NATO/CCMS
Third International Conference
Demonstration of Remedial Action Technologies
for Contaminated Land and Groundwater
?,EPA
Montreal, Canada
6-9 November 1989
-------
This report is not an official document 0 f the
NATO/CCMS Pilot Study program.
-------
NATO/CCMS
Third International Conference
Demonstration of Remedial Action Technologies
for Contaminated Land and Groundwater .
S-EPA
Montreal, Canada
6-9 November 1989
-------
Abstract
In November 1986, the NATO Corrrnittee on Challenges of Modern Society (CCMS)
formally adopted a United States proposal for a five—year pilot study to
demonstrate technologies for cleaning up contaminated land and groundwater.
The participating NATO countries are Canada, Denmark, Federal Republic of
Germany, France, the Netherlands, and the United States. Japan Is also
participating. Norway and the United Kingdom are observer countries. The
Pilot Study Director is from the United States; the co-directors are from
the Federal Republic of Germany and the Netherlands.
The Third International Conference was held In Montreal, Canada, on 6-9
November 1989. Reports on 12 projects (final and Interim) were prepared,
including the following types of treatment: pump and treat (2 projects),
microbial treatment (3 projects), thermal (2 projects), solidification!
stabilIzation (1 project), electrokinetic (1 project), chemical (1
project), and soil extraction (2 projects). The discussions at this
meeting also Included recent developments in the regulations and remedial
technology research and deve1opment in the attending countries. The next
meeting will be a workshop held in Oslo, Norway on 13-15 March 1990.
11
-------
Table of Contents
Page
ABSTRACT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . , , . . . . . . . . . . . , . • • • • • _ • • • • _ • , • j
INDEX TABLE OF PROJECTS, BY MATRIX AND TREATMENT TECHNOLOGY....... V
INTRODUCTION...... • . . •.. S S • • • . . 5S5•SS•SS Se • 5•II • • S S S S 1
BACKGROUND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
REPORT ORGANIZATIONS.................................S............ 7
RECENT DEVELOPMENTS IN NATIONAL PROGRAMS.......................... 9
Canada... . . . . . . . . . •5••eS . . . . . . . . .. . . . . . . . •5 • • • • • . . . . . . . . 11
Denmark. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Federal Republ 1 C of Germany. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
France... ........... . . •... . . .... S •••S • • . . .... . . . . . . . .. 17
Norway. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
The Netherlands................. ..... .......... .... ......... 18
United Kingdom.............................................. 20
United States.................... ...... •••••• ...•.... ....... 21
PROJECT REPORTS.................................... ••S•S••e • s 5s 33
Pump and Treatment
Pump and treat groundwater (Ville Mercier, Canada)...... 35
Photo/chemical oxidation (barrel and drum
reconditlorier; California, United States)............... 59
Microbial Treatment — A
In-situ enhanced aerobic restoration (jet fuel; Eglin
AirForceBase,Un ltedStates).......................... 83
Thermal Treatment
Thermal desorption and destruction — radiation heating
(herbicides and pesticides; Hamburg, Federal Republic
of Germany)........................ S55Se 55 5SS5• • •SSSSSS 99
Stabilization/Solidification
In-situ vitrification (Parsons site; Michigan, United
States)..... • .. ..... 555 S•S•S••S• S S S S S S • 5 5 •S 5 101
El ectroki neti c
Electro—reclamation (timber finishing plant; The
Netherlands)..... • •s.. .. . •SSSS c s s s ss . . .sss . . . . . . . . 115
111
-------
Table of Contents
(Continued)
Chemical Treatment
K-PEG technology In soils and liquids (chemical
recycling facility and road oiling; Gary. Indiana and
WideBeach,NewYork,UnitedStates)................... 137
Soil Treatment by Extraction
Vacuum extraction (well field; Verona, Michigan, United
States).. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
High pressure soil washing (Goldbeck-Haus, Federal
Republic of Germany)...............................,.... 187
Microbial Treatment - B
Aerobic/anaerobic In-situ degradation (chemical waste
disposal site;Skrydstrup,Denmark)....,..,.......,..., 205
On-site/in—situ reclamation; membrane—filtering and
biodegradation (Denmark)............... ................. 207
(Appendices A - 0 under separate cover.)
iv
-------
Index Table of NATO/CCMS Projects Reported on at the Third International
MATRIX
TREATMENT
ORGANIZATION/SITE
GROUND WAT ER
Biological
Enhanced Aerobic Restoration
U.S. Air Force, Battelle
Eglin Air Force Base, FL,
United States
Chemical / !hyslcal
rump ana ireat broundwater
Envi ronment Canada
Yule Mercier, Quebec
UV/Ox ldation
Ultrox
San Jose, CA
United States
SOIL
Enhanced Aerobic
U.S. Air Force,
Eglin Air Force
United States
Microbial Treatment
Fomer gas works
Fredensborg
Denmark
Restoration
Battelle
Base, FL,
on erence on tne not tuay on
-------
Index Table (continued)
Treatable Contaminants
HAIR IX
TREATMENT
ORGANIZATION/SITE
Chemical/Physical
K-PEG Process
U.S. Environmental Protection
Agency
Wide Beach, NY, United States
High Pressure Soil Washing
Scrap metal & copper refinery
Berlin, Federal Republic of
Ge many
High Pressure Soil Washing and
Oxidation
Goldbeck Haus, Hamburg
Federal Republic of Gemany
Soil Vapor Extraction
U.S. Environmental Protection
Agency
Verona Well Field
Battle Creek MI, United States
Stabl 1 izatlon/Sol Idificatlon
In—Situ Vitrification
Parsons Chemical Site
Michigan
United States
‘I
I,
‘I
yf
I ,
‘I
I,
1 ’
‘I
1 ’
1
Treatment
Status
of
Location
Technology Page
On-Site, Demonstrated
Mobile
137
On-Site, Conriercial App.
Mobile
D
In-Situ Demonstration 187
.In-Situ Demonstrated 165
In-Situ Experimental 101
I
/
I, I ’ I f V ’ I ’
PCBs, dioxin
Lead, PAHCs
Phenol, kresol
ecif lc
)ntami nants
Treated
Halogenated and
aromatic hydro-
carbons
Mercury
-------
Index Table (continued)
Treatable
Contaminants
pecific
ontami nants
Treatment
Location
Status of
Technology
Electrokinetic
Electro-reclamation
Loppe rsum
The Netherlands
The ma 1
Themal Desorption and
Destruction (radiation
heating)
Dekonta GmbH, Hamburg
Federal Republic of Germany
p /
‘I
Arsenic
In—Situ Conniercial 115
On-Site Experlmenta1 99
MATRIX
TREATMENT
ORGANIZATION/SITE
‘I
Page
Chlorobenzenes,
Chlorophenols,
Hexachlorocyclo—
hexane, dioxins,
furans
-------
Introduction
This Is a report of the proceedings of the third International
meeting for a pilot study under the NATO Corrinittee on Challenges of Modern
Society (CCMS): Demonstration of Remedial Action Technologies for
Contaminated Land and Groundwater. The meeting was held In Montreal,
Canada, 6-9 November 1989. ExhIbit 1 Is the agenda for the meeting.
The purpose of this conference was to present final or Interim
reports on 12 Pilot Study projects. The conference also included visits to
two c1eanup projects: at Ville Mercier, and at Stablex, both located In
the Province of Quebec.
Background
The problems of contamination resulting from Inappropriate hand1ing
of hazardous materials and hazardous wastes are faced to some extent by all
countries. The need for cost—effective remedial technologies to apply at
these sites has resulted In the application of new technologies and/or new
applications of existing technologies.
Building a knowledge base so that emerging remedial technologies are
identified is the impetus for the NATO/CCMS Pilot Study on “Demonstration
of Remedial Action Technologies for Contaminated Land and Groundwater.”
New technologies being demonstrated and evaluated in the field are
discussed. This allows each of the participating countries to have access
to a data base of applications of individual technologies without any
country having to cornnit a disproportionate amount of its internal re-
sources to a specific research activity. The technologies Include bio-
logical, chemical/physical, and thermal technologies for both soil and
groundwater. With few exceptions, they are in—situ or on—site tech-
nologies, and they are not containment technologies.
The study was approved in November 1986 and will last for five years;
It Includes nine countries. Projects are selected and their status moni-
tored during an annual administrative meeting held In the Spring. (The
next administrative meeting will be held in Oslo, Norway, 13-15 March 1990.)
There are currently a total of 24 NATO/CCMS Pilot Study projects.
The exchange of Information on developing technologies Is the prime
goal of this study. The presentation and discussion of In—depth Interim
and final reports on demonstration projects is one of Its key aspects.
These reports are presented at the annual International conference held in
the fall of each year and contain both technical and cost data. (The next
International conference will be held in France in November 1990.)
1
-------
EXHIBIT 1
AGENDA
Third International Conference
NATO/CCMS Pilot Study
DEMONSTRATION OF REMEDIAL ACTION
TECHNOLOGIES FOR CONTAMINATED LAND
AND GROUNDWATER
6—9 November 1989
Le Meridien Montreal
4, Complexe Desjardins
Montreal, Canada
Canadian Host: James W. Schmidt
Monday
6 November 1989 Day I
8.30 Registration
9.00 Welcome from Host Country — Keynote speaker
- Main Jolicoeur, Environment Canada
9.30 Opening remarks by Pilot Countries
— Donald Sanning, United States
- Klaus Stief, Federal Republic of Germany
— Esther Soczo, The Netherlands
9.50 Introduction of Attendees
10.10 Coffee Break
10.25 Tour de Table
o Canada - Jim Schmidt
o Denmark - Neel Stroback
o Federal Republic of Germany - Klaus Stief
o France - Rene Goubier
o Norway - Morten Hefle
o The Netherlands - Merten Hinsenveld
o UnIted Kingdom — Paul Bardos
o United States — Walter Kovalick
11.45 NATO Expert Guest Speakers (Canada): “Clean—up
Critera In Canada”
— Art Steizig and Brett Ibbotson
12.45 Lunch
2
-------
Monday
6 November 1989 Day 1 (Continued )
14.15 PresentatIon of Pump and Treatment Projects
o Pump and Treat Groundwater, Canada
(liquid industrial waste/Ville
Mercier) - (Final report*) - Mark Halvey
o Photo/chemIcal Oxidation (groundwater)
- United States (barrel and drum recon—
ditioner/California) — (Final report*) -
Norma Lewis
16.15 Coffee Break
16.30 Presentation of a Microbial Treatment Project
o In—sItu Enhanced Aerobic Restoration soil,
groundwater) - United States (petroleum
spill/Eglin) - (Updating Report* — Douglas
Downey
17.00 Adjourn
*Forty_five minutes per presentation followed by 15 mInutes discussion.
3
-------
Tuesday
7 November 1989 Day 2
8.30 PresentatIon on Thermal Projects
o Overview of thermal projects in Federal Republic
of Gemany, Including thermal desorptlon and
destruction: radiation heating — (Herbicides
& pesticides/Hamburg) — (Interim report+) — Klaus
Stief
8.45 PresentatIon of the Stabilization/Solidification
Project
o In-situ Vitrification — United States
(Parsons site) — (Interim report+)
- Edward Burk
9.15 PresentatIon of the Electrokinetic Project
o Electro-reclamatlon — The Netherlands
(timber finishing plant) — (Final
report*) — Relnout Lageman
10.15 Coffee Break
10.30 Presentation of the Chemical Treatment Project
o K—PEG Technology (soil and liquids) —
United States (chemical recycling/New
York) - (Interim report+) - Herbert King
11.00 Presentations by Continuing NATO/CCMS Fellows
— Thomas Dahi
- James Gossett
— Wayne Pettyjohn
12.30 Lunch
14.00 SIte visit to Ville I4ercler project
17.00 Return to hotel and Adjourn
+Twenty minutes per presentation followed by 10 minutes discussion.
*Forty_five minutes per presentation followed by 15 minute discussion.
4
-------
Wednesday
8 November 1989 Day 3
8.30 NATO Expert Guest Speaker (The Netherlands):
“Microbial Treatment”
— Rene Kleljntjens
9.30 Presentation of Soil Treatment by Extraction
Projects
o Vacuum Extraction — United States (Well
field/Verona) — (Final report*) — Margaret
Guerriero
10.30 Coffee Break
10.45 Completion of Soil Treatment by Extraction
Projects
o High Pressure Soil Washing — Federal
Republic of Germany (disinfectant/Goldbeck
Haus) - (Interim report 4 ) — Wolfgang Sondermann
11.15 PresentatIon by Continuing NATO/CCMS Fellow — Sjef Staps
12.45 Lunch
13.45 Site visit to solidification project (Stablex
Corporation; Blainville, north of Montreal)
16.45 Return to hotel and Adjourn
+Twenty minutes per presentation followed by 10 minutes discussion.
*Forty_five minutes per presentation followed by 15 minutes discussion.
5
-------
Thursday
9 November 1989 Day 4
8.30 NATO Expert Guest Speaker (Canada): “Anaerobic
degradation”
— Cohn Mayfield
9.30 PresentatIon of Microbial Treatment Projects
o AerobIc/Anaerobic In-situ Degradation
(soil, groundwater) — Denmark (refrig-
erator manufacturer/Skr.ydstrup) —
(Interim report ) - Steen Vedby
o On-sIte/In-situ Reclamation: membrane
filtering and biodegradation - Denmark
(former gas works site) — (Interim report+)
— Troels Wenzel
10.30 Coffee Break
10.45 Presentations by Continuing NATO/CCMS Fellows
- Aysen Turkman
— Resat Apak
11.45 Report from F1na Report SubcorrIT lttee
— Robert Olfenbuttel
12.15 Other Administrative Topics
12.45 End of the Conference
+Twenty minutes per presentation followed by 10 minutes discussion.
6
-------
The need to share the Pilot Study information with the technical and
scientific comunity is also key. The Infomation resulting from the Pilot
Study is assembled in Proceedings following each international meeting.
Information Is also submitted to International technical journals as data
warrants. A final report on the entire study will be published by NATO at
the conclusion of the Pilot Study.
NATO/CCMS Fellows have been Identified from participating countries.
They provide additional technical resources to the Pilot Study In their
expertise, information resulting from their projects, and access they may
have to information about other emerging remedial technologies. The
Fellows’ involvement in the Pilot Study Is primarily through their atten-
dance and participation in the annual international meetings. There are
currently 11 Fellows Involved with the Study.
Report Organization
This report has three sections. The first is a report of the “tour
de table” during which country representatives discussed recent develop-
ments In their national regulatory and research and development programs.
Reports on specific projects in this Pilot Study are second and form the
bulk of the proceedings. The reports Include both interim and final proj-
ect reports, and are arranged in the order in which they were included on
the program. Presentations by NATO/CCMS Fellows and by Guest Speakers, as
well as a final report on a NATO/CCI4S project which was presented at an
earlier conference are included in the Appendices, the third and last sec-
tion.
7
-------
Recent Developments in
National Programs
9
-------
CANADA
Much of the Information that would normally be covered in the “tour de
table” concerning legislative and Research and Demonstration initiatives,
was referenced by the keynote speaker for the meeting from Canada, Main
Jollcoeur, and therefore, will not be repeated here. In addition, the
first expert speaker from Canada gave a presentation concerning industrial
site decomissioning and the setting 0 f soil quality objectives. Again,
therefore, this material will not be provided here.
However, an update on Canada’s most extensive clean up project, the Sydney
Tar Ponds is provided as Exhibit A. It should be noted that in July, 1989,
a contract was awarded to Superburn Systems Ltd. of British Columbia to
design and build two revolving fluldized bed incinerators. Construction
has started on the Incinerators and It Is planned to begin excavating and
incinerating the PAH contaminated sediments In November, 1990. (A fact
sheet on the incineration system Is provided as Exhibit B).
DENMARK
Waste management in general
The amount of waste being produced in Denmark is 8 mIllion t per year muni-
cipal waste. As mentioned last year Is the goal for the year 2000, that
50% Is recycled, 25% incInerated and 25% goes to landfill sites. Today the
figures are 25%-25%—50%. The Strategy Plan for the waste management area
emphasizes the Issue of using clean technology In the industrial production.
The amount of chemical waste In Denmark Is today 120.000 t per year. This
amount is regarded as almost constant for the years ahead.
Two major soil treatment centers are established in 1989.
New regulation
The Danish Government has announced a major simplification In the laws of
envi ronmental protection.
As mentioned last year, there has been made a draft for a new act on reme-
dial action on chemical waste sites. The suggested act and the
appropriation, which has been estimated to 6—8 billIon Danish Kroner, will
be proposed In the parliament this year. The suggested act will deal with
11
-------
EXHIBIT A
clean- u
ate
Major Highlights of Year One of the
Sydney Tar Ponds Cleanup Project
T he spring of 1987 was the start of something big.
That is when the Sydney Tar Ponds Clean-up
officially began. The ten-year project is one of the
largest and most
complex toxic
waste clean-ups
in Canadian his-
tory, using new
and innovative
techniques for
toxic waste
destruction.
Much will
happen between
1987 and the esti-
mated comple-
tion of the clean-
up in 1997. To
keep people
informed of the
many activities
and accomplish-
ments throughout
the life of the Tar
Ponds project,
clean-update
will document
major highlights
as they unfold
each year at the Tar Ponds
site, and behind the scenes.
Most of the action in the first two and a half years
of the “clean-up” will actually happen away from the
Tar Ponds site. This first phase of the project entails
laying the planning
and design ground-
work for the future
destruction of the Tar
Ponds sediment.
When 700 000 tonnes
of toxic muck must be
moved and incinerat-
ed, it is crucial that
every base be covered
before the first shovel-
ful of sludge is scooped
from the ponds.
I
June, 1987 — March,1988
Now you see them...
now you don’t. Sysco’s
coal tar storage tanks
are demolished as the
Tar Ponds Clean-up
Project begins.
N . Sc
12 Dep.rtm.nt ol
th Envlronmsnt
I .1 Environnement Environment
Canada Canada
-------
y .r D N
From dismantling coal tar storage tanks to erect-
ing a Public Information Display Centre, the first
year of the Tar Ponds project was a busy one.
Here are the major highlights of the year:
At the Tar Ponds Site
• Thirteen abandoned coal tar storage tanks at the
Sysco Steel plant were dismantled and removed
between November, 1987 and February, 1988. Their
contents were transferred to one remaining tank,
where they will be stored until they can be safely
destroyed at a later stage of the clean-up process.
An incineration system will be designed and con-
structed to destroy both the coal tar from the tanks
and the contents of the Tar Ponds.
One of the primary aims of the overall project is to
• create much-needed jobs for Cape Bretoners: the
tank demolition employed 17 people, including 12
former Sysco workers.
• Numerous samples of the Tar Ponds’ contents
were analyzed to provide engineers and scientists
with a more complete picture of the chemical make-
up of the ponds. This information is vital to the
success of the project because the chemical proper-
ties of the sludge determine which combustion
method can best destroy the toxic material, and how
best to monitor the local environment during the
incineration process.
• Further studies of the Tar Ponds sediments were
made to get a more precise estimate of the amount
of waste at the site. New estimates put the volume
of toxic material at 560 000 cubic metres, although
the actual quantity excavated will likely be higher.
Other tests revealed that the levels of metals in the
sediments—notably lead and zinc—were higher
than previous estimates.
• Studies were also undertaken to explore the
amount of potential energy locked inside the Tar
Ponds sludge. The incineration process to come
will generate steam and electricity which will par-
tially offset the costs of the clean-up.
Behind the Scenes
Designing and planning for the
incineration and power generation
phases of the clean-up project contin-
ued at a steady pace throughout the
year. The technical, environmental
and economic details of the clean-up were each given
special consideration.
• Acres International Limited was hired by federal
and provincial Environment departments to pro-
vide project management services for the clean-up.
• A Quality Assurance and Control program was
established to ensure consistency in all laboratory
tests and analyses. In a project as complex as the
Tar Ponds Clean-up, scientific data must be con-
stantly monitored and reviewed, and rigorous labo-
ratory standards must be met and maintained.
Quality control is especially important at this early
stage because the successful and environmentally-
safe destruction of the Tar Ponds material hinges on
accurate and reliable test results.
• Various methods of combustion were studied to
determine the most suitable system for incinerating
the Tar Ponds sludge, and destroying harmful
polynuclear aromatic hydrocarbons (PAHs). New
and existing combustion techniques and facilities
were examined and evaluated during the year. As
well, a special laboratory contract was established
to pinpoint the various incineration times and tem-
peratures required to destroy the PAHs at a 99.99
percent rate of efficiency.
• To ensure the clean-up does not release further
pollutants into the environment, plans were devel-
oped to monitor the local air, land and water sys-
tems for signs of contamination from the excavation
or incineration of the Tar Ponds contents. And since
current and future monitoring systems will generate
large amounts of data, the appropriate computers
and software were purchased for the project.
• A five-yep Communications Plan was developed
to provide public information about the clean-up
throughout the entire project. As part of the plan, a
large trailer was purchased to house the future Tar
Ponds Public Information Centre. The centre will
contain a public information display for visitors,
and will serve as a meeting place for Sydney
residents throughout the project.
One of a series of publications produced for the
1987—97 Sydney Tw Pon
C Pro
ISSN 0847—10101
Aussi spomtIe i frar j
13
-------
EXHIBIT B
INCINERATION PLANT FACT SHEET.
1. Tar Pond Fuel Storage Tanks.
Tar pond material will be transported from Muggah Creek to concrete
storage tanks which are located adjacent to the incineration plant.
Three tanks will be constructed, one being used for unloading the material,
the second to allow water to separate and the third for feeding fuel to
the incinerator.Total storage in the three tanks will provide sufficient fuel
for 5 days of incinerator operation at full capacity.
2. Tar Pond Fuel Feed System.
Method of fuel feed from tank Clam bucket to screws feeding fuel pumps
Type of fuel feed system High pressure solids piston pumps (One/unit)
Total capacity through solids feed pumps Variable flow rate: 3.4-16.6 cu.nt/hr
3. Incineration Equipment.
Type of incinerator Revolving fluidized bed combustor
Supplier Superburn Systems Ltd.
Number of combusters 2
Incinerator input rates:
Total tar pond feed rate 14.4 tonnes/hr (Average)
Limestone feed rate 2 to 6 tonnes/hr
Sand make-up rate 0.2.3 tonnes/hr
The method of feeding limestone directly into the incinerator neutralizes the
effects of sulphur and chlorine in the tar pond material.70 % reduction of sulphur
oxides will occur in the incinerator.
The incinerator will be started and brought to operating temperature using light
oil. Once this temperature has been reached tar pond material will be fed into the
unit and the light oil feed will be turned off.Coal., supplied locally, will be used
as the auxiliary fuel and will be fed automatically into the incinerator in the
event that the heating value of the tar pond material should drop sufficiently
to cause the gas temperature to fall.
-------
Air flow rates:
Primary combustion air flow 76,950 cu.m/hr
Secondary combustion air flow 15,000 cu.m/hr
In order to ensure total destruction of PAH’S and HNC’S the following guaranteed
criteria will be met:
Gas temperature in incinerator 950 degC
Gas residence time in incinerator 3 seconds
Hot gases will flow through approximately 20 metres of refractory lined ducting
to an existing modified Sysco boiler. .Approximately 50,000 kg/hr of superheated
steam will be produced and supplied to an existing steam turbine generator which
will generate 10 11W of electrical power.
Gases at approximately 200 degC will be ducted to the baghouse. Also, a percentage
of this gas will be recirculated back to the incinerator to provide cooling to the
fluid bed if required.
Flue gas recirculation rate 8,900 cu.m/hr
The baghouse has overall dimensions of approximately 12m x 2.O.5m x 16.5zn high.
It comprises 6 compartments each compartment containing over 200 filter bags.
The bags are cleaned by means of compressed air.
Particulate inlet loading to baghouse 80,300 mg/Ncum
Particulate outlet loading from baghouse 27.3 mg/Ncu.m
Clean gases from the baghouse are ducted through a fan to the discharge stack.
-------
old o11 spills, old landfills and chemical waste sites and will thereby
take care of the clean up activities due to waste management In the past
(I.e. before 1972).
The suggested act Implies decentralization of a great part of the work done
today by the central authorities to the municipality administration level.
Chemical Waste Sites
The local authorities In Denmark has up to now recorded about 1750 chemIcal
waste sites in Denmark. The total amount Is estimated to 5—6,000 sites.
In urban areas the mapping of chemical waste site Is made from the
knowledge of the Industrial development In the areas during the past 50—60
year. Thereby a large list of potential polluted sites is recorded. The
actual polluted sites will be registered after pre1lminary investigations
on the sites.
Learning by experience has made the investigations more efficient and less
expensive. Research and development of In situ/on site remedial actions
are of high priority.
New Guidelines
A guide dealing especially with the risks of health, which occur when uti-
lizing chemical waste sites for residential areas will be available (In
Danish) In next month. Research In risks 0 f health is of high priority.
The series of publications concerning previous times Industrial production
Including kind of pollution derived from the specific production processes
are planned to be extended with pub1lcatlons on petrol stations, tar manu—
factories and tanneries. The publications will be available in Danish in
1990.
The Waste Division is planning to up-date the guideline on sanitary land-
fills. The existing guidelines from 1982 Is primary dealing with construc-
tion recomendatlons. Research and development on biological degradation
within sanitary landfills, exploitation of gas etc. will be Included in the
new guideline, which will be available In 1991.
Finally, It should be mentioned, that by Initiative of the Danish
Engineering Society a recorrinendatlon for Sanitary Landfill Liners has been
made. The reconvnendatlon is valid for liners made of clay, polymers or
combinations thereof.
FEDERAL REPUBLIC OF GERMANY
There Is not anything new to report on in the Federal Republic of Germany
since the Copenhagen Workshop In May 1989. However, I would like to say a
16
-------
few words which may be helpful to better understand what Is happening in
the Federal Republic of Germany.
Every politician mentions the Altiasten Problem (the contaminated sites
problem), as one of the biggest challenges of today and the future. Every
responsible party - municipality or private industry - Is afraid of the
.clean-up cost In particular because of lacking threshold values to deter-
mine how clean Is clean. Responsible parties are expecting more stringent
demands from the authority for better decontamination.
What Is really happening in the Federal Republic of Germany is the iden-
tification of suspect sites and the ranking of them on a state or regional
basis. There is no Federal ranking system currently as the Federal
Government has not taken any responsibility for clean-up and no money to
allocate for clean-up actions. Therefore, a Federal priority list would be
of no help.
In the Federal Republic of Gemany, 40,000 out 0 f 48,000 suspect sites are
abandoned waste dumps or former landfills assumed to contain more or less
hazardous waste. No one knows how to decontaminate these sites. Hydraulic
control of the plume of the polluted groundwater, and encapsulation with
caps and walls have been chosen to secure these sites to get time until
they can properly decontaminated.
Many more new techologles have been developed to clean soil from con-
taminated sites. Soil washing, thermal treatment, and biological treatment
facilities are available. However, only a few thousand tonnes have been
treated. This is due to high costs, difficulties in obtaining licences to
operate the plants, and the lack of enforcement to clean soil Instead of
landfllulng or doing nothing. However, there will be some pressure an soil
cleaning requirements In the near future as the selling of land in urban
areas Is severely hampered.
In the NATO/CCMS Pilot Study we Increase our knowledge on technologies.
However, we should not forget to exchange experience on how to get the
technologies used.
FRANCE
Since our last meeting In Copenhagen the main event which occured In
France In the field of contaminated sites question has been the publica-
tion, In July, by the Secretary of State for Envionment of a new report on
old dumps of industrial residues.” In this report:
- the policy of National and local authorities is explained
- It Is said that the registration of contaminated sites is far from
being completed and will go on
— a new list of about 100 of newly registered sites Is given.
17
-------
NORWAY
Till now, most of the effort on hazardous waste management has been
establishing the legal fundamentals and establishing a system for collec-
tion and disposal/treatment of hazardous waste. We have no central treat-
ment plant as In Denmark, but the parliament has now decided there shall be
a central treatment plant in Norway. The plant is expected to be starting
in 1992/93.
Concerning contaminated land and ground water, it took a long time before
we discovered these problems. Most of the drinking water supplies are sur-
face water from inland, but most of the industry Is located near the
coastline. Neither do we have many heavy chemical industrial plants. Some
problems showed up, however such as old gas works and wood preserving
industry. A special problem is leachate from tips/fills from mines.
Therefore we started a couple of years ago a project on methods for
locating hazardous waste sites, and now we are running a project on
locating the sites. The project is to be finished in 1990. In the mean-
time, we shall regard methods, criteria, priority, etc, for cleaning up and
we shall regard the economical aspects: who shall pay for the cleanup
actions.
Concerning clean up criteria, we have made a study by Dr. Robert L.
Siegrist on “International Review of Approaches for Establishing Cleanup
Goals for Hazardous Waste Contaminated Land.” I hope the report will be
available to the workshop In Oslo, March 1990.
THE NETHERLANDS
“Concern for tomorrow”, a National environmental survey 1985 — 2010, was
published by the RIVM in December 1988. Based on this report, the Dutch
government developed a National Environmental Plan for the next 5 years.
This plan emphasized that more attention has to be payed to clean-up opera-
tion. In persuance to this mandate, the Dutch Ministry of Housing,
Physical planning, and Environment established a steering con’wittee and
requested that it develop a “ten year scenario for clean-up of contaminated
sites.”
Number of Contaminated Sites
Survey
Potential
Sites
Urgent Sites
Estimated Costs (*lO 9 Dfl)
1981
4.000
1.000
1-2
1987
7.500
2.000
2-3
1989
100.000+
6.000
5
+Incl. current Industrial sites
18
-------
This ten year scenario was developed and based on a 1989 survey. This sur-
vey extimated the number of potential contaminated sites to be 100,000.
This includes former and current industrial sites (99%) and waste disposal
and automobile junkyard sites (1%). It Is expected that of the 100,000
sites, 6,000 are urgent sites.
The steering corrinittee recomended that the anticipated 6,000 urgent sites
be cleaned-up In the next 10 years. For this cleaning up, about 5 billion
Dfl must be appropriated. Up to now, the central government has financed
90% of the clean-up cost. In the future, it is proposed to finance the
work in the following way.
Central government 300 million Dfl/year
Local authorities 100 million Dfl/year
Polluter 100 mIllion Dfl/year
In the near future these recoinnendatlons will be discussed by our newly
formed government.
Work has also proceeded on the study “Alternative physico—chemical and
thermal treatment of contaminated soil.” This project Is carried out
within the framework of the Netherlands Integrated Soil Research Progranine.
The aim of this study is to give a survey and a systematic evaluation of
the physico-chemical and thermal techniques in the Netherlands and foreign
countries in search for new possibilities and cast—effective alternatives
for the treatment of contaminated soil. On the basis of the evaluation, a
research programe for the development of the most promising techniques
will be reconmuended for the Comittee for the Development 0 f Soil
Protection Techniques of the Netherlands Integrated Soil Research
Programe.
The project will be carried out in three stages. This report Includes the
results of the first stage of the study: a survey of alternative
techniques based on a literature study and a first selection of the most
promising techniques by means of a ranking system.
Extending the main target alternative techniques are also considered for
the possibilities for the cleaning of other waste streams like slurries and
sludges. Beside the cleaning techniques, a limited range of imobilization
techniques is also examined, because these techniques have a fair
perspective for the remediation of soils heavily contaminated with both
heavy metals and organic compounds.
The ranking system is based on four criteria:
(A) The range of application with regard to types of soil and pollution,
(B) The priority of this application (on the base of todays bottlenecks
of the operational treatment techniques),
(C) The stage of the development and
19
-------
CD) The market perspective.
In the present ranking system the criteria (A) and (B) are emphasized.
A rough selection of the most promising techniques was possible with this
ranking system. However for some techniques a further evaluation was also
necessary. After this followup, sixteen techniques were selected. In the
second stage these techniques will be more extensively described in eight
monographs concerning: supercritical oxidation, rotary kiln Incineration,
a plasma reactor, dechlorination techniques, particle separation
techniques, froth flotation, extraction with complexing agents and
electroreclamation.
The (six) selected particle separation techniques will be combined in one
monograph because they have many comon aspects. Tough Imobilization
techniques were also selected on basis of the ranking system no separate
monograph will be prepared on these techniques, because they are not
considered as treatment techniques.
In the third stage the techniques will be evaluated more extensively on the
basis of the monographs on environmental and economical aspects. Based on
this evaluation, the most promising techniques will be reconinended for the
Programe Comittee of the Integrated Soil Research Programe. A brief
evaluation of imobilization will also be made based on the information
collected in the first stage. The results of the second and third stages
of the study will be published together in a second report.
UNITED KINGDOM
Developments regarding contaminated land issues In the UK since the Third
Workshop, Copenhagen, include the following.
The House of Comons Select Comittee Inquiry into Contaminated Land has
now finished taking evidence. Department of the Environment officials gave
oral evidence on 22 May 1989 and the Department of the Environment Minister
with responsibility re. contaminated land, Mr Trippler, appeared before the
comittee on 1 November. The Coninittee has visited the Netherlands and
Italy to see how contaminated land is dealt with in other EC countries.
The Society for Chemical Industry (SCI) held a successful meeting ‘Cleaning
Contaminated Soils’ on 29 June 1988. Speakers included a Department of the
Environment Minister, Mrs Bottomley, and experts from the UK, USA and the
Netherlands.
At the SC! meeting Mrs Bottomley announced the extension of the Environment
Protection Technology (EPT) Scheme to include funding for research into
novel ways of cleaning contaminated land. The function of the EPT scheme
20
-------
Is to encourage new ideals in environmental protection by giving grants of
up to 50% for innovative research projects In selected areas. (Contact
Point: EPT office, Room B357, Romney House, 43 Marsham Street, London SW1P
3PY). Under the scheme about 10 mIllion of grant—aid will be available
over four years. Contaminated land is one of seven priority areas.
The Department of the Environment ‘Review of Derelict Land Policy’ was
published in September 1989. The review includes sections regarding the
nature and extent of derelict land in the UK based on the 1988 survey, a
review and evaluation of current organisatlonal mechanisms for the
prevention and reclamation of dereliction. Policy Issues are discussed.
Dereliction is not synonymous with contamination. Dereliction Is a wider
term encompassing, for instance, abandoned buildings and disused chalk,
gravel or clay pits. Derelict land Is reclaimed under a variety of
government progranines IncludIng Derelict Land Grants (DLG) City Grants and
the work of Urban Development Corporations. Derelict Land Grants have
resulted in the reclamation of 8,500 ha of land between 1982—1988 (much of
which will have been contaminated). Expenditure on DLG during this period
amounted to 650 millIon. (The Review Is available from Room C16/06, 2
I4arsham Street, London SW1P 3EB, Tel 01 276 4466).
Also of interest to the Pilot Study may be the joint Department of the
Environment/Welsh Office circular ‘Landfill Sites; Development Control’
Issued 26 July 1989 (DoE 17/89) which gives further advice to local
authorities about the use of their planning powers In relation to landfill
sites In England and Wales with regard to landfill gas.
The following figures are taken from the 1989 Derelict Land Policy Review
and Illustrate the changes In UK derelict land distribution from 1982 to
1988. It is again emphasized that dereliction is not synonymous with
contamination.
UNITED STATES
Program Progress
— Superfund Amendments and Reauthorization Act (SARA) passed Oct.
17, 1986. Three years under that law completed In October, 1989
with results (pre-and post-SARA) were shown on charts 1-3.
— Chart 4 shows the response of responsible parties in funding
early cleanups (removals), RI/FS (engineering studies) and RD/RA
(design/construction). Over $2 billion of responsible party
contributions are represented on this chart.
Technologies
— Charts 5 and 6 show the relative use of technologies In the past
2 years versus the first 50 records of decision (site reme—
diation choices). A possible trend toward more non—themal
choices is shown.
21
-------
DERELICT LAND RECLAIMED AND BROUGHT BACK INTO USE
1982-1988 BY END USE OF LAND
Hard End Use
Sport and Recreation
06%
Public Open Space -
310%
industry
13 O%
Other
0’3%
Residential
96%
Commerce
4 •0%
1
p - I
C
Agriculture/Forestry
210%
Sott End Use
-------
CHANGES IN AMOUNT OF DERELICT LAND AND AREA JUSTIFYING
RECLAMATION 1914-1988, BY TYPE OF DERELICTION
S.
4
Spoil Heups
Derelict
Reflway Lend
Other Forms
of Dereliction
1002
t Area Justilying Iteelanisfion
‘See
I d
I2
l0
S.
..
C,
0
I
0
C
0
I .-
I ’ S.
I S 52
1074 $082
1074
1002
lee.
$074
-n
C,
I,,
2
1974
1082
Excavations Military
and Pits Darellotlon
-------
CHANGES IN THE AMOUNT OF DERELICT LAND AND THE AMOUNT
JUSTIFYING. RECLAMATION 1074-1088
L ZJ Ar.a JustItyth R•olamailon
1062 Stock
a
a
a
U
a
- I
0
a
C
a
a
0
I-
40
10’
20•
*0’
-10
1006 Block
H.w
D•s.flciIon
C!..l.d
1002- 1060
SO
1074 hook
P1.w
D. .I odon
C,..t.d
gaI- 1 182
C,
r i - i
(A)
fl.oJ.Im.d
1014-1082
Recl Im.d
1082-1008
-30
1974-1982 1082-1988
-------
FIGURE 4
• FLOWS INTO AND OUT OF DERELiCTION — BY TYPE
1982 OTHER 1982 SPOIL 1982 MUJTARY 1982 1982 RAILWAY
QNOUSTRJAL) HEAPS DERELICTION EXCAVATiONS DERELICTiON
DERELICTION AND PITS
11109ha 8300ha 2452ha 6402ha 6015ha
RECLAiMED RECLAIMED RECLAIMED RECLAiMED RECLAIMED
1982-1988 1982-1988 1982-1988 1982-1988 1982-1988
5428ha ‘ ‘ha 1412ha 1939ha 2012ha
-49% —39% —58% —30% —33%
NEW NEW NEW NEW NEW
DERELICTiON DERELICTION DERELICTION DERELICTION DERELICTION
7008 ha 2388 ha 1021 ha —94 ha 1043 ha
+63% +29% +42% -1% +17%
DERELICTiON! DERELICTION DERELICTION DERELiCTION DERELICTION
1988 1988 1988 1988 1988
12689ha 7466ha 2061ha 4369ha 5046ha
14% -10% -16% —32% -16%
25
-------
FIGURE 5
FLOWS INTO AND OUT OF DEREUCTION 19741988
___ H P _______________
NOTE Thes. figures wiU not preasIdy maxo’ those in R;.8 because of the diflerencis in the basis of redamanon and dsreliotlon figures
szp ainsd in pats 3.4. All figures have been rouncea inc may not max exaCdy Vt. IesuI of the Derelict L.asid Survey.
