characterized for three classes of substances:

    1) Miscible or dissolved substances

    2) Immiscible substances with specific gravity of
       less than 1.0

    3) Immiscible substances with specific gravity of
       more than 1.0.

 According to the mass transport laws of advection and
 dispersion, miscible substances that enter the ground
 water move in the general direction of ground-water
 flow.  Advection  is the movement of a contaminant
 plume with the mean ground-water flow.  Dispersion
 refers to the spreading out and dilution of a contam-
 inant as  it occupies more of the saturated zone than
 can be explained only by advection.  The three levels
 of  dispersion are molecular diffusion,  microscopic
 dispersion, and macroscopic dispersion. Macroscopic
 dispersion, which is due to variations in soil permea-
 bility (heterogeneity), is the principal phenomenon in
 most field situations.
 Immiscible substances that are "lighter" than water (a
 specific gravity of less than 1.0) are usually found only
 in the shallow part of the saturated zone.  The rate of
transport probably depends on the local ground-water
gradients and the viscosity of the substance.

 Immiscible  substances that are  denser  than water
 (specific  gravity  greater  than 1.0)  move downward
through the saturated zone. A dense immiscible sub-
stance poses  a greater danger in terms of migration
potential  than less dense substances because their

                                                3-
  deeper penetration into the saturated zone increases
  the potential for its  solution and dissolved com-
  ponents then migrating with the prevailing  ground-
  water flow pattern.  Although some residual saturation
  (and therefore retention  in  the  soil) occurs  as the
  substance moves downward,  it  is less than in the
  unsaturated zone.  If the quantity of release exceeds
  the retention  capacity of the unsaturated and satu-
  rated zones, the nonaqueous-phase liquid continues
  its downward migration until it reaches an impermeable
  boundary. At this point it forms  a lens and spreads
  with the slope of the barrier and pools in depressions
  in the boundary.

  3.2.4  Vapor Phase Transport
  A liquid contaminant leaking from  an  underground
  tank enters the vapor phase in the unsaturated zone
  according to its specific vapor pressure (the higher the
  vapor pressure of the substance, the more likely it is to
  evaporate).  Once  in the vapor phase, the contam-
  inant moves by advection and by diffusion.  Vapors
  move primarily in a horizontal direction,  controlled by
  the  slope of  the water  table and the location  of
  "impermeable" barriers. Those vapors less dense than
  air, however, often migrate in a vertical direction, which
  results in their entrainment in sewer lines, basements,
  and like areas.

  Molecules of  gas may adhere to soil particles by
  adsorption and then be released  at a later tirhe after
  passage of the plume of fluid contamination.  Vapor
  movement can-be blocked by monolithic buried struc-
  tures, but  the vapors will  move  readily through  the
  backfill  surrounding  such structures (such as along
  the route of a buried pipeline).
1

-------
3.3  initial Assessment of Extent
of Release
When a leak  is discovered or suspected, an  initial
assessment must be made of the risk posed to human
health and the environment must be  made quickly.
This assessment should provide an approximation of
the extent (lateral and vertical dimensions) and sever-
ity (quantification of specific constituents present) of
the contamination problem.  The assessment results
will be used to design and implement programs for
rapid control of the  spread of contaminants (if neces-
sary), or to define additional data collection programs
before remedial action programs are developed.

The initial assessment may consist of  one or both of
the following: 1) data collection from existing sampling
points, and  2) installation of new sampling points for
data collection.

3.3.1 Existing Sampling Points
Existing sampling  points  include inventory records,
underground  structures,  water-supply wells,   and
surface water bodies. Frequently, existing sampling
points can provide a good estimate of health and en-
vironmental effects of a leak. Some provide a place of
environmental contact (e.g., a water-sampling well).
Identification of contaminants  in receptors usually is
sufficient to prompt  emergency response actions.

3.3.1.1   Inventory Records
Inventory records  are frequently the  trigger mech-
anism to indicate that a leak has occurred. They  also
can indicate the approximate quantity of  material re-
leased  by a leak.  Such records are especially useful
where both inflow and outflow are monitored (such as
 in retail gasoline facilities). Pinpointing the amount of
 loss may be  difficult, however, because of inac-
 curacies in the gaging and metering devices used to
 measure fluid movement into and out of the tank.

 3.3.1.2  Underground Structures
 Sizable releases   of  volatile  substances, including
 gasoline, may be  associated with gas migration and
 releases of toxic  or volatile  vapors  at or near the
 ground surface. Basements and sewers, for example,
 may accumulate vapors migrating through the soil from
 underground  tank  releases, and  such vapors some-
 times reach explosive concentrations. Monitoring in
 such underground areas  can identify symptoms of
 tank leakage.
 Several kinds  of  currently available  air monitoring
 instruments can locate and quantify vapor releases to
 confined underground spaces.  Explosimeters  and
 oxygen meters can determine the existence of an
 explosive atmosphere. These instruments report vola-
 tile vapor concentrations as a percentage of the lower
 explosive limit (LEL). The oxygen meters  that are
                                                3-72
normally included in the explosimeter instrument can
identify high concentrations of vapors by indicating
the displacement of oxygen by some other gas.
Among  a variety of  more sensitive  direct-reading
instruments currently on the market for detecting and
quantifying vapors are the organic vapor  analyzer
(OVA) and the P1101 Hazardous Waste detector.

The   Century  OVA,  a  lightweight  shoulder-borne
instrument manufactured by the Foxboro Corporation,
can  monitor volatile organic  compounds in the air
through the use of a flame ionization detector (FID).  ;
This instrument can be assembled as a total organic
"sniffer" or as a portable gas chromatograph (GC),
which can also function as a sniffer.

When used as a sniffer, a hand-held output meter
attached to a sample probe collects the air and in-
dicates the concentration of nonmethane compounds
in the air sample in parts per million. iWhen the OVA is
assembled as a portable  GC, a sample is introduced
into the GC column for quantitative determination of its
components. A strip chart recorder provides a printout
of the response in peaks similar to those obtained with
a laboratory-scale GC.

The Model P1101 Hazardous Waste Detector, a port-
able  gas  analyzer  manufactured by HNU  Systems,
Inc.,  was designed to detect  the presence  of poten-
tially harmful  chemical vapors.  It operates on  the
principle of photoionization and is sensitive to a wide
range of both organic and inorganic pompounds.

Model PI  101  uses different sample probes equipped
with a lamp of different energy levels (9.5 eV, 10.2 eV,
 11.7 eV).  Positive  identification of suspect com-
pounds   is  based  on  correlation  of  the  known
 ionization potential of the suspect compound with that
of the  appropriate  probe.  A   direct-concentration  ;
 readout provides results in parts peri million.

 Sorbent tubes can provide both qualitative and quan-
 titative information.  A known  volume of air is pumped
 through the  sorbent  tube, and  th6 contaminants  in
 this  sample are adsorbed and  concentrated on the
 sorbent material in the sampling tube. Some sorbent
 tubes can be taken into the  laboratory for treatment
 and  analysis to determine concentrations of a variety
 of constituents.  Others  are designed to register the
 presence of  only  a  single  compound  or group  of
 compounds based on a colorimetric chancie produced
 as the sample is drawn through the tube.

 3.3.1.3 Water Supply Wells
 One of the most serious threats posed by releases
 from USTs is the contamination pf  residential and
 municipal drinking water wells.  Such wells should be
 sampled  promptly when a  leak is known to have
 occurred. The locations  of wells in a particular vicinity
 can be obtained from several sources, including city or
 county engineers, local water companies, local health

-------
 departments, and direct communication with  nearby
 residents.

 Whenever possible, well-water  samples should be
 analyzed in a laboratory, particularly when the exact
 nature of the contamination has  not been confirmed;
 however,  several  usable  screening  techniques and
 test kits are  available for  field analysis  of  water
 samples.  Table 3-6 presents a  partial listing  of por-
 table field instruments for water monitoring.

 3.3.1.4  Surface Water Bodies
 Nearby surface  water bodies,  including  enclosed
 sewers and drains, should be sampled as they reflect
 points of  potential  release from  the subsurface
 environment. Like well-water samples, analysis should
 be done in a laboratory, but field  identification may be
 necessary  to  expedite initiation of a  surface-water
 management program.

 3.3.2  New Sampling  Points

 When no   existing  sampling  points  can  provide
 qualification of the risk posed  by a leak  or these
 sampling points fail to show any effects, new sampling
 points must be developed to define the problem.

 Both  direct and  indirect  methods are available to
 collect data from these new sampling points. Indirect
 methods generally comprise remote sensing tech-
 niques, whereas direct methods  involve extraction of
 soil and liquid samples for direct chemical analysis.
   /
 3.3.2.1  Indirect Methods
 As  defined herein,  indirect methods  include geo-
 physical surveys and soil vapor analysis.

 Several geophysical  survey techniques are  available
 to define subsurface water quality conditions. The two
 most  frequently used  are electrical  resistivity and
 electromagnetic  conductivity.  Both  methods  work
 best for leakage of  inorganic fluids  with high ionic
 strength, but they may be used for some organics with
 high isolating capabilities.

 Use of the resistivity survey requires  the introduction
of a known electrical  current  into the ground with
electrodes. The amperage of the current is measured
at other electrodes, and the resultant decrease (re-
sistance of the earth) is also measured.  If geologic
conditions are uniform,  high earth resistance  indicates
general absence of ionic contaminants and low resis-
tance indicates the presence of  ionic contaminants.
The procedure can be  used to measure both vertical
and lateral changes. The presence of metallic objects
                                               3-1
 (pipes, tanks, etc.) or electrical sources (power lines)
 can interfere with the procedure.

 Essentially a corollary to the resistivity survey, the
 electromagnetic conductivity (EM) survey has similar
 limitations and interferences. The EM survey is usually
 quicker, but it provides only limited vertical evaluation.

 Recently, developed procedures for direct measure-
 ment of vapors in soils involve the use of the following
 instruments: surface  flux chambers, downhole flux
 chambers, accumulator devices, and ground probes.

 Surface Flux Chambers

 The surface flux chamber is an enclosed device used
 to sample the gaseous emissions from a defined area
 on the ground surface. Clean, dry, sweep air is.added
 to the chamber at a controlled and measured rate, and
 the concentration of the  species of interest  is mea-
 sured at the exit of  the chamber.  Analysis  is by
 portable gas analyzer or GC.

 Downhole Flux Chambers

 The downhole  flux  chamber operates  on the same
 principle as the  surface flux chamber. It is either driven
 into the  ground or placed down an augered hole.
 Sweep air is added, and the exit gas concentration is
 monitored. Analysis is by  portable gas analyzer or GC.
 The downhole flux chamber is more sensitive than the
 surface  flux chamber because sampling  occurs  at
 depths below any surface interferents and where soil
 gas concentrations are normally at higher levels.

 Accumulator Devices

 Use of the  accumulator  devices involves collecting
 and concentrating  soil   gases  to  provide a  time-
 integrated sample. One variation is to collect gases on
 activated charcoal bonded to a curie-point wire. Anal-
 ysis is by mass  spectrometry.  A second variation is to
 pump gases from an enclosure or ground probe and
 then collect the gas  in an  absorbent.  The  primary
 advantage of this technique is that the sampling  time
 can be adjusted to reach the desired sensitivity."

 Ground Probes

 The use of ground probes involves placing a tube into
the ground to the desired sampling  depth.  Openings
 in the tube near the leading edge allow the entry of
 soil gases.  A port in the upper end of the tube allows
 sample gas to be extracted.  Analysis is typically by
 GC. The technique is relatively sensitive, and sam-
pling depth can be varied to reach below interferents
or low-concentration areas.

-------
3-
S
CO
CL
ra

O
f
o

i2

O)
c
o
a   3
           1
        o

        8
         8

         5
       8
       CX
       W
       X C
       a «
       p j=
e


I
                 as
                 CL
                                  s

                                  £


                                  1-8

                                  o §
                                  S o
                                  " Q.
                o>

                CO
                             I
                             o
                             o
                    I!
                    fi-o


                    > .2


                    1 8

                    If
                    O §
             o
             o"

             1
                 s
                 I
                      £
                      CO
                      I
                       .
                   ••5
                   o
                   in


                   o.
                        •§>
                                      CL O



                                      S"e-
                                      ffi 

                                                           s
                                                           o
                                8
                                00
                                                                         2 CO



                                                                         Oo"

                                                                         P<


                                                                         %2

                                                                        l
                                                                   J. o
                          l
                                                                 li  i
M

ter

ical,
                          	i ^5 ^

                     83  «ll

                     °a  l£§
                                                                      o

                                                                          S

                                                                          'c
                                                                          o
                                                                 :
                                                                 o
                                                                 *o

                                                                 II
                                                                                      s
                                                                                      CO
                                                                          O    —
                                                                            li
                                                                            o S
                                                                            &£
                                                                            *°
                                                            •11
                                                                                   1
                                                                     o
                                                                     t
                                                                          1

                                                                          CO
                                                                          Q

-------
3.3.2.2  Direct Methods

Another method of approximating the extent of both
liquid and vapor  contaminant  migration  below the
ground surface entails soil borings.  The most  com-
mon application of soil borings involves collection of
undisturbed soil core with an auger or drilling a  sam-
pling tube  into the ground. The collected sample is
sealed in  a container with  minimal head space and
sent to the laboratory for analysis.

An  effective field  screening technique for grab  sam-
                                               3-
   ples of soil cores involves the  use of the portable
   OVA (described earlier).  Example soil and/or water
   from the boring is placed in a septum vial so that it is
   approximately 3/4  full.  The sample is then heated,
   usually by placing the vial in hot tap  water.  Air is
   extracted from the headspace in the vial for evaluation
   by an OVA, either the sniffer type or the GC type.  A
   comparison of the GC output for the sample with that
   of a blank, or uncontaminated, soil, can identify areas
   of major contamination, lesser contamination, and
   uncontaminated soil.
15

-------

-------
                                              Section 4
                              Corrective Action Response Process
This section presents procedures and information for
the selection of appropriate initial responses and long-
term corrective actions when a leak occurs in an LIST.
Specific  technologies mentioned  herein are  dis-
cussed in detail in Section 5.

The responses to  releases from USTs depend on
several different factors,  largely site-specific.  Each
incident is unique.  The intent of this section, there-
fore, is to aid the reader in the  development of an
appropriate corrective-action response when such a
release occurs.

Corrective action usually involves two phases.  The
first involves the initial corrective  actions intended to
limit the  impact of a sudden  or newly  discovered
release. The second  involves long-term,  permanent
corrective measures.  Initial corrective actions (or re-
sponses) are those directed at immediately containing
and controlling a release.  Timing is critical. Therefore,
efforts  are focused on  source  control and  public
welfare. This includes the collection and containment
of released material. Permanent corrective measures,
Figure 4-1.   Typical initial corrective-action process.
on the other hand, involve comprehensive cleanup to
protect  human  health  and the  environment.  Both
require an assessment of the situation, which includes
the physical and chemical nature of the released sub-
stance, the environmental setting of the incident, and
the extent of the impact.

The options available for addressing both initial and
long-term corrective actions are discussed under the
appropriate heading.

4.1  Initial Corrective-Action Options
When a leak in an LIST is discovered or occurs sud-
denly, the initial corrective actions are directed toward
collection and containment of   the  substance  re-
leased. Initial efforts typically occur within a short time
frame, are  of brief  duration, and involve limited
resources. This often  entails deployment of field per-
sonnel and equipment to the  scene within  hours of
the occurrence to minimize  the impact of the release.
Figure 4-1 presents  typical steps in  the  decision-
making  process  for  selection  of initial  corrective
actions.
                                            endangeipent
                                             public health
                                               and the
                                             environment?
                                                      to

-------
4.1.1  Evaluation of Release
The first step in responding to an UST release usually
involves an evaluation of the situation. Such an eval-
uation entails  determining the substance released,
the  migration  pathways and receptors,  and  the
potential impact  on human  health  and the environ-
ment. Several  offices and field environmental  inves-
tigative techniques  used to evaluate  UST releases
were discussed  in Section 3; this information  is not
repeated here. This section focuses on incorporating
environmental   investigative  techniques  into  the
response process and identifying significant  data re-
quirements for the  planning and implementation of
response options.

4.1.1.1 Determination of Substance
Released

Underground  storage  tanks  are  used  to  store
petroleum  substances  and  selected   hazardous
chemicals. Table 4-1, adapted from a table in a study
conducted by Camp, Dresser, and McKee (1986), lists
the various petroleum products stored in USTs. Tabje
4-2,  from  the  same study,  lists  chemicals stored in
USTs.
4.1.1.2  Site Characterization
An  important  step prior to  initiation  of  a response
action is site characterization. How detailed the charac-
terization should be  depends on the extent of  the
release problem and  the amount of data required for
the  planning  and  implementation  of an  adequate
response action. Figure 4-2 presents a matrix of the
corrective  action  technologies   addressed  in  this
report and the typical site information necessary to
select,  plan, and use  a particular technology. The
amount of information and data collected during  site
characterization efforts is related to the complexity of
the  situation  and  of the  selected corrective action
response plan.
A site evaluation should cover hydrogeologic charac-
teristics, geographic  and topographic characteristics,
and water- and land-use patterns.

Hydrogeologic Characteristics

A hydrogeologic characterization  addresses both un-
saturated (soils) and saturated  (unconsolidated  and
consolidated) flow and contaminant \ transport condi-
tions. It involves collection of information concerning
the subsurface where  the problem lexists,. Some of
this information  can be obtained from  standardized
regional and  local reference sources; however, most
of  the  needed  data is  site-specific and must be
obtained by  collecting  soil, rock, water, and con-
 Table 4-1.  Petroleum Products Stored in Underground Storage Tanks

                      Product
                      Power fuels - civil use

                          Aviation gasoline and additives
                           Grade 80
                           Grade 100
                           Grade 100LL (low lead)
                          Jet fuel and additives
                           Jet A (kerosene type)
                           Jet A-1 (kerosene type)
                           Jet B (wide cut or naphtha)
                          Automotive (motor gasoline) and additives
                           Leaded
                           Unleaded
                          Diesel fuel oil and additives, Nos. 1-D, 2-D, and 4-D
                          Gas turbine fuel oils, Nos. 0-GT, 1-GT, 2-GT, 3-GT, and 4-GT

                      Heating and illuminating oils

                          Fuel oils Nos. 1, 2, 4, 5, and 6
                          Kerosene

                      Solvents

                          Petroleum spirits, types 2,3, and 4, and commercial hexane
                          Mineral spirits orStoddard solvent (Type 1. petroleum spirit)
                          High-flash aromatic naphthas, types I and II
                          VM&P naphthas - moderately volatile hydrocarbon solvents, types, I, II, and III
                          Petroleum extender oils, types 101,102,103, and 104
                       * Data from Camp, Dresser, and McKee 1986.

                                                    4-2

-------
I

I
•o
•s
I
eg
«

§
I


1
V)
,2
                        sisiit
                              (!)
                              i-
                               ** *» A c Ji  ^4 o c H 2 E *~ e — ^~^ ^^ —" —~ - == n
            li|ii|liijllfii
            -  ^^^l|illllof|-||||||||ig
53  s  tsl-o*  » I I s s





ooQQiuiijiiiCuGJCuiuiiluLifiajGj
OOOOOOOQ
 >,
 r
              I
     fiitil
                «
               o> ^~
               Son

               Hi
               ill
                           " «r«- es- " "1 c
                           = i>§11
                            -  (DO)*"

                               If
If
«":
                       ftl
                       CD m o
3
S
     f | 1 |
     ill!
     o o o o

-------
Figure 4-2.    Site Information needs for evaluation of corrective-action alternatives.

Technology
Romoval/exeaval'On of lank, soil and sediment
Tank removal
Soil excavation
Sediment removal
Oratta and olfsKe treatment and disposal of contaminants
Solidification/stabilization
Landfilling
Landtarmlng
SoN washing
Thermal destruction
Aqueous waata treatment
Doap well Injection
FraB product recovery
Dual pump oyatama
Floating Ittttr pumps
Surface oil/water separators
Ground-walar recovery systems
Ground-water pumping
Subsurface drains
SiibsurlncB barriors
Slurry walls
Grouting
Shoot piles
Hydraulic barriers
)q sMu treatment
Soil Hushing
Blostlmulatlon
Gnomical treatment
Physical treatment
Hydrogeologlc characteristics
Soil profiles
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Soil physical properties
•

•
•

•
•
•
•
•









Soil chemical properties


•
•












Death to bedrock
•




















Depth to ground water
•

•


















Aaulfer physical properties





















1
1
|
I
i
C3





















1
•o
s
E
I





















03
£





















Renharae rates





















Aquifer characteristics





















%
3
O
Sii
i
0
i



•
















Geographic and
topographic
characteristics
Precipitation





















Temperature
•
«
•














E
O
3
o
w
1
§
UJ











Tonooraohv
•
•
•



•
•



•

Accessibility
•
•
•



•
•
•
•
•
•


•
•
• '
•
•
Site size





•

•
•
•
•
•
•
•
•
Proximltv to surface water









1
rs
JC
o
S
E
ct




•
•


•
•
•
•
Land and water
use patterns
in
1
S
5
i
c

•


•

•
•
•
•
•
•

V.
I
a
c
a
a
|
S
i
LL

e


e

e
e
e
e
«
e

Current land-use patterns

«
«

9

•
«
•
•
•
•

Growth projections



•
«
•
•
•
•

                                                            4-4

-------
Figure 4-2.     (Continued)
       Technology
                                                                   Hydrogeologic characteristics
                                                                    i
Geographic and
  topographic
characteristics
Land and water
 use patterns
                                                                                                                                            (3
       Ground-water treatmant

           Air stripping
           Carbon adsorption
           Biological  treatment
           Preclpltatlon/flocculatlon/sedlmentatlon
           Dissolved air flotation
           Granular media filtration
           Ion exchange/resin adsorption
           Oxidation/reduction
           Neutralization
           Steam stripping
           Reverse osmosis
           Sludge dewatering
       Vapor migration control, collection, and treatmant

           Passive collection  systems
           Active  collection systems
           Ventilation of  structures
           Adsorption
           Flaring
       Surface water/drainage controls

           Diversion/collection systems
           Grading
           Capping
           Rovegetatlon
       Restoration of contaminated water supplies and sewar lines
           Alternative central water supplies
           Alternative point-of-use water supplies
           Treatment of central water supplies
           Treatment of point-of-use water supplies
           Replacement of water and sewer lines
           Cleaning/restoration of water and sewer lines
       Note:  Technologies in bold type are likely to be used in response to UST releases at gasoline stations,

-------
taminant samples for chemical and physical testing in
the laboratory or by conducting in situ tests of hydro-
geologic properties.

Geographic and Topographic Characteristics

An understanding of the  general  physical  charac-
teristics of a site is important.  Surface attributes of a
site should be identified because they can affect the
subsurface environment  both  directly and indirectly.
Climatic information (e.g., precipitation, temperature,
and evapotranspiration) is also needed.  Other factors
having  an impact on corrective actions  are topog-
raphy, accessibility, size of site, proximity to surface
water, and proximity to population centers.

Water- and Land-Use Patterns

An understanding of surrounding water- and land-use
patterns is essential to an assessment of the potential
risk of affected populations. Current water usage and
expected future requirements in the area of the  LIST
release are important factors in determining the  criti-
cality of the threatened or polluted water resource.

4.1.2  Initial Corrective Actions
After the release has  been  evaluated, appropriate
corrective actions should begin immediately.  Initial
corrective actions can be categorized as those actions
that are directed at stabilizing a newly discovered or
sudden release. During the implementation of initial
corrective actions, efforts are focused on source con-
trol involving  collection  and containment of released
material. The following is  a list of those options that
typically would  apply in any  initial corrective action
response.

     Removal of remaining product from the leaking
     UST
     Collection of readily retrievable released
     substance
     Containment of release to minimize immediate
     threat to human health and environment

 •   Repair or removal of leaking tank
     Restriction of access to the site

 Effective implementation of  initial corrective actions
 commonly dictates that field deployment occur within
 hours of the discovered release.  It is essential that
 early discovery and  containment of  any suspected
 leak occur as soon as possible so that recovery pro-
 cedures may be initiated and the  influence of any
 release can be minimized. Normally, the first action
 taken is the removal of  any remaining product in the
 leaking UST. In addition, measures are often taken to
 minimize imminent and immediate risk to human health
 and the environment. Figure 4-3 presents a listing of
 options available for the initial response to typical UST
 release situations.
4.2  Permanent Corrective-Action
Options
After the initial response, the focus of assessment
and investigation activities turns toward the need for
permanent corrective measures.   In situations where
the release poses  no danger to ground-water re-
sources, human health, and the environment, correc-
tive action  may be  limited  to pumping the tank and
removing significantly contaminated soils.  Figure 4-4
is a generic flow diagram showing the procedure for
deciding  what the   permanent  corrective-action
approach should be. The appropriate local, State, and
Federal agencies  should  be  consulted  for  more
specific corrective-action guidelines.
4.2.1 Assessment and Investigation to
Determine Surrounding Land Use and
Potential impact on Public Health

Selection of the  proper permanent corrective-action
approach requires consideration of the physical geo-
graphy of the site where the release occurred.  For
example, an underground tank leak  at a corner gas
station presents  different problems  from those en-
countered at a large industrial facility.  Because gas
stations  are on  small  plots  in  residential or  light
commercial developments, several factors must be
considered:

    Limited available land for implementing corrective
    action.

    Possible rapid offsite migration requiring
    easements to conduct offsite cleanup.

    High potential for public health impacts because
    of the proximity of densely populated areas.

•   Implementation  difficulties posed  by  the  com-
    plexity   of  adjacent  subsurface  public-service
    facilities (e.g., sewers, water lines,  and electric
    utilities).

 Figure 4-5 presents a matrix of site problems com-
 monly encountered and the appropriate  corrective
 action.

 4.2.2 Factors Affecting Selection of
 Corrective Actions

 After the appropriate actions for particular situations
 have  been  identified,  several  factors  should be
 evaluated before a commitment is made to a particular
 approach.   The extent of the evaluation will depend
 on the magnitude of the release and the sensitivity of
 the public to the available options.  The factors that
 should  be considered  are  broken down  in four
 categories: 1) technical, 2)  institutional,  3)  human
 health and the environment, and 4^ costs. The fol-
 lowing subsections address each separately.
                                                  4-6

-------
Figure 4-3.    Potential initial response situations and associated corrective actions.




Situation
Ground-water contamination
Existing public or private wells
Potential future source of water supply
Hydrologic connection to surface water
Soil contamination
Potential for direct human contact:
nuisance or health hazard
Agricultural use
Potential source of future releases
to ground water
Surface-water contamination
Drinking water supply
Source or irrigation water
Water-contact recreation
Commercial or sport fishing
Ecological habitat
Other hazards
Danger of fire or explosion
Property damage to nearby dwellings
Vapors in dwellings

•a
§
E
Tank repair/re

0

0

0

•








0

0

£•
O
SB
Free product i

0

0






0
0








1
o
S2
Ground-water
and treatment

•

•






•
e











*




















e
2
i
n
CD
i
,
ii o-
co 3
Alternative/tre
central water s










0
0







Q.
„_, Q.
§ «
E §
co to
Altemative/tre
point-of-use w










0
0








&1
:= ^
=3 0)
Restoration of
water, and sew










0
0








£
m
c
Evacuation of
residents
















0

0




Restricted
egress/ingress
















m
9
0
Figure 4-4.    Permanent corrective-action process.
                                     Is
                                  additional
                               corrective action
                                  required?
                                                             4-7

-------
Figure 4-5.    Matrix of permanent corrective action categories for specific site problems (EPA 1985a).










Volatilization of chemicals into air
Hazardous particulates released to atmosphere
Dust generation by heavy construction or other
site activities
Contaminated site run-off
Erosion of surface by water
Surface seepage of released substance
Flood hazard or contact of surface water body
with released substance
Released substance migrating vertically
or horizontally
High water table which may result in ground-
water contamination or interfere with other
corrective action
Precipitation infiltrating site and
accelerating released substance migration
Explosive or toxic vapors migrating laterally
underground
Contaminated surface water, ground water,
or other aqueous or liquid waste
Contaminated soils
Toxic and/or explosive vapors that have been
collected
Contaminated stream banks and sediments
Contaminated drinking water distribution
system
Contaminated utilities
Free product in ground water and soils
Corrective action categories


"o
=
11
1.1
X T3
8 <5
TB *°

o S
E .±
CC §

•-


•

•

•










•
•


•




tO
c c
0) CO
If
S c
I*
T3 to
C 0
Co u.
„ to
& '-5
w 73
6 i



















•
•

* .
•

•




0
£




s
u_






•



•

















to
S*
to
to
1
CO
3
-6
c
2
C3










*


•





•





•





o



M
JD
W










*


•



*

*





•





C
o
To


3
















•



•
•









1
o
i

;»
-6
c
1
O




•














•





•




•si
1 E
o SS
O P
c *-

.5" cf


fl
•
•















•




*



»

I
C
%'

S "cL
•2 <2-
2 «

(^ i

























•
•

                                                          4-8

-------
4.2.2.1  Technical Factors
Performance

Two aspects  of  corrective  actions  determine their
desirability from a performance standpoint: effective-
ness and useful life.

Effectiveness refers to how well the corrective action
technology accomplishes its intended purpose, such
as  containment,  diversion, removal, destruction, or
treatment.  The effectiveness of available alternatives
should be determined either through design  speci-
fications or by performance evaluation (EPA 1985a).

Useful life refers  to  the  length of time a corrective
action technology can maintain the  desired level of
effectiveness.  Some  corrective action  technologies
deteriorate with time. Such deterioration can some-
times be slowed down through proper system opera-
tion and   maintenance,  but  the technology  may
eventually  require replacement. Each corrective ac-
tion alternative should be evaluated in  terms of the
projected service  lives of its component technologies
(EPA 1985a).

Reliability

Two aspects of LIST corrective action  technologies
that provide information about reliability are their opera-
tion and maintenance requirements (i.e., frequency
and complexity) and their demonstrated reliability at
similar UST sites  (EPA 1985a).  Demonstrated and
expected reliability is a way of measuring the risk and
effect  of   failure. The  engineer should  evaluate
whether  the  considered technologies  have  been
used effectively at similar sites; whether a combination
of technologies has  been used effectively; whether
the failure  of any one technology has an immediate
impact on receptors; and whether the alternative has
the flexibility to deal with uncontrollable changes at
the site.
Implementation

Another  important consideration  in the  selection
process is whether the corrective action can be imple-
mented with relative ease,  i.e., whether it can be
constructed or installed within the time frame required
to achieve the desired results.

4.2.2.2  Institutional Factors (Regulatory
Compliance)

The effects of Federal, state, and local standards and
other institutional  considerations on the implementa-
tion and  operational  timing should be analyzed.
Regulatory  programs  under  the  Resource   Con-
servation  and Recovery  Act (RCRA) Subtitle  I, the
Safe Drinking Water  Act  (SDWA), the Federal Water
Pollution Control Act (Clean Water Act  or CWA), the
                                                 4-
  Toxic  Substance  Control  Act  (TSCA),  and  the
  Comprehensive  Environmental Response  Compen-
  sation and Liability Act (CERCLA) will often have some
  impact on UST corrective actions (EPA 1985a).

  4.2.2.3  Human Health and Environmental
  Factors

  Safety

  Each corrective action alternative should be evaluated
  with regard to safety.  This evaluation should  include
  threats  to  the safety of  nearby  communities and
  environment as well as those to workers during imple-
  mentation. Safety factors to  consider are fire, explo-
  sion, and exposure to hazardous substances (EPA
  1985a).

  Human Health and the Environment

  Consideration should be given to the effect on human
  health  and any beneficial or adverse environmental
  impacts in  the selection of  corrective actions. The
  objective should  be to improve public health and the
  environment  to  the  extent   site-specific conditions
  permit (EPA 1985a).

  4.2.2.4  Costs
  Both capital  and operation  and maintenance costs
  should be estimated.  The estimate should reflect the
  scale of the planned corrective actions. Estimates for
  large, long-term corrective action options will  require
  more detail. Application  of  a  particular option  is
  determined by conditions both internal and external to
  the site, such as the location of underground utilities,
  depth of the water table, heterogeneity of subsurface
  materials, and the  site itself  (i.e., whether it lies in a
  remote area or a congested urban area)(EPA 1985a).

  4.3 Risk Analysis
  Before a commitment is made to a specific corrective
  action strategy, the levels of acceptable risk should be
  established to assist in the determination of criteria for
  meeting cleanup objectives. Data obtained during site
  characterization  activities  and  technology  perform-
  ance evaluations are used in this determination.  An
  analysis could be made to forecast concentrations of
  the released substance in  the media  of  concern.
  Based   on  toxicty  data  and regulatory and envi-
  ronmental  standards,  the   relative  human  and
  environmental hazard posed  by  exposure to  various
  concentration  levels  can be assessed.  The com-
  bination of the likelihood of exposure to the released
  substance and the hazard posed by such exposure
  provides a measurement of the relative risk  involved.
  This relative risk factor could then be used to establish
  cleanup goals and to assist in the selection and
  design of appropriate corrective-action technologies.
9

-------

-------
                                               Section 5
                                         Technology Profiles
This  section  describes the  corrective action tech-
nologies used to clean up leaking USTs. Under each
technology, it also presents information on availability,
applicability,  design,  and  construction  implemen-
tation. When available, general cost data updated to
mid-1986  dollars  are included.  The technologies
covered  are appropriate for response to  petroleum
and  hazardous  chemical   releases.  Many  of  the
technologies  apply  to  both  initial  response  and
permanent corrective actions.
Figure 5-1.   Flow chart of activities involved in the initial response to an UST release.
Several factors affect the selection and application of
the  technologies  described.  Figures 5-1  and  5-2
present flow charts of the selection procedures for the
initial response and  permanent corrective  actions,
respectively.