NEW INDUSTRIAl. Etc.
DERELICTiON
(9. 00 hal
STO cK OF DERELICT LAND
1974
33000 ha
STOCK OF DEE.JCT LAND
1982
34.500 ha
PEE RESOURCES 1982-1988
E555 million)
i
UNIT COSTS
E?5. per ha)
I
NEW MINERALS
DERE.ICTiCN
2.SC0 hal
RGURES IN BMC’i TS ARE
ESTiMATES/AVERAGES
GRANT AJD
REC.AMATION
(9000 ha)
ADDITiONS TO STCCIC
OF DERELiCT LAND
(11,000 ha)
R UC71CNS TO STOCK
OF DEREJCT LAND
(13.500 hal
NCN.GRANT AJD I
RE ..AMAflCN
(4. 00ha)
STOCK OF DERELICT LAND
1988
32.X0 ha)
TREATED
OThERWISE
(—)
26
-------
FIGURE 6
FLOWS INTO AND OUT OF DERELICTION 1988 1 994
STOCK OF DERELICT I
1994
(28. cO-29.5C0 ha)
STOCK OF DERELICT LAND
1982
34.000 ha
NEW INDUSTRIAL. E .
DERE.JCTION
(8.000-4.000 ha)
I
L
PES RESCURC S 1988-1994
( 140miIIicnp .a .)
,
UNif COSTS
(V5.000 per ha)
STOCK OF D E.JCT LAND
1988
C0 ha
NEW MINERALS
0EELJCT ON
3.000—4.C00 ha)
N
I
GRANT AiDED
RECLAMATION
(11.000 ha)
ADOmONs TO STOCK
OF D EJC7 LAND
(1Looo-1a ha
REDUCTIONS TO STOCK
CF 0E EJCT LAND
(1 .5OO—16.S00 ha)
NCN-GRANT AIDED
______ REC LAMAflCN
I (4.000-5.000 ha)
FIGURES IN BRACKETS
ARE ESTiMATES
N ___
TREATED
OTHE.W1SE
(500 ha)
27
-------
CHART I
Most Sites in Superfund Inventory Have Been Assessed
CERCLIS Inventory
31,904 *
Awaiting Assessment
Site Inspection
Complete
10,610
No Further Action
(P,. nina,y As..ssmsnl Coinpisle)
1 4,230
Awaiting Site
Inspection
4,994
as of 10-27-89
CHART 2
WORK AT THE CONSTRUCTION END OF THE
REMEDIAL PIPELINE HAS INCREASED POST-SARA
• 314 CONSTRUCTION PROJECTS HAVE
BEEN INITIATED
00 • 175 RA MANDATE HAS BEEN ACHIEVED
• RI/PS STATUTORY GOAL HAS BEEN MET
400
• RODS AND RDS INDICATE GROWING
NUMBER OF CONSTRUCTION PROJECTS
200
Is
0 — AM C C ETED 1 3
• POST.SARA (3 YEARS)
• PRESARA (6 YEARS)
RI/FS
28
-------
CHART 4
CHART 3
NUMBER AND VALUE OF
SUPERFUND SETrLEMENTS
Trends in Superfund Accomplishments
1987 - 1989
FY1989
(Total • $980 milhonj
FY1988 (Total — $470 milllon
FYI9BO-1987 (Total.$626m 1111on)
4 00
348
p1841
295
Number of
Settlements
1987 1988 1989
Remedial Investigation!
Feasibility Studies Started 183 150 167
Fund (120) (93) (77)
RP (63) (57) (90)
Records of Decision Signed 76 153 138
Remedial Designs Started 57 99 153
Remedial Actions Started 35 72 108
200
0 —
Removal RUFS RDIRA
Data Source CERCLIS - Prelfminaiy Pt 1989 End .ot .Year Data.
Dale. October 20.1969
-------
CHART 5
EXHItIT I
Summary of FY 1987 ROD Sites Selecting Treatment
Technologies as Components of Source Control Remedies
CHART 0
SUMMARY OF FY 1989 ROD SITES SELECTING
TREATMENT TECHNOLOGIES AS COMPONENTS OF
SOURCE CONTROL REMEDIES’
NUMBER
OF SITES
TYPES
13
7
2
3
2
I
4
I ICINERAT 1ON
SOUDIF1CAflON1FIXAT1ON
STABIL ATIOP4 INEUTRALIZAT)ON
VOLATII.2ATIOWAERAT1ON
SOIL WASHING/FLUSHING
BIODEGRADATION
OTHER
-------
-
Sohdihcation/Stabllization (6)
Biodegradation!
Land Application (4)
Soil Washing!
Flushing (1)
VoIati zationI
Soft Aeration (5)
Other Treatment
Technologies (1)
VacuunWapor
Extraction (6)
‘Treatment t.chno ogy r*nrtes reprewl 49 ei nsd ROD, out clan ealinated total of
i so ROD, lobs s ned durii FY 19 .
°Mote than one trsatmsrt lochnology may b. auocâated with a site.
C Incineration inc ides an amended ROD which re lacss Outboard Maflne. 1.5/15 184 ROD.
BIODEGRADATION
(1)
(9
0
STABILZATIONI
NEUTRALIZATION
(2)
-40-
-------
Program Regulations/Studies
Two major regulations were required to be revised by SARA for the U.S.
Superfund program:
- National Oil and Hazardous Substances Contingency Plan (NCP) is
the °blueprint” regulation outlining processes and procedures
for remediation in U.S. It will be finally promulgated in
February 1990.
- Hazard Ranking System (HRS) Is used to rank sites on U.S.
National Priority List. HRS will also be revised in late winter
1990 to better recognize the nature of health and ecological
risks at sites.
- The new administrator of EPA conducted a 90 day study of the
U.S. Superfund program In February 1989. The final report with
50 major recoirrnenatlons as well as the work plan to accomplish
these reconvnendations is available from EPA.
- The 2nd Forum on Innovative Hazardous Waste Treatment
Technologies: Domestic and International will be held May
15-17, 1990 at the Wyndhani Franklin Plaza, 17th & Race Streets,
Philadelphia, PA. It will showcase U.S. and foreign vendors
with specific treatment technologies. The proceedings of the
First Forum held In Atlanta in June 1989 are available from EPA
— CERI etc.
31
-------
Project Reports
33
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTERIZATION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On-/Off—site Treatment Location:
Pre— and Post—treatment Requirements:
Physical/Chemical
Pump and treatment of aquifer
Envi ronment Canada
Experimental
Halogenated hydrocarbons, aromatic
hydrocarbons, iron
Groundwater
On—site
Sludge treatment, carbon regenera—
ti on
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Mr. J. Schmidt
Wastewater Technology Centre
Environment Canada
P.O. Box 5050
Burlington, Ontario
CANADA L7R 4A6
416-336—4541
Ville Mercier aquifer, Quebec,
Canada
1 ,2—dichloroethane, benzene,
various mono— and polycyclic
aromatics
Permeable sand/gravel ridge con-
taining lagoons with underlying
glacial till and fractured bedrock
Available March 31, 1990
Available March 31. 1990
Richard Martel
Envi ronment Quebec
2360 Chemin Sainte—Foy
Ste. Foy, Quebec
CANADA GIV 4H2
418-646-7688
11/89
5-1
35
-------
ASSESSMENT OF CONTAMINATED GROUNDWATER TREATMENT
AT VILLE MERCIER, QUÉBEC
by
Marc Halevy, Robert M. Booth and James W. Schmidt
Environment Canada, Conservation & Protection
Wastewater Technology Centre
Burlington, Ontario
Presented at the Third International Meeting of the NATO/CCMS
Pilot Study Demonstration of Remedial Action Technologies
Contaminated Land and Groundwater - Montreal, Quebec
November 5 - 9, 1989
36
-------
1
1 BACKGROUND
The Vile Mercier 8tudy site is located on the south shore of the St. Lawrence
river, 20 km south-west of Montreal, Quebec (see Figure 1).
In 1968, the “R gie des Eaux du Québec”, a Quebec Government agency,
authorized Lasalle Oil Carriers to temporarily store oily liquid wastes from petroleum and
petrochemical industries in a former sand and gravel pit located near Vile Mercier,
with the intention of recovering the oil components at a later date.
From 1968 to 1972, an estimated 40 000 m 3 of liquid waste was stored at
the Mercier site. The liquid wastes consisted of used oils and refinery sludges,
chlorinated hydrocarbon and industrial solvents, paint residues, insecticides, pesticides,
mercaptans, polymers, acids, alkalies and other undefined residues.
By 1971, the wells supplying local farmers with groundwater were
contaminated. This conts mination was directly attributed to the oily wastes stored at the
Mercier site.
By 1972, the permit issued to Lasalle Oil Carriers was suspended, and site
remediation was initiated by commissioning Tricil to build and operate a liquid waste
incinerator to treat the sludges and organic wastes.
By 1975, most of the liquid wastes were removed and incinerated. However,
it was not before 1981 that residual sludges, mainly consisting of non-combustible residues
were removed, treated and disposed of in a marine clay landfill site, 500 meters east of
the Mercier site. It was also know in 1981, according to studies conducted by Hydrogeo
Canada, that the contaminated groundwater extended over an area of 30 km 2 . Later on
that year (1981), the Ministry of Environment of Quebec (MENVIQ) decided to take action
by commissioning Foratek International Inc. and The SNC Group to conduct
hydrogeological and treatment technology feasibility studies respectively, in order to
determine the most technically and economically attractive option to cope with the
contiiniination and its progression.
Following the studies’ recommendations in 1982, it was decided that restoring
the aquifer by means of a pump-and-treat scenario set up in the most heavily polluted
zone (see Figure 2), would hold the most promising potential.
The range of concentrations of the contaminants of major concern found in
the raw groundwater during the 1982 feasibility studies’ pumping tests are presented in
Table 1, together with the plant’s effluent objectives set by the MENVIQ.
A treatment system consisting of three (3) purge wells with an overall
capacity of 76 Ifs would pump up enough groundwater to create a hydraulic barrier in
37
-------
2
FIGURE 1. LOCATION MAP OF THE VILLE MERCIER SITE (REF. 1)
VIII. Mercier sits
U.S.A.
Sal
•VILLE MERCIER SITE
38
-------
P -’ . ”-i Sand and gravel complex
__________ Narn. clay
Sscias landfill
Groundwstsr triatmint facility
inciniratlon plant of organic wastes
• Pivg. well
- Groundwater flow
— Groundwatar divide
- Treatment facility discharg. pipe
FIGURE 2. HYDROGEOLOGICAL SET 1NG OF
THE VILLE MERCIER SITE (REF. 1)
39.
0 O J00 4
-------
4
TABLE 1. RANGE OF CONTAMINANTS AND EFFLUENT OBJECTIVES
Parameters
Concentration ( xgfL)
Objective
Minimum
Maximum
Avg.
( .LgIL)
Phenols
167
1000
507
2
1,2 Dichioroethane (1,2 DCE)
76
517
187
50
111,1 Prichloroethane (1,1,1 TCE)
23
326
127
33
1,1,2 Trichioroethane (1,1,2 TCE)
16
326
-
50
1,1,2 Trichioroethylene (1,1,2 TCEe)
19
349
114
4.5
Chloroform (CC1 3 )
1
165
-
50
Chlorobenzene (-)
2
16
-
50
Trans 1,2 Dichioroethylene (Trans
6
139
-
50
1,2 DCEe)
PCB’s:
Arochior 1242, 1254, 1260
.01
.08
.05
.01
All V.O.C.’s
-
-
-
50*
Iron (mgdL)
.84
12.00
8.19
.30
Manganese (mg/L)
.06
.82
.18
.05
Suspended Solids (mgfL)
-
3
Turbidity (NTU)
-
5
* Total V.O.C.’s . tit.. in effluent should not exceed 1000 p g/L.
g’ y1 rrc
‘IT’
order to control the progression of the plume, and restore the aquifer within a five year
period.
The recommended treatment train consisted of the following sequence:
1. Aeration - induced stripping to oxidize iron and manganese, and to remove
volatile organic compounds (VOC’s).
2. Coagulation and precipitation , following alum and polymer addition, and
rapid sand filtration to remove iron and manganese;
3. Activated carbon adsorption to remove remnining organics.
The sludge treatment cycle would consist of the following scheme:
1. Dewatering through pervious tiles located above a negative-pressure
chamber (vacuum).
2. Disposal in a suitable landfill site.
40
-------
5
In addition to the need to rehabilitate the aquifer for ecological reasons, such
a decontamination effort would offer a unique opportunity to collect necessary information
on the treatment’s effectiveness and costs that such an exercise would call for. Hence,
in Canada, the first full scale program to restore a contaminated aquifer was
commissioned in 1983 in the town of Mercier, Quebec. The MENVIQ allocated at that
time a total of $5.7 million to restore the Mercier site as follows: $3.0 million would cover
the groundwater treatment plant’s construction cost and $2.7 million would pay for its five
year operation. In 1986, an additional $1.7 million supplemented the original operating
cost estimate in order to complete the restoration work.
2 DESCRIPTION OF TEE ORIGINAL TREATMENT TRAIN
This section briefly describes the original concept and design of Mercier’s
groundwater treatment plant. The following section (3) presents the problems encountered
and the solutions implemented. The problems were resolved through modifications to
some of the original design parameters and through the addition of new process elements
to the train.
The original treatment train consisted of the following five (5) unit processes:
• Aeration-induced air stripper
• Flash-mixer
• Pulsator clarifier
• Rapid sand filters (2)
• Activated carbon columns (3)
3 PROBLEMS ENCOUNTERED AND MODIFICATIONS IMPLEMENTED
A few weeks after plant start-up, serious operational difficulties were
encountered:
• Bacterial growth fouling throughout the plant;
• DNAPL’s entry into the plant resulting in coating the unit processes with oil;
• Excessive iron precipitation in the aeration-induced stripper;
• Incomplete removal of iron;
• Granular activated carbon loss during backwashes;
• Reduced capacity of the wells due to fouling of the stramers;
• Polymer incompatibility;
• increased load of contRminants;
41
-------
6
4 EXiSTING TREATMENT TRAIN
This section briefly presents the operating treatment train since its last
modification in January 1987. The existing treatment plant’s characteristics and process
diagram are presented in Figure 3.
4.1 Hydrogen Peroxide Addition
Hydrogen peroxide addition represents the first process element of the treatment
train. Its purpose is to oxidize iron and manganese. The estimated dosage of hydrogen
peroxide is 11 mg/L
4.2 Chlorine Addition
The purpose of chlorine addition is to control biological growth in the aeration-
induced stripper and flash mixer by creating a “free-chlorine” residual. Chlorine
requirement in the Plant varies between 10 and 20 mg/L, depending on the influent
pumping rate and the quality of the raw ground water.
4.3 Aeration-Induced Stripper
Stripping represents the third process element of the treatment train. No
design modifications have been carried out on that unit since conception.
4.4 Flash-Mixer
The flash-mixer represents the fourth process element of the treatment train.
No design modifications have been carried out on that unit since conception. Alum is
presently used as the coagulant. Its dosage varies between 50 and 85 mgfL. Aquafloc
464 (Dearborn Co. ) is the polyelectrolyte presently added in the flash-mixer to aid
coagulation. It is a high MW slightly cationic polyacrylamide. Its dosage varies between
0.7 and 1.2 mg/L with an average value of 0.9 mg/L.
4.5 Chlorine Dioxide
Chlorine dioxide addition represents the fifth process element of the treatment
train. The sole purpose of incorporating this process element in the train was to gain
control over biological activity in the GAC colmnns, chlorine dioxide being a stronger
oxidant thRn chlorine and a less adsorbable compound on GAC than chlorine is (Ref. 2).
Chlorine dioxide is added at 2.5 mg/L prior to the pulsating clarifier.
42
-------
7
PROCESS DIAGRAM
HYDROCEN
THREE IF R0 DE 1 SIRIPP(R I TW0 I STORAGE FIRST I SECOND
WELLS I C0 IcLARInERI S.ND RESERVOIR STAGE STAGE STORAGE I SUJDGE F1LTRAT 1ON
LCHLORINE FLASH I DIOXIDE
DDrn0HS 1 MIX J ADDf11ON FILTERS I GAC GAC TA AND DRY1NG SYSTEM
TREATMENT PLANT CHARACTERISTICS
SUBMERSIBLE PUMPS AND WELLS CLARIFIER SLUDGE DEWATERING
Dsdpe fls . 27LSi,, 1 1 1h (VACUUM FILTATION)
3 type Dy Is Owlfl.qII.ii
plo. e w i-. — 3. i/s •.i — a sa s
T.Id Need U iii Pw Iknr.bsr .1 fllw bs S
Molsr kIpI$ tøfIn 30 EW ____ Vie 5.5 M 2 12.2
7S4 - __ Ved. Usthi
M Ilee .1 Peipin
HYDROGEN P(ROX1D(
RAPID SAND FILTERS IdQ Pw’ S.$ IUI
C dVsnbus T 755 I
Yo mam PvU! 0.11 UI/W l
NI Rb ’ S f Vvdgs 27255 1/4
CHLORINE Dielpe flee s 35 I/s
__ (, )
-LM. - - .
fir Sed
AIR STRIPPER IOS.7 e
I V i.
Peekbi, Ns1 d 2.10 UI LI% 2.7 I/u/ OTHER CHARACTERISTICS
2.74 ii
Pas¼kiq sdIs P Va9s 20.5 see
VO IU,I. S I MS S 12.74 UI ACTIVATED CARBON FILTERS
2 . 13 m’/.ee 3 IMUS
IO U-S l aV
FLASH—MIX Iksi0.- $ IiUp
ut _ type WV-S Ca. tru s I eIAte eIs4 a,...r.La
Ve(vm. .1 OwMIbr S m fir 1 0.1 — 14 -._ a wvu . t. sw
4405011 SI C 50 5 11 10 1d/fl0 50IISVd 1 flIrs5 . 10 2 v.-ee’
350 SPy leel
flee Sa 75.7 l
Vdwee SM iui
CHLORINE DIOXIDE air’ 2 1 $
flhi 0.1
fls. 5 s Il l/ s
YdWUS 114 11
FIGURE 3. GROUNDWATER TREATMENT FACILITY
43
-------
8
4.6 Pulsating Clarifier
The pulsating clarifier represents the sixth process element of the treatment
train. No design modifications have been carried out on the unit since conception.
4.7 Rapid Sand Filters
Rapid sand filtration represents the seventh process element of the treatment
train. No design modifications have been carried out since conception other than the use
of a coarser sand as the granular filtration medium.
4.8 Activated Carbon Column
Activated carbon adsorption represents the eighth process element of the
treatment train. No design modifications to their set-up have been carried out since
conception.
5 TOIUCAL PERFORMANCE
The treatment objectives for the Mercier groundwater treatment plant were
presented previously in section 1. The SNC Group, responsible for the plant’s operation,
carried out bi-monthly samplings of the raw and treated water, and occasionally of
partially treated water sampled throughout the treatment train in order to verify the
plant’s ability to meet the objectives set by the MENVIQ. The characteristics of the raw
water are described by R. Martel (MENVIQ) in the document presented and submitted
to NATO, November 7-11, 1988 at Bilthoven, Netherlands (Ref. 3).
In order to gain a better understanding of the operation of the treatment train,
the analysis of the data gathered on the plant’s performance was subdivided into three
(3) populations:
1. The first time period ranges from July 1984 and ends in April 1985, and
corresponds to an evaluation of the original design of the plant.
2. The second time period ranges from May 1985 and ends in December 1986,
and corresponds to an evaluation of the partially modified design of the
plant following hydrogen peroxide and chlorine additions.
3. The third time period ranges from January 1987 and ends in August 1989,
and corresponds to an evaluation of the modified design of the plant
following hydrogen peroxide, chlorine and chlorine dioxide additions.
44
-------
9
5.1 Data tnnlysis: 1984 - 85
A limited number of samplings (4) were carried out during this period since
the plant’s operation was severely impaired (refer to section 3). However, when operating,
iron, manganese and 1,1,2 TCEe were removed effectively with effluent concentrations
meeting the objectives set by the MENVIQ. Removal of 1,1,1 TCE was nil but both
influent and effluent concentrations were below the objectives set by the MENVIQ. PCB’s
influent and effluent concentrations were below the detection level of 0.01 gL and,
therefore, met the objectives.
However, 1,2 DCE and phenols effluent concentrations averaged 1476 )LgfL and
76 p.gL and therefore did not meet the 50 t.g/L and 2 i.g/L respective objectives set by the
MENVIQ. Table 2 summarizes the information presented above.
TABLE 2. SUMMARY TABLE - HISTORICAL DATA 1984 - 1985
Parameter
% Removal
Iron
Mn
1,2
DCE
1,1,1
TCE
1,1,2
TCEe
Phenols
PCB’s
Raw
-
-
-
-
-
-
-
Effluent
95
94
43
0
96
84
n.a.
5.2 Data Annlysis: 1985-86
A more extensive (35 samplings) and detailed data base is available for this
period, thus allowing the evaluation of the performance of:
a) Hydrogen peroxide, chlorine additions; stripper; pulsating clarifier; sand
filters, and
b) Sacrificial and polishing GAC columns.
Overall iron removal was 96%: 84% by the first part of the train, the
remainder (12%) by the GAC columns. The average effluent iron concentration was
0.09 mgfL and therefore, met the objective of 0.30 mg/L.
Manganese, however, was not sufficiently removed (52%) to meet the
MENVIQ’s objectives of 0.05 mg/L. Its average effluent concentration was 0.07 mg/L
with most of it being removed by the first part of the train. Effluent concentrations of
1,1,1 TCE, 1,1,2 TCEe, and PCB’s all met the objectives set by the MENVIQ.
45
-------
10
However, one should note that the influent concentrations of 1,1,1 TCE and PCB’s
meet those requirements prior to treatment.
1,2 DCE and phenolic compounds did not meet the 50 pg L and 2 j .igfL
respective objectives since their average effluent concentrations were 872 tg/L (34%
removal) and 4 p.gIL (96% removal). One should note that 89% of the 1,2 DCE
removed was eli’ninsited prior to the GAC columns. Table 3 summarizes the
information presented above.
TABLE 3. SUMMARY TABLE - HISTORICAL DATA 1985 - 1986
Parameter
Cumulative
% Removal
Iron
Mn
1,2
DCE
1,1,1
TCE
1,1,2
TCEe
Phenols
PCB’s
Raw
-
-
-
-
-
-
-
Post-Filters
84
46
30
62
73
33
n.a.
Effluent
96
52
34
68
92
96
n.a.
5.3 Data Aniilysis: 1987-89
An even more extensive (65 samplings) and detailed data base is available
for this period, thus allowing the performance evaluation of the following combinations
of unit process(es):
a) Hydrogen peroxide, chlorine additions and stripper;
b) Pulsating clarifier, sand filters;
c) Sacrificial GAC column;
d) Polishing GAC columns.
Iron was removed effectively by the treatment train (97%) and its average
effluent concentration of 0.12 mg/L met the 0.30 mg/L objective set by the MENVIQ.
Most of the iron (95%) was removed by the pulsating clarifier and sand
filters, the remainder (2%) being filtered in the GAC columnfi.
Manganese was poorly removed by the train (32%), and its average effluent
concentration of 0.13 mg/L did not meet the 0.05 mg/L objective set by the MENVIQ.
46
-------
11
1,1,1 TCE was mainly removed (63%) by the H 2 0 2 -C1 2 -stripper combination
while the overall impact on removal that was offered by the GAC columns was nil.
Its average effluent concentration of 1.6 .tg/L met the 33 .g/L objective set by the
MEN VIQ. Its average influent concentration, however, also met this requirement.
The effluent concentrations of 1,2 DCE; 1,1,2 TCEe and phenolic compounds
did not meet the MENVIQ objectives of 50 pg/L, 4.5 ig/L and 2 p g/L respectively.
1,2 DCE was mainly removed by the H 2 0 3 -C1 2 -stripper combination (29%). An
additional 14% was removed by the GAC columns. Its average effluent concentration
was 1057 g/L. 1,1,2 TCEe was mainly removed by the H 2 0 2 -C1 2 -stripper (66%). An
additional 23% was removed by the GAC. Its average effluent concentration was
9.9 g/L. Phenolic compounds were well removed by the H 2 0 2 -C1 2 -stripper combination
(51%). An additional 40% was removed by the sand filters and GAC. Their average
effluent concentrations were 3.5 i.g/L. Table 4 summarizes the information presented
above.
TABLE 4. SUMMARY TABLE - HISTORICAL DATA 1987 - 1989
Parameter
Cumulative
% Removal
Iron
Mn
1,2
DCE
1,1,1
TCE
1,1,2
TCEe
Phenols
Raw
-
-
-
-
-
-
Post-Stripper
2
11
29
63
66
51
Post-Filters
95
16
24
n.a.
59
66
Post-SAC
96
21
52
73
84
79
Effluent
97
32
43
55
89
91
During the last operating year (1989), a series of eighteen (18) samples of
raw, post-sand filter and effluent water were analyzed for their content of the
following compounds: 1,1,2 TCE; 1,2, DCEe; 1,1 DCEe; 1,1 DCE; 1,1,2,2 TTCEe; CCI 3 ;
Toluene and Beuzene.
47
-------
12
1,1,2 TCE was poorly removed by the H 2 0 2 -C1 2 -stripper-sand filters
combination (29%). An additional 7% was removed by the GAC. Its average effluent
concentration of 547 g/L did not meet the 50 pg/L objective.
Average effluent concentration for the remninii g seven (7) compounds were:
19 pg/L, 19 tg/L, 15 jLg/L, 11 Lg/L, 5 Lg/L, 4 ig/L, 4 j.LgL respectively. These all met
the 50 gL objectives set by the MENVIQ. Table 5 summarizes the information
presented above.
TABLE 5. SUMMARY TABLE - HISTORICAL DATA 1989
Parameter
Cumulative
% Removal
1,1,2
TCE
1,2 1,1 1,1
DCEe DCEe DCE
1,1,2,2
rl’CEe
CC1 3
Toluene
Benzene
Raw
-
- - -
-
-
-
-
Post-Filters
29
63 51 57
46
47
53
67
Effluent
36
65 53 78
80
78
89
97
5.4 Observations
From the data collected by The SNC Group during the five (5) operating
years of the plant, and from the analysis presented, the following is observed:
1. 1,2 DCE and phenolic compounds do not meet the objectives for the first
time period.
2. Manganese, 1,2 DCE and phenolic compounds do not meet the objectives
for the second time period.
3. Manganese; 1,2 DCE, phenolic compounds; 1,1,2 TCEe and 1,1,2 TCE do
not meet the objectives for the third time period.
4. Although a combination of chemical oxidants were added to the train
during the second and third time period, manganese was more
effectively removed during the first operating year. Two possible
reasons for this may be:
a) Soda ash was substituted for alum during that period in order to
increase the pH of the groundwater from 7.0 to 8.5. The oxidation
48
-------
13
rate of manganese is known to be affected by the pH of the
groundwater with increasing values increasing the kinetics of the
oxidation reaction.
b) The retention time of the groundwater in the plant was greater
during the first operating year than any other, i.e. as much as
140% more, since the influent flow rate to the plant, when
operating, was well below the design flow rate of 76 us. Hence,
more time was available for the oxidation reaction to go to
completion.
5. Chemical oxidant addition permitted the plant to operate during the
second and third time periods more effectively and economically than
during the first (see Table 6). The impact of chemical oxidant addition,
however, on improving the quality of the effluent to the plant is
debatable. 1,2 DCE is still poorly removed by the train. H 2 0 2 , Cl 2 or
C10 2 addition does not seem to have any noticeable impact on its
removal/oxidation. The effluent concentration of 1,1,2 TCEe although
meeting the objective set by the MENVIQ during the first time period,
does not do so for the third time period by a margin of 5 .g/L. This
could be attributed to a two-fold increase in its influent concentration
during the latter period. Hence, H 2 0 2 , Cl 2 , C10 2 addition does not seem to
have any noticeable impact on its removalloxidation. 1,1,2 TCE does not
meet the objective during the third time period. However, no analysis of
that compound was carried out during the first (or second) time period.
The impact of chemical addition is harder to assess for phenolic
compounds due to the wide disparity (an order of magnitude) between
influent concentrations for the first and third time period. Average
removals of 329 pgL (84%), 86 p.g/L (96%), and 35 Lg/L (91%) were
observed during the three time periods, respectively. Can the increase in
removal efficiency be attributed to chemical oxiclsrnt addition or to a
lower content of phenolic compounds in the raw groundwater?
6 PERFORMANCE EVALUATION STUDY
In July 1988, a consortium consisting of The SNC Group, the University of
Sherbrooke and Laval University were awarded a contract to undertake the following:
49
-------
14
a) The mandate of the first program is to understand the physical chemical
treatment processes required to optimize the treatment of the
groundwater. This study is sub-divided into three (3) parts:
TABLE 6. OPERATING COSTS (x 1000’s CDN $)
1984/85
1985/86
1986/87
1987/88
1988/89
Fixed Costs
270.0
307.0
283.7
237.0
237.0
Variable Costs
- Maintenance
11.0
41.1
60.3
222.7
16.5
57.9
108.0
271.0
22.4
59.8
145.0
210.3
22.0
76.0
168.0
290.0
23.2
79.8
176.4
304.4
• Electricity
- Chemicals
- Activated Carbon
Supplementary Costs
172.7
65.6
152.6
109.7
115.0
Total Costs
777.8
826.0
873.8
902.7
935.8
Volume Treated (million
gallons)
181.8
284.0
369.1
420.0
472.5
Operating Time (%)
34.5
54.1
70.3
80.0
90.0
Costs ($ Cdn/million
gallons)
Costs ($ 1 1 5/rn 3 )
4.30
0.70
2.90
0.48
2.37
0.39
2.15
0.35
1.98
0.32
1. analysis of historical performance data collected by the Plant’s
operator, The SNC Group.
2. conduct a series of samplings (Full-Scale Plant Monitoring
Campaigns) around each process unit in order to fully evaluate the
treatment train’s performance.
3. direct research efforts at bench-scale and subsequently at pilot scale
in order to collect design and operational parameters for a
50
-------
15
treatment train which would meet the objectives set by the
MEN VIQ.
This treatment process evaluation study is being carried out by The
SNC Group/University of Sherbrooke project team.
b) The mandate of the second program is to study the hyrogeological impact
that the pump-and-treat decontamination scheme has had on the aquifer.
This hydrogeological impact evaluation is being carried out by The SNC
Group/Laval University project team.
A report on the results of the recently completed (September 1989) bench
and pilot scale treatment studies and the hydrogeological impact evaluation study will
be submitted to the NATO/CCMS Pilot Study Committee upon completion in 1990.
6.1 Full.Scale Plant Monitoring Campaigns
Two (2) monitoring campaigns were carried out on the plant. The first
campaign was conducted four (4) months after the last GAC replacement. Samples of
raw and treated groundwater along the treatment train were drawn daily for a period
of two (2) weeks. The second campaign began immediately after a GAC replacement.
Samples of raw and treated groundwater along the treatment train were drawn
regularly for a period of six (6) weeks in order to gather sufficient data to evaluate the
contsminant’s behaviour in the GAC columns.
The following parameters were evaluated:
Inorganic Parameters
1. Iron (Total/Dissolved)
2. Alknlinity
3. Hardness
4. Dissolved Oxygen
5. Turbidity
6. Suspended Solids
Organic Parameters
Chlorinated Hydrocarbons :
1. 1,2 Dicb.loroethane
2. 1,1,1 Trichloroethane
3. 1,1,2 Trichloroethane
51
-------
16
4. Trichioroethene
5. Tetrachloroethene
Aromatics
6. Benzene
7. Toluene
8. m-xylene
9. o+p-xylene
Sampling was carried out prior and after each of the following process
elements:
1. H 0 addition
2. Cl 2 addition
3. Stripper
4. Clarifier
5. Sand filters
6. Storage reservoir
7. Sacrificial GAC column
8. Polishing GAC columns
The data are presented below.
6.2 First Full-Scale Monitoring Campaign: Results
Iron
hon is completely oxidized following H 2 0 2 addition and is mainly removed in
the clarifier (92%), the remainder being filtered out by the sand filters (99%). The
average influent and effluent concentrations were found to be 6.7 mg/L and <0.1 mg/L
respectively.
1.2 DCE
1,2 DCE is mainly removed in the stripper (13%). An additional 5% is
eliminated in the rest of the treatment train. The average influent and effluent
concentrations were found to be 752 p ..gJL and 620 p.g/L respectively. This does not
meet the 50 pg/L objective.
1.1.1 TCE
Levels of 1,1,1 TCE in the influent were found to be extremely low, i.e.,
1 pg/L. Few conclusions can be drawn from the effectiveness of the train in removing
this compound. The average effluent concentration was found to be nil.
52
-------
17
1.12 WE
1,1,2 WE is poorly removed by the treatment train (19%). The average
influent and effluent concentrations were found to be 547 z&JL and 441 g/L
respectively. This does not meet the 50 tg/L objective.
TCEe
TCEe is partially oxidized (17%) following 11202 addition. An additional 30%
is removed by the stripper, the remainder being mostly eliminated by the polishing
GAC columns (92%). The average influent and effluent concentrations were found to
be 135 p.g/L 11 p.glL respectively. This does not meet the 4.5 p.g/L objective.
flCEe
TTCEe is partially oxidized (17%) following 11203 addition. It is moderately
removed by the stripper (additional 52%), the remainder being eliminated by the SAC
and polishing columns (99%). The average influent and effluent concentrations were
found to be 95 p.gfL and 1 jxg/L respcctively.
Aromatics
The impact of the treatment train’s unit processes on benzene, toluene,
m+o+p-xylene(s) was found to be surprisingly similar. Chemical oxidation was found
to remove 16 to 33% of the aromatic compounds studied. The efficiency of the stripper
for removing these compounds was observed to be 20-25%. GAC adsorption was found
to be particularly effective (100%) in removing all but benzene, which showed signs of
desorption from the GAC. The average influent concentrations for benzene, toluene,
m-xylene and o+p-xylenes were 109, 77, 26 and 12 p g ,L respectively. The average
effluent concentration for benzene was found to be 78 p.g/L which does not meet the 50
p.g/L objective.
Others
The following inorganic parameters were found to remain relatively constant
throughout the treatment train: nlknlinity, hardness, p11 while suspended solids and
turbidity increased following ferrous oxidation by 11203 addition and decreased following
ferric hydroxide removal in the clarifier and sand filters. Table 7 summarizes the
information presented above.
Discussion
1) 1,2 OCE; 1,1,2 TCE, TCEe and benzene did not meet their respective
objectives set for the effluent to the plant. Aside from benzene, the
53
-------
18
same conclusions were drawn, regarding the three chlorinated hydro-
carbons mentioned above, during our analysis of the historical data.
2) Looking at the removal efficiency of each process unit with respect to the
chlorinated hydrocarbons and aromatic compounds studied, the following
is noted:
TABLE 7. SUMMARY TABLE - FiRST FULL-SCALE CAMPAIGN
Parameters
Cumulative
% Removal
Iron
Benzene
Toluene
m-xyl.
o,p-xyl.
Raw
-
-
-
-
-
Post-H 2 0 2
0
17
26
23
5
Post-Cl 2
0
18
29
33
16
Post-Stripper
6
46
51
54
4-4
Post-Clarifier
92
39
53
54
n.a.
Post-Filter
99
51
53
58
41
Post-Storage
99
50
60
54
42
Post-SAC
100
64
83
78
99
Effluent
100
30
100
100
100
Parameters
Cumulative
1,1,1
1,1,2
% Removal
1,2 DCE
TCE
TCE
TCEe
TTCEe
Raw
-
-
-
-
-
Post-H 2 0 2
5
15
4
22
17
Post-Cl 2
0
15
-14
20
5
Post-Stripper
13
15
7
52
57
Post-Clarifier
15
31
8
55
57
Post-Filter
15
45
9
55
62
Post-Storage
15
46
9
57
63
Post-SAC
16
92
11
62
76
Effluent
18
100
19
92
99
HA Addition
The overall oxidation rates of the chlorinated HC and aromatics studied
were 13% and 18% respectively. One should note that the oxidation rate for all
54
-------
19
unsaturated molecules (presence of double-bonds) was higher than saturated ones, i.e.
TCEe (22%), TTCEe (17%).
Cl 2 Addition
The overall oxidation rates, by Cl 2 addition, of the HC’s studied was nil.
Instead, Cl 2 addition seemed to have an overall mild adverse effect, as is indicated by
an overall higher value (+10%) in the concentration of the chlorinated hydrocarbons
studied. The sole impact, therefore, of Cl 2 addition is to control biological growth in
the stripper.
Stripi ,er
The overall stripping rate of the HC’s studied was shown to be 29%. Of
particular interest, the overall stripping rate of the 1,2 DCE and 1,1,2 TCE, which
account for 75% w/w of the HC’s studied averaged 16%.
Clarifier/Sand Filters/Storage
Very little (<5%) removal of the HC’s studied was observed either in the
sludges or by volatilization/evaporation in the storage reservoir.
Activated Carbon Columns
The GAC is evidently saturated after four (4) months of operation. The
chlorinated HC’s, with the exception of the two (2) unsaturated (double-bonded)
compounds, TCEe and TTCEe, did not adsorb any longer onto the media. Aroxnatics
still adsorbed well, however, with the exception of benzene which showed signs of
desorption.
6.3 Second Full-Scale Monitoring Campaign: Results
The first campaign’s main focus was to look at the performance of each unit
process of the train, since this information was lacking from the data collected during
the past five operating years of the plant. The behaviour of a series of inorganic and
organic parameters was assessed, and was found to correlate very well with the
“overall picture” presented during the historical performance analysis of the plant.
However, no specific data had been gathered on the GAC’s performance with respect to
the contzinLinRnts of concern found in Merciefs groundwater; this fact was surprising
since the GAC adsorption step constitutes the heart of the treatment process.