The intent of Figure 5-1 is to  show the technologies
that are generally associated with an initial response
action to assist the user in selecting the appropriate
actions. Selection, of  course,  will depend on evalu-
ation of the site and the characteristics of the product
released.
                                                  Release
                                                 discovered
                                                   Jt
Remove product
from tank.
Ik-

Repair or
remove tank.
                                                   Does
                                             release pose significant
                                               threat to human
                                               health and to the
                                                environment?
                        Monitor to ensure
                     acceptable environmental
                           quality.

1-
3roduct
overy
on Trench
p Systems
ilter Pumps
Separators

Review available background
information; contact
proper authorities.

|
Vapor Phase
Control
Ventilation of
Structures

i r

|


Ground-water
Recovery/Treatment
Supply
Well Points
Trenches
Suction Wells
Injection Wells
hfydraulic Barriers

i
r
Soil and Surface
Water Control
Soli Excavation
Sediment Removal
Diversion/Collection
                                                   5-1

-------
Figure 5-2.    Flow chart of permanent corrective action in response to an UST release.
                                                Does site still pose
                                             threat to human health and
                                                to the environment?
                                             Establish corrective action
                                               goals, objectives, and
                                                   standards.
                                                Select corrective
                                                action alternative.
                 Removal/
                 Excavation
                 of Soil and
                 Sediment
 Onsite and
   Offsite
Treatment and
  Disposal
                                               I
  Free
 Product
Recovery
                                    Vapor Migration
                                       Control,
                                      Collection,
                                     and Treatment
Ground-Water
  Recovery
  Systems
                        Surface
                        Water/
                       Drainage
                        Control
Subsurface
 Barriers
 In Situ
Treatment
                    Restoration of
                     Subsurface
                      Utilities
                                               Evaluate effectiveness
                                                   of alternative.
                                                 Design and implement
                                                  corrective action
                                                    alternative.
 Upon  completion  of  the  initial-response  activities
 (sometimes  during), it often  becomes apparent that
 more comprehensive site cleanup is needed.  When
 this occurs, the flow chart of procedures presented in
 Figure  5-2  is  applicable for  permanent  corrective
 actions. Permanent corrective measures are usually
 more costly and require more data for effective design
 and implementation as they are directed at arriving at a
 final solution to the problems created by a  leaking
 UST.
                            Figure  5-3  presents  applicable  corrective-action
                            technologies  for  four  hypothetical   UST  release
                            scenarios, two  related to petroleum products and two
                            to hazardous substances.  The reader will note that
                            some technologies apply to  all four [scenarios, (e.g.,
                            tank removal),  whereas others apply only to one
                            scenario (e.g.,  oxidation/reduction).    Most  technol-
                            ogies are adaptable to the release of both petroleum
                            and chemical products.
                                                      5-2

-------
Figure 5-3.    Potentially applicable corrective action technologies relative to release volume and chemical
              characteristics.








Technology









Removal/excavation of tank sol] $n.d sediment
Tank removal
Soil excavation
Sediment removal
Onsite and offstte treatment and disppfaj
Solidification/stabilization
Landfilling
Landfarming
Soil washing
Thermal destruction
Aqueous waste treatment
Deep well injection
Free product recovery
Dual pump systems
Floating filter pumps
Surface oil/water separators
Ground-water recovery systems
Ground-water pumping
Subsurface drains
Subsurface barriers
Slurry walls
Grouting
Sheet piles
Hydraulic barriers
fn situ treatment
Chemical treatment
Physical treatment
Soil flushing
Blostimulatlon
Ground-water treatment
Air stripping
Carbon adsorption
Biological treatment
Preclpitation/flocculation/sedlmentatlo
Dissolved air flotation
Granular media filtration .
Ion exchange/resin adsorption
Oxidation/reduction
Neutralization
Steam stripping
Reverse osmosis
Sludge dewatering
Vaoor miaratfon control collection anri tro
Passive collection systems
Active collection systems
Ventilation of structures
Adsorption
Flaring
Surface water/drainaoe controls
Diversion/collection systems
Grading
Capping
Revegetatlon
Restoration of contaminated water supplier
Alternative central water supplies
Alternative polnt-of-use water supplies
Treatment of central water supplies



>f contaminants




























i








atment









and sewer lines


Treatment of polnt-of-use water supplies
Replacement of water and sewer lines 1
Cleaning/restoration of water and sewer lines
If
If

3 at
Is
s
fa

to 2.

,
0




0


,












•



•
•

,
0












^



•
e

•

0


•
If
fl
.i 1
S (A
*•! «
If
o to
j!

f i

e
^
•

t

0

^
,


,
_
•

•
•

•
•

•



•
a

v
^











;
^











•


0>g
jl
li
a ^
£ w
£ v>
ll

„
^
•

v

^

0'
„


s
^
«

a
•

•
•

•


•
•
•

^
^







e



;
_



*
«
0
•

•
^
•
*


g.
aim
P
8 £
Si =
n K
I3
ti
11

,
^
•

c
0
^
0

0


^



•
•

•
•

«


•
•
•



























•
                                                           5\-3

-------
Although release characteristics and facility design
variables are taken into  account  in Figure 5-3, the
most  important considerations are the volume and
type of substance released and constraining site
features that can hinder or prevent effective imple-
mentation  of  a  technology. To  a lesser extent,
consideration was also given to the financial ability of
responsible parties to  implement certain technologies
and to the impacts on facility production or service
operations.  Case  histories and  members of the
American  Petroleum  Institute provided valuable in-
formation  relative  to  corrective action  technologies
currently being implemented.
Figure 5-3 is intended as a guide for the review of
technologies  given certain  site  characteristics; how-
ever,  it is  not  a technology  selection  matrix  for
corrective action responses.  Far more site-specific
analysis must be conducted before corrective action
technologies  and alternatives  are selected for per-
manent cleanup of a release from a leaking UST.

5.1 Tank Removal, Abandonment,
and  Rehabilitation Removal/
Excavation of Soil and Sediments
The excavation  and  removal  of  leaking tanks and
contaminated soil and sediment will apply to almost all
leaking underground  tank incidents.  These actions,
however, can become cost-prohibitive at great depths
or  in  complex hydrogeologic environments.  Factors
that must be considered in the decision-making pro-
cess  include the mobility of the wastes, the feasibility
of  onsite  containment or in situ  treatment, and the
cost  of disposing of  the waste (or tank) or decon-
taminating, it once  it has been excavated. Tank
rehabilitation Is a more appropriate approach for small
leaks in relatively new tank systems or for nonleaking
but high-risk tanks.

5.1.1  Tank Removal, Abandonment, and
Rehabilitation

5.1.1.1 General Description
 Determining whether  to abandon a tank in place or to
 remove it  normally would depend on the location of
the tank, local regulations, availability  of equipment,
 labor, materials, and  associated costs.  Current regu-
 lations,  however,  preclude  consideration  of  the
 abandonment of a leaking underground storage tank.
 Such action would not meet the  minimum standards
 outlined in the  1984 RCRA amendments and sub-
 sequent  EPA Subpart  I regulations.  Inasmuch  as
 corrective action will be required in response to all
 releases  of  regulated substances, even  proper  de-
 contamination and in-place abandonment of a leaking
 UST seems  impractible  (A leaking tank  would  be
 almost  impossible   to  decontaminate   in  place).
 Furthermore, USTs must be sealed in  a  manner that
 will prevent future releases.  This may be an impos-
sible task for in-place abandoned tanks.  Therefore,
only tank removal and rehabilitation are discussed in
this section.

5.1.1.2  Application/Availability
The most common form of rehabilitation for steel tanks
entails lining the interior with an epoxy-based resin,
isophthalic polyester-based resin, or some other ac-
ceptable coating.  Lining,  however,  does not  add
significant structural strength to a tank and should not
be  used under  the  following  conditions:  1) open
seams more than 3 inches long, 2) perforations larger
than about 1 inch  in diameter, or 3) more than five
perforations  in a given square foot of surface area
(Cheremsinoff et al. 1986). Tanks having any of these
shortcomings  should be  removed.  Tank lining is
performed by numerous  companies  and company
certification may be required in some States.

The repair or  rehabilitation of fiberglass-reinforced
plastic (FRP) tanks typically involved patching holes or
cracks in  the tank shell or end caps or repairing or
replacing  crushed  or cracked  ribs. The primary cri-
terion  typically used to  determine whether an FRP
tank should be  repaired or replaced is economics,
unless  the damage is  so great thiat  repair  is not
feasible or practical. Before rehabilitation of an FRP
tank is attempted, the tank manufacturer  should be
consulted (New York DEC, undated).

5.1.1.3 Design and Construction
Considerations

 Tank Rehabilitation

 Proper procedures for preparing the jtank for opening
 and entry are  essential to ensure  the safety  of all
 persons involved  in the tank  rehabilitation process.
 The following activities are especially important (API
 1981):

    Controlling sources of ignition

 •   Removal of liquid product

    Removal of flammable vapors

    Testing of vapor concentrations

 Guidelines for these activities can be found in the
 referenced API publication.

 If  no  manway or hatch  exists, an; 18 in. x 18 in.
 (minimum) hole can be cut in the tank top (steel) or
 tank end cap (FRP) with explosion-proof tools. Safety
 demands that persons entering an underground tank
 for habilitation activities be equipped with positive air
 displacement equipment with full facfe mask and wear
 a safety harness attached to a safety line.

 Proper preparation  of the tank  interior is critical to
 lining or  patching a tank.  This  entails the following
 steps (New York DEC, undated):    :
                                                 5-4

-------
  1)  Sludge removal-Sludge that has accumulated in
     the tank must be collected and  properly dis-
     posed of.

  2)  Surface inspections-After the sludge has been
     removed, the tank must be carefully inspected to
     ensure that it meets the structural  requirements
     for rehabilitation.
  3)  Sand blasting-The entire internal surface of the
     tank must be sand blasted  completely free  of
     rust, scale, and other foreign matter.
The  lining  material must be applied  as soon as
possible after the tank surface has been cleaned and
sandblasted. The coating must be  applied in strict
accordance  with manufacturer's specifications.  After
application,  the new lining must be tested  for the
presence of air pockets and pinholes and allowed to
cure fully before the tank is put back into operation.
American Petroleum Institutes Publication 1631  pro-
vides details on the "Recommended Practice for the
Interior Lining of Existing Steel Underground Storage
Tanks" (See Bibliography).

After  the tank  contents and  sludge  have  been
emptied, cracks or holes in the shell or  end cap of a
fiberglass-reinforced  plastic  (FRP)  tank can  be
repaired by  the following steps (New York DEC, no
date):
  1)  Surface preparation

  2)  Patching the damage

  3)  Curing the patch

Preparation of the surface and patching  the damaged
area  should include extending the  area of  repair
beyond  the  area  of damage. In all  cases, the tank
manufacturer should be contacted prior to the repair
of an FRP tank.

Tank Removal (AP11981)
The safe removal of underground tanks can be ac-
complished by taking the following eight steps:

  1)  Draining and flushing the piping into the tank.

  2)  Removing all flammable liquid from the tank that
     can be pumped out.   A hand pump may be
     needed to  remove the bottom few inches  of
     product.

  3)  Digging down to the top of the tank.

  4)  Removing the fill (drop) tube; disconnecting the
     fill gage, product, and  vent lines; capping or
     plugging open ends of lines that are not to  be
     used further.

  5)  Temporarily plugging all tank openings and com-
     pleting the excavation; removing the tank and
     placing it in a secure location; and  blocking the
     tank to prevent movement.
   Before degassing measures are undertaken, it is
   normally necessary to remove the tank from the
   ground because any product that may have
   leaked into the ground could reenter the tank.
   Extreme caution should be  used during this
   procedure.

6) Removing vapors.   The tank should  be condi-
   tioned by one of the following methods, or as
   required by  local codes,  to  ensure that no
   flammable, hazardous, or toxic vapors remain.

   a) If water is available and a suitable means for its
      disposal are available, the tank may be filled
      with water to expel vapors. While the tank is
      being filled with water, vapors  will flow out of
      the tank and  may surround the area.  If the
      tank overflows, purged product also may flow
      out.  Hence,  all normal  safety and pollution
      precautions regarding flammable, hazardous,
      or toxic liquids and  vapors must be observed.
      When the tank is  to be removed  from the
      premises, the contaminated water  should be
      removed and disposed of in accordance with
      regulations.

   b) If  the  method  described  in  (a)   is not
      practicable, the vapors in  the tank may be
      made inert by  adding solid carbon dioxide
      (dry ice) in the amount of 1.5 pounds per 100
      gallons of tank capacity.  The  dry ice  should
      be crushed and distributed evenly over the
      greatest  possible  area  to  secure  rapid
      evaporation. Skin contact with dry ice  should
      be avoided because  it may produce  burns.
      As the dry ice vaporizes, tank vapors will flow
      out of the  tank  and may surround the area.
      Hence,  all  safety precautions  regarding
      flammable, hazardous, or toxic vapors  should
      be observed.  It must be ascertained  that all
      of the dry ice has vaporized.

   c) Another alternate method is to ventilate the
      tank with air by use of a  small gas  exhauster
      operated with compressed air or  by other
      suitable means. The flow of air in through an
      opening near one  end of the tank and the
      discharge of the vapor-air mixture out of an
      opening near the opposite end will  quickly
      remove the vapor.  The  vapor concentration
      in the tank  can be  checked with suitable
      instrumentation .to determine when the tank
      is gas-free. While the tank is being ventilated,
      flammable, hazardous, or toxic vapors may
      flow from the tank, and proper handling or
      disposal procedures must be followed.

7) Plugging or capping of all holes  after the tank
   has been freed of vapors and before the  tank is
   moved from  the site. Screwed (boiler)   plugs
   should  be used to plug  any corrosion  leak
                                                 -5

-------
     holes.   One plug should have  a 1/8-inch vent
     hole to prevent the tank from being subjected to
     an  excessive  pressure differential caused  by
     extreme temperature changes.

  8)  Finally,  if size permits, securing the tank on a
     truck for transportation to the disposal site. The
     tank should be secured so  that  the  1/8-inch
     vent hole is located at the uppermost point  on
     the tank.

If a  tank remains  at the  site overnight or  longer,
additional vapor may be  released from liquid held in
the scale or sediment in the tank. Consequently,
tanks  should be  removed from  the premises  as
promptly  as  possible after these  procedures  have
been completed.

5.1.1.4 Costs

Tank lining costs about $2000 for a 4000-gallon tank
and  $4000 to $5000 for a 10,000-gallon tank.   An
additional cost of $500 to $1000 per tank is required if
a  manway  has to be  constructed  for tank  entry
(Cheremisinoff et al. 1986). Tank excavation/removal
will  usually  range from  $1000  to  $2000.  Tank
preparation  and testing prior to excavation will cost
approximately $1000, and tank disposal could run as
high as $2000 (Tusa 1986).  All costs given are in
1986 dollars.

5.1.2 Soil Excavation (EPA 1985a)

5.1.2.1 General Description

The quantity of excavation required for a leaking UST
occurrence depends on the volume of the  leakage,
the extent of the plume,  and the remediation method
chosen.  The minimum amount will be for exploration
work on the tank itself, and the total effort will be site-
specific.

Excavation efforts not only expose the tank, but also
permit removal of the  pollution  resulting from  the
release.   Contaminated soils and earth materials  are
excavated with  conventional construction equipment
such as backhoes, draglines, front-end loaders, and
even  shovels.  The  type  and   size  of  the leak
determines the type of equipment  required to effect a
satisfactory cleanup.  Generally, excavation should be
considered   (1)  when leaks are  small  and involve
wastes with relatively low toxicity, (2)  when a high
hazard to drinking water supplies exists, (3)  when
insoluble wastes  cannot  be  removed  by  pumping
alone, or (4) when long-term treatment would be  too
costly.
The  first step  in the  cleanup  process   involves
 excavation and proper disposal of the  contaminated
 soils.  Disposal could involve hauling the material to an
 approved hazardous waste dump site, incineration on
or off site, or some  other treatment technique (such as
solidification, encapsulation, or solvent extraction and
drying).  Solidification of excavated  materials can
ultimately lead to reduced disposal costs if laboratory
tests prove the solidified waste to be honhazardous.

Excavation is ineffective as a total removal technique if
the contaminants have already leached from the spill
site. Factors such as soil types, ground and surface
waters, and amount of precipitation after a spill can
influence the degree to which a material has leached.
Excavation of all contaminated substances (including
ground water and strata through which it flows) is rarely
performed because of the high costs associated with
the transport  and  treatment or  disposal of large
volumes of soil.