Hence, a second full-scale monitoring campaign was scheduled to commence
immediately after a GAC change in order to gather data on the capacity and behaviour
55
-------
20
of the GAC. Since the data analysis is at an early stage, the following observations
are presented:
1) During the 40 days of sampling, the following compounds met the
objectives set for the plant:
1,1,1 TCE: It should, however, be noted that the influent concentration to the
plant averaged only 6 .tgL during the campaign.
TCEe, TTCEe, Toluene, m÷o+p xylene, Benzene: It should be noted that in
the case of benzene, after 9 days, the influent concentration to the sacrificial carbon
column equalled the effluent, and a concentration above 50 ptg/L was observed in the
effluent of the plant. This was due to a surge in the content of benzene in the raw
groundwater (100% increase). Once the surge passed, benzene’s effluent concentration
kept on meeting the objectives set for the plant.
2) The following compounds did not meet the objectives set for the plant:
1,2 DCE: Three (3) days after GAC replacement, the effluent concentration of
1,2 DCE was observed to be >50 p.g/L. Nine (9) days after GAC replacement, the
effluent concentration of the SAC was found to be above (146% increase) that of the
iniluent value, and less than two (2) weeks after replacement, the influent
concentration to the polishing columns was observed to be equal to the effluent
concentration.
1,1,2 TCE: Influent concentrations were found to equal effluent
concentrations in the sacrificial and polishing GAC columns, two (2) and four (4)
weeks, respectively after GAC replacement.
6.4 Conclusions
Based on the analyses presented, the following is concluded:
The unit processes of the treatment train do not perform as well as
originally anticipated. Hence, the physical-chemical treatment processes
msiking up the treatment train of the Mercier plant will require further
modifications in order for the treated effluent to meet the objectives set
by the MENVIQ.
• The status of remediation of the aquifer is not complete after five (5)
years of pump-and-treat as can be inferred by the presence and
concentrations observed of the contsminants in the groundwater.
56
-------
21
REFERENCES
1. Simard, G. and J.P. Lanctôt, ‘Decontanhination of Vile Mercier Aquifer for
Toxic Organics”. Presented at NATO/CCMS First International Conference,
Demonstration of Remedial Action Technologies for Contaminated Land and
Groundwater, November 11-13, 1987.
2. Pilon, J. Personal communication, 1989.
3. Martel, R., “Groundwater Contan,ination by Organic Compounds in Ville
Mercier: New Developments”. Presented at NATO/CCMS Second
International Conference, Demonstration of Remedial Action Technologies for
Contaminated Land and Groundwater ”, November 7-11, 1988.
57
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTER I ZAT ION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On—/Off—site Treatment Location:
Pre- and Post-treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Stephen James
Office of Research and Development
U.S. Environmental Protection
Agency
Cincinnati, Ohio 45268
United States
513-569—7877
Physical/Chemical
Photo/chemical oxidation
(ultraviolet, ozone, and hydrogen
peroxide)
Ui trox
Demonstration
Organic contaminants, Including
chlorinated hydrocarbons
Groundwater
On—site
Barrel and drum reconditioners,
Lorentz Barrel and Drum, San Jose,
California, United States
Variety of chlorinated organics
Norma Lewis
Office of Research and Development
U.S. Environmental Protection
Agency
Cincinnati, Ohio 45268
United States
513-569-7665
11/89
5-3
59
-------
A FIELD EVALUATION OF THE UV/OXIDATION TECHNOLOGY
TO TREAT CONTAMINATED GROUND WATER AT A HAZARDOUS WASTE SITE
Norma Lewis, M.A.
U.S. Environmental Protection Agency
Risk Reduction Engineering Laboratory
Cincinnati, Ohio
Kirankuniar Topudurti, Ph.D.
Robert Foster, P.E.
PRC Environmental Management, Inc.
Chicago, IL
ABSTRACT
This paper presents the field evaluation results of the
ultraviolet radiation (UV)/oxidation technology developed by Ultrox
International, Santa Ana, California. The field evaluation of the
technology was performed at the Lorentz Barrel and Drum (LB&D) site in
San Jose, California, under the Superfund Innovative Technology
Evaluation (SITE) program from February 27 through March 10, 1989.
The UV/oxidation technology uses UV radiation, ozone, and
hydrogen peroxide to oxidize organic contaminants present in water.
At the LB&D site, this technology was evaluated in treating ground
water contaminated with volatile organic compounds (VOCs). The Ultrox
system achieved VOC removals greater than 90 percent. The majority of
VOCS were removed through chemical oxidation. However, for a few
VOCS, such as 1,1,1-trichioroethane (l,1,l—TCA) and l,1-dichloroethane
(1,l—DCA) stripping also contributed toward removal. The treated
ground water met the applicable discharge standards (NPDES) for
disposal into a local waterway at 95 percent confidence level. There
were no harmful air emissions from the Ultrox system into the
atmosphere.
60
-------
INTRODUCTION
The Environmental Protection Agency (EPA) is finding better
solutions to hazardous waste remediatiori through its Superfund
Innovative Technology Evaluation (SITE) program. The SITE program was
created to demonstrate and evaluate technologies that may destroy or
permanently change the composition of hazardous waste in the
environment by significantly reducing the waste’s toxicity, mobility,
or volume. The SITE program also generates reliable performance and
cost data for these treatment technologies to be used in evaluating
alternatives under the Superfund site reinediation process.
In 1988, Ultrox International’s proposal for its ultraviolet
radiation (UV)/oxidatiori technology was selected by EPA’s Office of
Research and Development (ORD) and Office of Solid Waste and Emergency
Response (OSWER) under the SITE program. This technology was
demonstrated at the Lorentz Barrel and Drum (LB&D) site in San Jose,
California, through a cooperative effort between Ultrox International,
ORD, OSWER and EPA Region IX.
UV/OXIDATION TECHNOLOGY: EQUIPMENT AND PROCESS DESCRIPTION
The Ultrox tN/oxidation treatment system uses UV radiation,
ozone, and hydrogen peroxide to oxidize orgarlics in water. The major
components of the Ultrox system are the UV/oxidation reactor module,
air compressor/ozone generator module, hydrogen peroxide feed system,
and catalytic ozone decomposition (Decompozon) unit. An isometric
view of the Ultrox system is shown in Figure 1.
61
-------
The tJV/oxidation reactor used in the demonstration (Model PM-150)
has a volume of 150 gallons and is 3 feet long by 1.5 feet wide by 5.5
feet high. The reactor is divided by 5 vertical baffles into 6
chambers and contains 24 UV lamps (65 watts each) in quartz sheaths.
The UV lamps are installed vertically and are evenly distributed
throughout the reactor (four lamps per chamber). Each chamber also
has one stainless steel sparger that extends along the width of the
reactor. These spargers uniformly diffuse ozone gas from the base of
the reactor into the water. Hydrogen peroxide is introduced in the
influerit line to the reactor from a storage tank. An in—line static
mixer is used to disperse the hydrogen peroxide into the contaminated
water in the influent feed line.
The Decoinpozon unit (Model 3014 FF) uses a nickel—based
proprietary catalyst to decompose reactor off-gas ozone to oxygen.
The Decompozon unit can accommodate flows of up to 10 standard cubic
feet per minute and can destroy ozone concentrations in ranges of 1 to
20,000 ppm (by weight) to less than 0.1 ppm.
During the Ultrox system operation, contaminated water first
comes in contact with hydrogen peroxide as it flows through the
influent line to the reactor. The water then comes in contact with
the UV radiation and ozone as it flows through the reactor at a
specified rate to achieve the desired hydraulic retention time. As
the ozone gas in the reactor is transferred to the contaminated water,
hydroxyl radicals (OH° ) are produced. The hydroxyl radical formation
from ozone is catalyzed by till radiation and hydrogen peroxide. The
hydroxyl radicals, in general, are known to react with organics more
62
-------
rapidly than the oxidants ozone, hydrogen peroxide, and UV radiation.
They are also much less selective in oxidation reactions than the
three oxidants. Ozone that is not transferred to the contaminated
water will be present in the reactor off-gas. This ozone is
subsequently destroyed by the Decompozon unit before being vented to
the atmosphere. The treated water flows from the reactor for
appropriate disposal.
SITE HISTORY AND CONTAMINATION
The LB&D site is in San Jose, Santa Clara County, California.
ThiS site was used for drum recycling operations from about 1947 to
1987. The drums contained residual aqueous wastes, organic solvents,
acids, metal oxides, and oils. A preliminary site assessment report
for the LB&D site showed that the ground water and soil were
contaminated with organics and metals (I). In 1987, the LB&D facility
ceased operation due to a restraining order issued by California
Department of Health Services. EPA Region IX assumed the
responsibility for site reinediation.
The shallow ground water at the LB&D site was selected as the
waste stream for evaluating the UV/oxidation technology. Ground-water
samples collected in December 1988 indicated that several volatile
organic compounds (VOCs) were present in the shallow aquifer. VOC5
detected at high levels included trichioroethylene (280 to 920 .Lg/L),
vinyl chloride (51 to 146 j g/L), and 12—trans—dichloroethylene (42 to
68 ig/L). The pH and alkalinity of the ground water were about 7.2
and 600 mg/L as CaCO 3 , respectively. These measurements indicated
63
-------
that bicarbonate ion (HCO), which acts as an oxidant scavenger, was
present at high levels. Other oxidant scavengers such as bromide,
cyanide, arid sulfide were not detected.
TECHNOLOGY DEMONSTRATION
The objectives of the technology demonstration were to: (1)
evaluate the ability of the Ultrox system to treat VOCs present in the
ground water at the LB&D site at different operating conditions; (2)
determine the extent of VOC stripping, if any, from the bubbling of
ozone gas; (3) evaluate the efficiency of the Decolnpozon unit to
decompose reactor off-gas ozone; (4) deter nine the operating
conditions needed for the effluent to meet applicable discharge
standards (NPDES) for disposal into a nearby waterway; and (5) develop
the information required to estimate operating costs for the treatment
system, such as electricity consumption and oxidant doses.
Testing Atproach
Eleven test runs were performed to evaluate the Ultrox system
under various operating conditions. After these runs, two additional
runs were also performed to determine if the system’s performance was
reproducible. The operating conditions for the runs are summarized in
Table 1. All 13 runs were performed over a period of 2 weeks.
The study was designed to evaluate the Ultrox system by adjusting
the levels of five operating parameters: (1) influent p11, (2)
retention time, (3) ozone dose, (4) hydrogen peroxide dose, and (5) 0 11
radiation intensity. The initial operating conditions (Run 1), given
64
-------
in Table 1, were based on the treatability study conducted by Ultrox
on LB&D site ground water.
During the demonstration, a preliminary estimate of the Ultrox
system’s performance in bach run was obtained based on the effluent
concentrations of three indicator VOCs. The VOCs selected for this
purpose were trichioroethylene (TCE, a major volatile contaminant at
the site), 1,1 -dichioroethane (l,l-DCA), and 1,1,1—trichioroethane
(l,l,1-TCA). 1,l-DCA and l,l,l-TCA were selected because Ultrox’s
experience indicated that these VOCs are relatively difficult to
oxidize.
In the first three runs, the influent pH was adjusted by adding
sulfuric acid to evaluate the system’s performance and to determine
the “preferred” influent pH (“preferred” operating conditions are
those conditions in which (1) effluent concentrations of indicator
VOCs are below NPDES limits and (2) the relative operating costs are
the lowest). Once the “preferred” influent pH was determined, it
remained at that level for the remaining runs. The Ultrox system
performance was then studied by varying other parameters, one at a
time, as shown in Table 1, to determine the “preferred” values for
those parameters. The criteria were the same as those used in
determining the “preferred” value for the influent pH. After the
“preferred” values were determined for all five operating parameters,
two runs (12 and 13) were performed to verify the reproducibility of
the Ultrox system’s performance at the “preferred” operating
conditions. By duplicating the “preferred” operating conditions
determined during the previous eleven runs, the two verification runs
65
-------
served to ensure that the results could be based on repeated
observations, with comparable findings.
Sampling and Analytical Procedures
Air and water samples were collected fro n the Ultrox system at
the locations shown in Figure 2. For the critical parameters in this
study (VOCs in water), six replicate samples were collected.
Duplicate samples were collected for other parameters listed in Table
2. Sampling at the influerit port began about 15 minutes after each
run was started. At other locations in the reactor, sampling began
after three retention times to allow the system to reach steady state.
All the air and water samples for off—site laboratory analysis were
preserved as required before being shipped to the laboratory.
The analytical methods followed in this study are listed in Table
2. To obtain reliable data, strict quality assurance and quality
control (QA/QC) procedures were followed. Details on al]. aspects of
the QA/QC procedures are presented in the Demonstration Plan and the
Technology Evaluation Report (II, III).
RESULTS AND DISCUSSION
This section summarizes the results of the Ultrox system
demonstration and also presents an evaluation of the UV/oxidation
technology’s effectiveness in removing VOCs from the ground water at
the LB&D site.
SummarY of Results for VOCs
The purpose of the test runs was to evaluate the effectiveness of
the Ultrox system in removing 44 VOCs present in the ground water at
66
-------
the LB&D site. The removal efficiencies and concentration profiles of
all VOCs are not presented in this paper. Instead, a summary approach
is used to present the results.
The mean concentration profiles and the discharge standards
(NPDES) for the three indicator contaminants (TCE; l,l—DCA; l,l,l—TCA)
in each run for each sampling location are plotted in Figures 3, 4,
and 5. The VOC concentrations progressively decreased from the
influent to the mid-point and from the mid—point to the effluent
except for Run 3 (in Run 3, the concentration of l,l-DCA at mid-point
was higher than that in the influent. it is believed that either the
mid-point concentration or the influent concentration is just an
outlier). This progressive decrease is due to the ozone and the UV
radiation provided in the last three chambers (after the mid-point)
and the increase in the retention time from the mid—point to the
effluent port. Additionally, the effluent and mid-point VOC
concentrations are comparatively high in Run 7, which appears to be
due to the decreased ozone dose during that particular run.
The average effluent concentrations (determined during the
demonstration by analyzing only two of the six replicates) for each
indicator VOC with the discharge standard (NPDES) showed that the
effluent met the discharge limits in Runs 8 and 9. Since a lower
hydrogen peroxide dose was used in Run 9, compared to Run 8, RUn 9 was
chosen as the “preferred” operating run. However, based on a complete
analysis of the six replicates performed after the demonstratian, the
mean concentration of l,l-DCA was found to be slightly higher than 5
Mg/L, the discharge standard for the VOC. Since Run 9 had the
67
-------
“preferred” operating conditionS during the demonstration, the
verification runs (12 and 13) were performed at those conditions.
A comparison of 95 percent upper confidence limit (U.C.L.) values
for the effluent VOCs in Runs 9, 12, and 13 with the discharge
standards (NPDES) is presented in Table 3. The U.C.L. values were
calculated using the one-tailed student’s t—test. Table 3 shows that
the effluent met the discharge standards for all regulated VOCs at 95
percent confidence level in Runs 12 and 13. In Run 9, the mean
concentrations for 1,1-DCA and 1,2-DCA exceeded ttie discharge
standards. Although 1,1-DCA and 1,2-DCA were present at levels
slightly greater than the discharge standards, the difference in
performance among the three runs is negligible.
The mean concentration profiles for total VOCs are given in
Figure 6. A comparison of the VOC concentrations presented in
Figure 6 with those in Figures 3, 4 and 5 indicates that the
concentration profiles for total VOCs are similar to those for the
indicator VOCs. For example, the peaks present at the mid—point and
effluent for indicator VOCs are also present in the total VOC
concentration profiles.
The percent removals for the indicator VOCs and total VOCs are
presented in Figure 7. The figure shows that the removal efficiencies
for TCE were higher than those for 1,1-DCA and 1,l,l—TCA which is
consistent with the rationale used in selecting the indicator VOCs.
The percent removals for total VOCs and the indicator VOCs decreased
considerably in Run 7, which appears to be due to the decreased ozone
dose.
68
-------
Since ozone gas is bubbled through the ground water treated by
the J1trox system, the VOC removal could be attributed to stripping in
addition to oxidation. To determine the extent of stripping within
the treatment system, VOC samples were collected from the reactor of f-
gas. A total of 25 samples were collected during the demonstration.
Although 1, l—DCE, 1, 2-DCE, benzene, 1,1,2, 2—tetrachioroethane, and
acetone were present in two samples at concentrations close to the
detection limits, TCE, vinyl ch1oride l,l,1—TCA, and 1,1—DCA were
detected more frequently. To determine the extent of stripping, the
emission rates in the reactor off—gas for these latter four VOCs were
compared to the VOC removal rates (estimated by difference between the
VOC input rates at the influent and output rates at the effluent ports
of the Ultrox system). The results are summarized in Table 4. Since
the extent of stripping for any particular VOC is expected to be
proportional to the ratio of air flow rate to the water flow rate,
this ratio is also presented in the table. The ratio for Runs 1 to 5
is approximately 2; for Run 6 and Runs 8 to 13, it is about 4.5; and
for Run 7, it is 1. If stripping contributed to the total removal of
the four VOCs, the extent of stripping should be the least in Run 7,
and the most in Runs 6 and 8 to 13. The data presented in the table
follow this trend for three of the four VOCs (except for the vinyl
chloride in Runs 6, 7 and 9). However, a quantitative correlation of
the extent of stripping cannot be made because the operating
conditions were different in each run. For example, at a given air to
water flow ratio, when oxidant doses are varied, the extent of
69
-------
oxidation also varies. Therefore, the extent of stripping will be
indirectly affected.
Table 4 also presents the Henry’s law constants for the four
VOC5. By comparing these constants for the VOCs, their volatility is
expected to increase from left to right:
l,1-DCA — TCE — 1,1,1-TCA — vinyl chloride
However, a significant removal fraction for 1,l,l—TCA and l,1—DCA were
observed to be due to stripping. Conversely, the extent of stripping
was low f or vinyl chloride and TCE. This is because it is easier to
oxidize vinyl chloride and TCE than l,l-DCA and l,l,l,-TCA because of
the double bonds between the carbon atoms in TCE and vinyl chloride.
In other words, in the tJV/oxidation process, stripping is a
significant removal pathway for compounds that are difficult to
oxidize.
Performance of the Decompozon Unit
The ozone concentrations in the influent to and the effluent from
the Decontpozon unit were analyzed in each run. These concentrations
are presented on a semilog plot in Figure 8. The effluent ozone
concentrations were low (less than 0.1 ppm) for Runs 1 to 8,
approximately 1 ppm in Runs 9 and 10, and greater than 10 ppm in Runs
11, 12, and 13. The high ozone levels (greater than 1 ppm) in the
effluent are attributed to the malfunctioning heater in the Decoinpozon
unit. The temperature in the Decompozon unit should have been 140°F
for the unit to properly function 1 whereas the temperature for Runs 1].
to 13 was only about 80°F. The ozone destruction efficiencies greater
than 99.99 percent were achieved in Runs 1 to 10.
70
-------
Although the primary function of the Decompozon unit is to remove
ozone, the data presented in Table 5 indicates that significant VOC
removal occurred when the unit functioned as designed (Runs 1 to 8).
Summary of Results for Noncritical Parameters
In addition to the critical parameters (VOCs), many non-critical
parameters were also measured. The noncritical parameters for
organics included semi-volatiles, PCBs/pesticides, and total organic
carbon (TOC); the noncritical parameters for inorganics included pH,
conductivity, and alkalinity. Temperature, turbidity, residual
oxidants, and electricity consumption were also measured.
No semivolatiles or PCBs/pesticides were detected in the influent
or effluent. TOC removal was achieved only at trace levels indicating
that complete oxidation of ‘organics to carbon dioxide and water did
not occur. However, since no new VOCs were found by gas
chromatography/mass spectrometry (GC/MS) analysis and CC analysis of
the effluent, the oxidation products were not VOCs.
Metals such as iron and manganese were present at low
concentrations in the influent, and no significant metal removal
occurred. No changes in alkalinity and conductivity were observed
after the treatment. However, the pH increased by 0.5 to 0.8 units
after the treatment. The increase in pH is probably due to the
reaction between hydroxyl radicals and bicarbonate ion (the
predominant form of alkalinity at the ground water pH, which is 7.2)
in which hydroxyl ions are produced (XIII).
Turbidity increased by 1 to 4 units after the treatment, which
may be due to the insignificant amount of metal removal by oxidation
71
-------
and precipitation. The temperature increased by approximately 2 to
3° C after the treatment and was due mainly to the heat from UV lamps.
Ozone gas transfer to the ground water was over 95 percent, with 5
percent remaining in the reactor off—gas. After the reaction, the
residual ozone and hydrogen peroxide concentrations in the effluent
were usually less than 0.1 ppm. The average electrical energy
consumption to operate the Ultrox system was about 11 kwh/h of
operation.
CONCLUS IONS
The ground water treated by the Ultrox system met the discharge
standards for disposal into a nearby waterway at the 90 percent
confidence level at a hydraulic retention time of 40 minutes, an
influerit pH of 7.2 (unadjusted), an ozone dose of 110 mg/L, a hydrogen
peroxide dose of 13 mg/L, and with all 24 UV lamps operating.
There were no VOCs detected in the air emissions from the
treatment unit into the atmosphere.
The ozone destruction unit (Decoxnpozon unit) destroyed reactor
off-gas ozone to levels less than 0.1 ppm (OSHA Standards) with
destruction efficiencies greater than 99.99 percent.
The Ultrox system achieved removal efficiencies as high as 90
percent for total VOCs present in the ground water at the LB&D site.
The removal efficiencies for TCE were greater than 99 percent.
However, the maximum removal efficiencies for l,1—DCA and 1,].,1—TCA
were about 65 percent and 85 percent, respectively.
72
-------
The removals of 1,1-DCA and l,1,1-TCA are due to both chemical
oxidation and stripping. Specifically, 12 to 75 percent of the total
removals for l,l,l—TCA, and 5 to 44 percent of the total removals for
l,l-DCA were due to stripping. However, stripping for TCE and vinyl
chloride was observed to be less than 10 percent. For other VOCs,
such as 1,1—dichioroethene, benzene, acetone, and 1,1,2,2—
tetrachioroethane, stripping was found to be negligible. VOCs present
in the gas phase within the reactor at levels of approximately 0.1 to
0.5 ppm were removed to below detection levels in the Decompozon unit.
Based on the GC and CC/MS analyses performed for VOCs,
semivolatile organics, and PC s/pesticides, no new compounds were
discovered in the treated water. The organics analyzed by GC methods
represent less than 2 percent of the TOC present in the water. Very
low TOC removal occurred, which implies that partial oxidation of
organics took place in the system but not complete conversion to
carbon dioxide and water.
The Ultrox system’s average electrical energy consumption was
about 1]. kwh/h of operation.
AC KNOWLEDGEMENT
The authors sincerely thank Dr. Gary Weishans, PRC Environmental
Management, Inc. for managing the field demonstration, and reviewing
this paper.
73
-------
REFERENCES
I CH2M Hill, Preliminary Site Assessment Report for the
Loreritz Barrel and Drum Site , 1986.
II PRC Environmental Management, Inc., and Engineering-science,
Inc., Demonstration Plan for the Ultrox International
UV/Oxidation Process , prepared for U.S. EPA, February 1989.
III PRC Environmental Management, Inc., and Engineering—Science,
Inc., Technoloay Evaluation Reoort SITE Program
Demonstration of the Ultrox International UV/Oxidation
Process , in preparation for U.S. EPA.
IV Methods for the chemical Analysis of Water and Wastes , EPA—
600/4—79—020, Environmental Monitoring and Support
Laboratory, Cincinnati, OH, U.S. EPA, 1983.
V Test Methods for Evaluatina Solid Waste , Volumes lA-iC:
Laboratory Manual, Physical/Chemical Methods; and Volume II:
Field Manual, Physical/Chemical Methods, SW—846, Third
Edition, Office of Solid Waste, U.S. EPA, Document Control
No. 995—001—00000—1, 1986.
VI Boltz, D.F., and J.A. Howell, Hydrogen Peroxide,
Colorimetric Determination of Noninetals , John Wiley & Sons,
1979, 301—303.
VII Standard Methods for the Examination of Water and
Wastewater , Sixteenth Edition, APHA, AWWA, and WPCF, 1985.
VIII Bader, A., and 3. Hoigne, Determination of Ozone in Water by
Indigo Method, Ozone Science and Engineering , 4:169, 1982.
IX The National Primary and Secondary Ambient Air Quality
Standards, 40 CFR Part 50 , Appendix D -— Measurement of
Ozone in the Atmosphere.
X NIOSH, Manual of Analytical Methods , Third Edition, U.S.
Department of Health and Human Resources, DHHS (NIOSH)
Publication No. 84—100, 1984.
XI Operating instructions provided with the instruments.
XII Suoerfund Public Health Evaluation Manual , EPA 540/1—86/060,
Office of Emergency and Remedial Response, Washington, DC,
1986.
XIII Hoigne, J., and H. Bader, Ozonatiorl of Water: Role of
Hydroxyl Radicals as Oxidizing Intermediates, Science ,
190:782—784, 1975.
74
-------
OPERATING PARA2 ETERS XATRXZ FOR TBE ULTROX SYSTEM DEMONSTRATION
Run No. Retention
Ozone
H O UV Lamps Influent PH
Time
Dose
Dose
1 r 2’ All ON Unadjusted
2 X Y 2 All ON (Unadjusted - 1)
3 X I 2 All ON (Unadjusted - 2)
4 1. 5X I 2 All ON Preferred’
5 O. SX Y 2 All ON Preferred
6 Preferred 1.51 2 All ON Preferred
7 Preferred O.5Y Z All ON Preferred
8 Preferred Preferred l.5Z All ON Preferred
9 Preferred Preferred 0.5Z ALL ON Preferred
10 Preferred Preferred Preferred Only ON in Preferred
the first
three chambers
11 Preferred Preferred Preferred Only ON in Preferred
the last
three chambers
12 Preferred Preferred Preferred Preferred Preferred
Preferred Preferred Preferred Preferred Preferred
Notes:
• X — 40 minutes.
• Y—75sg/L.
2—25mg/ I ..
(X, I, and 2 values were determined by Ultrox International to be the
optimum conditions for treating ground water in the treataba.lity study
at the L8&D site.)
“Preferred” operating conditions are those conditions in which (1) the
concentrations of effluent indicator VOCS are below their respective
NPDES limits and (2) the reLative operating Coats are the lowest.
• Verification runs performed to check the reproducibility of the Ultrox
system’s performance at the “preferred” operating conditions.
75
-------
TABLE 2
ANALTTICAL XZTBODS
Method
Method
Analyte Matrix Type
Reference
Alkalinity Liquid Field MCMJW 310 • 1 N
Arsenic Liquid Lab SW—846 7060’
Liquid Lab SW—846 8270’
(Semivolatiles)
Chremiun Liquid Lab SW—846 7195’
(Cr )
Chloride Liquid Lab SM 429k ’
Chromium I.iquid Lab SW—846 7191’
Conductivity Liquid Field Manual tm
Hydrogen Liquid Field Beltz et al.
Peroxide (1979) ’
Metals Liquid Lab SW—846 6010’
(Barium. Cobalt,
Iron, Manganese.
Nickel, Zinc,
Potassium. Calcium.
Magnesium, and
Sodium)
Ozone Liquid Field Bader and
Hoigne(l9B2)
Ozone Air Field 40 CTR Part 5o
pH Liquid Field Manual’
Pestic ides/PCBs Liquid Lab SW—846 8080’
Silica Liquid Lab SW-846 6010’
Sulfate Liquid Lab SM 429k’
TABLE 2 (Continued)
MI7ILTT tOIL METEODS
Method
Method
Ax a1yte Matrix Type
Reference
Temperature Liquid Field Manual tm
Total Organic Liquid Lab SM 505Am
Carbon
Turbidity Liquid Field Manual tm
Volatile Organics Liquid Lab SW—846 8010
and 8020’
Volatile Organics Liquid Lab SV—846 8240’
Volatile Organics:
Vinyl Chloride Air Lab HXOSR 1007’
1,1—Dichioroethene Air Lab NIOSM 1015’
l.l-Dichloroethane Air Lab NIOSH 10031
l.2-D ich.loroethene Air Lab NIOSH 1003’
1.1.1—
Trichloroethane Lab NIOSH 1003’
‘rrichloroethene Air Lab NIOSH 1022’
Benzene Air Lab NIOSH 1500’
1,1,2.2—
Tetrachiereethane Air Lab NIOSH 1019’ 76
Acetone Lab NIOSH 1300’
-------
TABLE 3
COMPARISON OF EFFLUENT lOC CONCENTRATIONS
IN RUNS 9, 12, AND 13
Mean , g/L 95% UCL,i q/L RT, g/L Conclusion
Run nun er: 9
l.l.1—I ’CA 0.75 1.0 5 OK
1,1,2,2—PC?. 0.045 0.045 5 OK
1,1—DC?. 5.3 5.5 5 N
1,1—DC! 0.000 0.000 5 OK
1.2—DC?. 1.3 1.4 1 N
l,2—DCPA 3.3 3.4 5 OK
Senzene 0.023 0.026 5 OK
Chieroethane 0.000 0.000 5 OX
Ch lorofo rn 1.1 1.2 5 OK
PC! 0.24 0.63 5 OK
1-1,2—DC! 0.000 0.000 5 OX
ICE 1.2 1.3 5 OX
Vinyl Chloride 0.11 0.11 2 OK
Run nun .r: 12
1 1.1— IC?. 0.43 0.48 5 OK
1,1,2,2—PC?. 0.045 0.045 5 OK
1 .1—DC?. 3.8 4.2 5 OR
1,1—DC! 0.000 0.000 5 OX
1,2—DC?. 0.92 1.0 1 OK
1,2-DCP ? . 2.6 2.9 5 OK
Senzene 0.023 0.026 5 OK
Chioroethane 0.000 0.000 5 OX
Ch loroforn 0.74 0.82 5 OK
PC! 0.19 0.38 5 OK
1—1,2—DC! 0.000 0.000 5 OK
ICE 0.55 0.65 5 oX
Vinyl Chloride 0.11 0.11 2 OK
Run number: 13
1,1.1—IC?. 0.48 0.54 5 OK
1.1,2.2—PC ? . 0.045 0.045 5 OX
1.1-DC?. 4.2 4.5 5 ox
1.1-DC! 0.000 0.000 5 OK
1,2—DC?. 1.0 1.0 1 OX
1,2-DCPA 2.9 3.1 5 OK
Benzene 0.45 0.52 5 OK
Ch loroethane 0.000 0.000 5 ox
Chloroform 0.81 0.87 5 OK
PC! 0.091 0.17 5 OK
1-1,2-DC! 0.000 0.000 5 OR
ICE 0.63 0.73 5 OK
Vinyl Chloride 0.12 0.12 2 OK
TABLE 3 (Continusd)
COMPARISON OF EFFLUENT VOC CONCENTRATIONS
IN RUNS 9, 12, AND 13
Notes:
95% tJCL: tIpper 95% Confidence Limit
RI: Regulatery Threshold
OK: Effluent met the regulatory threshold
N: Effluent did not neet the regulatory threshold
Abbreviations:
1,1.1—IC?.: 1,1,1—Trichloroethane; 1,1,2.2—PC?.: 1,1,2,2—
Tetrachioroethane: 1,1-DC?.: 1,1, —Dichloroethane: 1,1-DC!: 1,1—
Dichleroethylerte: 1,2—DC?.: l,2-Oichloreethane: 1,2—DCPA:
1,2-Dichleroprepane: PC!: Tetrachlorcethylene: 1-1, 2-DCE:
Trans-i, 2-Dichieroethylehe: ICE: Trichloroethylene.
77
-------
TABLE 4
EXTENT OF VOC STRIPPING IN TEE ULTROX SYSTEN
Air flow
rate
çrcent
Strncina ontr bution For
Run Water flow rate
No.
1, 1-DQ
0.0043
0
TCE 1, 1 • l—TC?
.0091 0.O]4
VC
0.082’
1 2.1
7.4
2.0 43
o.oi
2 2.3
9.1
3.4 34
0.95
3 2.1
9.9
2.7 31
0.013
4 2.0
7.4
3.0 29
0.01
5 2.1
17
3.5 29
1.7
6 4.5
16
1.2 65
0.072
7 1.0
4.9
1.2 12
3.1
8 4.5
23
7.5 85
3.2
9 4.5
16
6.6 58
0.04
10 4.3
27
9.4 73
1.1
11 4.6
44
24 >99
13
12 4.4
34
7.0 76
8.9
13 4.3
37
26 75
1.8
Notes:
a VC: Vinyl
Chloride
b Henry’s law
constant
of the
VOC,
atm—m’/ntol.
TABLE 5
VOC REMOVAL IN THE DECOMPOZON UNIT
Run
No.
TCE.
Influent
ppm
Effluent
1 ,l-DCA,
Influent
ppm
Effluent
1.1,1-TCA, ppm
Influent Effluent
Vinyl chloride, ppm
Irifluent Effluent
1
0.15
<0.15
0.1
<0.1
0.15
<0.1
0.002
<0.002
2
0.15
<0.15
0.1
<0.1
0.1
<0.1
0.070
<0.002
3
0.15
<0.15
0.1
<0.1
0.1
<0.1
0.002
<0.002
4
0.15
<0.15
0.1
<0.1
0.1
<0.1
0.002
<0.002
5
0.15
<0.15
0.2
<0.1
0.1
<0.1
0.150
<0002
6
0.15
<0.15
0.1
<0.1
0.1
<0.1
0.003
<0.002
7
0.15
<0.15
0.15
<0.1
0.1
<0.1
0.517
<0.002
8
0.15
<0.15
0.1
<0.1
0.1
<0.1
0.041
<0.002
9
0.15
<0.15
0.1
<0.1
0.1
<0.1
0.002
0.061
10
0.225
<0.15
0.15
<0.1
0 .1
<0.1
0.064
0.078
11
0.55
0.325
0.25
0.2
0.2
<0.1
0.570
0.189
12
0.15
<0.15
0.2
<0.1
0.1
<0.1
0.271
0.006
13
0.45
<0.15
0.22
<0.1
0.1
<0.1
0.420
0.004
78
-------
PIGURE I OMETRIC VIEW 0 ULINOX 5YSTIN
Issa . On Gai
CAY LVTIC OZONE CECONPOSER —->
CII Gas
Mab l s.p
Wilif
L l a.rI
—
LLTROII
W/OXJDAI N REACtOR
GEJ’ ERAIOR
“—.- Camp.au.d *fl
- hyafaguR PWO.W.
WU F..ø t. Ik
GaWo IU.
V GI E I ULTRQX 81$TE*I MNPUIIG &DC flCPIS
—
Fssd lsrA
79
— WWW
-------
nouns we CONCeNTRATIONS IN DIPFERLNT RUNS
1 2 3 4 S 6 7 8 9 10 I I 12 13
RUN NUMOER
stn.am l a o Pam Ii i ] EIIa.am
90
80
70
60
50
40
30
20
10
0
Is
I I
I ]
12
II
l0
9
8
7
6
5
4
3
2
I
0
Atclwgs $uaøaid
mousse i.
&
I 2 3 4 5 6 7 8 9 I I I II 12 13
RUN NUMBER
b ase AbdPam [ J Easae — asaw 5 s SISCNO
80
-------
•cunu •.IJ4CA COI ICENIRATIONS IN DQFERENT Ra fts
g
200--
I S O-
180-
170 -
160-
150-
1 10 -
130-
120-
l I D -
100-
90-
80 •
70
60
50•
40
30-
20 •
10 -
- —r
I I I I I
2 3 4 5 6 7 8 9 ID I i 12 13
RUN NUMBER
ttnssni Lii tea Pawn [ j J Elitism actwge Sisidaid
6-
5-
S
1 hL
O rit
4,
S
‘4
‘S
‘ S
‘S
‘ S
5,
5 ’
‘ 4 !
‘ 4!
S i
i i
FIGURE S TOTAL CC CONCENTRATIONS IN DIFFERENT RUNS
t
5,
5’
4’
‘4
I i i ;
11
I I I I I I I I I I I
I 2 3 4 5 6 7 8 9 Ill II 12 13
RUN NUMBER
i i
-------
icime v a RSMOVAL$ IN WFLRtHT RUNS
I
8
I I I I I
I 2 3 4 5 6 7 8 9 10 II
RUN NLJMUER
ICE 1.1 OCA I 1.1 ICA isialvOCo
FlOURS I OZONE CONCENTRATIONS IN D IFFERENt RUNS
HUll NUMBE H
mrnan 82
I to -
100
90
80 -
70-
60-
50-
40 -
30-
20-
10 -
. 1
12
13
0
S
4
3
2
I
0
—1
-2
-3
-4
I
0
0
z
0
I
.
.q
%
,
.,
t,
.
s,
,
%
-
.
%
“
,
-
1
%
I
%
.
—
%
I ’
.,
I,
%
%
.,
q
I
I ,
%
1
%
Iq
I i
S
S I
S I
%
c
,
I ’
•i
%
%
,
?
‘, \
SI
., \
-‘ \
., \
‘ ¼,
“
.
%
“
%
\
\
Mt
¼, SI
\S1
-
.,
%
.,
I I I I U I I I I I I I I I I
1 2 3 4 5 6 7 8 9 Ui II
12 13
r
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTERIZATION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On-/Off-site Treatment Location:
Pre- and Post—treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Dr. Robert Hinchee
Battelle Memorial Inst.
505 KIng Ave.
Columbus, OH
United States
Biological
Enhanced Aerobic Biodegradation
USAF, Battelle Memorial Institute,
IT Corporation, EA Engineering
Experimental ful 1—scale
Aromatic and aliphatic hydrocarbon
fuels in soil and groundwater
Groundwater, soil
On—site, In—situ
Pump off free product, construct
aeration basins, water returned to soil
Eglin Air Force Base, Florida,
United States
Jet fuel (20,000 gallons)
Shallow groundwater, 4,000 cu. yd.
highly permeable sandy soil
30% removal achieved in 18 months
$1,700,000 minimum (non-research
cost) 5 years estimated to complete
cleanup
Douglas C. Downey
USAF Engineering & Services Ctr.
Tyndall AFB, Panama City, FL
32403-6001
United States
904-283-2942
11/89
7-a-3
83
-------
SUMMARY OF U.S AIR FORCE FIELD EXPERIMENTS
FOR IN SITU TREATMENT OF FUEL SPILLS
Presented to the Third International
Meeting of the NATO/CCMS Pilot Study
on Soil and Groundwater Remediation
Mr Douglas C. Downey
HQ AFESC/RDV Tyndall AFB FL
9 November 1989
84
-------
PERFORMANCE OF SELECTED IN SITU SOIL DECONTAMINATION
TECHNOLOGIES: AN AIR FORCE PERSPECTIVE
Douglas C. Downey, Michael C. Elliott
Air Force Engineering and Services Center, Tyndall ATh FL. 32403
INTRODUCTION
Every year the U.S. Air Force stores and transfers three billion
gallons of JP—4 jet fuel. Unfortunately not every gallon of fuel has been
consumed in flight. Fuel spills account for nearly half of the chemically
contaminated sites on Air Force installations and that percentage is
growing as underground storage systems are more closely inspected. The
Air Force Engineering and Services Center’s Environics Division is
responsible for developing and testing new and more cost effective
technologies capable of cleaning up fuel spills in a variety of soil and
groundwater conditions. Special emphasis has been placed on soil
decontamination because our sampling data clearly shows that the vast
majority of spilled fuels are adsorbed or occluded in the soil above the
water table.
This paper summarizes the results of several field tests which have
used a variety of in situ technologies to treat soil contamination.
Observations on the success and shortfalls of in situ soils washing,
enhanced biodegradation, soil venting and radio frequency soil heating are
presented in an abbreviated format. These field tests have shown that in
situ decontamination methods which use water as a contact medium to remove
hydrocarbons from the vadose zone have consistently fallen short of their
clean up goals. Technologies which use vented air and thermally enhanced
venting for contacting fuel residuals in the soils have met with greater
success. These results indicate that the physical accessibility of fuel
residuals to treatment media often determines the success or failure of
soil decontamination technologies.
LIMITED ACCESS
During the past two decades, researchers have primarily focused on
discovering and optimizing chemical and biological reactions which will
alter or degrade fuel hydrocarbons with little regard for the engineering
of effective application systems. Using batch and column experiments,
laboratory success has been achieved in biological degradation 1 ,
surfactant soils washing 2 , and chemical oxidation 3 . However, field
testing has shown that contact between injected or infiltrated treatment
chemicals and hydrocarbon contaminants is very difficult to achieve. Soil
structure, fuel composition, and the depth at which the release of fuels
occurred are just a few of the factors affecting accessibility that should
be examined prior to selecting any in situ technology.
85
Presented at the American Institute of Chemical Engineers 1989 Summer
? 7 t1ona1 Meeting. Philadelphia PA, August 20—23, 1989.
-------
Low permeability silts and clays slowly accept fuel hydrocarbons into
their available pore space. This can retard the advance of fuel spills,
but it results in a highly inaccessible fuel residual. Fuels become
trapped in soil micropores and form a thin film over the large surface
area of these fine—grained materials. Water attempting to convey a
chemical or biological treatment media to these fuels is limited both by a
low hydraulic gradients and channeling through only large pore spaces.
Even in larger—grained sandy soils, fuel blobs can become trapped in pore
throats and in small pores, causing water to channel through only the
large, unblockedpore spaces
The type of fuel also impacts accessibility for treatment. The
viscosity, solubility and bulk vapor pressure are important factors in
predicting fuel residual accessibility and response to water or air
contact. For example, gasoline typically contains a greater fraction
of water soluble compounds and has a higher bulk vapor pressure than jet
fuels. As a result, gasoline is very accessible to water and air based
treatment methods. Less viscous diesel and heating fuels are more likely
to completely fill available pore spaces and retard the advancement of
treatment media. Accessibility is also limited by the lower vapor
pressure and water solubility of heavier fuels.
The location of the initial fuel release relative to the water table
and capillary fringe also effects the distribution of fuel residuals in
the soils. Laboratory experiments indicate that when hydrocarbons are
released in water—saturated or near—saturated conditions, they form large
blobs which are trapped in larger pore spaces with little displacement of
water from micropores. However, fuels released in drier soils above the
water table tended to migrate into micropores and form thin films over
soil particles. 5 These residuals may have more limited contact with
treatment fluids. Our field experiments have clearly shown that fuel
accessibility, and the factors which determine fuel location in the soil
matrix, deserve far more attention by both the scientific and engineering
community.
IN SITU SOILS WASHING
In the early 1980’s the EPA Hazardous Waste and Engineering Research
Laboratory at Edison, New Jersey performed experiments using a variety of
surfactant solutions to remove crude oils and PCBs from contaminated soil
columns. 6 The results of these column tests showed that after passing
ten pore volumes of a 1.5 percent surfactant solution through the columns,
that 88 percent of the crude oil and 90 percent of the PCBs were washed
from the soils. With these promising results the EPA and AFESC initiated
a joint project to pilot test In situ soils washing at a contaminated Air
Force site.
Because soils washing is best suited for permeable soils, a sandy site
was desirable. An abandoned fire training area at Volk Field Air National
Guard Base (ANGB) in Wisconsin was selected as a research site.
Historical data indicated that the fire training area had been used since
1955 and that as much as 200,000 liters of .JP—4, waste oils and solvents
may have soaked into the soils. 7
86
-------
A sieve analysis of the soil indicated 95 percent medium to fine sand
with less than 5 percent fines by weight. The sand is uniform and
unconsolidated to a depth of 3 to 5 meters where a highly compacted
sandstone is encountered. The vertical permeability of the soil in the
unsaturated zone was measured in a laboratory permeazneter at 4 X i - to
5 x l0 cm/sec. The hydraulic properties of this soil, seemed
acceptable for a soils washing application.
A series of in situ test beds, each containing 0.2 in 3 of undisturbed
soil, were established in the fire training area and initial soil, samples
were taken to establish baseline contaminant levels. Soils were analyzed
using two methods, an oil and grease extraction was used as a general
indicator of hydrocarbon contamination and a gas chromatograph was used to
analyze the volatile aromatic and aliphatic fractions of the fuel.
Initial oil and grease values ranged from 1000 to 6000 mg/kg of soil.
Three synthetic surfactant solutions mixed in clean water were used in
this field test and recirculated groundwater was used as a control. 8
Wash solutions were applied at the rate of 70 liters/m 3 /day for six
consecutive days. An unexpected decrease in percolation rate was observed
in all of the test beds. After 14 pore volumes were passed through the
soil, the test beds were rinsed with clean groundwater to remove excess
surfactants. The test beds were then resampled to determine contaminant
removal rates.
Soil samples were collected from two depths (5—10 c m and 30—35 cm
below the surface) •in each test bed. These samples were again analyzed for
oil and grease. The results of the post—test analysis showed that
surfactant solutions did not provide a statistically significant decrease
in fuel and oil contamination at either depth. Despite the repeated
success of engineered surfactants to clean contaminated soils in
laboratory columns, the sandy soil at the Volk Field site was not cleaned
in situ. Within statistical limits, there was no significant difference
in pre— and post—wash contaminant levels.
Although the laboratory coluis were packed to simulate in situ soil
density and permeability, the reduction in permeability and incomplete
contact with fuel residuals encountered in the field test was not
predicted in the laboratory. This underscores the importance of pilot
testing on contaminated sites before committing to full—scale
decontamination technologies.
ENHANCED IN SITU BIODEGRADATION
Common sot]. microorganisms have the ability to degrade virtually all
of the hydrocarbons found in common fuels. Several scientists have
confirmed biodegradation under favorable . .erobic conditions. 9 ”°
Enhanced in situ biodegradation is an attempt to create these favorable
aerobic conditions in an environment of heterogeneous soils and delicate
geochemical balances. While several commercial firms have claimed
successful site remediations, published results often lack sufficient data
to determine the effectiveness of biodegradation in reducing fuel
residuals in the soil. For this and other reasons, AFESC decided to
conduct independent field tests of this technology prior to recommending
it for widespread Air Force application.
87
-------
In 1984, AFESC initiated a pilot—scale test of enhanced biodegradation
at a site on Kelly AFB TX. As this test progressed, problems with soil
permeability were encountered reducing the delivery of hydrogen peroxide
and nutrients through injection wells. This reduction in permeability was
attributed to both natural silt and clay soils and the precipitation of
calcium phosphates which formed as injected phosphates reacted with
calcium in the 50 ii. Permeability problems reduced the delivery of
oxygen and consequently little biodegradation occurred. Based on these
results a second site was selected at Eglin AFt FL for additional testing
under more favorable hydraulic conditions.
In 1984 a fuel leak was discovered in a 15 cm (6 inch) underground
JP—4 fuel supply line inside the base fuel storage compound on Eglin A lt.
At the time of discovery 1 an estimated 75,000 — 100,000 liters of fuel had
contaminated over 3000 m of soil and shallow groundwater. A series of
shallow, gravel filled trenches and skimmer pumps were used to recover
over 30,000 liters of fuel. In 1986, MISC initiated a field test of
enhanced in situ biodegradation. The site is located in an area of
unconsolidated coastal sands that extend from the surface down to 12
meters where a thick layer of clay is encountered. Groundwater is found
only one one meter below the surface and has a high hydraulic conductivity
of 6 X 10—2 cm/sec. These favorable soil and hydraulic properties made
it an excellent site for testing enhanced biodegradation.
Extensive soil and groundwater sampling preceded the test. Soil
samples were taken from 4 to 6 depth intervals at 12 sampling locations
across the site and analyzed for total petroleum hydrocarbons using EPA
Method 418.1. Soil and groundwater samples from four of the locations
were analyzed using a CC/MS and 43 representative compounds were
identified for special monitoring. Site characterization confined that
over 90 percent of the jet fuel remaining on the site was above the water
table, adsorbed and occluded in the soil matrix. Although the leak
occurred at or slightly below the water table, rising and falling ground
water had deposited hydrocarbons up to the ground surface.
A nutrient and hydrogen peroxide delivery system was designed to test
the relative effectiveness of three delivery methods in stimulating
biodegradation in the vadose zone and in the groundwater. Two shallow
injection wells, infiltration galleries and a spray irrigation system were
all installed for a side—by—side comparison. Four downgradient recovery
wells were installed and initially produced 150 — 190 liters/tin (40 — 50
gpm) for recirculation through the site. Due to the presence of 5—15 mgll
of iron in the shallow groundwater, an aeration basin and settling tank
were added to precipitate and remove iron prior to reinjection. Iron
fouling is a common cause of reduced permeability and failure of
reinjection systems.
Several important tests were completed in advance of full—scale
operation which began in June of 1987. Prior to nutrient/peroxide
additions, site hydrology was studied under pumping and delivery
conditions. The initial capacity of the three delivery systems was
measured to provide a baseline for site permeability. A conservative
chloride tracer was introduced into the infiltration gallery and its
transport monitored across the well field to insure hydraulic connection
across the site.
88
-------
Fuel contaminated soil and groundwater was observed in laboratory
microcosm studies to determine the general activity of existing fuel
degrading microorganisms and their response to nutrient and oxygen
additions. Simple bench scale microcosm studies confirmed that under
enriched oxygen and nutrient conditions existing soil bacteria could
degrade soluble, aromatic hydrocarbons in less than two weeks. Success,
however, is not measured in laboratory flasks but in the field where these
optimum conditions are much more difficult to achieve.
Oxygen supply and distribution is absolutely critical to field
success. Although estimates vary, complete biodegradation of fuel
compounds will require 2 to 3 grams of oxygen per gram of fuel degraded. (
An interesting discussion on the importance of the oxygen requirement is
found in a recent publication by HincheeJ 2
Hydrogen peroxide is the most frequently used oxygen source for
enhanced biodegradation. Peroxide is capable of releasing enough oxygen to
fully saturate injected groundwater to its oxygen solubility limit of
approximately 40 mg/L. This is a substantial improvement over the 10 mg/l
Oz solubility for standard aeration of groundwater. However, hydrogen
peroxide is inherently unstable and its use depends upon a gradual
breakdown of peroxide and time release of oxygen downgradient of injection
points.
Shortly after initiating 500 ppm hydrogen peroxide additions at the
Eglin site, gas bubbles were noticed coming up through the water in the
shallow infiltration galleries. Gas sampling showed that this was
virtually pure oxygen; the product of rapid peroxide decomposition. Iron.
was first suspected as the catalyst of this rapid decomposition. However,
laboratory tests showed that the rate of H 2 0 2 decomposition in iron
solutions was at least an order of magnitude slower than field
decomposition rates. Subsequent laboratory experiments found that
peroxidase enzymes produced by bacteria near the point of injection were
the most likely cause of uncontrolled 0202 decomposition. 13 As a
result of this oxygen off gassing, only an estimated 16 percent of the
potential oxygen supply was actually delivered to the contaminated soil
and groundwater. Although there was a slight increase in dissolved oxygen
levels downgradient of injection points, there was no evidence of 11202
transport.
After 18 months of peroxide and nutrient additions, aromatic
concentrations in groundwater monitoring wells had decreased from 8 ppm to
200 ppb. However 1 intense sampling of soils above and beneath the water
table did not show a significant removal of soil bound fuel
residualsJ 4 The failure of nutrient and hydrogen peroxide additions
to impact soil contamination was particularly evident in the unsaturated
zone of the spray application area. Over 190 pore volumes o’ treatment
water passed through these sandy soils with no significant fuel removal
measured. It appears that neither hydraulic washing or biodegradation had
an impact on this tightly bound fuel residual. We have concluded that
fuels trapped within the micropores of the soil were largely inaccessible
to the nutrients and oxygen that were being provided.
89
-------
Our experiences at Kelly AFB and Eglin AFB test sites have shown that
enhanced biodegradation can not be applied at sites with poor permeability
and that contaminant accessibility can be a problem even in more
permeable, sandy soils. This technology seems best suited for sandy or
gravel aquifers where the majority of the contamination is in the
saturated zone. A similar test of this technology conducted by the EPA ’s
R.S. Kerr Laboratory showed greater peroxide stability, particularly when
the peroxide was injected into the saturated zone. 15 In situ peroxide
stability is clearly a major design consideration that should be
determined through a carefully monitored pilot test on each site. If the
effective oxygen concentration achieved with peroxide is not substantially
greater than the 40 mg/i available through pure oxygen aeration, then the
use of hydrogen peroxide is not cost effective.
IN SITU SOIL VENTING
In situ soil venting is a soil decontamination technique which uses
vacuum blowers to pull large volumes of air through contaminated soil.
The air flow sweeps out the soil gas disrupting the equilibrium existing
between the contaminants on the soil and in the soil vapor. This causes
volatilization of the contaminant and subsequent removal in the air
stream. In situ soil venting has reportedly been successful for removal
of volatile contaminants such as gasoline 16 and trichloroethylene 17 .
Because of the reported success of the technology on volatile
contaminants in soils, the Air Force Engineering and Services Center began
a research project with Oak .Ridge National Laboratory to conduct a
full—scale test of in situ soil venting at a jet fuel (JP—4) spill site.
In. general, JP—4 has more heavy molecular weight hydrocarbons and is less
volatile than gasoline and other contaminants which have previously been
investigated for remediation by in situ soil venting.
The site chosen for test was in a fuel yard at Hill ATh Utah where a
100,000 liter JP—4 spill had occurred in January 1985. The JP—4 spill
site is in an area of medium to fine dry sands with thin interbedded
layers of silty clay. This soil type extends to a depth of approximately
180 meters below land surface. Perched groundwater has been encountered
on top of the silty clay layers. The regional aquifer of greatest
significance as a water bearing unit is at an average depth of 180
meters. After extensive soil sampling and multiple soil gas surveys, it
was determined that the spill had contaminated an area 37 meters by 37
meters to a depth of approximately 15 meters below land surface.
Information from the site characterization and the one—vent pilot
test 18 provided the basis for the design of a full—scale in situ soil
venting system for remediation of the JP—4 contaminated soil. The
full—scale venting system design consisted of the three subsystems: (1.) a
vertical vent array in the area of the spill, (2) a lateral vent system
under a new concrete pad for the recently excavated underground storage
tanl.s (USTs), and (3) a lateral vent system in the pile of soil from the
excavation of the USTS. This design includes features which permit
evaluation of several factors affecting contaminant transport and
subsurface air flow.
90
-------
The vertical vent subsystem consists of a 15 vents placed in the
contaminated area to depths of 15 meters. The venting subsystem under the
new concrete pad includes six lateral vents spaced 46 meters apart at a
depth of approximately 6 meters below land sur face. The subsystem for
excavated soil pile consists of eight vents spaced 5.5 meters apart at a
depth of 1.5 meters below the top of the pile. This pile is approximately
49 meters long, 13 meters wide, and 4 meters high. A blower/emission
control system was installed for inducing air flow to the three subsystems
and for treating emissions as necessary to meet regulatory requirements.
A condensation drum, flowmeters, and gas monitors were also included in
the system.
Operation of the full—scale, in situ soil venting system began in
December 1988. As of August 15, 1989, approximately 45,000 kilograms of
JP—4 hydrocarbons have been extracted in the gas from the soil.
Decontamination of the site can be seen by comparison of soil gas
concentrations in an area of the vertical vent system. The highest soil
gas hydrocarbon concentrations have dropped from 179 percent of the lower
explosion limit (LEE .) in February 1989, to 88 percent LEL in April 1989,
to 33 percent LEL in June 1989, to 5 percent LEL in August 1989. Also,
the concentration of hydrocarbons in extracted gas from the entire venting
system has dropped from 38,000 ppm hexane equivalent in December 1988 to
700 ppm hexane equivalent in August 1989. It is estimated the entire
venting system vapor concentration will decrease to below 500 ppm by the
end of September 1989. This is below the level required for mandatory
site cleanup in some states (e.g., Florida 19 ). The state of Utah has
not set standards for cleanup levels of petroleum contaminated soils.
Another important mechanism of remediation from in situ soil venting
besides volatilization and removal, is biodegradation. The increased
oxygen levels in the soil gas due to infiltration of atmospheric air may
considerably stimulate biological activity. To evaluate this factor,
carbon dioxide and oxygen are being measured in the extracted gas at the
soil venting test. Initially, high CO 2 (11%) and low oxygen (1%) levels
were measured in the soil gas. As venting continued, the CO 2 levels
decreased and the oxygen levels increased. Carbon dioxide levels have
continued to be an order of magnitude higher than atmospheric, which
suggests biodegradation may play a significant roll in the remediation of
the site.
The results obtained to date from the JP—4 in situ soil venting test
have shown that this technique is very effective in removing large amounts
of jet fuel from the soil in a relatively short period of time. Continued
testing is aimed at determining the importance of various factors in
hydrocarbon removal. We are continuing to sample the extracted gas to
determine both the total hydrocarbon levels and hydrocarbon distribution.
The effects of moisture on volatilization and bioactivity will be
determined by monitoring soil moisture and extracted gas humidity. The
effect of heat addition to the soil for enhanced volatilization will be
tested by routing heated air from the catalytic oxidation units to vents
acting as air inlets. In October 1989, the system will be shutdown for an
extensive soil sampling of the site to determine the extent of the JP—4
91
-------
hydrocarbon removal. Based on the data from the extracted gas, we project
between 70 to 80 percent of the 100,000 liter spill will have been removed
by the time of the sampling. This upcoming data should provide valuable
information for full—scale design of in situ venting systems for .JP—4
contaminated soils.
RADIO FREQUENCY THERMAL SOIL DECONTAMINATION
Radio frequency (RI) heating uses electromagnetic energy directed
through electrodes in the soil to create molecular vibration and rotation
which uniformly heats the soil. Radio frequency heating was first
developed for recovering oil from oil shale and tar sands in the l970 s.
Field tests proved the feasibility of heating rock formations from 200 to
400°C. As the energy crisis calmed, the developer of the technology,
the Illinois Institute of Technology Research Institute ( IITRI), sought
out alternative applications for in situ soil heating.
In 1985, AFESC and EPA began a joint research project with I ITRI to
explore the use of RF heating for in situ soil decontamination. Because
the majority of Air Force contaminants, including JP—4 fuels, have boiling
points less than 150°C, RF heating could be used to vaporize
hydrocarbons in situ and remove them from the soil. Laboratory
experiments using soils contaminated with fuels and solvents produced
excellent results with over 95 percent removal at temperatures of
100—150°C. Follow—on experiments in 1.5 meter soil columns proved the
feasibility of removing volatilized hydrocarbons using a minimum of soil
venting. These promising results led to a decision to conduct a pilot
test of this technology at a contaminated Air Force site.
The abandoned fire training area at yolk Field ANGB WI was again
selected as a test site because of its uniform soil and contaminant
profile. Although a description of the site is contained in the soils
washing section, it is important to point out that a great deal of
additional soil sampling was performed for this test to study the removal
of different compounds at different depths in the treatment volume. Over
90 different soil samples were obtained from three depths within the 500
ft 3 test volume.
A trailer—mounted 40 kW RE generator was available from past DOE
research and was transported to the test site. The test volume measured 4
meters long, 2 meters wide and 2 meters deep and was heated by 39
electrodes in 3 rows of 13 each. A vapor barrier was placed over the
heated area to collect escaping soil gas and to transport the gas to a
vapor condenser for separating liquid hydrocarbons and a carbon bed to
treat remaining volatile organics. A 560 liter/mit. (20 cfm) vacuum
provide a slight negative pressure to insure all vapors were collected and
treated. The test volume and gas handling system was heavily instrumented
to provide soil temperature data and hydrocarbon concentration data in the
escaping gas stream.
92
-------
R I energy was applied to the soil over a period of 12 days. After 8
days the 1500C target temperature was achieved throughout the test
volume and this temperature was maintained for a period of 4 more days.
During this heating period, careful records were kept on. the release of
hydrocarbons and water vapor from the soil. At one point an inert tracer
was injected into the soil outside of the treatment area to confirm that
migration was into the heated zone and to estimate soil gas velocity.
Power consumption was also monitored to determine the operating cost of
this process. After 12 days power was turned off and the soil was allowed
to cool prior to resampling.
The efficiency of the RI decontamination process was determined by a
careful comparison of pre—test and post—test soil samples. Samples were
analyzed to determine changes in moisture, volatile aliphatics, volatile
aromatics, and semi—volatile aliphatics and aromatics. The average
removal rates from the heated volume were impressive with 97 percent
removal of semi—volatile hydrocarbons and a 99 percent removal of volatile
aromatics and aliphatics. Closer examination of the samples showed that
contaminant removal at the 2 meter depth, the fringe of the heated zone,
exceeded 95 percent. 2 ° The total RI process consumed approximately 1000
kw—hr/m 3 . Use of a state—of—the—art RI generator for full—scale
applications could reduce the power input to less than 650 lcw—hr/m 3 .
FUTURE RESEARCH
Future research conducted by the Air Force Engineering and Services
Laboratory will emphasize new applications of the soil venting process to
remove or destroy fuel residuals in the unsaturated zone. The
demonstrated ability of air to access soil—bound fuel residuals represents
a distinct advantage over water based treatment systems. Although the
authors’ conclusions are based primarily on personal field experiences, we
have found no convincing evidence in open literature to suggest that in
situ water based treatment has consistently remediated fuel contaminated
soils.
Two enhancements to the soil venting process will be developed and
tested during the next year. One enhancement is the combination of soil
venting and RI heating to more rapidly volatilize fuel residuals and to
increase the volatilization of compounds with higher boiling points. The
uniform heating provided by radio frequency energy is also expected to
improve soil porosity and improve removal rates in clay and silt soils. A
full—scale test of the RI heating/venting system will be conducted on an
Air Force site in the summer of 1990.
A second enhancement of the soil venting process will attempt to
optimize the biodegradation of fueli that results when vented air provides
oxygen to subsurface bacteria. The goal of this research will be to
determine the optimum range of soil moisture, nutrients, and venting rates
to achieve in situ biodegradation while minimizing the emission of
volatile organics to the atmosphere. A pilot—scale test of this enhanced
biodegradation method is currently underway at a fuel contaminated site at
Tyndall A lE FL.
93
-------
Research is continuing in the use of in situ and aboveground
biological, physical and chemical treatment technologies to remediate
contaminated groundwaters. Because site remediation will generally
require two or more technologies, more emphasis vi].1 be given to systems
integration to impact the both source and dispersed contaminants at
minimum expense.
94
-------
REFERENCES
1. Downey, D.C., Hinchee R.E., Westray, M.S., Slaughter, J.K.,
Combined Biological and Physical Treatment of a Jet Fuel Contaminated
Aquifer”, Proceedings of NWWA/API Conference on Petroleum Hydrocarbons in
Ground Water . Houston TX, Nov 1988.
2. Nash, .1., Traver, R., Do iney, D.C., Surfactant—EnhanCed In Situ Soils
Washin& , AT Engineering and Services Laboratory Technical Report No.
87—18. September 1987.
3. Rauch, P.A., Watts, R.J. “In Situ Treatment of Pentachiorophenol
Contaminated Surface Soils Using Fenton Reagent”, Proceedings — 61st
Annual Conference of the Water Pollution Control Federation , Dallas TX.
4. Wilson, J.S., Conrad, S.H., “Is Physical Displacement of Residual
Hydrocarbons a Realistic Possibility in Aquifer Restoration”, Proceedings
of NWWA/API Conference on Petroleum Hydrocarbons in Ground Water . Houston
TX. 1984. pp 274—297.
5. Wilson, J.L., Conrad, S. ! !., Hagan, E., Mason, W.R., Peplinski, “The
Pore Level Spatial Distribution and Saturation of Organic Liquids in
Porous Media”. Proceedings of NWWA/API Conference on Petroleum
Hydrocarbons in Ground Water . Houston TX. 1988. pp 107—133.
6. Nash, Surfactant—Enhanced In Situ Soils Washing , p 20.
7. Nash, Surfactant—Enhanced In Situ Soils Washing , p 41.
8. HMTC, Installation Restoration program Records Search—
Volk Field Air National Guard Base . August 1984.
9. Atlas, R.M., Microbial Degradation of Petroleum Hydrocarbons: An
Environmental Perspective. Microbiological Reviews . V45, pp.180—199.
10. Lee, M.D. et al,” Biorestoration of Aquifers Contaminated with
Organic Compounds,” CRC Critical Reviews in Environmental Control . Vol
18., pp 29—89.
11. Wetzel, R.S. et al, In Situ Biolo&ical Treatment Test at Kelly AFBL
Vol II Field Test Results and Cost ModeL Air Force Engineering and
Services Laboratory Technical Report No. 85—52, Jul 1987.
12. Hinchee, R.E., Downey, D.C., Voudrias E., “The Role of Hydrogen
Peroxide Stability in Enhanced Biodegradation Effectiveness”, Proceedin&s
of lIWWA/API Conference on Petroleum Hydrocarbons in Ground Water . Houston
TX, 9 —11 Nov 1988.
95
-------
1.3. Spain J.C., Milligan J.D., Downey D.C., Slaughter J. ., “Excessive
Bacterial Decomposition of Hydrogen Peroxide During Enhanced
Biodegradation.” Journal of Ground Water . Mar—Apr 89.
14. Kinchee, R.E., Downey D.C., Westray, M.S., Slaughter, Enhanced
Biodegradation of Jet Fuels — A Full—Scale Test at Ealin AFB FL . Air Force
Engineering and Services Laboratory Technical Report No. 88—78. Aug 89.
15. Personal Communication with Dr John Wilson, EPA R.S. Kerr Laboratory,
Ada OK. Apr 1989.
16. Anastos, C. 1., P. 3. Marks, M. H. Corbin, and N. F. Coia,
In Situ Air StriD ina of Soils Pilot Study. Final Report ,
ANXTH—TE—TR—85026 , October 1985.
17. Thornton, 3. S., I L E. Montgomery, T. Voynick, and W. L. Wootan,
‘Removal of Gasoline Vapor from Aquifers by Forced Venting,” 1984
Hazardous Material Souls Conference Proceedings , April 1984.
18. Elliott, N. G. and B. W. DePaoli, “In Situ Venting of Jet
Fuel—Contaminated Soil,” presented at the 44the Purdue Industrial Waste
Conference, May 10, 1989, West Lafayette, Indiana.
19. Florida Department of Environmental Regulation, Guidelines for
Assessment and Remediation of Petroleum Contaminated Soils , January 1989.
20. Dev, K., Dovney, D.C., Stretsy, C., Bridges, J.E., “Field Test of the
Radio Frequency Soil Decontamination Process”, Proceedings of Superfund 88
National Conference and Exhibition , Wash D.C., 28—30 Nov 1988.
96
-------
BIOGRAPHICAL SKETCH
Douglas C. Downey, PE is Chief of the Air Force Engineering and
Services Center Laboratory’s Environmental Engineering Branch,,
Tyndal]. AFB FL. Mr Downey is responsible for the development and
testing of soil and groundwater decontamination technologies for
the clean up of contaminated sites at over 200 Air Force
facilities. His most recent projects include field demonstrations
of enhanced biodegradation, radio frequency thermal soil
decontamination and soil venting. He is a 1977 graduate of the US
Air Force Academy and attended Cornell University where he
received a Masters Degree in Civil/Environmental Engineering in
1981.
Captain Michael C. Elliott, PE is a research engineer assigned to
the Air Force Engineering and Services Laboratory. He is
responsible for the development of innovative air—stripping
techniques, emissions control technologies for volatile organics,
optimized pumping strategies for groundwater cleanup and most
recently a full—scale test of in situ soil venting. He holds a
bachelor of sciences degree in civil engineering from the
University of Illinois and a master of science degree in
environmental engineering from Southern Illinois University.
97
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTERIZATION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On-/Off—site Treatment Location:
Pre- and Post-treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Rol Roth
Dekonta GmbH
Lotharstr , 26
6500 Malnz
Federal Republic of Germany
06132-772211
Thermal
Thermal desorption and destruction
Dekonta GrnbH
Pilot
Soil
Transportable (not mobile)
Excavation
Former herbicide and pesticide
production facility, Hamburg,
Federal Republic of Germany
Chlorobenzenes, ch 1 orophenol s,
hexachiorocyclohexane, dioxins,
furans
11/89
1-2
99
-------
IDEkONTA1
Ge e :la I. . D,’3Ma’ ruI.3r .—
S I A I U S R E p 0 R T
i ia i’aBr 2
• 0132 7 ”
NATO/CCMS Pilot Study on Demonstration of
Remedial Action Technologies
12. Oktober 1989
Thermal treatment of contaminated soil by radiation heating
The pilot plant conceptual design and the goal of the intended remedial
action has been reported in the ‘Copenhagen—Workshop’.
Because of handling soil with the main contaminants chlorobenzenes, chlo-
rophenols, hexachiorocyclohexane, dioxins and furans we have to have a
permit by german ‘BlmSchG’-requirements; because of the dioxin problem, we
also need a ‘Sicherheitsanalyse’ (comparable to faile—safe-analysis) to
get a permit.
The expert-checking of the ‘Sicherheitsanalyse’ took more time than expec-
ted and the ‘green party’ has some objections to the permit — so up today,
we have a remarkable delay in getting the permit for start of operations.
In between, we use the ‘interim status’ for intensive operators training
by a lot of ‘dumny running’ and functional tests with and without clean
wet soil - in total more than 600 hours.
o we could do some mechanical development and improvements arc rEalize n
advanced computer control system, but we had no chance to demonstrate the
capability of the skidmounted, transportable pilot system.
Before start of decontamination each responsible staff-member will have
nearly 1000 training hours, of which nearly 50 % is ‘learning by doing’.
100
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTER I ZAT I ON
General Type:
Specific Type:
t4anufacturer/Researche r:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On—/Off—site Treatment Location:
Pre- and Post—treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Edward Burk
U.S. Environmental Protection Agency
Response Section I
9311 Groh Road
Grosse lie, MI 48138
United States
313-675-3146
In—situ vitrification
GeoSafe Corporati on
Experimental
Organics, inorganics, heavy
metals, radioactive wastes
Soil, sludges (with nondestruc—
tible solids)
in situ; mobile
Pesticides formulator, Parsons
chemical site, Grand Ledge,
Michigan, United States
DDT, mercury, arsenic, heavy
metals, chlordane
James E. Hansen
GeoSafe Corporation
303 Parkplace, Suite 126
Kirkland, WA 98033
United States
206—822-4000
11 / 89
6-2
101
-------
Vitrification of Organic and Inorganic
Contaminated Soils at the Parsons Chemical!
ETM Enterprises Site, Oneida Township, Michigan
Edward C. Burk, Jr.
U.S. Environmental Protection Agency
EXECUTIVE SUMMARY
The purpose of the application of the vitrification technology at the Parsons
Cheiiical/ETM Enterprises (Parsons/ETM) site is to mitigate an imminent and
substantial threat to public health, welfare, and the environment from
hazardous material that contains pesticides, heavy metals, and low levels of
2,3,7,8-tetrachloro—dibenzo-p—dioxins (TCDD). The chosen application shall
provide a permanent destruction of organic contaminants, an encapsulation of
inorganic contaminants, as well as providing significant volume reduction of
the treated waste. An estimated 2,000 cubic yards of contaminated material
will be staged in the most effective configuration for the vitrification
process. The on-site treatment utilizing vitrification technology will allow
detoxification of the complex waste stream at a competitive market price with
a system that has received positive community support.
INTRODUCTION
The Parsons Chemical/ETM Enterprises (Parsons/ElM) site is located
approximately one-quarter of a mile east of the intersection of M—43, Grand
Ledge Highway and Larson Road, Eaton County, Grand Ledge, Michigan (Figure 1).
During 1980, the present owner, EN Enterprises (ETM), had a hydrogeological
study performed to determine the source of contamination. In 1983, upon
recommendation from their consultant, EN completed the excavation and
disposal of the septic tank and leach field (Figure 2).
The near surface geology to a depth of approximately 15 feet can be generally
characterized as a glacial till deposit of stratified clay and loam with minor
sand stringers. A thin loamy top soil ranging from one—half to two feet in
thickness covers the site. Below this is found either a brown loam or clay
with variable amounts of granular material, ranging in thickness from seven to
eleven and one-half feet. This material contains thin, saturated sand
stringers. Overall moisture content varies from relatively dry, firm clay to
soft loam to the saturated sand stringers. The underlying layer is a gray,
firm till clay which dips to the south.
The source of groundwater appears to be perched. It is either originating
from the saturated loam or sand stringers. The elevation of these sources
vary across the site. Although there is some degree of interconnection
between the water bearing units, is appears to be slight. From lithologies
observed, it is likely that the water is under perched conditions and travels
within the more granular portion of the subsurface at a very slow rate.
Prior to ElM occupying the property, Parsons Chemical Works, Inc. (Parsons)
engaged in the mixing, manufacturing, and packaging of agricultural chemicals.
Materials handled during Parsons’ operation included pesticides, herbicides,
102
-------
103
-------
CLOSED
CATCH
BASIN
L HWAY 1
GRAND LEDGE DITCH
CULVERT
TO MILLEROOK PRINTING
OPEN CONCRETE
CATCH BASIN
SCALE IN FEET.
o 30 60 .0
•1 —
SNEoD
FIGURE 2 SITE MAP
(TM ENTERPRISES
GRAND LEDGE, KICHIGAN
ONEIDA SWEET
H ’ CULVERT
STORM DRAIN PIPE
4 CLAY DRAIN TILES
SEPTIC TANK
(REMOVEDI
POWER
POLE
104
-------
solvents, and mercury—based compounds. Concerns arose in 1979 and 1980 when
the Michigan Department of Natural Resources (MDNR) collected sediment samples
from the unnamed creek and the Grand Ledge ditch located on the north boundary
of the EN site. Analytical results revealed elevated levels of lead,
mercury, arsenic, and pesticides, including di chloro—di phenyl—trichloroethane
(DOT) and chlordane.
In light of historical information and previous MDNR sampling results, the
United States Environmental Protection Agency (U.S. EPA) targeted the
Parsons/ETM site for Tier 3 dioxin screening under the National Dioxin Study
conducted in 1984. Tier 3—dioxin screening began in November 1984. Nineteen
sample locations were selected from surface and subsurface soils on the ElM
property. All were analyzed for TCDD, volatile and semi—volatile organics,
pesticides, and metals. In addition, three samples were analyzed for dioxin
and furan isomer contamination. Detectable levels were present in all three
samples analyzed for these parameters. Results from samples collected in
November 1984 on ElM property indicated levels of ICOD to be 1.13 ppb at the
surface and 0.56 ppb 18 inches below the surface.
In November and December of 1988, the MDNR developed and implemented a
sampling program for the Parsons/ElM site. The sample collection was to
provide current analytical data on contamination levels present both on site
and off site. A sampling grid was utilized for on—site sample collection.
Samples were collected from the outfall to the unnamed creek, as well as
sediment samples from the creek and Grand River. Subsequent to the sample
collection, they were submitted through the MDNR contract laboratory program
for analysis. Samples were analyzed for pesticides and metals.
Results of the MDNR sampling program revealed wide spread contamination of
mercury, both on site and at the drainage outfall. Concentrations range from
1.9 ppm to 150 ppm on site and from 1.5 ppm to 36 ppm at the outfall. One
sample location in the stream revealed an elevated level of mercury at 2.9
ppm. Sample collection was achieved at or near the surface, as well as at
depth.
As with past sampling efforts, the recent MDNR sampling revealed elevated
levels of pesticides. Those most prevalent on site are 4,4—DDE, 4,4-DOT,
dieldrin, and chlordane. Sample locations utilized for pesticide analysis
were also analyzed for heavy metals.
Presently, ElM is continuing to operate at the facility. The two suspect
areas contaminated with pesticides and TCDD have been fenced; however, over
time, the erosion of soils has made both areas easily accessible. Also,
through the erosion process, migration of contaminants may continue. The
septic tank and leach field were excavated and disposed of, but the adjacent
soils and clay tile leading to the ditch and creek remain. In addition, soils
identified on site as being contaminated remain.
TECHNOLOGY
The in situ vitrification (ISV) process, as applied to contaminated soil,
requires the insertion of four electrodes into the soil in a square array.
The average size cell for vitrification is 20 feet square with electrodes
105
-------
placed to a depth of approximately 15 feet. A path for electric current is
established by using a small amount of a graphite and glass frit mixture
placed between the electrodes on the soil surface. Dissipation of power
through the starter material creates temperatures high enough (up to 20000 C)
to melt a layer of soil. This molten zone continues to grow downward,
encompassing the contaminated soil. Rocks, which are less dense, create a
floating rock layer near the molten surface. The rocks will eventually be
incorporated into the molten mass. As the molten or vitrified zone grows, it
incorporates non—volatile hazardous elements, such as heavy metals, and
destroys organic compounds by pyrolysis. The pyrolyzed products migrate to
the surface of the vitrified zone where they combust in the presence of
oxygen. A hood placed over an area being vitrified directs the gaseous
effluents to an off-gas treatment system. Remaining ash, along with other
noncombustible materials, dissolve or become encapsulated in the molten soil.