5.1.2.2 Application/Availability

Remediation of a leak site usually entails excavation
and removal of contaminated materials followed  by
land disposal  or  treatment. Although  no absolute
limitations are  placed  on the materials that can  be
excavated and removed, worker health and safety are
strong  considerations  in  the decision  to  excavate
explosive, reactive, or highly toxic ^materials. Other
factors that are considered include the mobility of the
material, the feasibility of onsite containment or in situ
treatment,  and the cost of disposing of or  decon-
taminating the soil after  it  has been excavated.  A
frequent practice  is to excavate and remove con-
taminant  "hot  spots"  and  to  use  other remedial
measures for less-contaminated soils.  Excavation and
removal apply to  almost  all  site conditions; however,
such actions  may  become cost-prohibitive at great
depths or in complex hydrogeologic environments.

5.1.2.3 Design and Construction
Considerations

Regardless of  the  kind  of  equipment  used  for
excavation and handling, certain  standard operating
procedures and safety practices should be followed.

 During the excavation of the contaminated soils, air
monitoring  should  be conducted  to determine the
presence of  unsafe  levels of various  hazardous
constituents.   Numerous  portable  direct  reading
instruments are available for this purpose.   These
 include:

    Combustible  gas  detectors for  measuring the
     lower  explosive  limit (lowest  concentration  of
     flammable gas that  will explode, ignite,  or burn
     when an ignition source exists).,

     Oxygen meters for measuring the percentage of
     oxygen.

     Photoionization detectors, flame ionization detec-
     tors, infrared  analyzers, and detector tubes for
     measuring gases and vapors.
                                                 5-6

-------
 As contaminated soils are excavated from the disposal
 area, they should be transferred to box trucks or to a
 temporary storage area, preferably a diked or bermed
 area lined with plastic or low-permeability clay.  A layer
 of absorbent material should be placed on the bottom
 of the temporary storage area. Gas analyzers are often
 used  to determine  the  approximate level of  con-
 tamination  in soils  so  they can  be segregated ac-
 cording to  their degree  of  contamination.  Pools of
 liquid wastes or solvent materials  should be removed
 promptly by the use of pumps.

 5.1.2.4  Costs

 Total  excavation costs  will depend on the amount of
 excavating  required. Table 5-1 presents some unit
 costs for excavating with different  pieces  of equip-
 ment.
 Table 5-1. Soil Excavation Costs*

 Machine               Daily Capacity (yd 3)
Cost per Day ($)
Backhoewith3.5-yd3
bucket
Crane with 3.0-yd 3
bucket
Wheel-mounted bucket
loader
Towed scraper, 10-yd 3
capacity
Dump truck, 12-yd3
capacity, 1/4-mile haul
1200
900
1480
440
356
2025
1335
1275
1800
515
  Data from EPA 1985a.
 5.1.3 Sediment Removal (EPA I985a)

 5.1.3.1 General Description

 Contamination of  the bottom sediments of streams,
 ponds, lakes, harbors,  estuaries, and  other  water
 bodies may occur via several different pathways. The
 most  likely  pathway  is via a  ground-water plume
 extending to nearby water  courses. Similarly,  con-
 taminated ground water may drain to surface  water
 courses,  and the transported  pollutants may  settle
 into or chemically bind with the bottom sediments.

 Remedial techniques for contaminated sediments
 generally involve removal  and subsequent treatment
 and disposal.   During the removal  of contaminated
 sediments, efforts must be made to minimize the
 threat  of  further  environmental  harm  through re-
 suspension of contaminants. Various techniques are
 available to  control this problem.  Control of turbidity
 (and contaminant resuspension) can be accomplished
 by modifying dredging equipment and using curtain
 barriers.  Temporary diversion of water  flow is also
, necessary. Measures such as containment and in situ
 treatment  offer an   alternative  to  removing  the
sediment; however, these measures are not yet well
demonstrated.

5.1.3.2  Application/Availability

The process of removing bottom sediments from a
water body is commonly  known as dredging.  This
process, which  has been used for many years to
widen or deepen harbors and navigable waters, more
recently has been applied to the removal of sediments
contaminated by hazardous substances.
Dredging involves four tasks:  1) the loosening or dis-
lodging  of  sediment by  mechanically penetrating,
grabbing, raking, cutting, drilling, blasting, or hydraulic
scouring; 2) a lift action accomplished by mechanical
devices such  as  buckets or by  hydraulic suction;
3) the transport of  dredged materials by pipeline,
scows, hopper dredges, or trucks; and 4) the disposal
of the dredged material.

Selection  of equipment and methods is  based  on
several practical considerations, including 1) type and
amount of  sediment to be dredged, 2) physical and
hydrologic characteristics of the dredging site, 3) dis-
posal options, and 4) availability of equipment.

5.1.3.3 Design and Construction
Considerations

Mechanical  dredging  involves  the  use  of  such
excavation  equipment  as  backhoes,   draglines,
clamshells, and bucket-ladder dredges. Draglines and
clamshells  used for dredging  are  usually  vessel-
mounted, but they can be track-mounted  and land-
based.

The main advantage of mechanical dredging is that it
allows removal of sediments at nearly in situ densities,
which maximizes solids content and minimizes the
scale of facilities required for transport, treatment, and
disposal   of the   dredged  material.    A major
disadvantage  of this  approach is  that   the direct
application of mechanical  force used to dislodge the
material can result  in high sediments resuspension
(and therefore  turbidity).   Dredging-of stream-bed
sediments may  be  an undesirable  corrective action
technique because  it will  almost   always  release
additional contamination   into  the  water.    Another
disadvantage is that mechanical dredging is relatively
ineffective in the removal of free or unabsorbed liquid
contaminants.   The  productivity of  this approach is
also characteristically low.
Mechanical  dredging generally  can  be  applied  in
relatively shallow streams and rivers with low  flow
velocities.  It can  also be used  to remove  con-
taminated sediments deposited on dry river banks or
in flood plains with less potential for resuspension of
the contaminants.

Hydraulic dredges  (including plain suction, dust-pan
cutter head,  and  hopper  dredges)  remove  and
                                                 5-7

-------
transport sediments in  liquid slurry form  by use of
diesel- or electricity-powered  hydraulic pumps with
discharge pipes. Slurries containing 10 to 20 percent
solids  by weight are  normally transported through
pontoon-supported  pipelines  for  distances up  to
several thousand meters.
A small  hydraulic dredge  known  as  the  Mudcat  is
readily available in the United States from the National
Car Rental Agency. The Mudcat can dredge a swath 8
feet wide by 18 inches thick in water as shallow as 19
inches and as deep as 15 feet. This hydraulic dredge
has been tested in a hazardous material cleanup role
and seems well suited to small cleaning jobs.

5.1.3.4 Costs
Estimated costs for dredging with a clamshell  ma-
chine, not including disposal of sediment, is $15/yd3
of sediment. The cost for dredging with a Mudcat is
estimated  to be  $5/yd3  of  material  removed, not
including disposal.

5.2 Onsite and Offsite Treatment and
Disposal of Contaminants
Onsite  and   offsite   treatment   and   disposal
technologies  discussed   in  this  section  include
solidification/stabilization, landfilling, landfarming, soil
washing, thermal  destruction, aqueous waste treat-
ment, and deep-well injection.  These technologies
are applicable  for both  onsite  and offsite   use.
Budgetary  constraints,  however,  will usually de-
termine whether onsite or offsite treatment/disposal is
more  feasible.   Transportation costs often exceed
actual treatment costs when a large volume of waste
(contaminated soil) is involved.

5.2.1  Solidification/Stabilization

5.2.1.1  General Description
Solidification and  stabilization methods  of treating
contaminated soils are applied to change the physical
or leaching characteristics of the waste or to decrease
 its  toxicity.   In the  solidification  process,  waste
 constituents are mechanically locked within a solidified
 matrix in the form of a crumbly soil-like mixture or a
 monolithic   block   similar  to  concrete.  Although
 solidification  improves  the waste handling or other
 physical characteristics of the waste, it usually has little
 effect  on  the chemistry of the  waste ;theref ore,
 leaching of waste constituents  may  occur in  the
 future. Solidification may be accomplished by the use
 of Portland cement or a lime-containing product (e.g.,
 fly ash,  ground  blast furnace slag or cement kiln dust).
 Solidification can also be accomplished by the use of
 an organic polymer. In this case, the waste is mixed
 with a prepolymer and solidifies  when a catalyst  is
 added.   In  a  third form, known as thermoplastic
 solidification, the waste is mixed with heated bitumin,
paraffin, or polyethylene; this mixture solidifies upon
cooling  (Radian  1983).   Stabilization  is a  chemical
reaction that fixes substances in a matrix that results in
a decrease in the solubility or toxicity of the hazardous
waste constituents  (EPA 1985a).  iWhereas stabil-
ization reduces  the solubility or mobility  of waste
constituents, it may  not improve the physical handling
characteristics.   Therefore,   solidification  and  sta-
bilization are often  combined to obtain the desired
effects of both processes.  This is accomplished by
com-bining alkaline earth materials  (such as  lime,
cement kiln dust, siliceous materials; or cement) with
organic polymers  and  proprietary chemicals (EPA
1985a).

Solidification/stabilization processes fall into  one of
the following categories:

  1) Cement-based
                                 t
  2) Lime-based

  3) Thermoplastic

  4) Organic polymer

  5) Self-cementing

The first four of these processes are  summarized in
Table   5-2,  along  with   their  advantages   and
disadvantages.

5.2.1.2  Application/Availability

Cement-   and   lime-based   processes capable  of
solidifying  a wide  range  of materials  are  readily
available.  The materials must be amenable to  forming
a slurry with water. Thermoplastic ana1 organic polymer
processes, on the  other hand, are often developed
on a waste-specific basis  and involve  the  use of
proprietary chemicals and procedures.

5.2.1.3  Design and Construction
Considerations

Cement-  and lime-based processes  require readily
available equipment such as chemjcal storage hop-
pers, chemical feeding equipment, mixing equipment
 (ribbon blenders and single- and double-shaft  mixers),
 and waste handling equipment (EPA 1985a). Thermo-
 plastic and organic polymer processes require special
 equipment and  trained operators, generally provided
 by vendors  who are  conducting some solidification
 proprietary processes.

 Lime-based  solidification also can be  conducted on a
 batch basis in drums.  In addition to chemical storage
 and batching equipment, drum  handling and mixing
 systems  are needed.  Some firms1 have  developed
 drum  kits. In this  type of  application, earth-moving
 equipment (backhoes,  shovels,  etc.)  are  required in
 addition to chemical storage and mixing equipment.
                                                  5-8

-------
 Table 5-2.   Solidification/Stabilization Processes*

 Process              Description
                                  Advantages
                                 Disadvantages
 Cement-based
Slurry of wastes and water
is mixed with Portland cement
to form a solid
                                                        Low hosts; readily available
                                                        mixing equipment; relatively
                                                        simp e process; suitable
                                                       foru;
                                      5e with metals
 Lime-based
 Thermoplastic
 Organic polymers
Waste is reacted with lime
and a fine-grained siliceous
material (fly ash, ground
blast furnace slag, cement
kiln dust) to form a solid
Waste is dried, heated, and
dispensed through a heated
plastic matrix of asphalt
bitumen, paraffin, or
polyethylene
Waste is mixed with a
prepolymer and a catalyst
that causes solidfication
through formation of a
sponge-like polymer matrix;
urea-formaldehyde or vinyl
ester-styrene polymers are
used
Low cost; readily available
mixing equipment; suitable
for power-plant wastes
(FGCi sludges, etc.) as well
as a wide range of industrial
wastes, including metals,
waste oil, and solvents

Less|increase in volume than
with cement- or lime-based
processes; reduced leaching
relatij/e to cement- or lime-
based processes; suitable
for radioactive wastes and
some industrial wastes
Suitable for insoluble solids;
very successful in limited
applications
Solids are suspended, not
chemically bound; there-
fore are subject to leaching;
doubles waste volume;
requires secondary con-
tainment; incompatible with
many wastes (organics,
some sodium salts, silts,
clays, and coal or lignite).

Increases waste volume;
may be subject to leaching;
requires secondary contain-
ment
Wastes must be dried before
use; high equipment costs;
high energy costs; requires
trained operators;
incompatible with oxidizers,
some solvents and greases,
some salt, and chelating/
complexing agents; requires
secondary containment

Pollutants are not chemically
bound, subject to leaching;
strongly acidic leach water
may be produced; requires
special equipment and
operators; some of the
catalysts used are corrosive;
harmful vapors may be
produced; incompatible with
oxidizers and some organics;
some resins are
biodegradable and
decompose with time
  Data from EPA 1985a.

5.2.1.4 Costs

Solidification/stabilization costs vary widely with  the
specific wastes  and  process  used; however, some
general guidelines  are  available.  Table  5-3 shows
approximate costs for some solidification processes.

Table  5-4  presents  another  cost  summary,  which
compares   costs for   in-drum  and  in  situ   stabil-
ization/solidification alternatives.

5.2.2 Landfilling

5.2.2.1  General Description

Landfilling  is  the  burial of wastes in  the  ground.
Hazardous waste landfills are constructed of lined cells
with leachate collection systems, gas control systems,
run-on/runoff controls,  and ground-water  monitoring
systems.  Landfilling has historically been the method
                                                          5-
                                         Table 5-3. Solidification Costs

                                         Solidification Agent      Design Basis
                                                 Cost ($)
                                         Cement-based



                                         Lime-based

                                         Thermoplastic





                                         Organic polymer





                                         Self-cementing
                           Cost of cement only;
                           transportation highly
                           site-specific

                           Site-specific
                           Cost for fixation of
                           FGD sludge
                58 to 81/ton
                0.03to0.15/ton

                Cost not available;
                contact Werner and
                Pfleiderer Corp.,
                Ramsey, or other
                vendors

                Cost not available;
                contact Todds
                Shipyards Corp.,
                Galveston, TX, or
                other

                2.00 to 2.75/ton

-------
Table 5-4.  Summary Comparison of Relative 1986 Costs of Stabilization/Solidification Alternatives*


                                                 	Plant Mixing §	
              Parameter
                                    In-drumt
              In situ?
              Pumpable
               Nonpumpable
              Metering and mixing
              efficiency

              Processing days
              required

              CosWton ($)

                Reagent


                Labor and per diem


                Equipment rental
              Used drums
              at $11/drum

              Mobilization-
              demobilization

              Cost of treatment
              demobilization

              Profit and overhead
Good
374
25.12
(9%)t

62.74
(23%)

45.63
(17%)

(59.25
(21%)

19.27
(7%)

$212.01
63.58
(23%)
                                                  Fair
21.84
(63%)

1.45
(4%)

1.47
(4%)
1.68
(5%)
6.91
(23%)
                            Excellent
                            10
21.84
(53%)

4.08
(10%)

4.18
(10%)
1.52
(4%)

31.62
9.54
(23%)
                                                                               Excellent
                                          14
21.84
(42%)

7.38
(14%)

8.03
(16%)
2.40
(5%)
                             39.65
11.90
(23%)
              TOTAL COST/TON
$275.59
$33.35
$41.16
                                                                               $51.55
              ' Data from Culiinane and Jones 1985; costs were updated to 1986 dollars by use of the 1986 ENR
                Construction Cost Index.                                                        ;
              t Assumes 49 gallons of untreated waste per drum and an average processing rate of 4.5 drums per hour.
              t Assumes wastes would be mixed by backhoe in a lagoon and left there. Site is located 200 miles from the
                nearest equipment.                               ,                            :
              § Assumes pumpable sludge has a daily throughput of 250 yd d/day and nonpumpable sludge has a i
                throughput of 180 yd 3/day. Site is located 200 miles from the nearest equipment.
              f Percentage of total cost/ton for that alternative.
 of choice for disposal of hazardous wastes; however,
 the 1984 Hazardous and Solid  Waste Amendments
 (HSWA) to RCRA  have reduced the  availability of
 landfilling as an option.

 5.2.2.2 Application/Availability

 The 1984 HSWA Amendments to RCRA established
 a hierarchy of hazardous waste management meth-
 ods. The legislation shows a strong  preference for
 treatment, recycling,  or destruction as opposed to
 landfilling. The Amendments require that EPA review
 wastes; if their disposal on land  is not safe, landfilling
 of  that  waste will be prohibited unless  no  other
 alternative exists. As this review of wastes proceeds,
 the landfilling of  more waste types is expected  to be
 prohibited.  Currently,  liquids are banned in landfills.
 Listed  dioxin- and  solvent-containing  wastes  also
 were prohibited  effective November 8, 1986.   The
                    "California List" of hazardous wastes goes into effect
                    July 8, 1987, unless EPA sets treatment levels.  This
                    list  includes   liquid  wastes  containing  specified
                    amounts  of free  cyanide,   metals,  polychlorinated
                    biphenyls, and halogenated organicjs.  In addition, the
                    limited number of permitted hazardous waste landfills
                    often requires that hazardous wastes be transported
                    over long distances for land disposal.

                    5.2.2.3 Design and Construction
                    Considerations

                    The 1984 HSWA Amendments to RCRA are specific
                    in  the design  requirements for hazardous  waste
                    landfills.  New or expanded landfills must have double
                    liners, leachate collection systems, and ground-water
                    monitoring systems.  The leachate collection system
                    must be above and between the liners.
                                                    5-10

-------
 5.2.2.4 Costs

 Landfilling costs vary with the type and toxicity of the
 waste and by disposal site. Many industrial wastes can
 be  landfilled for  $40 to  $85/ton.  Ignitable  wastes
 generally cost about $125/ton.  Fees for highly toxic
 wastes are about $245/ton (EPA 1985a).  These fees
 are in addition to transportation costs, which generally
 run about $3/mile for each load.

 5.2.3 Landfarming
 5.2.3.1 General Description

 Landfarming consists of applying  wastes to the soil
 surface, with or without mixing in the top layers of soil.
 The  application of  selected  wastes to  soil allows
 degradation or immobilization to  occur.   Additional
 agents may be applied to provide nutrients, to control
 moisture, or to accomplish chemical or biological reac-
 tions. Immobilization reduces the  rate of release of
 contaminants from soil to reduce exposure. The mech-
 anisms of sorption,  ion  exchange,  and precipitation
 produce immobilization (Sims et al. 1985). Metals can
 be immobilized through  sorption by the addition  of
 adsorbents (e.g., straw, sawdust, peanut hulls, bark,
 and activated  carbon) or metal complexing  agents
 (e.g., Tetran), which are  tilled into the soil with the
 waste (Sims et al.  1985).  Some organics can be
 sorbed onto soil by the  addition of organic material
 such as sewage sludge; this is most effective in soils
 such as sands or  gravels, which are naturally low in
 organic matter. Immobilization through ion exchange
 is accompished by the landfarming of wastes onto clay
 soils or by the addition  of  synthetic  resins (e.g.,
 Chelex 100 or Dower 1-X8) or zeolites to the soil.
 Zeolites are crystalline-hydrated alumino silicates used
 to adsorb heavy metal cations.  Precipitating  agents
 such as lime, limestone,  trebel superphosphate, and
 calcium sulfide can be added to heavy metal wastes
 and soils to form insoluble metal precipitates.

 Chemical  degradation includes  oxidation, reduction,
 and polymerization reactions to render the wastes less
 toxic or less mobile. Oxidizing agents (e.g., ozone
 and hydrogen  peroxide)  or reducing agents  (e.g.,
 catalyzed  metal powders  of iron, zinc, or aluminum)
 can be added to soils by spray, injection, or tilling. For
 example, ferrous iron can be added to surface soils to
 reduce hexavalent  chromium to the less toxic trivalent
form. Organics that can be polymerized (e.g., styrene,
vinyl  chloride,  isoprene,   acrylonitrile,  or  methyl
 methacrylate) can  be treated by mixing them in soil
with iron and sulfates.  Chemical  degradation tech-
niques in soil should be  tested in laboratory bench-
scale and  field pilot-scale  experiments  before full-
scale landfarming is attempted.

Some wastes may biodegrade naturally in  soils, or
biodegradation  may  be  accomplished  by  adding
                                                 5-11
 specific microorganisms selected for their ability to
 degrade a given  waste type.   In either case, land-
 farming treatment consists of applying the waste to
 the soil and optimizing degradation through addition
 of nutrients, water, air, or agents to control soil pH.

 5.2.3.2 Application/Availability

 Landfarming normally is limited to organic wastes that
 are biodegradable, such as petroleum sludges, and to
 metals that can be  immobilized in soil.  Hazardous
 waste regulations require  that a demonstration pro-
 gram  be conducted before a permit can be issued.
 For some wastes,  laboratory bench-scale testing must
 be  performed prior to field  studies.  Landfarming
 operations  also must  comply with monitoring reg-
 ulations.

 5.2.3.3 Design and Construction
 Considerations

 Landfarming can  involve  the application of wastes
 onto the surface only, or it can involve tilling the waste
 into the first few inches of soil.

 Solid or semisolid  wastes or reagents can be applied
 with truck-mounted,  hitch-mounted,  or tractor-drawn
 spreaders.   Liquids  can  be  applied  with  hydraulic
 sprayers or subsurface injectors.  Reagents can be
 mixed with plows, rotary  tillers, disc harrows,  spike
 harrows, or spring-tooth harrows (Sims et al. 1985).
 Moisture can be added by sprinklers, irrigation pipes,
 or trenches. Aeration is accomplished by tilling.

 5.2.3.4 Costs

 Costs  vary widely with the cost of the land, labor, and
 reagents required  at a given site.  Using a manure
 spreader costs about $5/yds.  Other significant  costs
 include:

     Purchase of the land

     Excavation of wastes

     Development and implementation of a health
     and safety program

     Sampling and analysis of the soil

 5.2.4  Soil Washing

 5.2.4.1  General Description

 Soil washing is the  extraction of contaminants from
 excavated soil by mixing the soil with water, solvents,
 surfactants, or chelating agents.  The contaminated
water  or  solution  is then treated  for  removal of
contaminants.  Heavily  contaminated soils are  com-
 monly   treated  several   times  in   a  multistage
countercurrent treatment system. A similar process
for in situ treatment of soils, referred to as soil flushing,
is described in Subsection 5.6.1.

-------
5.2.4.2  Application/Availability
Soil washing is applicable to the treatment of heavy
metals,   halogenated   aliphatics  (e.g.,  trichloro-
ethylene,  trichioroethanes,   chloroform), aromatics
(e.g., benzene, toluene, cresol, phenols), and volatile
hydrocarbons such  as gasoline. Surfactants  have
been  used  successfully to  remove  hydrophobia or-
ganics from  soils  (e.g., polynuclear aromatic hydro-
carbons, aliphatic  and aromatic hydrocarbons, poly-
chlorinated   biphenyls,   and   chlorinated   phenol
mixtures).
Soil washing is an appropriate technique  when a
single waste type is involved.  Soils that are most
amenable to soil washing are those with a low organic
content, a  low cation exchange  capacity, and a
permeability greater than  1x1f>4  cm/s.  A sandy
porous soil is more easily treated than clay or silt.

5.2.4.3  Design and Construction
Considerations

All soil-washing systems are mobile and are set up at
the contaminated  site, as the cost of moving the soil
would make this system uneconomical. A mobile soil-
washing/wastewater-treatment system developed by
the EPA has been demonstrated at hazardous waste
sites. A flow diagram of this EPA system is shown in
Figure 5-4.  Several variations on this design that are
currently being used or demonstrated for cleanup of
hazardous  wastes  at Superfund  sites have  the
potential for use at UST sites as well.
                5.2.4.4 Costs
                To date, much of the soil washing has been done on
                an experimental basis. The soil-washing system used
                at a Superfund site in Wisconsin cost about $150 to
                $200/yd3; much of this amount was for the  water-
                treatment system. (These are 1986 costs.)


                5.2.5  Thermal Destruction

                5.2.5.1 General Description
                Thermal destruction is the high-temperature oxidation
                of recovered product and wastes.  Under controlled
                conditions of  time,  temperature, and  turbulence,
                wastes can be degraded to simple end products such
                as carbon dioxide (COa),  water vapor, sulfur dioxide
                (SO2), nitrogen oxides (NOX), hydrochloric acid (HCI)
                vapors, and particulates.  Many types of incinerators
                can be  used for thermal destruction of  waste oils,
                recovered products from leaks or spills that cannot be
                reused, and other wastes.  The most  commonly used
                incinerators (liquid-injection, rotary-kiln, fluidized-bed,
                and multiple-hearth) are described inTable 5-5.

                In addition to these standard technologies, emerging
                technologies  for thermal  destruction  include  the
                following:  molten-salt, wet-air oxidation, plasma-arch
                torch,  circulating-bed,  high-temperature  fluid wall,
                pyrolysis,  supercritical water,  advanced  electrical
                reactor, and vertical tube reactor.
Figure 5-4.   Flow diagram for EPA mobile soil washer (EPA 1985c).
               Contaminated
                   Soil  —1
                                                  1
                 CleanAir
                 Discharge
                                                Air Cleaner
Drum Screen
Water Knife
Soil Scrubber
                                     ^Recycled | i
                                    Stripper Spray
                          Oversize
                           Non-soil
                          Materials
                          and Debris
                                MakeupWater
                                                      T
                                                  Spent Carbon
                                                  5-12

-------
 Table 5-5,   Summary of Commonly Used Incineration Technologies*


 Type              Process Principle
                             Application
                                                     i
                         Combustion
                         Temperature
                                                                                          Residence Time
 Single-chamber/
 liquid-injection
 Multiple-hearth
 Fluidized-bed
                   Waste is burned in a rotating
                   refractory cylinder
Wastes are atomized with
high-pressure air or steam
and burned in suspension

Wastes descend through
several grates to be burned
in increasingly hotter
combustion zones

Waste is injected into an
agitated bed of heated
inert particles.  Heat is
efficiently transferred to
the wastes during combustion
                             Any cohibustible solid,
                             liquid, or gas
Liquids
and slurries that
                                                can be pumped
                                                Sludges and granulated
                                                solid wastes
                                                Organic liquids, gases,
                                                and granular or well
                                                processed solids
1500° to 3000°F     Seconds for gases;
                 hours for liquids and
                 solids

1300° to 3000°F     0.1 to 1 second
                         1400° to 1800°F     Up to several hours
                         1400" to 1600°F     Seconds for gases and
                                          liquids; minutes for
                                          solids
  Data from State of California 1981.


 5.2.5.2  Application/Availability

 Liquid-injection incinerators can thermally degrade  a
 wide variety of wastes (pumpable liquids or sludges
 and gaseous wastes). A liquid or sludge waste must
 be atomized before it is burned.  This is accomplished
 by passing it through burner nozzles en  route to the
 incinerator. The burner nozzles are usually designed
 to  accommodate  specific wastes.   Liquid-injection
 incinerators have  been  used for the destruction of
 polychlorinated  biphenols  (PCBs),  solvents,   still
 bottoms, reactor bottoms,  motor  and industrial  oils,
 lacquers, chemical warfare agents, polymer wastes,
 and  pesticides (EPA 1985a).   Wastes with  a  high
 content  of heavy metals or other inorganics are not
 suitable for liquid-injection  incinerators;  neither  are
 wastes high in materials that can clog the nozzles or
 wastes  prone to violent decomposition (e.g.,  gaso-
 line,  peroxides,  and explosive  wastes)  unless  the
 undesired characteristic can  be  modified by  pre-
 treatment (e.g., blending or precipitation of the  heavy
 metals). A supplemental fuel is sometimes needed to
 ensure  complete  combustion,  particularly  for   the
 highly aqueous wastes  sometimes encountered at
 ground-water  cleanup sites.  About 219 liquid injec-
 tion incinerators are in service in the United States.

 Rotary-kiln incinerators are designed to handle a wide
 variety  of wastes  separately  and in combination
 without  any pretreatment,  including gases, liquids,
 and solids.  They have been used to destroy RGB's,
tars,  munitions, chemical warfare  agents, polyvinyl
 chloride  wastes, and  solvent  reclamation bottoms
 (EPA 1985a).  It is the method of  choice  for thermal
destruction of mixed (solid and liquid) wastes.
                                    About  42 rotary kiln  incinerators  are  currently  in
                                    service. The EPA has developed a mobile rotaiy kiln
                                    for destruction of PCBs.

                                    Multiple-hearth  incinerators  are widely used  for the
                                    destruction of  municipal  sewage  sludge and coal
                                    wastes. Although these incinerators can be used for
                                    all  forms of wastes, including solids, tars, sludges,
                                    liquids, and gases, they are best suited for sludges.
                                    Pretreatment of some wastes may  be necessary
                                    before incineration. Wastes containing large amounts
                                    of ash  create material-handling problems due to the
                                    formation  of  slag. These incinerators also  do not
                                    provide a  preferred alternative for wastes  requiring
                                    very  high  temperatures  or significant  amounts  of
                                    supplemental fuel. They can, however, handle  wastes
                                    that need to be evaporated.

                                    Fluidized-bed incinerators can be used for thermal
                                    destruction liquid, solid,  and  gaseous  combustible
                                    wastes; however, they are commonly used for slurries
                                    and sludges such as wastes from oil refineries and
                                    pulp and paper mills. Some wastes  must be dried,
                                    shredded,  or   sorted  before  incineration.   These
                                    incinerators are suitable for wastes high in moisture or
                                    ash.  Only about  nine  fluidized-bed incinerators are
                                    currently being  used to destroy wastes; thus avail-
                                    ability is limited by both geographical proximity and
                                    waste suitability.

                                    5.2.5.3  Design and Construction
                                    Considerations

                                    Thermal destruction of hazardous wastes is a complex
                                    technology that  requires  skillful operators,  and ex-
                                    perience is required for selection of a system.  Waste
                                    storage,  mixing, and  pumping equipment  is often
                                    needed, particularly  for liquid-injection incineration,
                                                  5-13

-------
which  requires  a homogeneous waste  and  steady
flow. Air pollution by particulates, NOX> SO2l and other
products must be controlled.  Afterburners are often
used to destroy organic byproducts. Ash disposal is
also necessary.

Mobile units may be the most practical where sites are
remote from stationary incinerators or when wastes are
particularly toxic or difficult  to  handle or transport.
Some  thermal destruction technologies may  require
testing on a lab- and field-scale prior to their full-scale
application.

5.2.5.4 Costs
Thermal destruction costs vary with the waste type
and volume; some estimated costs are presented in
Table 5-6.
Tablo 5-6.  Approximate Costs for Hazardous
          Waste Incineration
Type of Waste
Cost ($/ton)
 Drummed                           130 to 425
 Lkjuids                             55 to 425t
 Relatively clean liquids with high Btu value    14 to 56t
 Solids and/or highly toxic liquids           420 to 842
  1986 dollars.
 t Some cement kilns and light manufacturers pay for these
  relatively clean, high-energy-value wastes.

 5.2.6  Aqueous Waste Treatment

 5.2.6.1 General Description

 Aqueous waste streams  resulting from cleanup of
 leaking USTs may require treatment.  These include
 waste plumes from the tank, recovered tank contents
 or tank  residue,  ground  water or  surface  water
 contaminated by the leak, or water contaminated by
 tank or other equipment cleanup.  Treatment tech-
 nologies  vary  with  the  waste  type,  volume,  and
 concentration.   Aqueous  waste  treatment  technol-
 ogies include the following:

      Air Stripping
      Carbon Adsorption
      Biological  Treatment (Aerobic or Anaerobic)
      Precipitation/Flocculation/Sedimentation
      Dissolved  Air Flotation
      Granular Media Filtration
      Ion Exchange/Resin Adsorption
      Oxidation/Reduction
      Neutralization
      Steam Stripping
      Reverse Osmosis
      Sludge Dewatering

  Detailed   discussions  of  each  are  presented  in
 Subsection 5.7  (Ground-Water Treatment).
Aqueous waste treatment can be conducted on site
or off site.   Onsite  alternatives include the use of
mobile  treatment  units  or  the  iConstruction  of
stationary systems.  These  are  often  used for site
cleanup. The waste can be fully treated on site and
discharged to the ground or to surface water, or it can
be  partially  treated  on  site and discharged to  a
municipal wastewater treatment plant.  The feasibility
of the latter  alternative depends on'the toxicity and
treatability  of the waste  and  the capacity  of the
wastewater treatment plant  to accept the pretreated
wastewater.  Aqueous waste can also be collected
and hauled to an offsite treatment facility. The latter is
generally an expensive alternative for aqueous wastes
because of their weight and volume.

5.2.6.2  Application/Availability   !
See Subsection 5.7

5.2.6.3  Design and Construction
Considerations

See Subsection 5.7                I

5.2.6.4  Costs
See Subsection 5.7

5.2.7   Deep-Well Injection

5.2.7.1  General Description
Deep-well  injection involved the drilling of a well and
the injection of wastes into the well, the wells are dug
in  geologic formations selected with the purpose of
permanent containment of the wastes therein.  The
wastes are injected under pressure to displace native
fluids. The volume and porosity of the  geologic strata
used for deep wells must be sufficient to accept the
wastes, and formations capable of preventing migra-
tion of the waste must surround the well.

5.2.7.2 Application/Availability

In  addition to having a suitable geology to accept and
contain  the  wastes,  areas selected   for deep-well
injection must also have little seismic activity. Of the
 159 deep wells in the United States,  60 percent are in
 Region  VI  (Texas,  Oklahoma, New Mexico, and
Louisiana); availability is thus limited.

Only liquid wastes can be  disposed of by deep-well
 injection.  Wastes that are viscous or high  in sus-
pended solids are unsuitable. This disposal method is
therefore not recommended for disposing of wastes
from leaking underground storage tanks because the
wastes  are  often mixed  with  soil.    Contaminated
 ground water or surface water, liquids  recovered from
 a leaking tank, or wastes recovered in  liquid form with
 little  soil,  however,  may  be amenable to deep well
 injection.  Suitable waste  types  include  dilute  and
                                                  5-14

-------
concentrated acids or bases, heavy metal solutions,
inorganic solutions,  chlorinated  hydrocarbons,  sol-
vents, and high Chemical Oxygen Demand (COD) and
Biochemical   Oxygen   Demand  (BOD)   organic
solutions.
5.2.7.3 Design and Construction
Considerations

Deep wells are limited to specific geological forma-
tions.  Deep wells used for hazardous waste disposal
generally range in depth from a few hundred to a few
thousand feet.   Figure  5-5 shows  a diagram of a
typically constructed well.