Natural convective currents within the molten soil help distribute the
stabilized materials uniformly. The molten soil cools to a durable glass and
crystalline waste from resembling natural obsidian.
The process operation is based on extensive joule—heated melter work performed
at Battelle Pacific Northwest Laboratory for various nuclear waste
immobilization projects. The joule—heating principle operates by internal
resistance heating of the conducting material as an electric current passes
through the molten media. In ISV, the resistance decreases as the molten mass
grows. Therefore, to maintain a power level high enough (according to the
formula P = 1 2 R) to continue melting more soil, the current must be increased.
To accomplish this, the amperage is increased by using a power transformer
with multiple voltage taps.
At start—up, the ISV process requires a high voltage potential and low
amperage. As the melt progresses and resistance decreases, the lower voltage
taps on the power transformer allow increased amperage to the melt, thus
maintaining a high power level into the melt. The process will continue until
heat losses from the melt approach the energy delivered to the molten soil via
the electrodes.
Battelle recently demonstrated the effectiveness of ISV in destroying dioxin
in contaminated soil on an engineering scale test performed in a sealed metal
container. The results of testing done by Battelle, using a sealed metal
container and soil containing up to 100 ppb dioxin, are reported in a 1984
technical paper. Sanipling of the vitrified mass showed no detectable residual
level of dioxin. Also, no dioxin contamination was detected in the majority
of soil surrounding the vitrified block, indicating that migration outside the
vitrification zone was not a significant probl n. Battelle estimated that the
ISV process and the dual stage filter system have a combined Destruction and
Removal Efficiency (ORE) of between 99.9999% and 99.999999%.
During the vitrification process, metals are either dissolved in the glass and
incorporated in the vitreous matrix or escape with the off-gas. Since the
vitrified mass (melt) is reducing in nature, the most likely form of metals in
the melt is either in the pure state or the lowest oxidation state. Any
metals not retained within the melt, especially those present closer to the
melt surface, escape with the off—gas and are removed by the off—gas treatment
system. The sum of the metals retained within the melt and the removal of
metals released from the melt by the off—gas treatment System is the overall
106
-------
system Destruction Removal Efficiency (DRE). The results of pilot scale
testing indicated that the following metals’ removal efficiencies can be
obtained during the vitrification operation:
Percent
Percent Percent Removed Combined
Retained in in the Off-gas Removal
Type of Metals the Melt Treatment System Efficiency
Semi-volatile 99 gg•g 999999
metals (cobalt,
cesium, antimony,
tellinum and
molybdenum)
Volatile metals 90 99.9 99.999
(cadmium and lead)
Based on the above results, it can be expected that metal emissions from in
situ vitrification operations would be minimal.
The off—gas hood contains the gaseous effluents from the process, provides a
confined area for combustion, and directs the gaseous emissions to the off—gas
treatment system. The hood is grounded because it supports the electrodes via
insulators. The hood is sealed to the ground by a high-t iiperature flexible
skirt, which is bolted around the hood’s perimeter and covered with soil. The
hood is connected to the off—gas treatment system by two off—gas lines (two
are provided for redundancy). A combustion air inlet and a pressure relief
valve with carbon filters are annexed to the hood. The carbon filters provide
off—gas decontamination in the event of pressurization and backflow. The
combustion air and off—gas flows are separated by a heat shield that extends
below the hood’s ceiling.
The off—gas treatment system cools, scrubs, and filters the gaseous effluents
exhausted from the scrubber, and filters the gaseous effluents exhausted from
the hood. Its primary components include: a gas cooler, two wet scrubber
systems (tandem nozzle scrubbers and quenchers), two heat exchangers, two
process scrub tanks, two scrub solution pumps, a condenser, three mist
eliminators (vane separators), a heater, a carbon filter assembly, and a
blower system (Figure 3).
0ff-gases entering the off—gas treatment system can be expected to reach a
maximum temperature of 750° C. To keep the size of the heat exchange
equipment manageable for a transportable facility, a gas cooler is provided to
remove a major portion of the heat load from the off gases before quenching.
The gas cooler is a finned air-to—glycol heat exchanger. The gas cooler can
be bypassed by operating three pneumatically actuated butterfly valves.
From the gas cooler, the off gas is split and directed into two wet scrubber
systems operating in parallel. Because scrubbing efficiency falls drastically
with flow rate, two parallel systems are required for this range of flows. At
flows less than 60 standard cubic meters per minute (m 3 /min), only one system
will operate. The dual scrub system also provides redundancy in the event of
107
-------
iR INFU j
*
1
iGURE 3
c:) F c As TRE.ATMENT
SYSTEM SCHEMATIC
1
I
UTER 1
PO’i’JER
TO
— - — — — r f— : ‘
.1
— — — I £ — —
ULYCOL
000UNG
POWER
COND-
TIONINO
HOOD
QUENCH
SCRIJ3
\r I
HEAT
ADSORB
UTILITY OR
DIESEL
POWER
V
I —— I JI
II _ • •I r • ‘‘-
____ a-
108
-------
single component failure. Each system is composed of a quench tower, a tandem
nozzle scrubber, and a vane separator. The quencher reduces the gas
temperature from 3000 to 60° C, and provides some scrubbing action to remove
90% of particulates and semi—volatile components. The tandem nozzle
scrubbers primary functions are to remove an additional 90% of the particles
0.5 micrometer diameter (urn dia) and larger, condense remaining semi-volatile
components, and provide additional cooling of the off gas. The vane separator
that follows each tandem nozzle scrubber removes all droplets l2um and larger.
The exit temperature from the scrubber system is 60° C.
The scrub solution injected into the quenchers and tandem nozzle scrubbers is
cooled through two single—pass heat exchanges before being returned to the
process scrub tanks. Each heat exchanger can remove 120 kW from the scrub
solution. Each heat exchanger transfers the heat to a glycol solution.
Two independent scrub pumps recirculate the scrub solution from the process
tanks to the wet scrubbers. Each pump’s suction is located within 15 cm of
the bottom of the lowest portion of its respective tank. Each pump can
deliver 510 liters per minute (L/min) with a maximum deliverable pressure of
100 pounds per square inch (psi). In addition, the scrub pumps can flush out
the gas cooler and off—gas piping that are not wetted by the wet scrubbers.
Following the scrubber systems, the off gas is recombined and cooled further
from 60 to 520 C in a single—pass condenser. The condenser and mist
eliminator provide additional decontamination of the off gas (90 to 99%
removal) by condensation and removal of water droplets. The condenser
transfers 320 kW from the off gas into glycol flowing at 1500 L/min. The mist
eliminator, a van separator, removes droplets 12 urn and larger. Both the
condenser and mist eliminator are rated at 104 standard rn 3 /min at the outlet.
After the condenser/mist eliminator, the gases flow to the heater and carbon
filters. To prevent moisture condensation in the carbon filters, the off gas
is reheated by 25° C to 770 C in the heater. The heater is capable of raising
the off—gas temperature by a maximum of 50° C at 104 standard m 3 /min. Final
decontamination of off—gas is achieved in the two—staye carbon filter
assemblies. The first stage is composed of two parallel filters and one set
of secondary filters. T 0 avoid breakthrough of organic contaminants, off—gas
is continuously monitored as it exits the primary and secondary filters. A
valve between the primary filters allows for an immediate change over, if
needed.
The gaseous emissions are drawn through the off-gas system components by an
induced draft system. The driving force is provided by a blower capable of
achieving 104 standard m 3 /min at 90° C and —90 inches of water. A backup
blower rated at one-quarter the capacity is also provided in the event of
failure of the primary blower. The backup blower is intended to maintain a
negative pressure on the off—gas hood to prevent direct release of emissions
until the process can be safely shut down. The backup blower is automatically
activated by the process control system when the header vacuum is reduced
below a preset limit. After the blower system, the off gases are exhausted to
the stack, which is monitored continuously. The stack is removable and
extends high enough (3 meters) to prevent interference with the off gas and
control trailer’s heating, ventilating, and air conditioning (HVAC) systems.
109
-------
The advantages of ISV are:
o There is a significant volume reduction.
o Organics are destroyed with a demonstrated DRE greater than 99.9999%.
o Inorganics are permanently immobilized.
o Long term durability of the vitrified material requires little or no
long tern site monitoring.
o Materials handling is r, inimized. There is no need to crush or
separate components of the soil.
o Scrap metal up to 5% by weight and/or 70% of the linear distance
between electrodes can be processed with the soil; therefore,
separating metal from the soil is not required.
o There are no hazardous materials remaining for disposal.
The disadvantages of ISV are:
o The center of the vitrified mass takes approximately one year to reach
ambient temperature.
o Cost is affected by electric power rates and soil moisture content.
Portable generators are typically used if power costs exceed 8.25
cents per kilowatt hour. -
To vitrify a cell of 20’ X 20’ X 15’ (approx. 222 yd 3 ), would take
approximately 75 hours. Based on this configuration and the time to vitrify,
it would take approximately 45 days to treat an estimated 2,001) yd 3 of soil
The unit cost for ISV was obtained from Geosafe. The unit cost ranges from an
estimated $250 to $350 per cubic yard.
Metal scrap in the soil is not expected to interfere with the process,
therefore, removal -is not necessary. As a result, extensive material handling
is not required. Since this process has not been applied on any large scale,
maintenance requirements are not known. However, based on the operation, it
can be expected that no significant maintenance may be required.
iSV is a patented process which was developed by Battelle Pacific Northwest
Laboratory (PNL) for the U.S. Department of Energy as an in—place
stabilization process for treating radioactive wastes. A license to use the
process must be obtained from Battelle. The license is based on site—specific
issues such as waste type and region. It restricts the use of the process
outside of the established parameters. Lead time to obtain the license is
minimal
Vitrified soil blocks were analyzed to determine their chemical durability. A
series of leachate tests showed a leach rate for all elements to be less than
1 x i — grams per cubic centimeter per day. This rate is comparable to Pyrex
and much less than rates for marble or bottle glass.
110
-------
Another indication of the durability of the vitrified soil blocks is a
weathering study of obsidian. This is a material which is similar both
physically and chemically to the vitrified mass. In the natural environment,
the hydration rate of obsidian is approximately 10 square micrometers per
1,000 years. Based on this rate, the ISV block can be expected to maintain
its integrity for approximately 10,000 years. Due to the ORE of the process
and the durability of the waste form, the vitrified mass can be left on site
and covered with topsoil.
RE SULTS
Geosafe Corporation, in conjunction with Battelle Pacific Northwest, conducted
a treatability test on the sofis at the Parsons/EN site. The treatability
test was financed by the Michigan Department of Natural Resources. Both
contaminated and noncontarninated soils were shipped to Geosafe to conduct the
test.
Preliminary results of the treatability test have revealed destruction
efficiencies (DE) for dioxins and furans ranging from 76.3% to 99.9%. DEs for
pesticides range from 99.98% to 99.998%. When calculated with the efficiency
of the off-gas treatment system, the demonstrated removal efficiency (ORE) is
99.9763% to 99.9999% OREs for dioxins and furans and 99.99998% to 99,999998%
for pesticides. The calculation of the DE is limited by analytical procedures
and detection limits achievable for suspect compounds. Therefore, the DRE for
Hx Furan (HxCDF) may be 99.9763, but actual efficiency may be greater. The
calculation is limited by the analytical detection limit for HxCDF.
A ORE for mercury was not calculated. Mass balance calculations only
accounted for 5.4% of the original mercury concentration. The assumptions
made at this point are that mercury is bound up or encapsulated into the glass
block. Any remaining volatile fractions of the mercury is assumed to “plate
out” in the off—gas system piping. The mercury combines with the nickel
contained in the stainless steel piping and coats or plates the inside of the
piping.
Total cost or unit prices for the ISV technology can not yet be calculated due
to the fact that no full scale application of this type has been completed.
Cost estimates have been stated in previous sections.
CONCLUSIONS
The preliminary results of the treatability study have indicated excellent
applicability of the ISV technology for rBnediation of the Parsons/ETM site.
An overall DRE has been calculated at approximately 99.9763% efficiency for a
worst case situation. Application of the technology will result in
approximately 30% volume reduction, destruction of organic contaminants and
destruction or encapsulation of inorganic contaminants. The glass monolith
generated by the process should result in a delisting of the site from the
National Priorities List (NPL). Stability of the block is comparable to
unreinforced concrete (stability should be able to be measured on a geologic
time scale). To date, the proposed application of the ISV techno1ogy at the
Parsons/ETM site has received positive community support.
111
-------
Major elements for consideration of applicability of the ISV Technology are:
moisture content, waste type, soil type, waste configuration, and volume to be
treated.
The moisture content of the material dictates, to a large degree, the cost and
efficiency of the ISV technology. Water contained in the treated material is
the first component to be driven off, in the form of steam. The greater the
water content the greater the amount of energy that is needed to drive the
process. As a result, the cost per ton for treatment will increase.
The ISV technology can handle a large number of waste stream types. Most
importantly, ISV can handle mixed waste streams. Complex waste streams with
both organic and inorganic contaminants can be effectively treated. In
addition, with slight modification, the ISV unit designed for hazardous waste
can treat mixed waste streams that include low level radio nucliatides.
Soil or material characteristics are important for estimating subsidence
(volume reduction). Sandy or loosely packed soils will subside much more than
tightly packed clay soils. The subsidence or volume reduction is directly
related to the amount of moisture and interstitial void space contained in the
material to be treated.
There are several possible waste arrangements for ISV processing. There is
the classical in situ processing arrangement where little to no site
preparation is needed. Such applications would apply to landfills or general
soil contamination. Waste material may be staged below grade or partially
above grade if desired. The material is located in a trench for treatment.
Use of staging options should be considered when: 1) contamination is
shallow, 2) contamination can not be treated where it is presently located, 3)
it is necessary to ada soil to or above the contamination, and/or 4) to
minimize the area of treatment by consolidation.
Special staging applications of the ISV technology may include treatment of
containers such as drums and even underground tanks. ISV may even be
considered in some cases as a containment technology. This would be achieved
by constructing a barrier wall surrounding a waste impoundment or to produce
an impermeable layer (cap) over contaminated soils.
The volume of material to be treated via ISV must be large enough to offset
certain capital costs. The cost of mobilization/demobilization should be
offset by the volume of material to be treated. Utility hookups and usage
cost should also be considered when evaluating the applicability of the ISV
technology. The more settings or applications required for waste treatment,
the more economical ISV becomes. Volumes of less than 1,500 cubic yards
become uneconomical for ISV treatment.
It is always recomended to conduct a treatability test on the proposed
material for treatment. The test can answer any questions for technology
applicability, efficiency for waste removal, physical or operational settings
needed for treatment and project cost. The cost for a treatability test to be
perfomed is approximately $40-550,000.
112
-------
COIITACTS FOR MORE INFORMATION
Edward C. Burk, Jr. (313) 675—3146
On-Scene Coordinator
U.S. Environmental Protection Agency
Response Section I
9311 Groh Road
Grosse lie, Michigan 48138
James E. Hansen (206) 822-4000
Geosafe Corporation
Kirkland Parkplace
303 Parkplace, Suite 126
Kirkland, Washington 98033
Robert J. Bowden (312) 353—9295
Chief, EERB 5HS—11
U.S. Environmental Protection Agency
230 South Dearborn
Chicago, Illinois 60604
Vincent F. Fitzpatrick (206) 822—4000
Geosafe Corporation
Kirkland Parkplace
303 Parkplace, Suite 126
Kirkland, Washington 98033
REFERENCES
1. Geosafe Corporation. “Application and Evaluation Considerations for In
Situ Vitrification Technology: A Treatment Process for Destruction and/or
Permanent Immobilization of Hazardous Materials.” April 1989. GSC 1901.
2. Edward C. Burk, Jr. “Engineering Evaluation and Cost Analysis for the
Parsons Chanical/ETM Enterprises Site, Oneida Township, Michigan.” U.S.
Environmental Protection Agency, Grosse Ile, Michigan. April 1989.
3. Edward C. Burk, Jr. “Action Memorandum — Request for Removal Action at the
Parsons Chemical/EN Enterprises Site, Oneida Township, Michigan.” U.S.
Environmental Protection Agency, Grosse lie, Michigan. September 1989.
113
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTER! ZAT ION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On—/Off—site Treatment Location:
Pre- and Post—treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Esther Soczo
M.Sc.
RI Vt4
P.O. Box 1, 3720 BA Blithoven
The Netherlands
(31) 30-74—30—65
Soil treatment by extraction -
In—situ
Electro-Reclamati on
Geoki neti Cs
Comercial
Heavy metals
Soils — mci. clay and river sludges
In—situ/on—site, off—site
Former timber finishing plant,
Loppersum, province of Groningen,
The Netherlands
Arsenic
Ave. of 10 pm arsenic limits
$45,000 for 450 tons of soil
Dr. Relnout Lageman
Geokinetics
Poortweg 4
2612 PA Delft
The Netherlands
(31) 15—61-06-37
11/89
4-a-3
115
-------
NATO/CCMS Pilot Study
De nstration of Ren dia1 Action Technologies
for Contaminated Land and Gro dwater
Montreal, Canada
6 - 9 November 1989
Electro—Recla ation : State—of—the—Art
drs. Reinout Lag iwt
Ceolcinet I
Deift, Cronin.gen
the Netherlands
co—author3
drs. W.Poo].
drz. C.A.Seffinga
Ceoktheticz
116
-------
Electro—Reclamat ion : State—of-the—Art
Electroki net i cal phenomena
During the last 4 years Geokinetics has been developing a method to
remove heavy metals and other contaminants from soil and groundwater.
The method is based on electrokinetical phenomena, which in one way or
another have been made use of since the end of last century. These
pher mena occur when the soil is electrically charged with direct
current by means of one or several electrode arrays
1. Electro—osmosis : Movement of soil moisture or groundwater from the
anode to the Icathode.
2. Electrophoresis : Movement of soil particles within the soil
moisture or groundwater.
3. Electrolysis : Movement of ions and ioncomplexes within the soil
moisture or groundwater.
Electro-osmos is
the electro—osmotic transport depends on the following factors
- the mobility of the ions and charged particles within the soil
moisture or groundwater;
— the hydratation of the Ions and the charged particles;
— the charge and direction of the ions and charged particles, which
cause a net water movement;
— the ion concentration;
- the viscosity of the pore solution, depending on the capillary
size;
- the dielectrical constant, depending on the amount of organic and
Inorganic particles in the pore solution;
— the temperature.
From existing literature and own experiments the average electro—osmo-
tic mobility has been calculated to be In the order of 5.10-’ ms/U.s,
where U = potential drop (V).
To drain 1 m 3 of soil by electro—osmosis, the following parameters
should be ) own
— the porosity;
— the moisture content of the soil to be treated;
- the conductivity of the pore solution;
Apart from these, other factors like the desired time period, the use
of the soil after treatment and safety requirements regards maximum
voltage and current are also of importance.
Electrophores Is
Electrophoresis (c.ataphoresls) involves the movement of particles under
the influence of an electrical field. This definition Includes all
electrically charged particles like colloids, clay particles floating
in the pore solution, organic particles, droplets etc.
117
-------
—‘..
•contaminated
sail
-
I.
j. .
I.:
.1
Y. : - ,. :.
4
a — a ..
_______ circulation system
current supply
— — e - boundary of etectrokineticat treatment
Fig 1 Schematic representation of ER—fie(d unit and
electrokineticat transport in the soil
118
+
AC/DC
CERTER G
— GOCATOR
OR MA I l
-------
The mobility of these particles corresponds with that of ions. Within
the pore solution these particles transfer the electrical charges and
affect the electrical conductivity and the electro—osmotic current.
Clay minerals as such have .2 electrical polarity possibilities. One
consists of the stucture based dipole moment, which depends on the
atomic masses and has an orientation parallel to the longest axis of
the clay particle. The second polarity stand.s at right angles to the
first and is caused by the external electrical field. It depends on the
way of polarization of the electrical double layer. The mobility of
clay particles is an interplay between these two nioments and is, ther-
fore, less than the electro-osmotic mobility. It varies between
1.10 ° and, 3.10-’ m 2 /U.s.
Electrolysis
In the case of electro-osmosis and electrophoresim one considers only
water transport or particle transport respectively; with electrolysis
only the movement of Ions and ioncomplexes is taken into consideration.
The average mobility of ions lies around 5.10’ m 2 /tI.s, which is ten
times greater than that of the electro—osmotic mobility. Therefore, the
energy necessary to move all ions over an average distance of 1 m
through a croessectional area of 1 m 2 of soil is ten times less than
with electro—osmosis.
To calculate the energy necessary to dispose of the contaminants within
1 m 3 of soil, the following factors are of importance
— chemical form of the contaminants;
— concentration of the contaminants;
— required concentrations of the contaminants;
— behaviour of the contaminants at different pH levels;
— pH control around the electrodes within the soil;
— removal of the contaminants and particles at the respective elec-
trodes;
— supply of a conditioning solution to replace the removed contami-
nants and other particles at the electrodes;
— processing of the contaminated solution removed at the electrodes.
The application of electrokinetical phenomena In practice
The effectiveness of electro—reclamation is largely determined by the
chemical composition of soil and groundwater. In many soils for exam-
ple it depends mostly on the kind of clay minerals and the . calci and
magnesli.sn bearing minerals like carbonate (e.g. lime) and sulphate
(e.g. gyps ).
Another important element is iron, whose concentration in groundwater
depends a.o. upon pH.
Generally speaking, the concentrations of the different metal Ions
depend on the C0 3 2 content and pH, while the following chemical equi-
libria are of importance
Me- (Clay—mineral) Me + Clay—mineral”
Me(OH). 1 Me + n(OH)
Me (CO 3 ). n(Me) + m(COs)
HC0 3 H + C0 3 ’
119 -
-------
A) Remediation of residential areas
B) Remediation of industrial areas
C) Remediation/fencing of hazardous waste sites
0) Preventive electrokinetical fence around potentially hazardous industrial complexes
Fig. 2 : Some applications of in situ ELectro ..Reclamatjon
;•$• ‘1
:. . . .. ..• . :.
C 0
120
-------
The element Me can be any metal element like sodium, potassium, calci—
um, magnesium, iron etc , but also heavy metals like copper, nickel,
lead, chromium, cadmium etc.
The concentration of the different metal loris In the groundwater depend
on the solubility-products of the hydoxides and carbonates of the
metals concerned and the pH (H concentration) of the solution. At
lower pH levels (higher H concentrations) metal concentrations will
increase.
When groundwater Is contaminated with salts of heavy metals like lead,
copper, nickel, chromium, zinc etc., the metal ions will influence the
original chemical equilibrium. At the £lewly formed chemical equilibria
part of the heavy metal Ions will exchange with the original metal .lons
in the mineral phase (like carbonate, hydroxide, clay mineral), whereby
one heavy metal Ion will exchange more easily with the original metal
ions than the other. The mobility or the displacement of a heavy metal
ion In the groundwater or the soil depends on its exchange capacity
with the original exchangable ions.
When soil Is electrically charged, the following processes can a.o. be
observed
At or near the anode
— The positively charged particles move into the direction of the
negatively charged cathode. As a consequence the concentration of
the metal Ions in the liquid phase (soil moizt e or groundwater)
will decrease. The decrease in concentration of the displaced
ions In the gro*mdwater will be restored by exchange with the
solid phase (mineral phase). This ion displacement and ion-exchan-
ge will continue as long as the electrical field is maintained.
The final concentration of a certain heavy metal thus depend.s on
the initial concentrations in the liquid and solid phase, the
electrokinetic mobility and the mutual exchange capacity with the
other metal Ions.
- At the anode moreover, H Ions are being formed through electroly-
sis of water. These positively charged Ions move via soil moisture
or groundwater into the direction of the cathode. As the H ions
exchange rather easily with the (heavy) metal ions of the mineral
phase and lower pH of the groundwater, the concentrations of
(heavy) metals will increase, thus accelerating the processes of
exchange and displacement.
At or near the cathode
— At a certain point near the cathode, the (heavy) metal ions move
Into the direction of the cathode, but total concentration of the
(heavy) metal Ions will stay the same, because an equal amount of
ions Is being supplied from the point situated at the anode side.
Total concentration of (heavy) metals will decrease when supply
from the anode side Is less than transport to the cathode side.
121
-------
1000
800
600
00
200
0
160
time (hours)
Fig 3a Decrease of Copper during e(ectrokinetical treatment
of contaminated pottery clay
C
0
I-
C
•1
1
C
0
•0
250
200
100
50
time (days)
Fig. 3b : Decrease of Cadmium during electrokineticat treatment
of contaminated fine argillaceous sand
E
0
0.
C
0
I-
C
w
1
C
0
I . ’
‘ -I
0 20 40 60 80 100 120 160
150
0 5 10 15
122
-------
The efficiency of electro-reclamatlon is less for soils with a high
cation or metal ion exchange capacity and an acid buffering capacity
(like marie). If feasible the efficiency of in situ remediation can be
increased by irrigating the soil with slightly acidified (pH : 3-4)
water
When treating the soil on or off site, the soil material can be mixed
with a slightly acidified solution after it is being excavated. The
amount of acid to be added depends mainly on the cation exchange capa-
city of the soil.
Electrolcinetical ln.stallation
The core of an electro)cinetlcal installation (fIg. 1) consIsts of the
electrode series and their housing. These can be installed in principle
at any depth, either vertically or horizontally. The cathode and anode
housings are interconnected but form two seperate circulation systems
(one for the cathode, one for the anode), filled with different chemi-
cal solutions. In these solutions the contaminants are captured and
brought to a connected treatment system, installed in a container
together with the solution tanks and measuring and monitoring devices.
The energy Is supplied by a generating set or taken from the main.
As explained before, electro-reclamatlon can be applied both for In
situ remediation of contaminated soil (fig. 2) and for on or off site
remedlation of excavated polluted soil or river andlor industrial and
sewage sludges. For the latter application a (se.mi)permanent installa-
tion would be most effective.
Laboratory experiments
The method of electro—reclaintion has been tested on the basis of ntme-
rous laboratory experiments. They focussed on important parameters like
kind of current, strength of current, voltage, moisture content, chemi-
cal additives and the like. Besides, the effectiviness of the method as
regards certain soil and heavy metal types has been examined with the
help of several simulation experiments (clay, peat, fine argillaceous
sa.rid polluted with As, Cd, Co, Cr, Cu, Hg, Mn, Mo, Ni, Pb, Sb Zn).
These experiments also provided good insight into the energy demand
and the time duration. Some results are presented in table 1 and figu-
res 3a and 3b.
123
-------
4a. ______
I Iii
II
Cu >500 ppm
•
O 505000 ppm
600
-------
Soil type Metal Conc. before Conc. after Decrease
(ppm) (ppm) (per cent)
Peat Pb 9000 2400 73
Cu 500 200 60
Pottery clay Cu 1000 100 90
Fine argil— Cd 273 40 85
laceous sand
Clay Az 300 30 89
Fine argil— Cd 319 < 1. 99
laceous sand Cr 221 20 91
Ni 227 34 85
Pb 638 230 64
Hg 334 110 67
Cu 570 50 91
Zn 937 180 81
Average : 83
River sludge Cd 10 5 50
Cu 143 41 71
Pb 173 80 54
NI 56 5 91
Zn 901 .54 94
Cr 72 26 64
Hg 0.5 0.2 60
As 13 4.4
Average : 69
Table 1 : Some results of laboratory electro-reclamatlon.
Field ex periments
Site 1
The first field experiment took place alongside part of a waterbearing
ditch, on one side bordered by a former paint factory and on the other
side by open grassland. The bank on the latter part was heightened by
sediment dredged from the ditch. This sediment was heavily polluted
with metals in the form of paint residuary. The raised sediment layer,
height 20 - 50 cm, length 70 m and width 3 m, contained Pb and Cu
concentrations up to 10,000 ppm and 5,000 ppm respectively. The origi-
nal peat soil underneath was contaminated by leaching of this overlying
layer with Pb concentrations ranging from 300 ppm to more than 5,000
ppm, while Cu concentrations were in the order of 500 to 1,000 ppm.
A preceding electrokinetic laboratory test with a sample of the sedi-
ment reduced the concentration of Pb from — 9,000 ppm to — 5,000 ppm
and that of Cu from 4,500 ppm to 1,600 ppm, all within a time
period of 320 hours.
For the field experiment one cathode and one anode array were installed
both with a length of 70 m and a mutual distance of 3 m.
125
-------
The cathode was installed horizontally, while the anodes were implaced
vertically into the soil about 2 m apart. On the basis of the energy
cons nption during the laboratory test the field experiment was conf i—
ned to 430 hours.
The changes in Pb and Cu concentrations were monitored at 26 sampling
locations, sampled at regular Intervals at several depths (10,20,30,40
and 50 cm below ound surface). The following table lists part of the
results for Cu and Pb within the peat at a depth of 30 to 40 cm below
round surface. The spatial distribution of the pollutants at the
beginning and at the end of the test are shown In figs. 4a and 4b).
Sample point Metal Conc. before Conc. after Decrease
(30-40 cm) (ppm) (ppm) (per cent
1 Pb - 440 110 75
Cu 185 35 81
2 Pb 3900 700 82
Cu 540 220 59
3 Pb >5000 560 89
Cu 1150 580 50
4 Pb ) 5000 2450 51
Cu 475 250 47
5 Pb 5000 610 88
Cu 1170 230 80
6 Pb >5000 300 94
Cu 580 45 92
7 Pb 3780 285 92
Cu 410 30 93
8 Pb 380 160 53
Cu 35 15 57
9 Pb 340 90 74
Cu 50 15 22
Average : 74
Table 2. Electro—reclamation, field results, site 1
Site 2
The second field experiment was carried out on the site of a galvani—
zin,g plant. According to preceding Investigations the soil (sandy clay)
around the plant was contaminated with Zn to a depth of 40 cm below
groundsurface. In the upper 10 cm Zn concentrations were reported to
have a maxim of 3,000 ppm. At greater depths Zn concentrations were
indicated as being In the order of 500 ppm.
For the experiment an area was selected with dimensions of 15 m x 6 m x
1 m. Two cathode drains were installed at a depth of 50 cm below groun-
dsurf ace, whIle 33 anodes, divided along 3 rows were implaced in holes
of 1 m depth with a mutual distance of 1.5 rn. The distance between the
cathode and anode series was also 1.3 m.
Energy was supplied by a 100 ICVA generating set. The resistivity of the
soil was 5 ( . The installation was calculated for a DC supply of 8
Amps/rn 2 of soil, which should result in a potential drop of 40 V/rn.
This potential drop could not be maintained during the whole period.
As a result of some material problems it was neither possible to main-
tain a 24 hour energy supply to the soil.
126
-------
Within 2 weeks temperature rose from 12 °C to an average of 40 °C. As a
consequence soil resistivity decreased to 2.5 n and the potential drop
to 20 V/m with an average strength of current of 8 Amps/in 2 . The effec-
tive ener ’supp1y per 1000 kg of soil amounted to 160 kWh during the 8
week period.
Changes in Zn concentration were monitored at 12 samplIng locations,
which were sampled at 3 different depth intervals (10, 30 and 50 cm).
Changes in oundwater concentration were monitored in 2 observation
wells. In table 5 and fig. 5 the results are given for the depth inter-
val of 30 cm.
Sample point Metal Conc. before Conc. after decrease
(30 cm detpth) (ppm) (ppm) (per cent)
1 —, 5120 4470 13
2 2030 1960 3
3 1600 800 50
4 2320 2320 0
3 2450 2450 0
6 > Zn 4390 2360 48
7 I 1960 940 52
8 I 3250 1960 40
9 2400 2000 17
10 I 70 30 57
11 150 120
—3 Average : 20
Table 3 : Electro—reclamation, field results site 2
The ener ’ demand for this test amounted to 160 kWh/ton. At the begin-
ning of the test the highest Zn concentration amounted to 7,010 ppm
with an average of 2,410 ppm over the whole area. At the end of the
test the highest Zn concentration was 5,300 ppm and the average had
been decreased to 1,620. The concentrations of Zn, Pb and Cd in the
&ound ater and the filtercake are presented in the tables 6 to 8.
A total of some 1000 kg of filtercake was produced with an average Zn
content of 117 g/kg. This comes to a total removal of some 50 kg of
zinc, ass ing an average moisture content of the filtercake of 60 %
A rough mass balans can be sim rized as follows :
— treated vol e of soil : 13 x 6 x 0.5 x 3/4 = 34 m 3 (1/4 of the
area did not show increased Zn concentrations).
— weight : 34 in 3 x 1.8 = 61 tons.
— weight of filtercalce : 1000 kg.
— average moisture content : 60 Z
— total dry matter : 400 kg.
— average Zn concentration : 117 g/kg.
— amount of zinc removed : 47 kg
— removed per 1000 kg of soil : 47 x 10’/61 x 1O = 770 ppm.
The last value is in the same order of ma itude as the average decrea-
se in Zn concentration (2410 ppm —1620 ppm — 790 ppm).
127
-------
An important outc.o of the test was the relatively high ener ’
consi=ption against a rather low Zn— bility. This was the result of
the high buffering capacity of the soil, caused by the presence of NH
and NH 6 C1 (as was establi& ed later), which is used d zin.g the
galvanizing process. Dt irig a following laboratory test with a soil
sample from the area it was found that the ener ’ necessary to reduce
Zn—concentrations below the 200 ppm level would amount to 500 kWh/ton
of soil. With art unchanged power-supily of 100 WA this would an
tripling the ti period of B weeks to 24 week. .
• 4
.::.: •.•.•:•:.•.•.
•:•:•:•:•::•:•:: : ..
o cathodes
• anodes
( 4 •
0
H
HE
Zn—concentrations 30 cm below g.s.
(24/10/88)
Zn> 4000 ppm
Zn—concentrations 30 cm below g.s.
(16/12/88)
2000
-------
Metal : Zn Pb Cd Zn Pb Cd
sample treatment : not acidified acidified
date
24—10—88
01—11—88
09—11—88
17—11—88
25—11—88
30—11—88
obs. well,
ppm
1
1
1
1
1
1
200 0.09 0.06 270 1.4 0.07
120 0.07 0.00 140 0.09 0
130 0.06 0.02 160 .0.17 0.02
172 0.07 0.03 198 0.11 0.04
130 0.17 0.03 180 0.22 0.03
120 0.13 0.03 150 0.16 0.03
24—10—88
01—11—88
09—11—88
17—11—88
25—11—88
30—11—88
2
2
2
2
2
2
10 0.17 0.02 LeO 0.34 0.02
1.5 0.09 0.09 2 0.15 0
2.5 0.06 0 3 0.15 0
6 0.03 0 8 0.07 0.01
2.8 0.03 0.01 5.8 0.18 0.01
4 0.09 0 5.5 0.14 0
Table 6. Zn-content (ppm) of the oundwater in obser-
vation wells 1 and 2.
Metal. : Zn Pb Cd moisture cnt
date
g/kg In %
24—10—88
09—11—88
17—11—88
25—11—88
30—11—88
136.9 1.9 0.34
199 1.1 0.18
99 2 0.12
89 1.5 0.16 78
61 0.58 0.11
Table 7. Zn—content (g/)cg) of the filtercake during
Electro-Reclamat ion.
Metal : Zn Pb Cd
date ppm
30—11—88 30 0.6 0
Table 8. Zn—content (ppm) of the solution in
the anode—circulation system.
129
-------
Re dia1 action
The first ‘official’ electro—reclamation project started at the end of
January 1989. It involved a site of a former timber impreg ation plant,
containing arsenic levels up tO 400 — 500 ppm. After a fire in 1984,
which destroyed a large part of the plant, it was decided not to re-
build the plant. After dismantling the same a ‘statement of unpolluted
soil’ was needed in order to allocate the land to building plots. A
following investigation established the presence of As concentrations
up to several 100 ppm in part of the heavy clay soil to a maxim n depth
of 2 m. Cause of the pollution was attributed to Superwolmansalt
D’(Na 2 HAsOs.7Hz0), used for impre iation.
In April 1988 Ceokinetics was requested to investigate the possibility
of remed.iatlng the soil by Electra—Reclamation. A following laboratory
test with a soil sample reduced the As concentration of 300 ppm to 30
ppm against an ener ’ cons ption of 115 kWh/ton. An additional field
investigation delineated the pollution to an area of 10 m x 10 m,
contaminated to a depth of 2 rn and an adjoining area of 10 m x 5 rn,
contaminated to a depth of 1 m. Total voli .m e of polluted soil : Z50 m’
(= 450 tons).
The project started In January 1989. Along the length of the polluted
area 4 x 2 cathode drains were installed : one at a depth of 1.5 m and
the other at 0.5 m. The cathode arrays had a mutual distance of 3 m.
In between 36 anodes were implaced in the soil, divided along 2 rows of
14 and 1 row of 8 pieces. Within the area of 10 m x 10 m the anodes
were installed to a depth of 2 m below ground surface. In the other
area of 10 m x 5 m the depth of the anodes was limited to 1 a depth.
All anodes were placed at a mutual distance of 1.5 m.
On the basis of both the laboratory test and the field investigation
the duration of the Electro-Reclamation period was calculated to last
50 (24 hour) days, using an energy supply of 200 kVA (= 44 kW effective
into the soil.
At the beginning the resistivity of the clay was 10 C and soil tempe-
rature at a depth of 0.5 a was 7 °C. After 3 to 4 weeks temperature had
risen to an average of 50 ‘C, while the resistivity decreased to 5 n.
The original potential drop of 40 V/rn decreased accordingly to 20 V/m
with an average current strength of 4 Amps/rn 2 (total crosssectiona].
area being 110 a’).
Changes in As concentrations were monitored at 10 fixed sampling loca-
tions and n .m erous randomly distributed sampling points. Of the fixed
locations, 2 were sampled at 0, 0.5, 1, 1.5 and. 2 rn depth. The others
at 0, 0.5 and 1 m depth. The analysis results from samples taken at the
beginning and on 30 April at a depth of 1 a below ground surface are
listed In table 9 and fig. 6.
When starting the project, the average As concentration over the whole
area amounted to 1.15 ppm, which cou s to a total As content of an ample
50 kg.
During the rernediation process It was observed that at one particular
area the decrease in As concentration proceeded much more slowly than
at other locations. After April 30” there was almost no reduction ob-
served anymore.
130
-------
• . ( :.•
• : :. •. •.
: #\J.
• . . • • .. . .