Figure 5-5.   Diagram of a conventional industrial injection well
      Fresh-Water-Bearing
          Surface Sands
             and Gravel
       Impermeable Shale
    Confined Fresh-Water-
       Bearing Sandstone
       Impermeable Shale
       Impermeable Shale
    Permeable Salt-Water-
      Bearing Sandstone-
          Injection Zone
       Impermeable Shale
               5.2.7.4 Costs

               Approximate  costs for deep-well  injection  of  oily
               wastewater (in 1986 dollars) are $0.06 to $0.16/gal or
               $17 or $43/ton.   Approximate costs for toxic rinse
               water are $0.05 to $1/gal or $141 to $281/ton.

               5.3  Free Product Recovery

               Sometimes the material released from a leaking UST
               can be  recovered.   This is especially  true'when
               gasoline and petroleum products are involved.  The
               success of any recovery effort hinges on the hydro-
               geology of the site,  the  properties of the material
               involved, and the  quantity released.   Technologies

             completed in sandstone (EPA 1985a).
                                                             Presure Gauge
                                                            Wellhead Pressure
                                                       Pressure Gauge
                                                            Opsi
                                   Domestic
                                     Well
Public-Supply
   Well
                    Surface Casing Seated  -Xvlvl
                    Below Fresh Water     .".".v.v.
                    and Cemented to Surface v-Xv'v
                    Inner Casing Seated in or
                    above Injection Zone
                    and Cemented to Surface

                    Injection Tubing ;"-.".--",

                    Annulus Filled with Xvlvlv
                    Noncorrosive Fluid .v.'lvlv;
                                                            Packers to Prevent Fluid"-]
                                                            Circulation in Annulus   "~
             Open-Hole Completion:
             in Competent Strata
.v.v.v.v.v.v.v.jL-.v.v.vs Injected Contaminated':
vXvXvXOXvOH™;vi Fluid              -
                                                  5-15

-------
applicable to the recovery of materials released from
USTs are discussed here.

5.3.1  Dual-Pump Systems

5.3.1.1  General Description
Dual-pump systems can be used successfully when
the leaked material floats on the water table.  In these
systems, one pump draws down the water table and
creates a cone of  depression into  which the  free-
floating material flows. A  second pump transfers the
free product to the surface for recovery (Figure 5-6).

5.3.1.2 Application/Availability
Dual-pump  recovery systems  have  been applied in
numerous free-product incidents.  This arrangement
reduces or eliminates the volume of water that must
be separated from the product and permits pumping
of uncontaminated  ground water to a point of free
discharge (AP11980).

Dual-pump  systems are  readily available and their
success depends primarily  on the  plume  definition
and the hydrogeology of the leak site.
                                    5.3.1.3 Design and Construction
                                    Considerations (AP11980)

                                    The objective of the drawdown well is to establish a
                                    depression in the water table that prevents the oil from
                                    spreading and concentrates it for removal. The rate at
                                    which fluid is withdrawn and the permeability of the soil
                                    determine the  size  and rate  of development of the
                                    depression.                        ;

                                    Because permeability varies,  the depression-forming
                                    process differs in each area. When enough infor-
                                    mation is known about an aquifer, the behaivior of the
                                    depression cone some distance from the well can be
                                    calculated quite accurately. Within  100 to 200 feet of
                                    the well, however,  such calculations  are  unreliable
                                    because of local characteristics of the aquifer.

                                    The flow rates of the pumps used should meet the
                                    maximum gallons  per minute expected to be required
                                    to establish the depression cone for the desired prod-
                                    uct recovery rate.  Flow rates can be reduced by valve
                                    controls, if necessary.

                                    When  ordering  pumping  equipment, the  supplier
                                    should be informed of its proposed use so  that the
Figure 5-6.   Dual-pump recovery system (OH Recovery Systems, Inc., undated).

         Water Table
          Depression

                    Water Table
                    Depression
                   Pump Controls
                                                  Oil/Water
                                                 Separation
                                                    Probe Scavenger
                                                    Control Assembly
 Clean
 Water
Output
                                                                                  Oil-Recovery
                                                                                     Tank
                                                             Perforated Well
                                                             Casing Permits
                                                              Row of Oil and
                                                              Ground Water
Oil-Contaminated Soil
               Water
               Table
           Depression
               Pump
                                                             Probe
                                                             Scavenger
                                                             Pump
                                                 5-16

-------
 proper seals,  impellers, gears, etc., are  provided.
 Most fractional horsepower pumping equipment has
 plastic, fiber, or metal impellers and gears, which may
 not be compatible with certain hydrocarbons.

 When positive-displacement suction pumps are used,
 protective screening or a filter ahead of the pump is
 needed to protect the pump from dirt or debris. High-
 speed  rotary pumps should be avoided when  com-
 binations of oil and water are to be pumped because
 such pumps encourage formation of oil-water emul-
 sions, which are difficult to separate.

 When recovery operations are expected to  be ex-
 tensive, an automated system should be  designed
 and installed.  Automated systems will reduce the
 number of man-hours required to operate and monitor
 the recovery operation.  They also can maintain the
 continuous  depression  cone necessary to contain
 and recover the product.

 Precautions  should  be taken  in  the  design and
 installation of recovery systems to eliminate the risk of
 explosive vapors in the area  of pumping operations.
 One technique is to keep the pump unit submerged
 in liquid at all times. The electrical cable for pumps and
 automated float control  systems must be  made of
 materials that are unaffected by the liquids or  vapors
 that will be  encountered.  All cables connecting the
 pumps  and  the power source should be continuous
 (no intermediate connections).   The power  source
 connection should be at least 5 feet from  the well
 opening.

 Electrical pumps or gasoline-powered engines  should
 not be used within 5 feet of the recovery well opening
 unless they are classified as explosion-proof for Class
 1, Division I, Group D, use in accordance with  NFPA-
 30.  Suction pumps used on volatile products  from
 test or  recovery  wells  should  be explosion-proof
 regardless of their distance from the well.

 Depending on the extent of the spill, pumping may be
 required for an extended period of  time.   Ideally,
 pumping  should continue  through  several  fluctu-
 ations of  the water table and should be abandoned
 only after the mobile product has been reduced to an
 acceptable level.

 Casing  is necessary in recovery wells. Any  steel,
 poly vinyl  chloride  (PVC),  or other strong  pipe  or
 tubular product  may be used.  The use of PVC  pipe is
 usually recommended, as it is readily available, light,
 easy to handle,  relatively inexpensive, and available in
 diameters of 1  to 18 inches.  This  pipe can be pur-
 chased  factory-slotted  (known commercially  as "well
 screen"),  or perforations may be  cut  in the field.
Acrylonitrile-butadiene-styrene (ABS) pipe should not
 be used, as it tends to become brittle with weathering
 and dissolves  upon contact with petroleum  hydro-
carbons.
                                                 5-
  Proper installation of a recovery well is critical to its
  success.  It should be large enough in diameter to
  accommodate the pumps and controls  designed for
  recovery operations and extend to a depth below the
  water table.  The depth will depend on the aquifer and
  soil characteristics, the area of contamination, and the
  area of  the cone of  depression. To eliminate ex-
  penditures for wells  or  equipment that   may  be
  unnecessary or may not perform as planned,  recovery
  well design should be left to professionals.

  5.3.1.4  Costs (EPA 19853)

  Some of the factors that determine recovery well costs
  are the  geology, the characteristics  of the con-
  taminated and naturally occurring ground water, the
  extent  of contamination,  the duration  of  pumping
  periods, local wage rates, the availability of  supplies
  and equipment,  and the  electrical power  required.
  Costs  associated  with   a   well  system   can be
  categorized  as mobilization  costs,  installation  and
  removal costs, and operation and maintenance costs.
  Because of these variables, the cost of well installation
  can vary considerably from site to site. For an accurate
  estimate of the costs associated with the installation of
  a well system,  as much  information as is  available
  should be obtained and evaluated prior to system
  design.

  Removal costs probably will be incurred at all sites at
  some  point  when pumping  is  no  longer  required.
  These costs  can be offset somewhat by the salvage
  value of the removed equipment.

  The  cost ranges  for  well  pumps and   screens
  presented in Table 5-7 give  some  indication of the
  cost of recovery wells.
  Table 5-7.  Product Recovery Costs*

  Item              Description
                      Cost Range ($)
  Submersible pumps
   4-inch pump
   Control boxes

   Magnetic starters

   Check valves

   Well seals

 Well screens
Pumping depths: 900 ft
Horsepowers: 1/3 to 3 hp
Capacities: 50to2000gph
                  Stainless steel, 1 1/4- to
                  36-inch diameter

                  Low-carbon steel, 1 1/4- to
                  36-inch diameter

                  PVC plastic, 1 1/4- to
                  12-inch diameter
435 to 1540




7510155

16510260

1510430

20 to 125

34 to 555/ft



18tOl61/ft



10to62/ft
                                                      Data from EPA 1985a.
17

-------
5.3.2  Floating-Fitter Pumps
5.3.2.1  General Description

Floating-filter pump systems are  a variation of dual-
pump systems. Dual-pump systems utilize two pumps
(one located well below the water table to form a cone
of depression  and  a contaminant recovery pump
floating on the water surface).  In floating-filter sys-
tems, the contaminant recovery pump is replaced with
a floating filter which  acts as the intake for a suction
pump  located  at ground  level   (Figure 5-7).  The
floating-filter  system  has a product-recovery pump
with a moving inlet that is free to follow the level of
product floating in the cone of depression. The inlet
is on a buoy with an oil/water filter built into it. A water-
level  probe,  which controls the ground-water level
depression  pump, is the  only  water-level control
required.

5.3.2.2 Application/Availability
Floating-filter  pumps  are  used  to  recover floating
contaminants from ground  water.  They are widely
commercially available.
           5.3.2.3 Design and Construction
           Considerations

           Design and construction considerations for floating-
           filter pump systems are the same as those for dual-
           pump systems; however,  larger well  casings  are
           required to accommodate the size of the floating filter.

           5.3.2.4 Costs
           Costs of floating-filter pump systems are comparable
           to those of dual-pump systems.

           5.3.3 Surface Oil/Water Separators

           5.3.3.1 General Description
           Leaked material brought to the surface will be mixed
           with water and must be  separated.  Simple gravity
           separators are used for  this purpose, and this  is
           usually the first step in an  above-ground  product-
           recovery process.   Further purification of recovered
           product is normally done offsite.
Figure 5-7.   Floating-filter recovery system (Oil Recovery Systems, Inc., undated).

          Water Table
          Depression

                    Water Table
                    Depression
                   Pump Controls)
                          Oil/Water
                         Separation
                     Scavenger
                     Control and
                   Pump Assembly
                  Clean
                  Water
                 Output
                Oil-Contaminated Soil
   Perforated          \
/ Well Casing \  XpH
                                              'Probe
                                        0
      Water Table
   , Depression Pump
                                               Oil-Recovery
                                               Tank
                                                                          Water Table;
                                                   5-18

-------
5.3.3.2 Application/Availability
Gravity separators are simple pieces of equipment that
provide a space where a water/product mixture can
separate naturally by gravity. This equipment can be
purchased from suppliers or can be field-fabricated.

The efficiency of these separators is a function of the
difference in densities between the leaked material
and the solubility of the material in water. The larger
the density difference and the less water soluble the
material is, the better and quicker the separation will
be.
5.3.3.3  Design and Construction
Considerations

A typical oil/water separator is shown graphically in
Figure 5-8.  Although designs vary, they all will have
the amenities shown  here, e.g., siphon break, vent,
drain spigots.

If necessary, some simple bench-scale tests can be
run to determine separability of the recovered prod-
uct/water mixture. The separation tank can then be
sized from tests that measure settling time for a given
volume based on the flow rate for the mixture coming
from the recovery well. Separators should be sized to
accommodate the maximum flow rate expected.

Once the product has been  separated, a means of
removing it and the water from the separator will be
required. The recovered product can be drummed or
pumped into tank trucks and moved offsite for further
processing.  The water also will have to be disposed
of, and it may require  further processing before it can
be used or returned to the well.
Figure 5-8.   Oil/water separator (AP11980).
                                                     5.3.3.4 Costs

                                                     Oil/water separators  are  available on the market in
                                                     various sizes and materials of construction.   Simple
                                                     field-fabricated units, which cost a few hundred dol-
                                                     lars,  represent  one  extreme.  Elaborate coalescing
                                                     plate separators, which can cost  up to several hun-
                                                     dred  thousand dollars  represent  the  other extreme
                                                     (Environmental Law Institute 1985).

                                                     5.4   Ground-Water  Recovery  Systems

                                                     Control of ground-water contamination involves one
                                                     of four options: 1) containment of a plume; 2) removal
                                                     of a plume after measures have been taken to halt the
                                                     source of contamination; 3) diversion of ground water
                                                     to prevent clean ground water from flowing through a
                                                     source of  contamination or to prevent contaminated
                                                     ground water from contacting a drinking water supply;
                                                     or 4)  preventing a product or substance  from con-
                                                     tacting the formation by lowering  the  water table
                                                     beneath  a source of contamination.  Each of these
                                                     options is appropriate for initial response and perma-
                                                     nent   corrective  actions  for  remediation  of  con-
                                                     taminated ground water.

                                                     Technologies for  controlling ground-water contam-
                                                     ination problems  generally  fall  into  one of  four
                                                     categories: 1) ground-water pumping, which involves
                                                     extraction of water from or injection of water into wells
                                                     to capture a plume or alter the direction of ground-
                                                     water movement; 2) subsurface drains, which consist
                                                     of gravity  collection systems  designed  to intercept
                                                     ground water;  3) low-permeability  barriers,  which
                                                     consist of a vertical wall of low-permeability materials
                                                     constructed underground to divert ground-water flow
                                                     or to minimize leachate generation and plume move-
                                                      . Vent
2-in.-diameter Pipe
                         Siphon Breaker of
                         1/4-in. Pipe
                                 250-Gallon or Larger Tank
                       * 2-in.-diameter Nipple
                              X,-\ Separated Gasoline' l\\\\f^\
                                Separated Water
                                  Pipe 2 to 4 in. from
                                  Bottom of Tank
Separated Water Exit
                                                 5-
   ' Spigot for Product Removal
    (Closed During Operation)

' Spigot for Tank Drainage or for
-Pump Cooling Water
                                                   19

-------
ment (see Subsection 5.5); or 4) in situ treatment
methods  for  biological  or  chemical  removal  or
attenuation of contaminants in the subsurface (see
Subsections 5.6 and  5.7).  These technologies can
be used singularly or in combination to control ground-
water contamination.  Pumping and subsurface drains
(EPA 1985a) are addressed here.

5.4.1 Ground-water Pumping

5.4.1.1  General Description
Ground-water pumping techniques involve the manip-
ulation and management of ground water to contain or
remove a plume or to adjust ground-water levels to
prevent formation of a plume. Types of wells used in
management of contaminated ground water include
wellpoints, suction wells,  ejector wells,  and  deep
wells.   The  selection of  the  appropriate well type
depends upon the depth of contamination  and the
hydrologic and geologic characteristics of the aquifer
(JRB1985).

5.4.1.2 Application/Availability
Well systems are versatile and can be used to contain,
remove, divert, or prevent development of  plumes
under various site conditions. The hydrogeologic con-
ditions of a site are the primary controlling feature of a
ground-water recovery system.

Pumping is most effective  at sites where underlying
aquifers have  high  intergranular hydraulic  conduc-
tivity.   Pumping has been effective at  sites where
hydraulic conductivities are moderate and  contam-
inant movement is occurring along fractured or jointed
bedrock.  In fractured bedrock, the fracture patterns
must  be traced  in  detail to ensure   proper well
placement (EPA 1985a). The effectiveness of pump-
ing can be limited in the  highly populated residential
and commercial zones typical of the locales of the
corner gas stations, as pumping may influence other
contaminant sources or nearby water supply wells.

Where plume containment or removal is the objective,
either extraction wells or  a combination of extraction
and injection wells can be  used.  Figure 5-9 depicts
the use of a line of extraction wells to halt the advance
of the leading edge  of  a contaminant plume and
thereby prevent contamination of a drinking  water
supply. Use of extraction wells alone is best suited to
situations where contaminants are miscible and move
readily with water,  where  the  hydraulic gradient is
steep and hydraulic conductivity is ihigh, and where
quick removal is not necessary.  Extraction wells are
frequently  used in combination  with  ground-water
barrier technologies to prevent ground water from
overtopping the barriers,  to  minimize contaminant
contact with the wall, and to reduce the  volumes to be
pumped.

A combination of extraction and injection  wells is
frequently  used for plume containment or removal.
Use of this technique, however, is limited to hydraulic
gradients  that  are  relatively  flat  and  hydraulic
conductivities that are moderate.  The function of the
injection well is to direct contaminants to the extraction
wells.  This method has been used with  some success
for plumes that were not miscible with water.  Figure 5-
10 illustrates an extraction/injection well system. One
problem with such  an arrangement of wells is that
windows (i.e., areas where water movement is very
slow or nonexistent) can occur. The size of the dead
spot  is  directly related  to the amount of overlap
between adjacent  radii of  influence; the greater the
overlaps, the smaller the  dead spots will be.  Another
problem is that  injection  wells can suffer from many
Figure 5-9.   Containment by the use of extraction wells (JRB Associates 1985).
             Underground
                Tank     To Treatment
                                     Domestic
                                       Well
                             Domestic
                               Weil
                Cross-Sectional View
                                                           Extraction Wells with
                                                           Radius of Influences


                                                                    Plan View
                                                  5-20

-------
 Figure 5-10.   Extraction and injection well patterns for plume I enioval (JRB Associates 1985).


                                                  ^J
                                                                           •   Extraction Well

                                                                           A   Injection Well

                                                                               Plume Boundary

                                                                               Radius of Influence
                                         Ground-Water  Flow
operational problems, including air locks and the need
for frequent maintenance  and  rehabilitation  (EPA
1985a).

Extraction  or injection  wells  are also  suitable for
adjusting ground-water levels, although they are not
widely used for this  purpose.   Plume development
can  be controlled  at sites where the  water table
intercepts the contaminant source by lowering  the
water table  with extraction wells.  For this pumping
technique to be completely effective, infiltration must
be reduced and free liquids must be removed. Unless
these conditions are achieved, the potential exists for
development of a plume of contaminants. The major
drawback to using  well systems for lowering water
tables is the costs associated with continuing main-
tenance of the system (EPA 1985a).

Injection wells can be used  to create ground-water
barriers that change both the direction of a plume and
the speed of plume migration. Figure 5-11 shows an
example of plume diversion accomplished by installing
a  line of  injection wells to protect domestic water
sources. Creating an area with a higher hydraulic head
can force the plume to change direction. This may be
a desirable technique for use during initial response,
when  short-term  diversions  are  needed or when
diversion will provide the plume with sufficient time to
degrade naturally so that containment and removal are
not required (EPA 1985a).
                                                  5-21

-------
Figure 5-11.
Plume diversion using injection wells
(JRB Associates).


             Future Plume
             Movement
                          Movement ^
                           -—*
         Injection Wells
                             Domestic Wells
Each of the well types used in ground-water pumping
have their own specific applications  and limitations.
Table 5-8 summarizes the  site conditions for which
each is most applicable.  The components of these
well types are described later in this section.

Wellpoint systems are effective in almost any hydraulic
situation,  but they are best suited for shallow aquifers
where  extraction is not  needed below  more than
about 22  feet.  Beyond this depth,  suction lifting (the
standard   pumping  technique  for  wellpoints)   is
ineffective.  Suction wells  operate in a similar fashion
          to wellpoints and are also limited by depth.  The only
          advantage  of  suction wells over  wellpoints is their
          higher capacities.  During the operation of suction lift
          wells, it may be necessary to treat air released from the
          vacuum pumps.   Under  negative  pressure, volatile
          organics are often released to the air.  For extraction
          depths greater than 20 feet, deep wells and ejector
          wells are used. Deep well systems are better suited to
          homogeneous aquifers  with  high hydraulic conduc-
          tivities and  in situations where large yolumes of water
          must be pumped.  Ejector wells perform better than
          deep wells in heterogeneous  aquifers with low hy-
          draulic conductivities. A problem with ejector systems
          is  that  they  are  inefficient  and  are  sensitive to
          constituents in the ground water that may  cause
          chemical  precipitates  and well  clogging  (Powers
          1981).

          5.4.1.3  Design and Construction
          Considerations

          Several factors must be considered in the design of a
          pumping well system.    These include the  hydro-
          geologic setting, the objective of the well system, and
          expected installation difficulties. The flow associated
          with ground-water pumping is  not  addressed herein;
           however, such information is  available from sources
           listed in the References and  Bibliography.  The ex-
          traction of  water from wells can be for removal of
          contaminated  ground water, control of ground-water
           (plume) flow, or both.  The discharge from wells that
           are to be used to extract contaminated ground water
       Table 5-8.   Criteria for Well Selection*
                  Parameters
Wellpoints
                                              Suction Wells     Ejector Wells
                                                                                        Deep Wells
                  Hydrology

                   Low hydraulic conductivities    Good           Poor            Good
                   (e.g., silty or clayey sands)

                   High hydraulic conductivities   Good           Poor            Poor
                   (e.g., clean sands and gravel)

                   Heterogeneous materials      Good           Poor            Good
                   (e.g., stratified soils)

                   Proximate recharge          Good           Poor            Good to fair

                   Remote recharge           Good           Good           Good

                  Depth of well                Shallow         Shallow         Deep
                                            (<20ft)         (<20ft)          (>20ft)

                  Normal spacing              5 to 10 ft

                  Normal range of         .     0.1 to 25 gpm
                  capacity (per unit)

                  Efficiency                   Good          Good           Poor
               20 to 40 ft       10 to 20 ft

               50 to 400 gpm    0.1 to 40 gpm
                                                                            Fair to poor


                                                                            Good


                                                                            Fair to poor


                                                                            Poor
                                                                             i

                                                                            Good

                                                                            Deep
                                                                            (>20 ft)

                                                                            :>50 ft

                                                                            25 to 3000 gpm


                                                                            Fair
                   Data from Powers 1981.
                                                    5-22

-------
 must be directed to some treatment facility.  Water
 removed from wells used for ground-water control can
 be discharged remotely or used locally to enhance the
 ground-water control system.  Treated contaminated
 ground water also can be used as recharge water to
 control grou nd-wate r flow.

 A complete hydrogeological understanding of the site
 should be established prior to the  design  of a  well
 system. The design must address the following items:
 the number and type of wells needed, their pattern,
 and their spacing; the radius of influence of the wells;
 the required pumping rates; and whether equilibrium
 or nonequilibrium pumping will be used.

 Determination of  the  proper  spacing of  wells for
 complete capture of a ground-water plume is probably
 the  most  important item  in  system design. Field
 practitioners have long had a standing "rule of thumb"
 for estimating well spacing: adjacent cones of depres-
 sion should overlap (i.e.,  radii of  influence  should
 overlap).   This  method is  reasonably accurate for
 aquifers that have low natural flow velocities, but it will
 not  be valid  for  aquifers  with  high natural  flow
 velocities.

 Determining the radius of influence for a well in a given
 aquifer is critical in corrective-action design because it
 affects well spacing, pumping rates, pumping cycles,
 and screen lengths. The radius of influence of a well
 increases  as  pumping  continues  until  equilibrium
 conditions  are reached  (i.e., when  aquifer recharge
 equals the pumping rate or the discharge rate).

 The most accurate method for estimating the radius of
 influence is by pumping  test analysis.  Pumping tests
 can identify recharge boundaries,  barrier boundaries,
 and slow  storage release conditions.  The pumping
 test should be performed until  equilibrium conditions
 are reached.  Typical test durations for a confined
 aquifer are about  24 hours, whereas they may be
 several days for an unconfined aquifer.

 When  pumping test data are lacking or incomplete,
 rough approximations of the radius  of influence can
 be obtained by use of equations and the values of
transmissivity  or  hydraulic conductivity,  pumping
times,  and coefficient of  storage.  Because these
 estimates  are approximate and do not take recharge
 into  account, the  value of R should be  adjusted
downward  so  greater  overlap   of  the  cones  of
depression will occur.  This lowers the probability that
contaminants will escape between the wells.

 For  a  confined  aquifer,  capacity  (and  therefore
pumping rate) is directly proportional to the drawdown
as long as the aquifer is not dewatered. Increasing the
pumping rate will not affect the radius of influence, but
it will  affect how  long  pumping  is necessary.   In
situations  where  the  contaminated  plume  floats,
 drawdowns and pumping rates will probably be small.
 Large drawdowns and  high  pumping  rates  are
 desirable where contaminants are dispersed through-
 out the aquifer, quick  removal is desired, and natural
 ground-water flow rates are large.

 For an unconfined  aquifer, the maximum efficiency for
 well operation occurs at about  67 percent of the
 maximum  drawdown;  pumping rates  should be
 adjusted accordingly.

 Equilibrium pumping is used in  most plume manage-
 ment systems because it allows greater well spacing,
 which reduces the  number of wells required. It is also
 easier to design and usually entails lower O&M costs.
 Nonequilibrium pumping  may be  a better alternative
 for  aquifers  with  low  hydraulic conductivity, for
 npnmiscible plumes such as gasoline, and for sites
 with ground-water barriers or scant recharge.
 Once the general approach to the remedial action and
 the  time constraints  have been established,  the
 design of the system as a whole can proceed. Optimal
 system designs may include the integration of many
 different  techniques that involve  not only strategies
 other than ground-water pumping, but combinations
 of the various ground-water pumping strategies dis-
 cussed herein.  Initial well locations, estimated pump-
 ing rates, and desired drawdowns for both extraction
 and injection wells can be established. The proposed
 system can be modeled to predict the response of the
 aquifer, and subsequent  system refinements can be
 made until the system is deemed to be workable. This
 procedure can be done manually for small systems, or
 several computer programs are  available for this use
 for both small and large systems.

 The following subsections describe the four types of
 extraction  wells  typically  used  in  ground-water
 pumping  systems.  All  extraction wells share the fol-
 lowing common components: 1) access to the aquifer,
 2) openings that allow ground water to enter the well,
 and 3) a means of moving the ground water to the
 surface.  Access to the aquifer is by means of a hole
from the ground surface to the depth  at  which the
target ground water exists.  This hole can be created
 by pushing materials out of the way of an  advancing
tool  or by drilling a  hole and removing geologic
 materials from the ground ahead of the  advancing
tool.   (These  procedures are  discussed  in greater
detail later in this  subsection, under Well Installation
 Methods.)

The  hole to the depth of the target ground water is
 necessary so that  ground water can move from the
aquifer into the well. Typically, a perforated or slotted
section of pipe, known as a well screen, is placed at
the bottom of the hole.  Well screens can be made of
 heavy wire mesh, continuous wire wrappings, slotted
or drilled pipe, or expanded and rolled sheet-stock.
                                                  23

-------
They can be made of inexpensive plastic and metal
materials or of relatively expensive chemically resistant
or inert materials.  Well screens are usually attached to
a solid-walled pipe, known as  the  well riser, that
extends to the surface of the  ground and serves to
hold the hole open and  sometimes to convey the
ground water to the surface.  If wells are installed in
bedrock  or where there  is little danger of the hole
collapsing, the screen and riser pipe are sometimes
omitted from the  deeper portion of the hole.  Other
details of pumping wells vary with  the particular well
type and are discussed in the following subsections.

Suction Lift Wells (Pumping Less Than 22 feet of Lift)

Suction-lift wells can be used to remove ground water
that is less than 22  feet below the pump,  usually
located at or near ground surface.  The suction pump
reduces  the  pressure of  air (or water) in the pipe
above the potentiometric surface of the ground water
to a point where atmospheric pressure forces the
ground water up to the pump intake, where it is drawn
into the  pump  and expelled  through  the  pump
discharge. Special caution must be observed during
the pumping of ground water contaminated with high
concentrations of volatile or explosive compounds, as
the reduced  pressure used in the  extraction process
may produce hazardous conditions.  Also, if the  re-
duced pressure is likely to lead to chemical changes in
the  water and  cause  precipitation  of  solids  and
subsequent clogging of the well screen, other meth-
ods may have to be applied.

The suction  can  be applied either to a pipe  with its
opening at the bottom of an open-topped well (Figure
5-12a) or  directly to the well casing (Figure 5-12b).
The first method merely removes the water standing in
 Figure 5-12.   Suction wells (Drlscoll 1986).
                                   \
                                       •Riser Pipe
                                          ction Line
                                        . Well Screen
  the well and relies on gravity alone to force water out
  of the aquifer to refill the well.  The second method
  removes the water standing in the well and uses both
  gravity and atmospheric air pressure to refill the well,
  which provides a somewhat higher recovery rate. This
  enhanced recovery entails greater installation cost, as
  the well must be airtight and  sealed tightly in the
  ground.  As in any suction-type process, the system
  must remain closed for proper operation.  If the water
  level  in an open-topped well drops ibelow the water
  intake pipe or leaks occur in the well casing, efficiency
  can drop greatly, maybe even to zero. The same is
  true if the aquifer water level drops below the top of
  the well screen in a sealed-top well. Some situations
  may  call for a' special well   design  that  includes
  drawdown protection  to  prevent  loss  of  suction.
  These  are design  considerations that should  be
  considered before the selection of any well system
  that depends on suction pumping techniques.

  Individual Suction Lift Wells.  When only a few wells
  are required  or  when wells are widely spaced, indi-
  vidual shallow wells with diameters ranging from as
  small as 1.5 in.  to very large ones 0an be equipped
  with a suction pump that will lift the desired volume of
  water from the ground. Wells of this type can typically
  remove from 5 to 400 gal/rnin of water from properly
  designed and constructed wells. Each well must have
  its own pump, power supply (either an electric motor
  or a combustion engine), and water discharge.

  Multiple Suction Lift Wells (Wellpoint Systems).  When
  a large number of relatively closely spaced shallow
  wells would be  advantageous,  multiple suction wells
  can be connected to a common suction pump via a
   manifold  (Figure 5-13).   Such systems, commonly
   referred to as wellpoint systems,  have been used
                                                                       ^ Pump
                  /\
                                                                     Well Screen
                   (a)  Open-Topped Suction  Well
   (b)  Sealed-Top Suction Well


5-24

-------
 Figure 5-13.   Multiple suction lift wells (Well point system) (Driscoll 1986).
widely  in the  construction  industry  to  dewater
foundation excavations, are readily available, and are
adaptable  to  a multitude of  situations and config-
urations.  Valving can be installed in the  manifold
network to allow controlled production from individual
wells and to prevent overpumping and loss of system
efficiency.  Systems of this type can extract  large
volumes of ground water, the amounts limited only by
the capacity  of available  pumps and header pipes.
Typical systems can extract 25 to 25,000 gal/min.
Deep Wells (Pumping Greater Than 22 feet of Lift)

Ground-water pumping systems that have to lift the
water  more  than  about 22  ft must  depend  on
techniques of pumping other than those described in
the previous  subsection.  Because these other  tech-
niques  work  on  the  principle  of  increasing the
pressure on the water at the bottom of the well, they
are  known  as  positive-pressure  lifting   methods.
Devices must be lowered into the wells and powered
so as to increase the pressure  on the water at the
bottom of the well to  overcome the gravitational pull
on the water  and push it up the discharge pipe to the
ground surface. Many such devices are available, but
relatively few  of them fulfill the requirements of ground-
water pumping sufficiently to be considered  here.
The  two general types of  deep-well pumping dis-
cussed here are 1) mechanical pumps lowered to the
bottom  of  the well,  and 2)  hydraulic  pumps that
depend on venturi-effect  velocity/pressure changes
to generate  sufficient pressures to  lift the  ground
water to the surface:

Submersible  or Downhole  Turbine-Pumped Wells.
Deep wells having a 4-inch diameter or larger riser pipe
can accommodate readily available submersible water
pumps  (Figure 5-14).   These  pumps,  which are
manufactured for the water-supply industry, have the
electric motors attached directly to the water pump
and  require  both  electricity supply cables  and an

                                                 5-
  attached pipe large enough to accommodate  pump
  discharge.  They are  available in capacities ranging
  from 5 to well over 1000 gal/min and have working
  depths ranging from 0 to more than 2000 ft below
  surface.  These pumps depend on water flowing past
  the motor to cool and  protect it from  overheating;
  therefore, care in installation and  maintenance  of
  adequate space  around  the  pump   are  critical.
  Selection of casing size (riser and screen diameter)
  must satisfy two  requirements: 1) it must  be large
  enough to accommodate the pump and allow enough
  clearance for efficient operation, and 2) it  must ensure
  proper hydraulic efficiency of the well at the intake
  section of the pump  (Johnson Division, UOP, Inc.

  Figure 5-14.   Submersible pump.

           Discharge Pipe        Electric Power
                             Static Water Level
                                 - Core of Depression
                               .Pump
                                Motor
25

-------
1975). The controlling factor is usually the size of the
pump selected for the desired capacity and head. As
a good rule of thumb, the casing size should be two
standard pipe sizes larger (and not  less than one
standard pipe size larger) than the nominal diameter of
the pump. Submersible  pumps  are  efficient and
relatively economical to purchase and operate.  One
drawback to this type of pump is that the discharge
pipe, pump, and motor must be removed from the well
to service any portion of the system.