E
:•. :::::::...::::::
.::::ó:::: ::::::•..: :
••.. •. :: : :: ::: • : :ø•:
15
As—concentrations 1 m below g.s.
(24 /1/89)
: •:‘ : •r : •:
.•. . . ..
a.
a....
.
: : : •u •: •:
As—concentrations 1 m below g.s.
(28/4/89)
As>250 ppm
100 As 50 <20 > 150
6 75 30
7 f 40 <20
8 I 175 <20
9 40 <20
10 .-.—J 60 <20
131
-------
A total of some 40 n of soil had to be excavated. Periodical treatment
of the electrode solutions resulted in some 800 kg of filtrate. A rough
mass balans can be stnmv trized as follows :
— treated volLmle of soil
— weight of soil
- average As concentration
- amount of As before remediation
- weight of excavated soil
- average As concentration of excavated soil
- amount of As in excavated soil
— amount of filtrate
- average moisture content
— total dry matter
- total amount of As removed by electro-reciamation
- total amount of As removed after remediation
— amount of As remaining in the soil
- average As concentration of remediated soil
250m 3
450 tons
115 ppm
52 kg
: 71 tons
200 ppm
14 kg
800 kg
70 %
240 kg
34 kg
48 kg
4 kg
10 ppm
Other a licat ions I fut e develo ents
Electrokinetical fencing
The electro}cinetic.al phenomena occurring when the soil is electrically
charged, can also be used for fencing purposes. These so—called elec—
trokinetic fences can be installed either at refuse sites/factory
complexes, where soil pollution has already been ascertained, or where
soil pollution is likely to occur. Depending on the local geohydrologi—
cal situation and the character of the soil, the elctrode configuration
can be such that
- the elctrokinetical transport is directed towards the source of
the pollution (fig. 7a). The cathode series Is situated nearest to
the source of pollution. Such a set-up should be applied in less
permable soils without substantial groundwaterf low C < 1 m/year).
- the contaminants, which are carried along with the groundwater
flow are diverted, collected around the electrodes and periodical-
ly removed. In this case the cathode series is farthest away from
the source of pollution and cathode and anode series are installed
perpendicular to the direction of groundwater flow (fig. 7b). Such
a set-up should be applied when the soil and/or subsoil is relati-
vely permeable (groundwaterf low velocity > 1 m/year).
Desalination of arabic land
Salination of arable land is a conon problem in those countries, where
precipitation is generally low and evapotranspiration high. In combina-
tion with relatively high groundwater levels (coastal areas and river
lowlands), soils of low permeability and irrigation water with high
total dissolved solids, the acc ulation of salts in the top layers
prohibits further agriculture.
132
-------
source of contamination
• .
o o
150 m
anode—series (vertical)
cathooe—serues (vertical)
flow paths of positively charged contaminants
Fig 7a Set—up of electrok,ne?ical fence in soil with low permeability
o)
• .
o 0
anode—series (vertical)
cathode—series (vertical)
flow paths of positively charged contaminants
> direction of groundwater flow
N
a
+
/
S
.
IL
source of contamw ation
S
Fig. 7b : Set—up of electrokinetical fence in soil with moderate to high permeability
133
-------
The most common technique for landreclamation consists of the lowering
of the groundwatertable and/or drainage of the soil by means of wells
and/or horizontal drains. The soil is then frequently irrigated with
relatively fresh water, thus leaching the salts from the soil. However
the low permeability of the soil hampers more often than not the perco—
latioh of the leachate to the deeper layers.
By applying electrokinetical processes these problems can be overcome,
as clay and argillaceous soils are specifically suited for electro-
reclamation.
The As remediatlon project mentioned earlier showed, that after elec—
trokinetical treatment, the permeability of the heavy clay soil had
increased significantly. When furthermore gyps = is added at the anode
solution, the soil structure will be improved even more and higher crop
production will be obtained (Collopy, 1958).
Removal of organic contaminants
An R & D project Investigating the possibilities of removal of orga-
nic pollutants like PCAs and PCB’s from soils and sludges is currently
in preparation. The first research objective is to examine wether ex-
tension and/or adaptation of the electrical parameters will result in
the removal of organic contaminants.
The second research objective Is to examine the possibilities of combi-
ning electrokirietical techniques with biodegradation, more specifical-
ly
- can micro-organisms be distributed more evenly in the soil, or
can micro—organisme be added to the soil,
— can pH and soil temperature be maintained at preferential levels
(pH : 6 — 7, and T : 30 C),
— can addtional oxygen and nutrients be brought into the soil ?
Cost estimates
El ectro-Recal amat Ion
Fig. B shows a set of graphs depicting remedlatlon cost per ton of soil
as a function of remedial action time (fig. Ba) and as a function of
the measure of contamination (fig. 8b), ass ing a polluted area with
dimensions of 500 m x 100 m x 1 m. From the graphs It is evident that
short remediation periods and highly polluted soil (low resistivity)
require a high amount of energy, having the greatest effect on the
costs.
In practice, however, there Is a limit to the electrical current which
can be put Into the soil. For every specific case, therefore, an opti-
mum must be calaculated for energy supply and time duration.
134
-------
a.
b.
U,
C
0
I .-
0.
UI
UI
0
‘I
Assumptians:
— Area dimertsions 500 m x 100 m x 1 m
— Weugth of soiL: 90.000 ton
— Distance between C— and A—series: 10 m
— Mutual distance anodes: 10 m
remetha action time (days)
- Soil resisTivity 10 Ohmm
Fig 8 Cost estimate of electro—reclamarion
a costs per time—period
b. costs versus grade of pollution
El ectrokinet I cal fencing
365
U,
C
0
I-
0.
UI
UI
0
soil resistivity (Ohmm)
— Duration of ER—proces 60 days
100
In fig. 9 the energy costs per year are given as a function of groun-
dwater flow velocity (Zig. 9a) and as a function of the rate of pollu-
tion (fig. 9b), ass ing an electrolcinetical fence of 500 m length and
10 m depth.
In areas of low groundwater flow velocity (clay, argillaceous sand) and
low soil pollution, the yearly energy costs of an electrokinetical
fence are insignificant. This changes rather quickly, when the soil
becomes more permeable (sandy formations) and the groundwater flow
velocity increases together with the concentration of the contaminants.
For relatively high groundwater flow velocities a combination of hydro-
logical measures and electrokinetical techniques will render the most
economic results.
Desalination of arable land
Preliminary cost estimations amount to US $ 1000 to 2000 per ha.
0
10 30 60
150
50
0
5 10 20 30 40 50 75
135
-------
a.
150
50
0
groundwater flow velocity (rn/year)
Assumptions.
- Length of fence 500 m
- Depth of fence 10 rn
- Average ion mooiliry. 1.56 m 2 /V. year
— Soil resistivity. 50 Ohmm
b.
x
I , ,
E
l v
V I
a
VI
0
a’
I-
C
1 1
200
150
100
50
0
soil resistivity t0hmrn)
— Groundwater flow velocity: 10 rn/year
Fig 9 Cost estimate of e(ectrokinetucal fencing
a enerqy costs versus groundwater flow velocity
b energy costs versus grade of polLution
October 1989
Geolcineti cs
Poortweg 4
2612 PA DelZt
the Nether land3
300
250
200
a
a
a
VI
E
C
C
I V
4-
‘p
C.
VI
VI
0
‘p
C
SI
0 10 20 30 40
0 20 40 60 80 100 120
136
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTER I ZAT ION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On—/Off—site Treatment Location:
Pre- and Post-treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
David Lopez
U.S. EPA, 05-210
Emergency Response Division
401 M Street, SW
Washington, DC 20460
United States
(202) 382—2471
K-PEG process
Galson Research Corp.
Demonstrated pilot scale
Halogenated hydrocarbons, PCBs,
dioxin
Soil, liquids, sludge
On—site; mobile
Excavation of soils to chemical
treatment mixer
Wide Beach, Brant, New York,
United States
30,000 — 40,000 gallons waste oil
containing PCBs
20,000 Cu. yards contaminated
clay/silt soil; residential area
<2 ppm PCBs soil; <10 ppb PCBs
for liquid treatment
$300/cu. yard
Herbert King
U.S. EPA
26 Federal Plaza
New York, NY 10278
United States
(212) 264-1129
11/89
6-1
137
-------
Chemical On-Site Treatment Utilizing KPEG Process at Wide Beach, New York
Presented by
U.S. Environmental Protection Agency
November, 1989
138
-------
Executive Summary
From 1968-1978, PCB (Aroclor 1254) contaminated waste oil was used by the
Wide Beach Homeowners Association for dust control on local roads. During this time
period, approximately 30,000 to 40,000 gallons of the oil were applied. In 1980, a
sewer line was installed and the highly contaminated soil along the roadways and
surrounding areas was excavated. The surplus soil from the installation activities was
ultimately used as fill in several residential yards and a community recreation area.
In 1981, an odor complaint investigation conducted by the Erie County
Department of Environment and Planning (ECDEP) led to the discovery of 19 drums,
some containing PCB contaminated waste oil. Subsequent sampling revealed
widespread PCB contamination in the air, homes, private water supplies, drainage
ditches and soil. Sampling activities in 1983 confirmed the presence of PCBs in
groundwater and soils.
Six remedial alternatives were considered: no action; excavation, landfill, and
soil replacement; excavation, on-site treatment, and soil replacement; in-situ biological
treatment; in-situ chemical treatment; and immobilization. Of these alternatives, three
(no action, landfill and chemical treatment) were chosen for further review. The no-
action alternative was found to provide inadequate protection to public health and the
environment in addressing the threat from the contaminated soils and sediments at the
site. The excavation and landfill alternative was considered too costly and did not
provide a permanent solution; rather, it only served to relocate large quantities of PCB
contaminated soils.
Based upon the technology assessment data, on-site treatment using the KPEG
process was selected. The KPEG soil treatment demonstrated a high degree of
effectiveness in reducing the concentration of PCBs present in the soil at the Wide
Beach site. During the pilot operations, soils with initial PCB concentrations ranging
from 30 ppm to 260 ppm were reduced to less than 10 ppm. Final PCB levels ranged
from 0.7 ppm to 1.7 ppm.
In addition to its destruction efficiency, the KPEG process is a cost-effective and
permanent solution. The process involves treating the PCB waste with a solvent
consisting of an alkaline reagent and polyethylene glycol which produces an innocuous
ether and a potassium chloride salt as end products. The total capital cost for this
alternative is estimated to be $6,017,480 with no operation and maintenance costs.
139
-------
Introduction
A. Site Description
The Wide Beach site is located in a residential area in the Town of Brant,
approximately 48 kilometers south of Buffalo, New York. Wide Beach is bounded on
the south by wetlands and the Cattaraugus Indian Reservation, on the west by Lake
Erie, and on the east and north by residential and agricultural property.
Approximately 274 residences were affected in the Wide Beach community and
surrounding areas. The community accommodates an estimated 120 people in the
summer months and 45 people year-round. Approximately 39 people reside year-
round in the Snyder Beach area, south of Wide Beach. During the summer months,
Snyder Beach is also used by campers.
Wide Beach lies within the Erie-Niagara Basin in the Central Lowlands
Physiographic Province, characterized by flat terrain of low relief. The Erie-Niagara
Basin is underlain by a series of layered sedimentary rock of Paleozoic Age, striking
roughly eastward, and dipping gently to the south. The Paleozoic strata, formed of
fine-grained sediment deposited in a shallow sea which covered the area during the
Silurian and Devonian periods, are overlain but unconsolidated deposits of glacial
origin. The low relief of the area is the result of glacial scour and lacustrin deposits.
The site itself is virtually flat, sloping southward to the woodland bordering the east,
and then dropping sharply to the beach.
Surficial soil, a silty sand, is found throughout the site except in the wetlands
area. The thickness of this layer varies from 0.6 - 1.2 meters (m) thick. In some
locations, thin lenses of this soil alternate with layers of a brown silty clay (till). This till
is the next significant soil horizon and contains small rounded rock fragments. Its
consistency varies from stiff to very stiff.
The overburden at the site consists primarily of till and glacial deposits and is
approximately 3 m thick. The till is composed of dark gray and brown silty clay with
some rounded rock fragments. In several soil samples obtained at the site during the
remedial investigation, fractures were found with oxidation staining of the surfaces,
associated with the percolation of surface water through the overburden to the
bedrock.
Weathered bedrock is only a few centimeters thick at the site and is up to 1 m
thick in the eastern portion of the site. The bedrock is described as a black to gray
decomposing shale with interbedded light gray siltstone and sandstone, dipping in a
southerly direction. Throughout the bedrock, zones of calcareous concretions are
found which may also contain some pyrite and marcasite. A discontinuous fracture
zone with shallow tension cracks was found in the upper surface of the bedrock.
Groundwater can flow through these rock joints and fractures at several liters per
minute.
1
140
-------
Based upon the data gathered during the remedial investigation, the overlying till
acts as a confining layer, imparting on the bedrock aquifer a confined or semi-confined
condition. On this basis, the aquifer of concern is the east shallow bedrock aquifer,
including: the basal 0.3-0.6 m of coarser grained till; weathered bedrock and the zone
of shallow tension cracks; and the upper few meters of open joints and fractures.
Field observation, indicate that recharge to the on-site private wells occurs
predominantly through the weathered/fractured zone and open fracture in the shallow
bedrock.
An estimated overall average gradient for the site is 0.0009. The wetlands at the
south of the site appear to constitute a ground-water discharge divide between the site
and the land to the south of the wetlands. Based on contours mapped in December
1984, rough’y 80 percent of the site’s ground-water discharge is by stream and
wetlands, with the remaining 20 percent being discharged directly to Lake Erie.
B. Site History
From 1968-1978, PCB (Aroclor 1254) contaminated waste oil was used by the
Wide Beach Homeowners Association for dust control on local roads. During this time
period, approximately 30,000 to 40,000 gallons of the oil were applied. In 1980, a
sewer line was installed and the highly contaminated soil along the roadways and
surrounding areas was excavated. The surplus soil from the installation activities was
ultimately used as fill in several residential yards and a community recreation area.
In 1981, an odor complaint investigation conducted by the Erie County
Department of Environment and Planning (ECDEP) led to the discovery of 19 drums,
some containing PCB contaminated waste oil. Subsequent sampling revealed
widespread PCB contamination in the air, homes, private water supplies, drainage
ditches and soil. Sampling activities in 1983 confirmed the presence of PCBs in
groundwater and soils. Testing for the presence of dioxin was conducted; however, it
was not detected.
Until long term remedial actions could be implemented, an immediate removal
action was conducted in 1985 to protect the public. At this time, the roadways,
drainage areas, and driveways were paved to prevent public exposure to dust and
runoff. Rug shampooing, vacuuming and replacing air conditioner and furnace filters
were procedures used to decontaminate the homes. Particulate filters were installed to
protect individual private wells from sporadic PCB contamination. Potentially
responsible parties were identified and notified.
2
141
-------
Extent of Contamination
a. Soil
Soil contamination ranged from .18 to 390 milligrams/kilogram (mg/kg),
1.0 to 226 mg/kg and 0.2 to 1026 mg/kg for samples collected from residential
driveways, roadways and drainage ditches, respectively. PCB concentrations
exceeding 10 mg/kg were detected at depths of approximately 0.15 m in the
yards, 1 m in the drainage ditches adjacent to the contaminated roadways, 0.5
m in the roadways, 0.3 m in the driveways, and 0.2 m in the wetlands. About
20,000 cubic yards of soil were contaminated.
b. Surface Water
Before 1985, surface water from roadways and drainage areas were the
primary PCB transport mechanism off-site. These areas were covered in 1985,
but the wetlands located south of the site were already contaminated with high
concentrations of PCBs. PCB concentrations in this area ranged from
nondetectable to 126 mg/kg. PCB levels were higher in the top section of the
sediment cores collected. Significant concentrations were found in the
immediate vicinity of the storm drain outfalls.
c. Groundwater
Well sampling results indicated that 21 of the 60 residential wells, six
sewer trench wells and one monitoring well were contaminated. While the
residential wells had the lowest and most sporadic concentrations, the sewer
trench wells had the highest. Groundwater contamination may have occurred
by either PCB leaching from the surficial soils or by migration. Migration may
have occurred when contaminated surficial soils were redistributed during the
sewer installation or other excavation activities.
d. Dust/Air
Vacuum cleaner dust samples from 47 of the 60 residences showed PCB
levels ranging from 0.35 to 770 mg/kg. No apparent PCB distribution pattern
was observed at this site. Linear correlation analysis for vacuum cleaner dust,
yard soil, driveway soil, and roadway soils revealed no statistically significant
associations.
Ambient air particulate samples indicated PCB levels ranging from 0.04 to
0.30 mg/rn 3 . Based on these measurements and meteorological conditions, the
concentrations of airborne PCBs, both in vapor and sorbed phase, were
modeled for conditions prior to the roadway paving activities. For a worst case
scenario, the on-site concentration was 0.29 lhg/m 3 and the concentration at
Lotus Point was 6.3 x 10-3 I.Lg/m 3 . For an average case scenario, the
3
142
-------
Because the road surfaces are now paved, the ambient dust concentrations are
likely to be significantly less.
C. Groundwater and Surface Water Impact
a. Groundwater
Migration of PCBs to saturated groundwater is possible through
several pathways. After the PCB-contaminated oil was applied to the
ground surface, it is believed that the PCBs migrated through the
unsaturated zone toward the water table. PCB movement may have
initially been in the form of a plume, but as they migrate PCBs bind tightly
to surrounding soil particles. As water moves through the soil, PCBs are
adsorbed and desorbed by organic matter. This mechanism slows or
retards the movement of the PCBs.
Migration may have resulted frbm excavation activities, in 1980,
installation of the sewer system resulted in backfilling the sewer trench
excavation with PCB contaminated soils removed from the surface of the
trenches. The bedding material (Number 1 stone), low specific surface
area, and organic carbon content of the backfill will facilitate the
transportation of PCBs into the bedrock or off-site to Lotus Point. The
results of ground-water analysis of samples from the sewer trench wells
confirm the presence of PCB5, and reinforce the theory of high transport
potential in the sewer trench.
Another potential route may have been from dust. Since high
levels of PCBs have been detected in house dust, it is possible that PCB
contaminated soils and other materials may have been disposed of in the
now inactive septic systems. General house cleaning, as well as
laundering and bathing, may have facilitated PCB transport to the
groundwater by the septic leach field. Also, many drinking water wells
may not be properly grouted, possibly resulting in a rapid conduit from
the surface for surface water carrying PCB contaminated soil particles.
Compared to the other more obvious transport pathways, migration by
this pathway is difficult to quantify.
Further migration may have occurred by PCB solubilization in
water infiltrating the upper water table. The factors controlling
solubilization typically are the soil organic carbon, water partition
coefficient (4.25 x 10 4 for Aroclor 1254) and the soil organic content
(approximately 1.3 percent at Wide Beach).
Groundwater PCB concentrations derived from soils can be
estimated by calculating the equilibrium state of a soil/water mixture from
soil PCB concentrations, the partition coefficient, the soil organic content,
4
143
-------
and the soil/water content. Using typical and high soil PCB
concentrations of 50 and 500 Mg/I, respectively, the resulting ground-
water PCB concentration can be expected to range from 3.2 to 32 Mg/I at
equilibrium. Although it is very difficult to fix a precise value to the PCB
migration rate through the unsaturated soils onsite, it is possible to draw
the following conclusions:
1. The surficial soils will act as a long-term source of PCBs.
2. Migration by this route may have resulted in low level
ground-water contamination in the saturated zone.
3. PCBs have migrated downward through the vadose zone.
4. The potential exists for more significant ground-water
contamination by this route in the future.
Unlike the saturated and unstaurated zones, PCB transport in the
fractured bedrock is less likely to be retarded. This is attributed to the
lower surface area and the organic matter available for adsorption. In
bedrock, the PCB velocity is highly dependent on the nature of the
fracturing. Water velocities in the meter per day range are possible.
Under such conditions, transport of PCBs would be quite rapid.
Although there is the potential for some northerly movement
through the sewer trench bedding, groundwater migration from the Wide
Beach Development site appears unlikely. Based on the estimated area
of ground-water drainage from the site of 4 hectares (ha) into Lake Erie
and 19 ha into the stream/woodland system, a total average annual
ground-water discharge of 4 million liters (I) to Lake Erie and the
stream/woodland system was calculated. Assuming an average PCB
concentration of 50 mg/kg across the site, the maximum PCB ground-
water discharge would be approximately 0.014 kg/yr to Lake Erie and
0.058 kg/yr to the stream/woodland system, totaling 0.073 kg/yr.
b. Surface Water
Before the road paving, PCB loading to the stream/woodland
system and to Lake Erie, was based upon estimated runoff volumes and
PCB concentrations. PCB concentrations in runoff were calculated to be
19.34 micrograms/liter (Mg/I) for the particulated fraction and 0.8 Mg/I for
the dissolved fraction, totaling 20.20 g/l. For the 22 ha site, the total
drainage to the stream/woodland system was estimated to be 19 ha; 1.5
ha appears to drain off-site to the north.
5
144
-------
In a worst case scenario, assuming that all of the estimated storm
runoff of 62 centimeters/year (cm/yr), is surface flow, the average
surface water flow was estimated to be 7.2 x 10 to the lake and 8.5 x 10
liters/year to the stream/marsh; representing a maximum loading
potential of 0.14 kg and 1.7 kg of PCBs to the lake and stream/marsh
systems, respectively. Assuming that the total PCBs discharged to the
stream/marsh will reach Lake Erie, a loading of 1.8 kg/yr to Lake Erie
would be expected by stormwater runoff. An estimated 0.13 kg/yr of
PCBs would be transported offsite to the north in storm runoff.
0. Technology Selection
Six remedial alternatives were considered: no action; excavation, landfill, and
soil replacement; excavation, on-site treatment, and soil replacement; in-situ biological
treatment; in-situ chemical treatment; and immobilization. Of these alternatives, three
(no action, landfill and chemical treatment) were chosen for further review. The no-
action alternative was found to provide inadequate protection to public health and the
environment in addressing the threat from the contaminated soils and sediments at the
site. The excavation and landfill alternative was considered too costly and did not
provide a permanent solution; rather, it only served to relocate large quantities of PCB
contaminated soils.
Based upon the technology assessment data, on-site treatment using the KPEG
process was selected. It has been successfully used in the treatment of dioxin
contaminated liquid waste in EPA Regions X and VIII at the Western Processing and
Montana Pole Superfund sites, respectively. The dioxin concentration was
approximately 120 ppb at Western Processing and ranged from 147 ppb to 83,923 ppb
at Montana Pole. Data obtained from these sites, following the treatment process,
indicated that the average dioxin concentration was less than 1 ppb.
In addition to its destruction efficiency, the KPEG process is a cost-effective and
permanent solution. The process involves treating the PCB waste with a solvent
consisting of an alkaline reagent and polyethylene glycol which produces an innocuous
ether and a potassium chloride salt as end products. Total capital cost for this
alternative is estimated to be $6,017,480 with no operation and maintenance costs.
Actions at this site will include: excavation of soils in the roadway with PCB
concentrations greater than 10 mg/kg; excavation and disposal of contaminated
asphalt material, retaining uncontaminated material for reuse in repaving; chemical
treatment of the PCB contaminated soils using the KPEG process and reuse as fill in
the excavated areas; repavement of the roadways and driveways; and treatment of the
perched water in the sewer trench.
II. Technology
6
145
-------
A. Implementing the Technology
The existing asphalt roadway, driveways, yards, drainage ditch and storm drains
had to be excavated to depths of 0.5 m, 30 centimeters (cm), 15 cm, and 1 m,
respectively, to remove all PCB contaminated soil with a concentration of 10 mg/kg.
Soil sampling and onsite analysis will be conducted to determine final excavation
depths.
Trees and shrubs in the yards will be removed only when absolutely necessary
to reduce contamination. Removal in certain areas will require clearing and grubbing.
Large stumps, which are expected to retain a large percentage of soil on their root
structures, will be excavated with a bulldozer and disposed.
The removal of various lengths and sizes of drainage pipe is also anticipated.
These pipes are to be considered contaminated and will be excavated and removed
with a bulldozer and loader.
The existing sanitary sewer in the Wide Beach site consists of approximately
1380 m of gravity line and a 150 m section of force main. Due to the high porosity of
the fill material around the sewer, the sewer trench area represents a possible conduit
for leaching contaminated water off-site. The perched water in the sewer trench has
been found to have PCB concentrations up to 10 g/l. The water will be extracted
utilizing the shallow wells installed in the trench, and then treated using granular
activated carbon. The soil surrounding the sewer line will be excavated and treated. A
hydraulic barrier will be constructed in the sewer trench to prevent future off-site
transport of any residual contaminated ground water.
The wetlands do not contain large areas of contaminated soils. Areas identified
as having contamination are those found at the storm drain outlets which are
discharge points for much of the sitewide roadway runoff. By excavating to a depth of
approximately 20 cm, approximately 150 m 3 of contaminated sediments with PCB
concentrations greater than 10 mg/kg, will be removed. Prior to excavation, clearing
and removal of roots and tree stumps may be necessary.
The excavated soil from the roadways, drainage ditches, driveways, yards,
open lots, and wetlands will be fed into a continuous chemical treatment reactor. A
heat source will remove any inhibitory water from the slurry during detention and
accelerate the reaction. Soil will be continuously charged to a mixer by a backhoe. In
the mixer, the contaminated soil will be slurried with reagents and then will be pumped
to a rotary kiln where it will be heated to about 100°C for a detention time of two hours.
After reacting, the decontaminated solids will be separated from the reagents by
sedimentation. The solids will then be water-washed and separated. Water washings
will be combined with used solvent, and the solvent separated. The purified solvent
will be recycled to the mixer while the bottoms will go to waste. The treated soils will
be fertilized and returned as fill for the excavated locations. The roadways and
driveways will be regraded and paved. The excavated storm drains will be replaced.
7
146
-------
Because the excavation activities will generate significant levels of dust over the
site, house decontamination, including the same cleaning activities undertaken for the
immediate removal action, will be conducted once the remedial activities have been
completed.
B. Process Chemistry and Personnel Requirements
This relatively simple process requires the presence of chemists, lab
technicians, and chemical plant operators. Process/chemical engineers are not
required onsite. The process relies on nucleophilic aromatic substitution to degrade
PCBs. The first step in the process is a reaction between potassium hydroxide (KOH)
and the polyglycol (ROH) with subsequent dissociation to form an alkoxide anion
which is an effective nucleophile. The alkoxide then attacks the aromatic ring,
displacing a halogen. This reaction will continue as long as there are active groups
present. The addition of DMSO enhances the reaction by overcoming the inhibitory
effects attributed to the formation of hydroxide anions. Waste streams generated by
this reaction are polyethylene glycol, chlorine compounds and wash water.
C. Safety and Health Precautions
The workers’ health and safety requirements should be detailed in the site
safety plan. This plan should address Level C (Tyvek clothing and respirator), dermal
and respiratory protection requirements. In addition, decontamination and air
monitoring activities should be outlined to ensure adequate protection of personnel
and the general public.
A QA/QC plan with the following components should also be developed:
Project Description/Identification of Personnel & Areas of Responsibility
Sample Collection and Tracking
Analytical Procedures
QC Checks/Data Validation and Reporting
Project Documentation
Procedures to Assess Precision & Accuracy
Corrective Action
8
147
-------
Ill. Results and Conclusions
At this time, full scale operations have not begun. These operations are
scheduled to begin in early 1990 and to end in early 1991. Preliminary process
information from the pilot-scale operations conducted from 3/88 to 9/88 is
detailed below. Initial soil and groundwater sampling data are contained in
Appendix A.
The KPEG soil treatment demonstrated a high degree of effectiveness in
reducing the concentration of PCBs present in the soil at the Wide Beach site.
During the pilot-scale operations, soils with initial PCB concentrations ranging
from 30 ppm to 260 ppm were reduced to less than 10 ppm. Final PCB levels
ranged from 0.7 ppm to 1.7 ppm.
Reaction times for the process varied depending upon the cleanup level
requirement. To achieve a cleanup level of 10 ppm, approximately 4.75 hours
were required; while 6.25 hours were necessary to attain a 2 ppm level.
Operation costs for the Wide Beach site are based upon the following
assumptions:
1. 20,000 cubic yards of soil to be decontaminated
2. 24 hour/day, 7 day/week, 12 hour/shift processing schedule
3. 20% of capital amortized over course of job (assumes 4 months
operation, 5%/month depreciation charges)
4. 70% process equipment availability (30% downtime due to equipment
problems)
5. Maintenance costs at 10% of capital cost over project (30%/year, 1/3
year)
6. Electrical power at $0.1 kW-hr
7. Fuel oil costs at $10/million BTU delivered
8. Staffing by crew of 5 - manager/operator on night shift,
manager/operator/chemist on day shift. Crews rotated on 7 day on/7
day off schedule
9. Salary levels -
manager $22/hr
chemist $17/hr
operator $1 5/hr
10. Incoming soil moisture level of 20%
9
148
-------
Based upon the above assumptions, the estimated cost for a 2 ppm clean up
level will be:
Costs for processing (no profit) Cost/unit total $/cubic yard
1. Reagent $65 $1,300,000 $65
2. Waste disposal $0.22 $14,300 $1
3. Fuel costs $33 $660,000 $33
4. Electric power and phone service $0.10 $351,482 $18
5. Field labor
manager $22 $114,286 $6
chemist $17 $44,156 $2
field operator $15 $77,922 $4
6. Per diem $75 $84,416 $4
7. Personal protective equipment $40 $45,022 $2
8. Travel, Syracuse to/from WB site $100 $16,035 $1
9. Depreciation $1,000,000 $50
10. Maintenance $296,507 $15
Subtotal $4,004,125 $200
Allocated home office overhead costs $400,413 $20
Subtotal $4,404,538 $0
Supervision and Administration (8%) $352,363 $18
Subtotal $4,756,901 $0
Contingency (10%) $475,690 $24
Subtotal $5,232,591 $262
Profit (15% of total cost) $784,889 $39
TOTAL COST, WIDE BEACH CLEANUP $6,017,480 $301
IV. Contact for More Information
David Lopez
U.S. EPA, OS-210
Emergency Response Division
401 M Street, S.W.
Washington, D.C. 20460
(202) 382-2471
10
149
-------
V. References
1. Record of Decision - Remedial Alternative Selection (Wide Beach
Development Site, Brant Township, New York)
2. Engineering Feasibility Study for Remedial Action at the Wide Beach
Development Site, Town of Brant, Erie County, New York, August 1985.
3. Paul E. des Rosiers, APEG Treatment of Dioxin- and Furan-Contaminated
Oil at an Inactive Wood Treating Site in Butte, Montana.
4. Richard T. Cartwright, Dioxin Dechlorination at the Western Proccessing
Superfund Site.
5. Pilot Plant Report, February 1989.
11
150
-------
Appendix A
Initial Soil and Groundwater Sampling Results
151
-------
ft!32Lfl 0 1 10 AZ4AL 313 ON sc:. stp: zs
r’c— CVtJ;T! C C! !!C _ %t J ‘X A.L_ —
ve :f : :c.: : i5 ! ..a :.: : ‘.as :: A ::: . :
SE A:,
— !
: tnr:a 190 .4 30
!ar:a E: ; ,rs —3.0
5 Ira:: 69 —a Ic:t. 8.4
6 L1 :.L L a .9 Epics 4.6
7 t 1 1i : 13 50 aa .: 11
a stkstt 35* 5. Lytcrc 70
9 larcac 16 52 r ;r.y 6.3
It PLn. 16 53 Nnta 12
11 dtckiv 24 5 No .b sc SCa )
1 : Sc at:: 11 55 t.n.t 72
HOts. . y5(.} so Pu?ILC :L
Pajor 5 4 57 MsL&tc.. , ts(a)
15 fl 1 .:.LL: 390 23 A; :.: ‘a
it tayL:: NS C 39 Ca : :.L :. 10
P.r sc a: 6 lcn: 63
13 Ge. 7 50
:9 ?auoc 3.’
3 Gs 1L .;
M ec ca :
41
:3 Aur U: ‘a:: C
0.18
: 2 ..nt:t.:; N E l l)
:t tar,:v 12
On..: 3. ;
P: c. 0.40
:9 3a 1 1 13
6.1
31 .atac:: a 130
3: Gainn
33 Nur;:e 190
3 I%ar;ny 8 ’
32 Plrnk 1 1
Pnsos a no
35 Cusri $7
39 Naysr Is
‘—3 Lain NE ’ ’ )
‘4 Hsi lsan 8.
42 Grabs:.cat:.r :6
‘ .3 Treat 55
(a ho c aacpL.d b.c. aS . dr vn.y ii pavsd.
153
-------
azs. ...:s or :L:L . £: 3t Is Os SOIl.
rp p ! 5 _ ::E 5V. A Z5T
ia :S * ::.:: :: — £:a: : , •s
__________ — .
— be. n
3 Xa1e::e .5 —: Cra:. s :act,g
7 aoz
S Y—s : •8 Lo s
6 M 11&:sLlo 6.0 45 .i.• 1cr 1.9
7 M .L1er 7.0 3ai.:
8 ALlen 5.2 7 lo;srs
9 Sarton 39 48 Iur&. 3.0
10 PL.va 2.6 ‘9 Speeb 13
11 1.4 30 Sagass 1.8
12 5cb Ltz ‘9 51 . ivIord 9.3
13 o1.a 14 52 77
3.0 33 :.L
13 P. 7j:,Uo 100 34 Nes biscb 6.8
16 TayLor 1.5 35 sr.gsr 1.3
17 P P3C6 :s 36 ? eru1th 1 6.3
18 Cr., 6.1 57 . .:.11o 11
19 1.7 38 L; . r 3.7
3 13 59 Caor. l s s. 5.8
: 16 60 3ov.
.L::srt 1’
:3 A ?S&&OIPJCb 21
: S ar.aP.ao 1.1
:s L. :Serg
:s 11
Os &e 3.5
0.05
:9 8a 1 1 0.06
30 )l ..e: 1.8
31 .a;acone
32 Ca evsb& 2.50
33 poy 10
34 hy 1.1
33 Ptwak 12
36a 91
363 p, 05 1 ,(b) 0.6a
37 ?roo.b&a 33
38 G srta 600
39 Meyer 9.3
(ii S pLe erllseted 15 1.eld sex: to P::sooii ,.sid.u. roadva..
(6) Sa pL. c LL.ctsd &s f sLd seat to Pross .i tea &d.scs 65 :
Ito. r:a;vay.
? IIHHANT 1W 111511116 up •‘ I ii 1 11111111*1 uI ItI•. Pill SI), I. SAMII P S li ii 111.111i
P Mlll_ TAll !I!l ! ‘, 11116 IN V 11.1 Sl iM II N 111 TlIlIt
L.,lIeelIo’i S.. 5.lf) Ib) AIIC 101 1734
_______!I!!IIOPJPE!L ).!Lfl_________ _!‘J’__ ... !! rii’_.— _ . us.— — 1
M,., snie. 79 Out 19 MAt 87 SI P Y.id-P 311.3
O 11.4
lionS. 38 Uv.l — P.. of joule it HA? 81 ?.Id-S 1.3
I ) U.S
Mii i . c i So 60 Ov.l It HAT I I I 611 ?.id—S 1111.6
II 5.6
Mii i . ,. SO O..l It HAT 112 50 Tend-P I 4
I I I I. .
i’rd..ci. Ill II ,il 59 MAT I I I 611 T iJ I 7 II
I I I.e
l.nup. 87 (lcd It HAT 87 611 t .,J - P 4.4
I I I.e
IIucIan to o ,.i 70 HAT 112 511 ?.id-F L I
i i eu.uS
ilocIsan 90 0 ,.i 70 HAT 87 I.,d 1.4
I I 3.11
H.srphy 174 O..l 74) HAT 82 M I T.rd-P
I I 5.1
Vsceni 1.1,1 711 HAT lii 611 lul 1 1.6
I I 31.3
M..y.u. 161 0 ,.I ill HAY 111 611 Y.sJ P 3.5
I I 1.7
I . . ., I. I 5..sI I. SI . IN HAT II I 611 V . ,d- S I 11.16
I )
Cr.t.ruiaI.SIt, I il ,.I IS HA? II? T.sd- I :.i.
C,.b.o., .ii., I Il..I IS MAT 117 611 9.,d-B 1 1.4
I I ‘11.03
154
la)Su;lici.I u.pI .s (511) 4.5,. ci U -6 I ..; drip •.. .ie (Ii) elm •t Sit depth.
(bi I l,.rI F ho... S • dde.
-------
3 !0tUT or 2:3 :z:nn:x*:::n costcc:© on Sc:L san.xs
c: ..t :: rtcp. tti4&tS cr w strcz as: £3.1 ::tnn.
fit, t?’iC. c.?PIt !qtI
: : .r
3
3
6
7
6
9
10
11
12
13
1’
13
16
11
1$
19
20
21
2 3
2’
23
29
30
1 ’
Cs_car 0—6 a.
A:a_31 Sc-..... 0—s a.
A:._31 Nor. 0—6 is.
Cant: 0—6 a.
Vast 0—6 a.
Last 0-6 a.
Can:.: 3—6 in.
Cant.. 6— 12 i a.
Cancer 0-3 a.
Last 03 a.
Last 3—6 a.
134_,Q7 Last 6—L i a.
heat 03 it.
V 15 : 6—fl a.
Vast 3—6 a-
A..539 Nor:t 0—3 a.
A36_39 Mort 3—6 a.
Soon 0—3 a.
Sou: 3—6 a.
‘ ‘36—39 Cancer 0—3 in.
“36—39 Canter 3.6 a.
3O03 Vast 0—3 a.
Canter 0—3 in.
Last 0—3 in.
C cci: 0—3 in.
San 0—3 a.
C04, 7 3n: 0—3 a.
100,03 Last 3—6 a.
G 0 7 Caesar 36 in.
oo—03 Can:.: 3—6 in.
Ar::.:: .: —
‘.61
9.0?
29.34
s .
7.41
11.97
4.33
5.46
5.23
7.42
59.44
s
6.02
9.42
25.9 1
53.79
17.13
:5.:?
I .62
20.37
26.37
11 6.03
2 . 13
16.19
12.53
119.11
‘.31
16.6 3
11.01
26.32
e..iLna Laratlon
Collection Samply 1 W Asorlo, 1734
.......P?IL...— PfP!IL- 3flUS3.....—. ..... 1’l’&i.._.
Franc. 6 ova l
I C
NAY
82
50
0
YsrdS
11.3
•u l.0S
U ns, I ? S mith St.
Il l
NAT
62
LU
I I
Yard-F
10.7
2.S
S.s ikt II South Si.
IS
NAT
•i
S u
i i
Yard-S
coin
0.9
loosen, 41 South St.
16
NAT
62
50
I ,
Yard-F
5 ,41
16.0
Satisen. 43 Soot!. St.
lit
KAY
(12
s o
Ii
Yard-S
(
11.4
lylord, 10610 lakeshore
19
hAT
112
50
I I
Yard-F
711.3
12.0
Neuasn 30 to. Si.
I I
NAT
12
50
I I
Yard-F
12.0
3.6
Nenan 30 F .. St.
Ill
ISA ,
Ill
III
Yard-S
0.9
ii
1 10
Ze.4er 76 Smith. Si -
us
NAY
i i
S O
Yard-S
ii. ,
Ii
‘0.03
Militel I.. 20 Smith Si.
IS
NAT
62
Lu
0
Yard-F
3.4
1.6
UnWed Are.
IN
NAT
82
Lu
C
Lull
1.6
U.S
lot Weau
I I
slAT
62
50
U
Soil
LI
1.1
Vacau.i ml
16
NAY
62
50
C
Loll
71.1)
7.8
WoWed Area
70
NAT
62
LU
p
Soil
56.0
1.43
Lot across road
70
NAT
67
LU
U
Soil
7 1 1.6
111.7
Pint Sl.ieJ
IC
NAY
II ?
LI I
1 1(1 )
Fill 1 ) 1st
11.8
II. ?
as..... F ,.. P.m.
I V
NAT
87
SU
I ’
Sail
31.0
1.1
155
-------
Si::L. £rsc :r ::sa
t .. •94
.L... S:n 3—a :.
:2 ‘ ._. NOfl 03 . 6.37
fl A.. - . Ciat., 0—3 La. IL.3’
u —— v.
3au: 0—3 &. 6.L9
3! A 34 _ 07 C.:tsy 3—6 t:.
39 Morn 3—6 i: 4L.05
.0 C 04 barth 0—3 it. 3.45
‘4 East 0—3 is. 2.40
Vaj 3—6 i:. 40.13
.3 Piraic Cray. 3.30
..‘. k05_ 1 C.::st 0—3 i:. ‘4.5.
43 Kasn Vs fan 3.73
0 A. ._ C.n:n 0—6 a. 5.44
— ? L 8 . 1 East 0—3 s:. 69.fl
4 $ C 3 ... C.t:r 3 4 &:. 36.36
Cntn 3—6 i:.
20 5QUt 03 :. fl.00
21 L .. Corn 0—3 is. 9.55
Saith 3—6 s: 1.96
23 C 04 4 07 Saith 3—6 is. 55.50
24 A:&..v Soijtb 0 4 i s. 37.54
c 04 . M orn 3—6 is. :s.oo
56 £24.27 Morn 0—6 a. 3 1.0
57 C 04 07 3.atb 0—3 is. 4Z.Z ’
56 L g Mirth 0—3 a. 8.09
59 Cast., 0—3 . 39.16
60 A..0 lath 3—6 a. 1.37
61 G 0447 Mirth 3—6 a. 15.71
6: 8 • Vast 3—6 is. 59.40
63 100..03 Last 6—12 a. 23.09
156
-------
A.rscLsV 15’
t= ‘•( T•
1 :.. 7
10.14
1.7’
114.45
1 • 05
$3 • 5,
.— iJ
.-
S
.1—S
a •a’
.1
I ;. .
H: .
I I
Locat .ort .o
•
•IS.’
— --—-— . . ..
SUNNAST 01 PCI I)LT( INATIIINS 1011 SO, I. SANPLLS COLlECTED I 11* 1 lIIPIiVAl ,
- —
3It1.AI!’114f1’ 1 ° °
Norgsate. 7 Oval
biller. SO Oval
Irep. $7 Oval
Macloan, U Oval
l .rpbp. 134 Oval
N .ra, 141 Oval
Gral.,.sIattef, I Oval
l.yI...4 , 101110 l.abe.kore Rd.
N.oaais 30 los SI.
1 g . . ,, Soath Si.
1.1 ersectiss .1 Oval
ad Fo p Si.
lai .,a.cI ion .1 Oval
auJ Access I I.
I WAY $2 Soil
IY NAY $7 Soil
IY NAY $2 Sail
30 HAY $2 Suit
20 MAT $3 Ss.il
3D MAY $2 Soil
IN NAT 12 Roil
l ’ NAY 02 Soil
I I NAT I? Roil
ii HAY 117 Soil
t ,0 . 3
6.e
1 1. 3
‘I..
S..
7.31
131.6
,,. 3
1 5.3
113.6
44.4
23.3
I .0
(a) Surlicial .a.pl.a taken at 0-6 us.; deep •aisplr s taken •1 3 I i.
157
S, Is
:.a: —, s.
“—.3
t. .. - L i i :
_g3
:6 1., - .s:
67 L3Ø, Ws•g 3—6 is.
61 E0003 Vest 0—3 jo.
69 100_03 Cotter 6—12 it.
70 tOO ,03 Center 3—4 L*.
71 30-O3 Center 0-3 it.
Collection Ss.ple SoaPlf 1 )
___ ‘jt .J3PL. hP..L . ... .
Aroc lot 1234
—
Rot lace
Sot (SC.
Sot I a c.
S ot (ace
Sot lace
Sot lace
Rot lace
Suit lace
Sot lace
S u it (ace
Deep
l.•d
l.ad
load
load
load
load
— “ ‘SI
I.’. ,’
load
load aide
IN HAY 5! Suit %o,lace luau
30 HAY Ii Soil So,lac,
load
-------
Su,uauT or REsults or rca tikyL 4lwATiONS tiN IIIL SANPI.ES CuI.IICTUP
Collection S.ple Sa_pi Ai.. lur 1754
________ lute •jflj D i pt h Sou fC i _ J.apJI gJ
I OCT SI Soil LU Ditch 91.9
19 N(iV SI Soil SU Pitch 1 ,076
I) IS.
19 NOV SI Soil SO Ditch 162
0 S
I V NI1V SI Soil SO Ditch 1.9
0 ilI.S
IV NOV SI Soil SO Ditch 7S
II
19 NOV SI Soil SU Ditch 1 5
U 71.5
IV NOV SI Soil SO Ditch 119
U 12%
ii NOV SI Soil SO Ditch 7.0
Ii 4.61
IV Mliv UI Soil SO Pitch 2 5.5
I ) 5. 0
a5;uuicial ..pl.u (SO) latin I I 0-6 ln. deep .a.ple. ( Ii) aim at 3 Ii.
(hI ..plr. latin at drasn. e dud.., located us, I,u.,u ni ,e.u.i ,-n. mu.
(c) ta.ple taimn at vault-oil barirl sluu ,a e airs.
• 5tethoj ’
lioct.sn, 90 Oval
Cup, 52 Oval
Vinnert, Oval
Plevai, 115 0.•l
Gi.benutstl .r , I Oval
Neu,.ai ,. 30 Fo. Si.
lien, 9 South Si.
lien. Rudyard on tuu tu.
SE Corner of Ion 6 Smith St.
5!!L!.L!tE.!L
i!2!L._.
J3ri..
S• PIf)
!n’JJ_
rc.
(b)
Vial aide .1 Oval, NI collier
19 lliiV SI
Soil
LU
I)
Ditch
61.5
340
G.nract 1 .1st.
Grey, 52 Oval
19 NOV I I
WAR 51
Soil
Soil
LU
I,
511
Ditch
Ditch
0.04
U_US
I ll
Whit silo. 60 Oval
19 HAT 51
Soil
LU
U
Ditch
205.9
59.1
killer 50 Oval
IV HAY 52
Soil
SO
I i
hitch
4.4
1.6
liucI.n , 90 Oval
30 NAY RI
Soil
LU
I)
Ditch
I t l i.9
A IU
Pleat, 175 Oval
30 HAT 51
Soil
SO
U
Ditch
0.)
0.2
IIell an, 7 South St.
IS NAY SI
Soil
LU
hi
Ditch
19.0
2.2
Pitch on sooth side of
lou Rd.
III MAT S2
Soil
LV
I i
Ditch
114
1.4
hutch no,lh side 01 Iou N.h.
IS HAT 52
Soil
Su
0
Ditcl,(c)
4Sf
158
-------
sininasv iii r c a tiiTl11NlNATluN lii 5 (1 1 1. SaNri I $ (SPill 5iiiiiw ANtI 511111 i1 S) iii tiLl Iii
I ’iiii.
Station Colli,ciiun Sa.p I Aruclor 1254
jfjjj n local inn _j!p .... .. j, . . . . . . _ j Ji&) - .
I N-i $ur(iciui (1—4 in.) 4 Sir 84 5,111 I I
1* 1-7 S..iliciai (1-4 in.) S Sir I ’S 5..ii 610
(*1—3 Sail •cu.l (1-4 In.) 6 Sir Ii’. Sail I .
1* 1-4 SarI ICIni (1-4 in.) b sir ‘is b..ii
(* 1-5 S.u,Iici.l (1—4 in.) 1 SiP 114 s ai i 0.6 5
IN-S Surlicial (1-4 in.) I D 511 114 Suit I ?
IN- i S..t(iciii (i-S in.) ii MP Ii’. Sail 1111 1
Sri icual II sir Iii. s.. ,l 72
1 7 . 1-9 Surf ici.l (u-S in.) I) sii Ii’. Soil 0.90
IN-ID Surf icial (0-4 in.) I ? 511 114 Sail 5.1
IN-Il 5..,fici.i (1 1—4 in.) Il SIP 5’. Soil 6.3
(*1—12 Suit cii i (0-6 in.) 711 Niiv iii , Sail 0.1 171
1* 1- f l S..,(icial (Il—S in.) lii iiiiV Ii’. Soil 0.0111
Latch him 3 11 5 1P 114 Sail 1110
Catch Ia . , . 6 11 SIP 84 Sail 5 .3 Db
Catch lasimi 1 11 SIP 114 Suit 31u
Catch ham I I? SF1 ’ 114 Sail 64
Latch basin I I I SIP 54 Soil I7IJ
S..rficial Onifall I Ii SIP 114 Soil l iii
S..rliriai (hush 3 I I sir 114 S..ii 110
IIp..n 1.11 I II Ii ) I I 511 II’. S..i I 34
iil.ifl 1.1.1 I ( 1 Ii ) I 7 511 114 Simi I 31
I Ip.II I.ot 3 (I Ii) 17 SIP 54 Sail
lipen lut 3 (7 Ii) Il 511 114 Suit 9.1
Ui .rn l ..i 3 (I it) 9 SiP ii’. Si.il 16
(ii . . ’. Ioi 4 (I Ii) I I SIr Ii ’ . Sail 95
upon i ui 4 ( .‘ It ) Il SIP Ii ’ . Su.i I 1.2
(ii.en 101 4 ( I Ii ) I I Si I 1(4 Si.i I I .6
open Lot 4 (4 It) 12 sir 114 Sail 0.04
Su,l icial (11-4 in.) I) SIp 116 I I II 13
111—7 S..,lici.l (11-4 in.) I ? sIr I I’ , :.uii tO
111—3 Surf i. il (0-4 ii..) 74 511 Ii. a 5. .i I IS
Station ColI. ction S..pI. Arocior 1754
- - - - Ii iIC
Surf c lal (0—4 In.) 24 $ $4 Soil 110
I N-S Sur(icjal (li—S ii.) 24 SI 84 Sail 0.71
5mph 4( 10-5.5 Ii) 24 SiP 54 Soil NI)
IN-S Sa.pl. S (l0. 5—ILD It) 25 SIP IS Soil Nil
N 16 Surlicisi (0-6 in.I 25 SIP $4 Ssid II
I l l- i S..rficiai (0-6 I..) 76 SIP 114 Suit 0.03
IN- I Surf icisi (0-6 in.) 39 NiiV 51. Soil 0.016
5 11—I Surlicial (li-S in.) II SiP 54 Soil 170
5 1 1— 2 Surf ciii (0—4 is.) II siP 114 5i .sl I)
51 13 5..,ficiai (0-4 in.) 19 SIP 114 Soil 0.55
511-4 Surf Ic si (0-4 in.) I V 511 54 Suit 3 11
S..ilici.i (0-4 in.) 1? Sir 84 Soil 34
Surf ciii (0-4 In.) 24 SIP 114 Sail 1110
S-i Su .rliciat (0-6 in.) I I SiP 54 Soil 0.1)
•1 Sa.pi. 9 (1.S—I. Ii) Ii 51.1 54 Soil (1.VlS
I-I Ssspi. 10 (51.0-9.5 It) Ii 511 54 Soil 0.12
5 1 Surf nil (U- I in.) I I SiP $4 Soil 1911
h- i Sapi, s i fo) II SIP 54 Suil 0.01111
I-I Sa.pIe S (9. 5-10.0 It) II SIr 114 Soil ( l.Uii ’ .
I) Surf icial (0-6 in.) 12 SIP 114 Soil 45
• I S..iI. 5 (.‘.S-S.u Ii) ii SIt 114 Soil 0.008
1-3 Sa .ple 6 (9.5- 10.0 It) U SIP 54 Soil 0.71
Surf mist (li-s in.) 13 sir 51. Soil 3 w
54 ...pl. S (?.u-i. 5 ii) I) SiP 114 Soul 0.011
I-S Sa.pl. 5 (5.0-5.5 It) Ii 51.1’ 114 Suit U.U09
I-S Surf ciii (li—S in.) 7 5 511 54 Soil 41
S-S Sa.pl. S (9.0—9.5 Ii) 24 SIP 54 Soil 0.14
I-S Surf ciii (0-6 in.) 3? SiP 114 Soil 75
I-b 5a.ple (9. 5-10.0 It) 7? Sir I’. Soil 0.06
held Slink ii SIr 114 Soil liii
ii.id Slams
‘iii
I meld husk
NI)
159
-------
I
I
I
I
I
‘i is
/
I
(LI
. 4
. ._. . e .• .S•s
.4 . .—
_. . I—. •II • S• .I•
p.... S. S.S .
On-Sit, Soil Sample. Location Wap
$U1IPIAIT OP IF.SVLTI OP PCI OPTCMIINATI0NS ON NAISM SEll WNT COILS
- —______ ______
Overall
$ei .t
ltatl.a
C.llecil..
Core Leath
Leagik
$.g.at
*rcIr 1214
Nuh e
IcptL.n
Pate -
30 AlE 14
(La.)
4.23
4.23
0,.it.s.tJ..
LI I
(r&hit)
III
CI
-
Wids leach WetIsad
C2
Wide leach We ,a
30 LIE $6
2.1?
2.17
C)
C)
Wide leach Wetla,d
Wide leach WetIaad
30 LIE $4
30 LIE $4
14.0
1.0
1.0
l.p
l.e.
5. 57
0.13
C 6
C6
Wide leach WetlaNd
Wide leach We(Iaad
30 LIE $4
30 ME $4
1 1.1 5
4.21
4.7 1
T.p
NiddI.
Ill
ND
C4
Wide leach WetlaNd
30 ME 14
S.?S
lott
ND
CS
CS
Wide leach Wetlead
Wid, leach Wet laud
30 MG $4
30 ME $6
11.31
1. 11
SI?
tap
l.tt
2.4
5.52
CS
C l
Cl
C l
C?
CI
CI
C
C
dO
CIO
CII
CI ?
CI ?
CI )
CI )
Wide leach Wet lead
WEd leach Wet laud
Wida leach Vetlaud
Wide leach Vetla,,d
Wide leach Wetlaud
Wide leach Wetlaud
Wide leach Wetlaud
Wide leach W,lla,d
Wide leach Wetlaud
Wide leach Wetlaud
Wide leach Wet laud
Wide leach WetlaNd
Wide leach Wetlaud
Wide leach Wet laud
Wide leach Wetlaud
Wide leach Wet ausI
30 LIE $4
30 LIE $4
30 LIE 14
30 LIE $6
3D LIE $4
30 LIE $4
30 LIE $4
30 LIE $6
30 MG 16
30 LIE I I
30 LIE 14
30 LIE $4
30 M C $4
30 LIE 14
30 LIE $4
30 AlE 14
13. 11
16.12
10.75
10.11
II.) ?
4.13
43
14.43
SI?
6. 4
6. 4
1.04
1.06
4.) ?
4.3 1
5.44
1.64
S. 4
4.S
4. 11
4.11
1 .3 1
1.31
Niddle
l.p
latt
Tep
l.tt .,
t p
lotIa..
l.p
lop
lote e
LII
Tap
latt
lop
lott
1.01
10
. 5
3.1
•.ui
13
0.10
1.4
0.12
4 4
1.40
200
O•
2.2
0.02
0
a a. . ... S . ..S .
160
-------
SWOIASY 0? KURT! 0? TOTAl. a! DISS.VW PC)
A*XWK 1254 0TT2 WAT2ONS ON STOS-VAfl1 MflLfl
COLL!CTW 30 A ST 195k. VWE S l ICE. WVJ YOU
D i i ao l cod Total
Cokl.csjs. Aroclar 1254 aroctor 1254
Station ti n ( u eIt ( ut/I )
Outfall 1 1133 0.92 93
OutfaLl 1 1203 O.M 5.0
OutfaLl 1 1235 0.47 6.
Outtalk 1 1333 0.75 3.2
Outtalk 1 1433 1.0 4.0
Outfall 1 1330 1.4 4.6
tarab 1 1215 0.05 0.25
Marsh 1 1313 0.30 2.9
?‘.arsh 1 1413 ND 0.26
Marsh 1 1313 0.04
at: Bait: 1 1213 0.51 14
Ca:cb Sass: 1 1313 0.93 13
:at:h last: 1 1415 1.’ 11
Canh lasto 1 1515 1.3 11
SUNN IlY D l YOIAYIL ! DICANICS DEYESNINATIONS (us/hi) Fop SOIL SANPLES
_ _ _ _ _ _ _ _ _ _ _ — CQg J JDE UA M. tilt COp fl _ fl jpp!LJ P !LI! Ø.J!S 4
Collection Simple Sample Nosluplens Tetra— Fluoro—
fl1 123 0555 Depth Type çftj osthene &etone tricluloroeethsos
ii(sll 2 I i SEP $4 0—6 Is. Soil 60 21
I-I I ? SIP 54 9-9., It Soil I SO — — DO — —
5 .2 I I SEP $4 1.5—S ft Soil 60 — — — — 12
1-4 I I SIP $4 1—1.3 it Spit — — — — - — 4$
5-5 34 SEP $4 0—6 In. Soil 26$ — — — — — —
5- 5 24 SIP $4 9-9.5 It Soil 40 — — — — — —
5-6 2) SIP $4 0-6 I.. SoIl 300 — . — —
1 4 21 SIP $4 9.5—10 ft Soil 230 — — — — — —
N W- S 24 SEP $4 0-6 in. $ il 200
Nw-S 24 SEP 54 S—U.S ft Soil 350
1fW6 iS SEP 54 0-4 in. Soil 260
•ois: A dished line lndicstes that coepouod was not detected in the sample.
SUNN I lY OF IAS IFNEUTflL CONPOUNDS DETETED IN SOILS COLLECTD P504 IONITOKIC WELLS
ANSI C lOCATIONS. VlDj _ jjj f ..i. j.. !QU
si.-t..pi 5.... ,. ) m a u I s...a .n. I ) Stat )..stpIieopII
P’sIln JetmtIn.. aSI’eJ.ps 5.Si.p.ne FI.,i,’ttlit, !I’ n r1 ”trt Pr , fl.. ...snslan..— .......tl4S ItJP—...
S 5..iIuI.I —. .. .. .. .. —. .. S..
S S
ai s.. l.a -- - -
• p s. pa -. .- .. — - - -
5.. 5..p 5.1 0 ) -- .. —. -.
• 5..Ul.EsI 0.1 0.5 oP o.a op s. 161
as 1.111.1.1 I P
-------
,-&
0 i
5 ’ )
SUNNA T OF NETAL, P1IEIIO1., AND CTANIIJg CONCENTRATIONS DETE fINED IN SOIL SAHPLES
COLLLCflD ! !i!tJ!I!YJCL$I1T U J1IE T(MNOFIPffl J!j! 2!!!jj.1 1tM TOM
SIJHNART OF HETAI.S, CYANVOg, AND PIIE1ftII. DETENNINATIONS (mg/I) OF AQuEOUS SAHPLF.S
ps.ut.. I . .. . I . .. ._ .
P u—’ ..
1.11.,...
. p. 1’....
a_il....,
I .is1’ .L
, . .. 5q
I’sf’d
S .pI II ...
JNI’sD_
t.1. I. . .
I .s I ’rJ
VS...I—s
j.p1 15?
I .p,s
..pflpJ
I..I
, IpP
1.. . ,
I.,IlpI
51.1.1
1S1’II
S•l•.i ..
Jpsl’aI
• 5,. .
I .tIPtI
u .s Ii . .
.i,sIP.J
SI..
l, .IKP
, ...si,.
J f’.L
I .lI,I.I fl..P.SS S
llP pl . lt. )S.SOl)
• S • pI• SUP!) I !)
• I P—,l• S 11.1 P IsP
• I a.. I. P 11.1 a. Is)
• I l . pl . PIP S 5.5 II)
• 1 l..pI , I 11.1 P.S II )
• a • pI. Sill I.) i )
l 5i .I.III 5l.1IuP
I S.,Il,I.l is S PIP
• IS..pI. lIPS).) ISP
S l ,II.IiI a S liP
1 5 l..pI. I I I. ) 55.5 III
I l. .ll.I .i IS I Is)
•ia..,I.s ls.s-s.issP
S S..pIa P lIP.) IS SIP
S S..SIiIiI IC $ III
1....
C ....
i..
i—s.
(I.••
V...
C..•
C...
1...
C...
C ...
Iui•
1....
P...
I...
(a•i
51 HP 54 • is S
SHIPS a ll
I) SirS! .,.l
u PIP 55 ‘ 5.5
l ISP P S as
SI III IS ‘ UI
IPPIPSS ‘ 5
II Ill SI SI
l ’IIPSS •l
14 PIP 5 5 1 I
PS5IP SI S I
15 liP H I P
19 lIP • . II S
SI Ill 55 I i
ISISPPI PS
II Ill 55 II. )
II IS? Si
H
H
H
II
5)
IS
H
P1
15
SI
PP
II
U
I
II
P
II
S 55
•SP
esi
S PP
S SI
SIP
5.55
555
5.53
5 5
CII
S II
5 U
S II
PSI
I 5P
S IS
5I 5
555
SIP
5 aS
S I)
S •S
•
aSS
ale
5 P
I It
5 P5
S 5
S SI
515
C PS
I P S
Il
pP
PP
I
a
IS
,
II
p.
I I
I
S.
II
P
IS
I
IS
H
I I
‘ S
PP
Ia
1)
U
I I
II
Ii
PS
P
5
U
a
II
PP
II
51
Il
S’.
It
It
Ii
,‘
‘S
SI
I .
I’
1
ss
U Nil
‘‘itS
• ‘
a NI,
S N5e
C II I
‘s”
• •
UNIS
a pie
a ut.
U NIP
P •‘U
S ala
u’
P
a a .a.
55
U
P s
pp
p
•
SC
sp
SI
j,
pp
p
P
II
1
gg
PS
s.,
p p
p p
SI
. 5
I p
IS
• p
p
p p
5 5
a S
II
IS I
•,
.e ,
551
s.d
, •
p •p
ass
e .
as . 1
.51
• ,s
•p,
• •p
p •p
• • ,
555
555
• •
s
I.e
,•
•p
p
• p
si
•s
55
s P
p
p •
S.,
• •
5S
5 5
...
pp
• 5
P,
••
pp
•
,s
I. ,
SN
N
pp
IS)
,
551
555
.
. p
4
ii ;
•
• •
..
, •
S
‘ 5.5
as
,, p
•
, ,
55
pp
...,
•
..
..eP
• 5• .
,. ,
. 5.5
• 55.
• a.p
•,
•
•.
5 5 5
.• •,
‘er I.
• !!‘. I_
..__P 5 ’l_
jilIJ_
Jm:i
•‘ ‘ ..
. ?_
TeS s) Cj. .i ,44 1
MIIeii .,p
A .a. .u
PsivIll..
(a.I.j,. .
lulal LS.,..la,.
LIsp..
I ea
1 5.uv ,
041,S
S I 1I I (ue
5•p,,•
1S.AI Is . ..
liøc
PI .rnoI .
0•555
VaI. 5
V.p,
V.i,
11.1,,
0.1, ,
V . 1. 1
Vul. ,
VaI ,
VIe.,
V.1 ,.
Va , (
V .I u
0 . 5
Vu 11
1) 1)5
0.0 1 1
‘0.00?
(0.IIUSfl
0.IIOU
0.001
0.11111
0.1101
‘0.01101
0.&IjlJ
(0.001
‘0.11001
(0.001
0.015
0.01
415.51
D.IP?
40.007
0.00I
0.00??
‘0.001
0.1 10)
0.001
(0.0001
0.111
(0.001
1) 0(901
40.0(15
0.054
0.01
40.01
0.0)0
0.Il0
‘I.U0U
0.1101.7
U 1101
0.1111,
0.11111
40. 51*01
1.11
‘O.I )O l
aO.000I
40.001
0.0 ) ,
(51.05
‘0.5 ),
0.00?
0.Ilu
11.15115 ,
0. 1 1 151 1
10.1101
0.11111
(U 11 151
‘11.11110,
0.15,
(U (lOP
IJ.lIIIU
‘Il.I’IIl
0.01?
0.0 1
40 UI
00)1
40.01 11
0. 111 11?
0.11011
40.11111
U.HISI
11.1111,
U.IIlIIlI
OPuS
1O. 1 5,ip
0.511109.
10.1101
0.061
(0.01
40.01
U.IIP’ ,
(11.0151
lD.IPIIu
0.1)011
0.1111)
U IPSO
0.1111
10.111101
0. 114
0.I1U2
40 110111
0.001
0.09.1
‘0. 111
411.11 ,
U.lw1
(0.111 11
‘U.lilIIIp
I I 0 5)1( 5
0.11111
II II I;
II. Out
(11.0110?
0.1)5
0.uo2
(0.11001
(0.11115
0.0155
0.0l
1J 01
U us;
0 5155 5
,1j.UlIul
II I)lI6
0 IluS
0 1141
Ii UP?
‘Ii 11(1117
0 411?
‘51.1,111
‘0.11055
alP (Pill
11.415
115.115
-------
.,uIIIIAR UP PitlAt. tONS EN1ISATPUN5 I)E1fldIINFU IN AQUU JS SAPIPLES COI.I.ECTED
.,i,r.... ..
- l ,. i 5 .n
t..il. ,lii.. 1 pi.
•._ P.iu. .. .. ‘ire...
I..i.’.p
.I.,I’P.
•....I,
•5.. I,
P..pIl.—.
—
l. ...
•I.., l 5
•l...•
5 .pfl I
‘is ’l liii
5., l 1 5.pfI
..a..p
%..pf5)
•1.S.i
. -pJl)
SumS..
J,sfi?_
Iii. .,
I .sI’I
P 5 . 5 5g ..
.I,atp _
ii ..
Ssaf’I
•
..l
. .•. 5
...
.1
I ,
i i
.1
I I .
.1
l...l.
. 5
is l mil.ius S I
al
. 1
I
I
I
.1
•l
I
. 5.
‘
.
• • . ad
ii, Pd
• i ii
I U
I Ii
OS
‘ .
I II
1 ..l.nl.
li ii
Pk...l..
I.1I..
L.l...
P ...l
OI .b•p
C I i5Ii. CI.
Il,lel.Pt.
P.....
P.. ..
P. ,.,.
t . ..êIS
ipluid
P.S.
•.p I .
II I , .
0451..
0. 5..
1i 5. ..iII
ri..i
05111.11.
a’. . , lp
P . ..4. l
C’ ...p .. , .
NI__in
1” .,.i
P.,.h
t.,I.. ui
I’ .’P’,
l..l..a S.,
1.1.. l.p
S ri. ).
1 .pd .i 5 .sqIs
_,d . , I.. .).
II .. . ..
• t• a1
NiU ,. pI .
.: ..,
5.1.11,
•..•••i
N ...
L .i 5Iefli ,).
0 , Ils a
l’—li •l.,
55.....
P. 5. .
I ’.p..a
1 ’ 1. ’ lillep
‘i ’l .Il.a
5...
1. 1 1 . ..
• i i 1 .
u 5.55,
I.•I..i PI
l I i 0p
‘1
I I.. Pill
•,,g... •, pp a’.: pa 5. ,.,
5 11 .1., sip Pp au. I I P.l.
Ull.k... l.p H •ui : I I P.1.1
P. l.IA. a. pp a;’; Si Nil.,
11.1... l.p P3 •11 .4 Nil.
I 11.1... lip ii *1’: S I S. C..
I..,....i l.p PP Ii .. Si P. ..,
1, 1. 1. .. III. pp SIC 51 Nil,.
•h,P. . l.p I . a,.: 54 W ii. .
15 1,1 .. . lip I’. 1 1 1 II hi..
SI,,i.. . lip pa ass; 5; P.,,•
0.1.51. isp II L IE I I Nil,,
111.1... 5. p H •It 51 Nil.,
lilul.. lip VI a..; I I
1 , 1 , 1.. . asp 4 psi; 54 v. ..
I 5il.. . lip pa a..: II V. ,.,
l ’—l .iI. Isp P1 as’; II N , .,
511,1.. lip I i SI’ PS Nil..
•lI ,l,,i . pi a,.: ii v. ..,
Iiialei lip IS ala; 5% P . s.,
P 1.-i 11 P11 PS V . 5 ,•
111.1... sip I I a.,: SI Niu.
IIl.l... lip II PIE PS Nil.,
Si*uI.,. Sip II al.: Si •I’I
5 1 1.1. . Sep IS ii .: • ; s....
111.1... lip I D ii. ; 54 N. s,.
115.1. . lap PP PlC 54 PsI.,
115 11.,. lap PP li. SI Nil..
Illil... Sap S PIP SI Nil. 1 .
II , , , . . . a .,. II PIE SI Nil,.
Pa.. .... l.p I i all. SI Nil,,
lit. 1.. a.,. PP au: Si Nil . 1
II..I . 5 1. lip PP all; 54 N . e .,
P .isl4. lip PP a’.; •a v....
l4t ,h, asp Pp MI: SI Nil,,
P1 1 .1... lip II PlC SI W.s.
544.1.. •. PP a’.; i i Nil.
111,1.. lip PP alE 54 Nil..
W;.1..... ... l , P a... UI Nil..
5,1.5... lip II P1 1. II Nil,.
PIlliua i.p 1, pci; S v.g,,
Iil h.nl.p 11511. 5 ’. Vsl.a
Pii•i...l lip 51 all: I, Wisp,
511.1.,.. s.f . PP as’: Ia Nil..
11 1.1.0 l.p ii as.: S’ Nil. 1
555.5... lip $1 M l. ft N...,
5 11 , 1 ., , lip II 511. 55 S.l.
PSI, 1.. . lip II a..: SI Nil,.
1...i IS LI.. II
t...1 I I ill: . 5 N.,.,
U. ;,l..,. lip IS MI. SI Pal .
‘l,.l., . lip I 5II •,
I•..l ill, lip PP p.s ; 5’, s,,.,
5l..(.jJ Isp PS all: SI Nil..
111.1... I. II 1..: •a s.,.
‘ r Vi i.,
NI p I ’ll
• 5I ,
llu,l.. . , s.f. Ii all UI V . 1 .,
P
P
p
S
I
I
5
P
I
I
P
P
i i
‘1
.1
. 5
• 5
aS
.p
‘I
•
up
IS
a
ap
p
1 5
15
•9
5
I
5
•
4
I
I
.
P
.5
p
I ,
I
I
p
5
,
P
,
‘P
‘1
•
a)
‘I
P
a
•I
‘P
5
.1
.5
Ii
‘P
Ip
I
p
up
j
.I
IP
‘P
II
i
.p
i i
I
I I
85
•5
‘I
ap
‘1
•1
1
.,
l
•
I I
‘1
I
p
P
I
p
‘p
• ,
Il
.p
•
,
p
‘1
.
• I
n
‘I I 3
•n 3
• . 1
.15 3
.N I
‘0 I
‘N 3
‘N ‘.
.ui I
in 5
IN 3
Il S
‘ II 5
III 5
iii I
.n i
.n
N I
Ill I
an 1
P 5
.0 5
.n I
I, S
•M 1
I I 5
p 5
I
1N5
•
•a 5
• ,
lt I
‘I I s
...
...
.o S
1 5
I 3
•SI
15. 5
‘ii %
III S
. ;
.
...
III I
•“ s
, ,,
• I
0
III S
.,,
I S
•
• S
• P
0 I
15 1
N I
N I.
N I
N I
N 5
0 a
I’ I
a” p
all p
N 5
II S
s
is P
N
N 1
4’ I
‘N P
II
I. S
N S
0
5 1
p
I
PS
,. p
5. P
.
n I
o p
..
N I
5
fl 5
.115
, p
N I
N $
• , S
. p
... ,
.t. i
o
I, I
II I
. •
, ,
II ‘.
a
I
‘
I
I
I
1
‘
I
-5
I
I
p
I
,
5
I
I
P
I
5
P
P
I
p
I
I
5
p
-I
.1
.,
.5
.
‘P
p
..
5
‘
5
.5
p
P
p
p
,
p
I
.
‘
.
t
-I
•
‘
Ca
I?
II
I
I, . ,
I I
S
II
Sin
n
IN
i i
a
5
IN
C C
I
I
‘.5
‘
P
p
ii
35
II
5 1
. ,
P
IS
l .
LI
pp
.
p
‘
5
•I
II
is
i
S
,
.
I .
a.
,s
C’. ..
‘ -•
•
.II
ii
P
I
I
I
5
I S
I
a
I
14
IS
U
P
P
P
S
iS
S
I
ii
I
P
.5
ii
I
1
S
5
p
P
I
•
p
p
.
I
s
P
.5
g
P
p
SI
a
S
1
, ,
Il
5
9
11
“ P’ .
‘I P3
‘S II
.5
aS Ii
‘5. 11
‘nil
to I’
.0 )
1W II
‘U 5 5
5 5 P3
‘S 13
‘ 1)
a 53
Ill 15
‘5.11
15.11
.• 1 3
ii PP
‘ 11
‘i II
.5 H
is is
‘0 H
•5 II
5 55
‘S 15
•0 SI
.p • .
‘N P s
. Is
, ,
•.
...n
••fl
.,.,,
s P 5
. p ,
‘s n
. p
‘N.H
I 5
. , ,.p,
II I Ii
. 1
aN Is
‘U ii
.
•.. Ii
•.
‘U I’.
a,, 9%
•
H
PP
SI
p•
Pt
5)
i i
S
p
SI
pp
1
31
5
p,
PP
PS
PP
45
ii
PP
pg
pa
PS
PP
15
15
I,
p
5
,•
•
is
PP
as
5
pp
•
pp
,p
Pp
p5
IS
I;
su
ii
,
,
pp
•p
•P
‘I
ii
.g
p
.p
IP
ip
ii
P
a,
•
a,
p
•P
up
ii
‘p
‘p
si
5
‘
,,
•p
i
ii
C)
‘p
p
P
a,
,
,
p
.p
a5
p
ap
I)
, ,
p
,p
• ,
II
II
Ii
•
,p
•
up
•
•
,
p
p
• P
‘. I
. 5
a.
p
• p
u p
I. ,
11.1
• p
• P
‘. ‘
5 p
• I
i 5p
•.‘
‘S.i
•.p
• I
s.s
p.,
•.i
a• p
•‘ s
‘S P
ap 5
p 5
•• p
S.,
• p
isp
iS 5
, 5
.
, ,
• p
p
,
., , ,
., p
h.p
••
. p
• p
P P
a S
I5 P
• P
p p
, p
• i
. .
,, ,
, ,
.g I
p I
•
ii
.
•
,p
.p
‘I
a 5
•
• ,
ig
•,
•p
II
ip
•
•
.
ig
is
ip
i 5
ap
.
ii
ip
a,
ip
a,
.,
.5
II
g
•
,•
ep
ii
,,
,p
•
‘5
,g
•p
•p
‘I
I
•5
5
‘P
•
‘S .
•
.,
.,
ap
e
a ss
S
•
V.H
,
.,
•
...
p.p
•.s.
s.s
V SI
•
VU
•_
pp
•,,g
•.ss
•p•
...s
• •p
. ,,
• ,ia
• ,
• p
p p
•,.a
551
•,,
P 4
...
,
• si
• ,
I SI
5 IV
5.5,
.5 ,
•
p.,,
•
a •1
v
p
• I
.
• ,,
• It
. .
, ,.
. .,
p a
•p
S at •,. I II P -i 5% 1 151 P5
I ‘p .• 5 I p •
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTER! ZAT ION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contami nants:
Treatable Waste Matrices:
On-/Off-site Treatment Location:
Pre- and Post—treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Margaret Guerriero
U.S. Environmental Protection Agency
Region V
Waste Management Division
230 S. Dearborn
Chicago, Ii 60604
United States
(312) 886—0399
Volatilization
Vacuum extraction of soil vapor
U.S. Environmental Protection
Agency
Demonstrated
Petroleum fuels, solvents, halo-
genated and aromatic hydrocarbons,
mercury, heavy naphthas
Soil
In—situ
Extraction well Installation,
groundwater treatment, activated
carbon treatment of vapors
Verona Well Field Superfund Site,
Battle Creek, Michigan, United
States
5,800 to 7,500 kg halogenated
hydrocarbons, aromatics
Fine to coarse grained sands, 7
meters to water table
99.8% removal efficiency
$50 to $60 per cubic meter soil
11/89
3-2
165
-------
In-Situ Soil Vacuum Extraction System
Verona Well Field Superfund Site
Battle Creek, Michigan
Draft Final
Final Report for NATO/CCMS Pilot Study on Remedial Action
Technologies for Contaminated Land and Groundwater
Presented at the Third International Meeting
November 6—9, 1989
Margaret M. Guerriero
U.S. EPA Region V
166
-------
I. INTRODUCTION
Site Descrit,tion
The Verona Well Field is located in Battle Creek, Michigan in the
south central portion of the State. The Verona Well Field (VWF)
site consists of four distinct problem areas within approximately
100 acres. The municipal veil field is located in the northeast
corner of the City and lies on both sides of the Battle Creek
River (Figure 1). The veil field consists of 30 production wells
that supply drinking water for 50,000 residents arid several major
food industries. The Thomas Solvent Raymond Road (TSRR)facility,
the Thomas Solvent Annex (TSA) facility, and the Grand Trunk
Western Railroad (GTWRR) have been identified as sources of well
field contamination. Figure 2. shows the location of these
sources relative to the well field. The site is located in an
urban setting which is primarily residential with some light
industry.