An  advantage  of downhole  turbine  (or   vertical
lineshaft) pumps is that the pumping mechanism can
be placed at  the bottom of the well while the power
unit is kept  above  ground  (Figure 5-15).   These
pumps are driven by surface-mounted electric motors
or combustion engines, which can be easily  repaired
or replaced in the event of breakdown or failure. The
power is  delivered to  the turbine  pump through  a
driveshaft house within  a riser pipe that serves as both
pump discharge pipe  and driveshaft stabilizer.  To
achieve discharge capacities equivalent to that from
submersible pumps may require slightly larger casing.
Vertical lineshaft pumps have been used primarily for
large-capacity  municipal  and  industrial production
wells with the ability to pump thousands of gallons per
minute, which are powered by sources ranging from a
few to more  than a thousand horsepower and from
depths reaching as much as 2000 ft.

Figure 5-15.   Downhole turbine (Drlscol 1986).
                               Pump Motor
                              , Pump
                               Driveshaft
                             Pumping Water
                             Level
                              " Casing

                              • Pump Turbine
                               Assembly
Ejector  Wells.    Ejector wells,  or deep-well  jets,
function on the principle of changing  velocities and
subsequently changing pressures to produce lifts of
up to 125 ft of water in wells that are not subject to
being pumped dry.  Figure 5-16 demonstrates the
basic operating parts of this system as well as its main
drawback-poor  efficiency.  In  spite  of  the  poor
efficiency (15%), the benefits of this type of pumping
system can be great when closely spaced wells with
lifts ranging from greater than 22 ft up to  125 ft are
required. The advantage of having all moving parts of
the system kept above ground is obvious, and as in
the case of suction well systems, ejector wells can be
operated  by a  single  pump  per well,  or multiple
extraction wells  can be driven by a  single pump if
properly designed distribution manifolds are used.

Figure 5-17 shows two commonly used versions of
the ejector well.  A closeup view of a two-pipe system
is presented in Figure 5-16. The single-pipe system,
in which the well riser is part of the pumping system, is
shown in Figure 5-18.

 Well Installation Methods           i

Each well type  discussed in the preceding sub-
sections can be constructed in a  range  of capacities
and sizes.  The selection of appropriate techniques to
install an  extraction well in a specific  setting will de-
pend on the geologic materials present, the depth to

Figure 5-16.    Deep well jet pump (Drlacol 1986).


                                   • Discharge
                                                             Nozzle
                                                                             Intake Pipe
                                                                              Foot Valve
                                                  5-26

-------
Figure 5-17.    Variations of ejector wells (Powers 1981).
                                              Discharge
                                                    ^M
                                     Return
                                     Header
                                                                                       Well Casing
                                                                                       and Screen
                                                i j" ^ 1 1/4-in. (32-mm) Riser
                                                     with Turned Couplings
                                        Typical single-pipe
                                        ejector installed in
                                          a 2-in. (50-mm.)
                                               well
Figure 5-18.    Single-pipe ejector.
            11/2-in.-
            Suppiy Line
                                                                                      " Ejector Body

                                                                                        Foot
                                                                                       Valve
 Typical two-pipe
ejector Installed In
 a 6-in.  (150-mm.)
      well
                                    Pressured
                                    Annular
                                    Tube
                                      Intake
                                      Pipe
                                                                                     uiscnarae
                                                                                   Pump
          Lin Pipe

          Well Riser
                                                                                  Venturi Tube
                                                                                   Well Screen

-------
the target zone, the  desired well  type, the  required
capacity, and other considerations. Three basic well
installation techniques (Figure 5-19) are discussed in
the  following  subsections.   Table 5-9  summarizes
these methods. The selection of the materials of con-
struction dictates which installation techniques can be
used.  Materials and some installation limitations they
present  are  discussed  briefly  in,  a  subsequent
separate subsection.
Figure 5-19.   Types of wells (Johnson Division, UOP, Inc. 1975; Powers 1981).


             Well Casing
                                            Driving Pipe
                                            Driving Bar
                          Driven Well
   Jetted  Well
                                     Illllilllll
                                  Bentonite Seal'

                                k   ,vi Filter Sand
                                   "   ~~
                                                          "Aquifer
                                                Drilled  Well

                                                     5-28
                       (from Pump)

                       Well Casing    ;


                       Ring Seal of
                       Semirigid Plastic


                      Temporary Wash
                      Pipe

                      Well Screen

                      Coupling on Wash Pipe
                      Rests in Conical Seal

                      Combination Back-Pressure
                      Valve in Open Position

-------
Table 5-9.  Common Methods of Well Installation

        Method
Basic
Driven
Jetted
Drilled





Variations


Solid stem auger
Hollow stem
auger
Bucket auger
Conventional
hydraulic
(and reverse)
Air rotary and
reverse
Cable tool
Geologic Material
Unconsolidated materials
free of boulders
Unconsolidated materials
free of boulders
Unconsolidated materials
without excess boulders
Unconsolidated materials
without excess boulders
Unconsolidated materials
without excess boulders
Any type
Any type
Any type
Size Range Diameter
1-3 inches
Up to 24 inches
Up to 18 inches
Up to 18 inches
Up to 72 inches
Up to 72 inches
Up to 72 inches
Up to 72 inches
Depth Range
Up to 100 ft
Up to 200 ft
Up to 100 ft
Up to 100 ft
Up to 100 ft
Unlimited
Unlimited
Unlimited
Installation
During hole opening
During or after hole
opening
After hole opening
During or after
hole opening
After hole opening
After hole opening
After hole opening
After hole opening
Driven Wells.   Driven wells are, in a general sense,
pipes driven into the ground by pile-driving hammers.
They can  be significantly more complex than this
simple definition implies, but, on completion, all driven
wells consist  of a  riser pipe to which a screen is
attached at the bottom of the well.  These wells are
usually 1.5 to 3.5 inches in diameter, have a conical
tip, and are of a material sturdy enough to withstand
the forces of the installation technique.  Driven wells
are usually pumped by  suction or  ejector pumping
systems.  Their installation is usually limited to areas
comprising relatively soft,  Unconsolidated materials;
and their maximum depth is approximately 100 ft.

Many suction  wellpoint systems use driven wells.  In
these  systems the  tight well-to-formation seal often
eliminates the need for installing artificial well seals to
prevent loss  of suction,  and  the  cost of  installing
driven wells is significantly less compared with other
types of wells.

Jetted Wells.  Jetted wells  rely on fluid  pumped
through the well riser and screen and its forceful exit
from the bottom  tip of the well  to  dislodge uncon-
solidated  geologic  materials immediately below the
specially designed well tip. The well screen and riser
then  sink  under their own  weight, often  with  the
assistance of  a drive hammer, which is used to over-
come friction along the length of the well pipe.

Jetting wells  into place  has certain advantages over
driving  wells.   Jetting  allows installation of small-
diameter wells to greater depths than driving does in
the same geologic  setting.  Also, somewhat larger-
                                                     Applications/Limitations
                                                  5-29
diameter wells can be installed, which permits the use
of some deep-well pumping systems.   In  addition,
because of the washing action of jetting and removal
of fine materials, only minimal well development is
necessary  to  achieve  optimal  production  charac-
teristics in these wells.  One major drawback to this
installation  technique in  certain  settings  is  that
relatively large volumes of water are required. This may
dilute contaminant  concentrations and  increase  the
total volume of ground water that ultimately must be
extracted from the well. Jetting can  also significantly
alter ground-water flow patterns in the vicinity of  the
well during installation.

Drilled Wells.  Both driving and jetting wells into place
are cost-effective and technically advisable in many
settings. In certain hydrogeologic settings,  however,
neither is advisable. In this case, a well borehole must
be  constructed in which to install the ground-water
extraction well. Wells installed in drilled boreholes can
be of nearly any type  construction, material,  depth, or
diameter, and can be designed to produce a specific
discharge rate and drawdown to facilitate  ground-
water control and enhance recovery of contaminated
ground water. Ground-water pumping from drilled
wells can be accomplished by suction pumps (when
lift  is less  than 22 ft)  or by any of the deep-well
pumping techniques mentioned previously.

Auger drilling, rotary wash, air rotary, and cable tool are
some of the techniques used  to drill boreholes for
installing wells. Each general drilling technique has its
own set  of variations, including solid- or hollow-stem
augering, normal or reverse circulation in rotary wash

-------
and air drilling, and a cased or open borehole during
rotary  or cable tool drilling.  All of the drilling tech-
niques and their variations have advantages and dis-
advantages when used in specific settings or under
given requirements. In some settings,  two or more
drilling techniques will be used to construct a single
borehole.

Wells  installed  in drilled boreholes can vary from
simple screened iron or PVC  pipes placed in the hole
to multiple-cased stainless steel, PTFE, or PFA well
pipe with O-ring joint seals, artificial screen filterpacks,
and borehole seals of volume-expanding materials.

Well Construction Materials

Materials used  in the construction of wells  can be
grouped  according  to  the  function  they  serve.
Materials should be selected  for risers, screens, filler
packs, and well sealing materials that minimize corro-
sion, chemical deterioration,  or encrustation.  Like-
wise,  the proper material type, wall thickness, and
screen  slot-size must be selected to  withstand the
physical stress placed on the well during installation
and operation.

Not all wells require all parts but all wells consist of, at a
minimum, a riser pipe  and a  screen, which is usually
attached at the bottom. These two components allow
physical access to and removal of ground water from
the aquifer. They  can be  constructed  of black iron
pipe,  PVC, or  other construction-grade plastic pipe,
stainless  steel, or  fluorocarbon  (Teflon)  plastics.
These pipes are commercially available in standard
sizes  ranging from 1 to 72 inches.  Specially manu-
factured or machined pipe has threaded  ends to allow
easy connection of pipe sections, whereas other pipe
requires welding,  gluing,  or  some mechanical con-
nection during the installation  process.

Artificial  screen filter  packs  are  often installed  to
provide  a  uniform,   controlled-size material  sur-
rounding the well inlet.  This is often advisable to allow
full production capabilities and to eliminate unwanted
fine materials from entering the well.   Artificial filter
packs can consist of sand or gravel poured or pumped
into the borehole after the pipe has been placed so
that it settles around  the  well  screen, or specially
designed bags  can be attached to the screen before
installation  and  filled  with  the  desired material.
Sometimes the natural geological  formation material
can be  developed by a  variety of techniques  to
produce the desired characteristics.  In this case, an
artificial filter pack will not be necessary.

It is often desirable for the well screen to be isolated
from the  geologic formation above  it. This is most
often  accomplished by  installing a borehole seal
above the screened interval or filter  pack.  Borehole
seals vary in  design  and material and can  include
inflatable  packers,  swelling  bentonite clay,  cement,
concrete,  mixture of cement and bentonite or organic
polymers, and expanding cements  (often  used  in
combinations to achieve the desired result).


5.4.1.4 Costs

Costs of well systems for plume management can vary
greatly from site to site. Some  of  the factors that
determine these costs are the geology, the charac-
teristics of the contaminated and naturally occurring
ground  water,  the extent  of  contamination, the
periods and  duration of pumping, local wage rates, the
availability  of  supplies  and  equipment,  and the
electrical power required.  Costs associated with a well
system  can be categorized as  mobilization  costs,
installation and  removal  costs, and operation and
maintenance costs.  Tables 5-10 through 5-12 pro-
vide typical costs for well pumping system installation
and operations.

The complexity of developing costs for ground-water
pumping systems  led  Geraghty and Miller, Inc., to
develop  a method for estimating total capital and
operating costs for deep wells based on the  use of
existing hydraulic models (Lundy and Mahan  1982).
They have applied the model to a number of scenarios
and the resulting cost estimates provide considerable
insight into how aquifer characteristics affect cost and
how the  cost of well system components compares
with total capital costs.

Table 5-13 summarizes seven recovery system cost
scenarios to which the cost methodology was applied.
For these scenarios, the plumes were assumed to be
moving in unidirectional flow fields and to  have the
following dimensions:  250 to 2500 feet wide, 500 to
5000 feet long, and 25 to 250 feet deep.

A high transmissivity (100,000 gal/day/ft) is assigned
to four low-flux scenarios. A low transmissivity (5000
gal/day/ft) is assigned to three  high-flux scenarios
(EPA 1985).
                                                 5-30

-------
Table 5-10.   1986 Costs for Selected Pumps and Accessories*
                      Pump/accessory               Description
                      Jet pumps
                          Shallow well

                          Deep well

                          Jets and valves

                          Seals
                          Foot valves
                          Air volume controls
                      Submersible pumps
                          4-inch pump

                          Control boxes
                          Magnetic starters
                          Check valves
                          Well seals
                      Vacuum pumps
Pumping depths:  25ft
Horsepowers: 1/3 to 1-1/2 hp
Capacities:  60 to 27,000 gph
Pumping [depths:  320ft
Horsepowers: 1/3 to 2 hp
Capacities:  60 to 1,000 gph
Single-pipe jets
Double-pipe jets
Single- or double-pipe seals
Pumping depths: 900 ft
Horsepowers: 1/3 to 3 hp
Capacities: 50 to 2,000 gph
                      * Data from JRB Associates 1985.
                                                         5-
                                   Cost Range ($)
210 to 520

260 to 680

40 to 105
30 to 80
15 to 40
10 to 50
10 to 30

445 to 1,565

80 to 155
165 to 260
15 to 440
20 to 125
Diesel motors
Electric motors
800 to 7,C
420,000 £
800 to 7,(
420,000 j
00 gpm (48,000 to
ph)
00 gpm (48,000 to
Ph)
14,000 to 52,000
9,000 to 37,000
        31

-------
Table 5-11.   1986 Costs for Well Screens and Wellpolnts*

                               Type                     Description
                               Driven wellpoints
                               Well screens
                               Jetting screens
                               (fittings)
Stainless steel,
1-1/4 to 2 inches in
diameter

Low carbon steel,
1-1/4 to 2 inches in
diameter

PVC plastic,
1-1/4 to 2 inches in
diameter

Stainless steel,
1-1/4 to 36 inches in
diameter

Low carbon steel,
1-1/4 to 3 inches in
diameter

PVC plastic,
1-1/4 to 12 inches in
diameter

Cast iron or mijd steel,
2 to 12 inches in
diameter
                               ' Data from Johnson Division, UPO, Inc. 1975.
Table 5-12.    1986 Costs for Completion of 2- to 4-lnch-DIameter Wells*
                           Cost range ($)
35 to 44/ft
                                                                                   16 to 31/ft
                                                                                   5 to 6/ft
34 to 555/ft
                                                                                   18to161/ft
                                                                                   10to62/ft
31 to 277/ft
                                    Drilling Method
                  Range of Well
               Completion Costs *($)
                                    Conventional hydraulic rotary

                                    Reverse circulation hydraulic
                                    rotary

                                    Air rotary

                                    Hollow-stem auger

                                    Cable tool

                                    Hole puncher Getting)*

                                    Self-jetting or driven*

                                    Mobilization
               30 to 50/ft

               40 to 60/ft


               30 to 50/ft

               30 to 50/ft

               25 to 75/ft

               30 to 50/ft

               20 to 30/ft

               200to1000/rig
                                     Data from PEL
                                    ' Includes drilling, well material, and installation costs. Can be
                                     significantly higher in some settings and in highly contaminated
                                     environments.
                                    + Includes rental of all necessary equipment, e.g., wellpoints,
                                     pumps, and headers.
                                                              5-32

-------















m
_o
5
§
o
CO
•5
o
o
1
5.
»
•*»
0
>
o
£
I

0>
co
"o
2*
co
W










8
o_
&
'w
8







WTO IBIOJ.


SJSOQ IBljdBQ IBJOJ.


t (WTO) Buuojjuo^j
$ (WTO) JuatuiBaij.

•t (W8O) su!BJp/s||sy\/y

4- XJJIJOBJ }ueiujBejj_
4- sjnpnjjsBjju! QOBjjng

4- surejp/s||9M

4.u6!Saa

.uoi.eunea








1
1

1
o
.2
a.
T3
'5
cr
<
co -M- co
3 10 - ^
* •» IO T-
S 2 S § CO CO fe
W "^ IO CO CO S T—
S S S J2 S S S
S5 SP ^"s* 35t co O) ^f
S CO § ip Si ^ S
s s s i i i 8
O) O>
a>
CO CO CO CO CO' CO CM
^- CM »- CM ^- CM CO

Ss w l
CO -^ '

> O> CO CM
t to iB ;-:
% § 1 § § § 1
o> a>
CD CD CO C
»- 10 i- C
00 CO CO 0

J S § 3
> CO CO CO
s s s 4 s s s
» CM O4 O
tx x g
o o ^
LO u5 x

C.*. £S £i Si £L 10 EM. CM. of
ts-
w "D
2 x
III i. 1 I
.B.

1 E E 1.
; § & S »
*• X ^^ ^^ CD cs' *3 ^3 «r^ J1^
= E CM - CM CM CM1 O=T* •«- —
3 §
3 mS
— 1 ~^
JC
.DJ
X






DC
o
O)
'to
£•
!2
J
•s
'?
'E
1
1
i
_
"8
s
Q.
£ co'
Q) CO
5 o>
«i 1

Ii ^

O) T- g

II 1
li "s

— ' .0 
-------
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

-------

-------