Site History
The contamination problem at the VWF site was first discovered in
August 1981, during testing of the City water supply. Test
results revealed that 10 of the City’s 30 supply wells were
contaminated with volatile organic compounds (VOCs).
Concentrations ranged from 1 to 100 ugh. During the same
period, private residential wells were also tested. Several of
these wells were found to contain VOCs in excess of 1,000 ugh.
The highest level found in a private well was dichloroethylene
at 3,900 ug/l. A bottled water program was implemented in this
area and residents were connected to the City’s water supply
system.
In the Fall of 1983, the first phase of a Remedial Investigation
(RI) was initiated to determine the extent of contamination in
the well field and potential sources.. Sample results from the
initial RI work confirpied the existence of a contaminant plume
with VOC concentrations ranging from 1 ug/]. to 100 ugh in the
well field. The investigation also identified the three major
sources of contamination.
Remedial Measures
In May 1984, U.S. EPA signed a Record of Decision (ROD) calling
for an Initial Remedial Measure (IRN) to implement a blocking
167
-------
(c
0..
VERONA
WELL FIELD —
GRAND
TRUNK WESTERN
RAILROAD
MARSHALLING
YARD
I ,
F,
I,
I,.
THOMAS
SOLVENT
RAYMOND
ROAD
FACILiTY
/
MAS
SOLVENT
ANNEX
/ S
/
/
/
,
I
/ I’
F
I —
I—
‘I
168
FIGURE 1
VICINITY MAP
-------
well system in the well field. The blocking system consists of a
line of converted supply wells that extract contaminated water
from the southern end of the VWF. The system prevents
contaminants from migrating further into the well field. An air
stripper with vapor phase carbon treatment was also installed to
treat contaminated water prior to discharge to the Battle Creek
River.
In August 1985, U.S. EPA signed a second ROD that addressed the
most contaminated of the three sources, the Thomas Solvent
Raymond Road facility. The ROD included a groundwater extraction
system to remove contaminated groundwater, treatment of extracted
water utilizing the existing air stripper at the well field,
demolition of the existing warehouse and loading dock, and a soil
vapor extraction system to remove VOCz from contaminated soils.
Site Characteristics and Sampling Results
The Thomas Solvent Raymond Road (TSRR) facility is a former
solvent repackaging and distribution facility that operated from
1970 to 1984. Solvents were stored in 21 underground storage
tanks, of which 19 were later discovered to be leaking. There
has also been documented reports of surface spillage during
operation. The site contained an office building, a warehouse
with a loading dock, and the 21 underground tanks (see Figure 2).
Site geology consists of a fine—to coarse—grairied alluvial—
glacial sand with traces of silt, clay, and pebbles underlain by
a fine—to medium—grained sandstone with minor lenses of shales
and limestones. The unconsolidated sand unit ranges in
thickness from 10 to 50 feet and the sandstone varies from 100 to
120 feet in thickness. The hydraulic gradient is primarily north
to northwest from the identified sources toward the VWF. The
depth to water is approximately 20 to 25 feet. The hydraulic
conductivity of the unconsolidated material ranges from 2.7 x 10
to 4.0 x io2 cm/sec. The hydraulic conductivity of the
sandstone ranges from 7 x i0 to 2 x iO2 crn,’sec.
Samples collected at the TSRR facility indicate that both soils
and groundwater are highly contaminated with a variety of organic
compounds. Table 1 lists organic compounds detected in soils at
TSRR. Groundwater samples showed concentrations as high as
100,000 ugh VOCs. The total estimated volume of organics in
groundwater and soils was 3,900 lbs., and 1,700 lbs.,
respectively.
It should be noted that these total mass estimates were based on
sample data obtained using an accepted soil sampling procedure
which is now known to produce VOC results lower than actual
values. The total mass in groundwater and soils is now estimated
169
-------
0
0
0
z
0
——
U:
TYPICAL UNOERGROUNDJJr
STORAGE TANKS
(APPROXIMATE LOCATION$ (
OFFICE BUILDING
FIGURE 2
LOCATION OF
UNDERGROUND TANKS
TABLE 1
PRINCIPAL Sort CONTAMINANTS DETECTED AT TSRR
I0RINATED HYDROCARBONS MAX • CONC. (ua/ 1
- ) THYL E RIDE 60,000
- IflROFORX 2,000
- 1,2-DI LoROErMANE 27,000
— L,1 ,1—TRI WROETIiANE 270,000
- TR CHLOROETHYID1E 550,000
- TETRA iI ROETHYL E 1,800,000
AROMA?I S
— TOWENE 730,000
— XYLENE 420,000
- ETHYL BENZENE 78,000
— NAPHTHALEKE 9,400
MES
- ACETONE 130,000 170
- METHYL ETHYL KETONE 17.000
ThOMAS SOLVENT
SUILDING
LOADING DOCK
-------
to be significantly greater based on the results of the operating
groundwater extraction system and the soil vacuum extraction
system.
Conventional subsurface soil sampling procedures involving the
use of split spoon samplers require the sampler to be opened and
the sample transferred to a bottle prior to shipment to the
laboratory. This allows for significant amounts of VOCs to
volatilize before analysis. This problem, coupled with the lack
of sampling in the capillary zone and beneath the former
warehouse building, resulted in estimates of VOC—contaminated
soils being considerably lower than actual. This has had a
significant effect on the operation of the soil vacuum extraction
system. These effects will be discussed in detail in a later
section of this report.
Technology Selection
Due to the significant mass of contaminants in the soil and
groundwater at the TSRR facility, alternatives that employed both
groundwater and soil remediation were developed in the
feasibility study. A two step approach to remedial action was
used at the TSRR facility in which each alternative developed for
the feasibility study included both a groundwater and a soils
portion. The selected alternative for the site includes a
groundwater extraction (GWE) system and the soil vacuum
extraction (SVE) system.
The groundwater extraction system includes 9 extraction wells
which pump a total of 300 gallons per minute. The extracted
water is pumped to the existing air stripper at the well field
and discharged to the Battle Creek River after treatment. Figure
3 shows the layout of the GWE system. Initially, VOC
concentrations in groundwater were so high that extracted water
was passed through pretreatment carbon units prior to being
pumped to the air stripper. The system has been operating since
March of 1987.
Several alternatives for soil cleanup were evaluated, including
SVE, excavation of soils with on/off site disposal, site capping,
and soil washing (flushing water through the unsaturated zone
with subsequent groundwater extraction).
SVE was chosen based on a number of reasons. Although it was
considered an innovative technology, it was felt that it had a
good likelihood of success given the site conditions and
contaminants. Excavation was considered unacceptable due to its
potential to release high concentrations of VOCs into the air,
which would create a health hazard to residents in close
proximity to the site, and significantly violate Michigan Air
Quality Standards for VOCs. Therefore, alternatives that would
171
-------
A EW’
MONITORING BUILDING
A
EW,
TYPICAL
L
THOMAS SOLVENT
BUILDING
(DEMOLISHED)
LOADING DOCK
(DEMOLISHED)
SVE PROCESS BUILDING
L’
OFFICE BUILDING
FIGURE 3
GWE SYSTEM LAYOUT
172
-------
not disturb soils were favored. Capping was not considered to be
consistent with future actions because of the high level of
contamination present at the site and because the underground
tanks would eventually have to be removed.
Of the two soil treatment alternatives, SVE was calculated to
take less time to remediate the site than soil flushing. It was
estimated that SVE, in conjunction with GWE, would reduce the
groundwater contamination to 100 ugh within three years. The
contaminant mass would be reduced to 2% within 1 1/2 years (based
on 1700 lbs. of VOCS). Soil washing was estimated to require 8
years to reach 100 ug/l in groundwater and 8 years to reduce
contaminant mass to 2%. This was significant in the selection of
SVE because it was not less expensive than soil washing. The
estimated capital cost of SVE was $413,000, with operation and
maintenance (0&M) of $90,000. Soilwashing capital cost was
estimated at $58,000, with O&M of $6,000. SVE was, however,
considerably less expensive than the excavation arid capping
alternatives.
Since SVE is an innovative technology, the procurement of a SVE
contractor was accomplished using a performance specification
which contained certain minimum requirements but left the major
design details to the discretion of the bidding SVE contractors.
Contract documents called for the construction, operation, and
maintenance of the SITE system. The performance standards
require that the SVE would operate until all soil samples showed
VOCs below 10 mg/kg, with no more than 15 percent of the samples
above 1 mg/kg total VOCs.
II. TECHNOLOGY
Process Description
The soil vapor extraction process is designed for use in
removing VOCs from the unsaturated zone in soils. The mechanism
by which SVE operates is fairly simple and straight forward. The
system is designed to create negative pressure in the unsaturated
zone using wells that are connected to a vacuum extraction unit.
The vacuum induces a flow of air through the soils, thereby
volatilizing VOCs that are absorbed on soil particles and
extracting the contaminants in the vapor phase.
A vacuum extraction system generally consists of several
extraction wells screened throughout the unsaturated zone or
within discrete units of the unsaturated soils. The wells are
connected by transfer pipes which are manifolded to the vacuum
extraction unit. The vacuum applied at the welihead creates a
negative pressure or vacuum in the subsurface which cause VOCs to
volatilize and migrate to the extraction wells. A vapor/water
173
-------
separator is also incorporated to remove water from the
contaminated air stream.
The materials used for an SVE system are generally readily
available and not specialized. Equipment requirements include
PVC/stainless steel wells and piping: conventional vapor phase
carbon treatment units; a conventional air/water separator; and
an induction blower.
The design of the SVE system is critical in order to insure
adequate cleanup of the soils. The design requires expertise in
modeling vapor flow, understanding site lithology, determining
contaminant mass and areal extent of contamination. These
factors are used to determine system variables such as well
spacing, number of wells, and depth of screened interval.
Site conditions, soil properties, and the contaminant’s chemical
properties are the important factors to consider in determining
whether to use soil vapor extraction. Information on soil
permeability, moisture content, and porosity is needed to make a
determination on whether the soils have sufficient air—filled
porosity. Insufficient air-filled porosity results from the
presence of excess water in the pore spaces which reduces the
effectiveness of vacuum extraction. Depth to groundwater is an
important cost consideration because if the vadose zone is less
than 10 feet, it may not be cost-effective to use SVE (excavation
of less than 10 feet may be less costly).
In order for a contaminant to be stripped from the soil using
SVE, it must have a Henry’s Constant of 0.001 or greater. The
higher the Henry’s Constant, the easier the compound is removed
by vacuum extraction. Figure 4 shows relative extraction rates
for compounds found at the TSRR facility.
Design and Construction of the TSRR System
Prior to full-scale construction of the SVE system at TSRR, a
preconstruction investigation was performed. The investigation
included a geophysical survey, a soil sampling program, and a
soil gas survey.
The geophysical survey was conducted to confirm locations of
underground tanks and to check for additional buried objects in
the effected area. The soil sampling program was conducted to
investigate the horizontal and vertical extent of contamination,
and to estimate the mass of VOCs in the vadose zone. As a result
of this activity, a revised estimate for VOCs ranging from 13,000
to 16,500 pounds was calculated. This did not include estimates
for a floating product layer that was discovered during the
sampling. The objective of the soil gas survey was to
investigate the extent of VOC contamination in shallow soils in
174
-------
11
1 ,2DCA
47.
MIBK
Toluene
FIGURE 4
Relative extraction rates.
Compound
316
1500.
1000-
500
0.
17
A tone
11
11-DCA
MeCI
TCA
ICE
42
PCE
Xylenes
E-Benzene
Other
VOCs
175
-------
areas not previously investigated at the site. Results confirmed
that the major area of contamination was in the vicinity of the
underground tanks and loading dock, however, contamination was
also detected along the northeast and southern parameters of the
site.
Results of the investigation were used to determine locations of
additional SVE wells, revise estimates of the mass of VOCs in the
soils, and to make determinations on system parameters.
Following the preconstr-.iction investigation, a pilot phase SVE
system was installed. The system consisted of 11 wells that were
operated for a total of 70 hours. The objectives of the pilot
phase were to verify the radius of influence of the wells and
determine the vapor flow rate/vacuum pressure relationship at
each well, investigate the effect of the underground tanks on the
vacuum pressure distribution in the vadose zone, and identify the
VOC loading rates form individual wells as a function of vacuum
pressure and flow rate.
Results of the pilot phase were used to determine process
variables and locations of wells. Extracted airflow rates range
from 60 to 165 standard cubic feet per minute (scfm) from
individual wells, with welihead vacuums of 3 to 4 inches of
mercury. VOC extraction rates vary between wells with the
highest measured concentration at 4,400 lbs/day during the pilot
phase. The radius of influence for the wells was measured to be
greater than 50 feet.
The full-scale system consists of a network of 23 4—inch diameter
PVC wells with slotted screens from approximately 5 feet below
grade to 3 feet below the water table. The wells are packed with
silica sand and sealed at the screen/casing interface with
bentonite and then grouted to existing grade to prevent short
circuiting. Each well has a throttling valve, a sample port, and
a vacuum pressure gauge. The wells are connected by an above
ground PVC piping manifold system. Figure 5 shows the location
of the SVE wells and the piping layout.
The piping manifold is connected to a centrifugal air/water
separator. This is connected to the vapor phase carbon
absorption system which is followed by the vacuum extraction
unit (VEU). The VEU is responsible for inducing extraction of
soil vapors from the subsurface, through the extraction wells and
into the treatment unit. After treatment, air is discharged
through a 30 foot stack located on site. Figure 6 is a schematic
of the SVE system.
The carbon absorption system consists of eight stainless steel
carbon canisters with four in the primary system and four in the
secondary or backup system. The primary carbon system is the
main unit for absorption of VOCs, with the secondary carbon
176
-------
FENCE LINE
MONITORING’
BUILDING
0
z
0
>.
AIR/WATER
SEPARATOR
FIGURE 5
SVE WELLS AND PIPING LAYOUT
INCH HEADER PIPING
SOIL VAPOR
CCTRACTION WELL
SVE PROCESS
BUILDING
177
-------
system acting as a backup when breakthrough of the primary system
occurs. Each carbon canister holds 1000 pounds of vapor—phase
granular activated carbon and are connected to the header piping
with flexible hoses and couplings that are easily disconnected
for ease in canister change outs. Figure 6 shows the various
sample ports, pressure gauges and temperature probes located
before, between, and after the carbon units. A carbon monoxide
monitor is installed between the carbon units to detect
combustion in the primary carbon unit and trigger an automatic
system shutoff upon detection.
The carbon system was installed on the negative pressure side of
the VEU to minimize leaks and eliminate the potential for
emissions to the atmosphere. During the pilot phase of operation
it was determined that carbon adsorption efficiency was
equivalent under positive and negative pressure.
Breakthrough of the primary carbon system is measured by an in-
line KNu photoionization detection meter. Four contaminants are
used as indicator compounds, tetrachioroethylene,
trichioroethylene, methylene chloride, and benzene. The
breakthrough point was determined using the relationship between
total VOCs measured by the HNu and the compound-specific
concentrations measured in the air flow. When breakthrough
occurs, the primary carbon canisters are changed out and replaced
with those from the secondary system and a new set of four
canisters are put into backup in the secondary system. This
allows maximum loading of the primary carbon system prior to
rotating the carbon units, while minimizing the possibility of
breakthrough in the backup system.
Samples are collected from both carbon systems as well as at the
individual weliheads. Results are used to determine VOC loading
rates and predict rates of breakthrough. Sample analysis is
performed on—site using a gas chromatograph with dual flame
ionization detectors and capillary columns.
III. RESULTS
The SVE system began full operation in March 1988. Results
discussed in this report are for the period of March 1988 through
September 1989.
Vacuum Extraction System Performance
To date, approximately 40,250 pounds of VOCs have been removed
from the soils. On—site gas chromatography has been used to
monitor VOC concentrations extracted by SVE. Off-site analysis
of spent carbon has confirmed that on—site monitoring is accurate
to within approximately 5 percent.
178
-------
PRESSURE INOICA TOR
1EMPERA lURE INDICA TOR
SAMPLING PORT
FLOWME ER
FIGuRE 6
SCHEMATIC OF SOIL
VAPOR EXTRACTION
SYSTEM
SI lENCER
—S
0
L EGEMD
I I
I I
I I
L --.—-J
VACUUM EXIRAC liON
UNIT
-------
The initial loading rate of total VOCs has declined over the
period of operation from an initial level of approximately 45
pounds per hour (pph) to below 10 pph. The floating product
layer that was detected during preconstruction and during the
initial operation period has not been detected since October
1988; however, at that same time, a 0.5— to 1.0 foot increase in
the water table was recorded.
The average VOC concentrations measured at the air discharge
stack is approximately 1.35 mg/l, with an average flow rate in
the stack of 1060 standard cubic feet per minute (scfm). This
has dropped from an initial VOC concentration of approximately 23
mg/i. Over the course of operation of the system, an average
efficiency rate of greater than 99.8 percent removal has been
measured.
Technoloay Evaluation and Performance Monitoring
Since SVE is an innovative technology, careful consideration was
given to the method by which the system’s performance would be
monitored and to confirm that the performance objectives would be
met. A Quality Assurance Project Plan (QAPP) and Sampling Plan
were developed for the sampling events. Three soil sampling
episodes were planned. One prior to startup, one at the mid—
operation point, and the last to confirm that performance
objectives have been met.
Since conventional soil sampling methods cause volatilization of
VOC5 prior to analysis, a special sampling and analysis
procedure was developed for collection of samples. Samples are
taken by driving a split spoon sampler fitted with four, 3-inch
brass liners, through hollow stem augers. To prevent excess
handling, and thus volatilization of contaminants, one brass
liner is removed from the split spoon, immediately wrapped in
aluminum foil, sealed, and sent to the laboratory for analysis.
Samples are analyzed using a core of the undisturbed sample for
extraction.
To date, the pre—operational and mid—operational sampling events
have occurred. The pre—operational samples verified that the
volume of VOCs in the soils had been underestimated during the
remedial investigation at the site. Based on sample results,
VOC concentrations were estimated to be between 12,800 and 16,500
pounds. This did not include estimates for the floating layer of
product that was identified during the startup work.
Data from the mid—operational sampling event have not yet been
received from the laboratory. It is hoped that this data will be
available for incorporation into the final version of this
report.
180
-------
Process factors
Carbon handling requirements have been the limiting factor in
performance of the SVE system. Because the estimate of VOCs
present in the soils was significantly underestimated, the
amount of carbon needed was also underestimated. The amount of
contaminants extracted to date has resulted in the use of 250,000
pounds of carbon in the treatment system at a cost of $541,000.
It is estimated that a total of more than 400,000 pounds will be
needed to complete the propect at a cost of approximately
$886,000.
In addition to the increased costs, the additional carbon
requirements have caused delays in the operation of the system.
Although the system has been operational for more than 18
months, actual number of days of operation is approximately 100.
This due to the need for frequent carbon change outs and
transporting the spent carbon off—site for regeneration. It is
estimated that an additional 50 days will be needed to attain the
levels required in the performance objectives. It is also
expected that carbon change outs will become less frequent as the
loading rates decline.
The equipment needed to operate the system has proven to be very
reliable and down time due to equipment failure has not been a
factor in SVE operation. As discussed, the materials used to
operate the SVE system are conventional and easily replaced if
necessary. Although Terra Vac, the vendor, has been required to
be on site for 8 to 10 days per month due to the frequent need
to change out carbon and monitor the system, the system was
designed for unattended operation. It is expected that as
loading rates decline, Terra Vac will be required to spent less
time at the site per month.
Instrumentation and controls have been installed to monitor the
system and trigger shut down if necessary. These include the
carbon monoxide monitor in the carbon system, a high water—level
shut down in the air/water separator, high temperature shut down
triggers, and an on—line HNu with a shutdown mechanism for
detecting VOC breakthrough of the primary carbon system. In
addition, the system contains an automatic dialer that contacts
Terra Vac when any of the shutoff mechanisms are triggered.
Costs
A summary of the costs to install and operate the SVE system,
current and projected future carbon costs, and the unit costs for
operation of the system are listed in Table 2. It was not
possible to separate out the cost of the pilot phase portion from
the cost of the full—scale system because the bid was received as
181
-------
TABLE 2
SOIL VAPOR EXTRACTION COSTS
SVE Lump Sum Bid — For $1,265,535
Construction & Operation
(excluding cost of carbon)
Cost of SVE/Cubic Yard of $22.50
VOC-Contaminated Soil
(excluding cost of carbon)
Unit Cost/pound for Carbon $ 2.16
(removal/transport/regeneration)
Carbon Costs to Date $541,000
(250,000 pounds used)
Projected Total Carbon Costs $886,000
(estimated 400,000 pounds)
182
-------
a lump sum for the project. The original cost estimate was
revised to account for the additional days per month Terra Vac
is required to be on-site, due to the increased contamination at
the site.
As previously discussed, carbon costs have been quite high due to
the increased level of contamination found at the site.
Initially, it was estimated that 20,000 pounds of carbon would be
needed to remediate the site. To date, 250,000 pounds have been
used and it is estimated that an additional 150,000 pounds more
are needed to complete the project. Table 2 lists the actual and
projected future carbon costs for project completion. No long
term maintenance costs are expected.
Lessons Learned
Vapor Treatment
As has been discussed throughout this report, the underestimation
of the total mass of VOCs in the soils at this site has
complicated the remediation of the site. The increased levels
have effected the project expense, the time to remediate, and the
operation of the technology.
During evaluation of the treatment options for the project, it
was determined that, based on a total VOC volume of 1,700
pounds, carbon absorption was the least expensive treatment
option. If contamination estimates were closer to the actual,
carbon absorption would not have been the least costly and would
likely not have been considered. In addition, the cost of
operating the system is more expensive because Terra Vac
must be at the site many more days per month than estimated.
The underestimation of VOC mass has also effected remediation
time and the operation of the technology. Due to the frequent
number of change outs required during operation, the system is
operational as little as five to ten days per month. This has
resulted in only 100 days of system operation in a period of 18
months.
In order to prevent this situation, it is imperative that
accurate estimates of subsurface contamination be obtained prior
to design of the system. Specifically, accurate mass estimates
must be obtained for amount of floating product present, and the
amount of VOC contamination in the capillary fringe, the zone
immediately above the water table, and in the smear zone, the
zone within which the water table fluctuates. Based on data
collected during operation of the SVE at TSRR, it was estimated
that 70 percent of the mass of VOCs occurs as floating product
and in V0C saturated soils in the smear zone and capillary
fringe.
183
-------
Radon Gas
A somewhat unexpected contaminant detected was radon gas, which
occurs naturally in the site soils. Measurements of the carbon
vessels indicate the presence of radioactivity on the spent
carbon. The presence of radon gas is not too unusual since it is
readily volatilized and activated carbon is a good collection
medium for radon. Concentrations measured to date at the TSRR
facility are not considered to present a public health or worker
hazard. However, the handling, transportation and regeneration
of radioactive spent carbon may need to be considered for SVE
operation in areas where radon occurs at high levels.
Future Plans
U.S. EPA’s contractor, CH 2 M Hill, is currently evaluating the use
of a catalytic oxidation (CATOX) system for the destruction of
VOCs in the soil vapor. This would replace the carbon absorption
system. While other treatment options have been looked at during
the life of the project, the cost for removing the carbon system
and installing another treatment system has not been shown to be
cost—effective. However, cost—effective CATOX systems have
recently been developed that can treat chlorinated VOCs without
generating dioxins or suppressing catalyst performance.
In addition to the reduction in cost to treat contaminants, two
other major benefits from switching to a CATOX system are the
destruction of contaminants on—site, which eliminates the
transporting of wastes off-site, and the ability to run the SVE
system continuously, thereby attaining site cleanup faster.
IV. CONCLUSIONS
Since the project is still in operation, certain conclusions
sould not be considered definite. However, based on evaluation
of the operating data from the site and on the recent literature
regarding SVE, the following conclusions have been drawn:
* SVE is a viable technology for the removal of VOCs in
unsaturated soils. The fact that over 40,000 pounds of VOCs
have been removed from the soils at TSRR indicates that the
technology works.
* SVE will operate in a wide range of soil types. Based on
work at the site, SVE is very effective in removing
contaminants from sandy soils. Recent literature on SVE
performance indicates that it is effective for soils with
measured permeabilities of i0 to i cY 8 cm/sec.
* The major considerations in determining the technology’s
184
-------
applicability are soil properties, depth to ground water,
and the contaminant’s chemical properties. Soils that have
a low air-filled porosity and high moisture content may not
provide adequate conditions. In addition, at sites where
groundwater is encountered at less than 10 feet, it may be
more cost—effective to excavate contaminated soils.
Chemicals with a Henry’s Constant of less than 0.001 may not
be sufficiently volatile for the SVE process.
* The SVE system has operated well in all weather conditions
at the site. Cold weather operation has not proved to be a
problem. The system has operated through an entire winter
in the midwest with temperatures that range from 0 degrees
celsius to —26 degrees celsius.
* The SVE system can be designed to operate for the majority
of the time without an on-site operator. Under most
circumstances, the system would be sized to provide
unattended operation with vendor personnel on—site 1 to 4
days per month depending on the size of the system and the
monitoring requirements.
* Based on experience at the TSRR facility, SVE appears to be
the only technology that can effectively remove the VOC
saturation remaining after free product is removed from the
capillary fringe and smear zone of VOC-contaminated soils.
* The costs of SVE at the site for 1 cubic meter of soil is
approximately $50.00 to $60.00. This includes the cost of
treatment of vapors using carbon absorption. If treatment
of vapors is not required, costs could be as low as $20.00
per cubic meter of soil.
The overall prognosis of the SVE process is that it offers an
economical, reliable, and rapid cleanup technology for
remediating soils contaminated with volatile organics. The
technology enhances groundwater extraction systems and greatly
reduces the time and cost for groundwater reinediation. The
process works on most soil types and has a limited number of
factors for consideration in determining applicability. There is
no limit on size of the site, or on the level of VOC
contamination (except in considering the need for treatment of
off—gases). The system is easily installed and removed, and does
not require specialized equipment for operation.
V. CONTACTS FOR MORE INFORMATION
Information useful to potential SVE technology users can be
provided by the following sources:
185
-------
Government :
Margaret Guerriero
Remedial Project Manager
U.S. EPA
230 S. Dearborn 5HS—1].
Chicago, Illinois 60604
(312) 886—0399
Site Engineer :
Joseph Dartko
Project Manager
CH 2 M Hill
2300 N.W. Walnut
P.O. Box 428
Corvallis, Oregon 97339
(503) 752—4271
Vendor :
James Malot/Ed Maliuanis
Terra Vac Inc.
4897-J West Waters Ave.
Tampa, Florida 33634
(813) 885—5374
VI. REFERENCES
1. Danko, J., Soil Vapor Extraction at a Superfund Site, CH 2 M
Hill, Corvallis, Oregon, undated.
2. Danko, J., Soil Vapor VOC Removal System at the Verona Well
Field superfund Site City of Battle Creek, Michigan, CH 2 N
Hill, Corvallis, Oregon, March, 1989.
3. Tanaka, J., Verona Well Field Superfurtd Site Battle
Creek, Michigan Soil Vapor Extraction System, U.S. EPA,
First International Meeting of the NATO/CCMS Pilot Study
Demonstration of Remedial Action Technologies for
Contaminated Land and Water, November, 1987.
4. U.S. EPA Office of Research and Development, Terra Vac In
Situ Vacuum Extraction System, Applications Analysis Report,
July, 1989.
186
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTERIZATION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On-IOff—slte Treatment Location:
Pre- and Post—treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Herr Wolf
Freje und Hansestadt Hamburg,
Umwel tbehorde
Amt für Aitlastensaneirung
Hermannstrabe 40
0-2000 Hamburg
Federal Republic of Germany
Soil treatment by extraction —
in—situ
High pressure soil washing and
oxidation
Arbel tsgemel nschaft Bodensanierung
Demonstrated
Aromatic hydrocarbons
Soil
In—situ
Soil—crete solidification of soils
underlying foundations
Disinfectant factory site,
(Goldbek Haus) Hamburg, Federal
Republic of Germany
Phenol, kresol
Sand and peat layers, sandy layers
filled with groundwater in contact
with adjacent canal
98% removal of phenol
Not available
Dr. Sondermann
GKN Keller GmbH, Spezlaltleflau
Kalserlelstr. 44
D-6050 Of fenbach 12
Federal Republic of Germany
069/8051-213
11/89
4-a-i
187
-------
NATO/CCMS PILOTSTUDY
International Meeting
6. — 10. Noveiber 1989
Montreal. Canada
High Pressure SoilWashin
and Soil Treatment by Extraction
— INTERIM RE PORT
by order of
U weltbehörde Rai burg
— A t fUr Aitlastensanierung —
Arbeitsge einschaft Bodensanierung
GKN Keller GmbFI, Spezialtiefbau
S + I Schlammentwässerung GmbH + Co. KG
WIJE Umwelt Engineering GmbEl
November 1989
188
-------
1. Indroduction
1.1 Site description
1.1.1 Geographic location and setting
1.1.2 Soil classification and geology
1.2 Site History
1.2.1 Responsible parties
1.2.2 Dates of contamination
2.. Technology
2.1 High pressure Soil Washing (in situ)
2.2 Soil treatment by extraction (on site)
3. Tests and Results
4. Conclusions
4.1 Applicability and limitations
4.2 Prognosis
5. Contacts for more information
Enclosure
Diagram 1 Plan view of the area
2 Subsoil conditions
3 Contamination within the subsoil
4 Spread of contamination
5 Technique of treatment and jet cutting methods
6 Location of test area
7 Parameter of execution
8 Results of washing
9 Results of treatment
10 Results of core—drilling and tests
189
-------
1. Introduction
1.1 Site description
1.1.1 Geographic location and setting
The site is located in the city of Hamburg in the
district of Winterhude. The site, with an area of about
5000 m 2 , is bordered by the canal “Goldbek—Kanal” on the
south side; and three old factory buildings are now
built on it (diagram 1). One of the former three fac-
tory buildings is now used as a community centre (build-
ing at the side of the Goldbek—Kanal) and the other two
are rented to various commerical enterprises.
Only the ground—floor—level of the east—building—wing
is not in use (at the moment).
The east—wing is now also to be used as the part of the
community centre, so it is planned to reconstruct and
renovate this part of the building.
1.1.2 Soil classification and geology
The area is nearly horizontal and varies between Nor-
mal—zero CNN) + 4.2 m and NN + 5.0 m.
The subsoil conditions can be described as follows:
Beginning with an approximately 4.0 m thick layer of
filling material, normally sand, a thin “upper peat”
layer follows with a thickness of 0.5 to 1.0 m. This
peat changes into a sand layer. Underneath, this is
followed by the second lower peat, mud or lime—mud layer
with a thickness of between 0 and 8.0 m. The ground is
made up of sand and glacial—drift/boulder clay and
marlaceous soil (diagram 2).
190
-------
The sandy layers are filled with ground water. The
ground water level was measured between NN + 2.8 and
about NN + 3.0 m, that means 1.4 xn to 2.0 below ground
level.
The ground water level is in contact with the Goldbek—
Kanal. The boulder clay acts almost as a seal and sepa-
rates the upper ground water from the second ground
water horizon, which begins about 50 m below ground
level. The subsoil conditions and a description of the
soilproperties are given in diagram 2.
1.2 Site History
1.2.1 Responsible parties
On the site between 1889 and 1963 a chemical factory
manufactured disinfectants.
Because of the previous chemical production on the area
and the improper handling of chemical materials over a
longer production period phenol, a chemical substance
which was used to produce disinfectans, penetrated the
subsoil and the ground water and caused contamination.
Today the factory area is owned by the city of Hamburg.
1.2.2 Dates of contamination
The contamination was measured as the concentration of
phenols (mg/kg = ppm), mainly phenol and the three
isomere of kresol, within ground water or soil (dry
mass).
With 32 borehole to a final—depth between 2 and 10 m and
6 borehole as gauges (depth between 9 and 62 m), the
contamination was recorded.
191
-------
The recorded phenol concentration within the different
boreholes is shown in diagram 3.
The spread of contamination at different levels under
the ground surface is shown in diagram 4. The contami-
nation is concentrated between two buildings and under
the buildings. Because of the contamination under the
buildings it was also necessary to find methods of
treatment for the contaminated soils without changing
the ground conditions for the foundations.
1.3 Technology Selection
Because of the high phenol contamination and the consi-
derable odour of the chemical substance, on—site techni-
ques using an open pit are not practicable and workable.
Biological treatment, which were carried out as a trial,
proved unsuccessful, because of the very different soil
conditions (sand peat, mud with different organic con-
tents and thickness) and rapid changes in contamina-
tion.
For these reasons it was necessary to create in—situ/
on—site methods of treatment also for the soil
underneath the buildings.
The joint venture “Arbeitsgerneinschaft Bodensanierung”
has developed a process of treatment for these contami-
nated soils with the following qualities:
— no direct contact with contaminated material
— no odour during decontamination process
whole process self—contained
— no pit or opening to treat contaminated material or
earth works
192
-------
— treatment of soil possible also under buildings
— no ground water lowering, because peat and mud layers
achieve settlements for the buildings
— no transport of contaminated material necessary
All these advantages open a wide field of application
for this technique of treatment.
2. Technology
2.1 High Pressure soil washing (in situ)
After drilling down a borehole to the final depth using
normal drilling technique, a jet cutting technique is
used to erode and wash the surrounding soil. To cut,
water with a pressure of about 300 — 600 bar is used.
The water is pressed through a nozzle and reach a speed
of about 200 rn/sec. With this water—jet—stream the
surrounding soil is eroded, mixed with the water while
the jet—cutting monitor rotates and is pulled up at a
constant rate.
The mixture of soil and water follows up the borehole
to the top level. The volume of the treated soil as a
column depends on the withdrawal rate and rotation.
The contaminated mixture is leaded into a completely
self—contained system.
After the treatment (see chapter 2.2) the soil is sepa-
rated into two classes. Class 1 material is brought back
into the jet—cut columns.
Therefore the class 1 material as cleaned soil is put
into a suspension of filling aggregates. After a column
has been jet—cut 1 this mixture oE soil and suspension is
placed into the column using the tremie method.
193
-------
After separation the class 2 material will be placed
into a filter press. The water is pressed out and a
nearly dry soil with a high unit weight is then left.
The water which was pressed out of the soil is without
of contamination and can be given to further use.
2.2 Soil treatment by extraction (on site)
This suspension of contaminated soil and water produced
by jet—cutting, completely self contained, is treated
and cleaned of phenolic contamination by oxidation.
To achieve the decontamination by oxidation chemical
substances were mixed into the suspension according to
the degree of concentration. The quantity and sequence
depends also on the degree of contamination.
Also the air coming up the borehole is absorbed and
cleaned so that the surroundings are not subjected to
any fumes or smells.
With this technique the contaminated mixture of soil
and water within the column is displaced by treated
soil within a suspension with a higher weight. The
contaminated soil/water which is forced out, is also
treated (diagram 5).
Underneath the foundations and buildings a variation of
processing is necessary. At the same time as the soil is
eroded by jet—cutting, a fill with suspension and har-
dening aggregate is mixed with the water soil compo-
nents below. This process is characteristic of the
SOILCRETE—method.
194
-------
3. Tests and Results
On the areas of the “Goldbek—Haus” between 31.3.88 and
8.9.88 five tests were executed with different execution
parameters.
The location of the test field is shown on diagram 6.
The boreholes beneath this test area and the beginning
concentration of contamination are shown on diagram 3.
With the technique described already five columns have
been executed.
The following table (diagram 7) shows the paramters of
execution for the different test columns.
Column
Execution
technique
Date
of
Execution
A
SOILCRETE
6.
9.88
B
1. washing,
eroding
31.
8.88
5.
9.88
C
2. Treatment
3. Filling
D
7.
9.88
E
8.
9.88
Diagram 7 Parameters of execution
Column A was executed as a normal SOILCRETE—column with
a stabilisation at the same time, while for the columns
B to E erosion and washing was executed as a frist step
and afterwards the remaining material in the colums was
a displaced with a clean mixture.
The results of the erosion and washing process for co—
lumn A to E and the quality control during the process-
ing is shown on diagram 8.
195
-------
RESULTS of HIGh PRESSURE SOIL WASHING
‘ -I
‘ .0
o ’
Test Co1u n
A
B
C
D
E
DRILLING
Waler
m 3
1,1
0,8
1,1
0,6
0,5
HP SOIL!
Water (D)
m 3
8,8
3,8
1.
2,0
2.
2,0
3,6
WASUING
Water (M)
m’
3,9
1,7
0,9
0,9
1,7
SOILCRETE
Water
m 3
2,9 + 4,2
W/2
0,67
Suspension
Water
“ Cement
m 3
m 3
to
3,75
2,4
3,8
FILLING
Consumption
m 3
4
8,5
5,0
6,0
Water
Sand
Lime St.
Fly Ash
Cement
Ca.Bent.
m 3
to
to
to
to
to
1,6
5,5
1,0
—
0,25
—
3,3
11,0
1,8
—
1,0
0,4
2,0
6,4
1,1
—
0,6
0,25
2,4
7,7
—
1,3
0,7
0,3
VOLUME for TREATMENT
m 3
12,0
17,5
15,0
11,4
11,8
Diagram 8 Results of Washing
-------
The result of the following treatment of the contamina-
ted soil is shown on diagram 9. These results show that
an about 98 percent removal of the phenol—contamination
was possible, both in the water as well as in the soil.
The time of reaction and decontamination depends on the
degree of contamination the pH—value the contents of
organic—material and the water—contents. With lower wa-
ter—contents longer time for decontamination was needed.
After the treatment the separation of the decontamina-
ted material follows.
Soil class 1 material as the coarse grain component
show a higher concentration of phenol as the finer one.
This fact is a result of the technique of separation.
For this trial, the separation was done only mechani-
cally, so that also a higher contents organic coarse
material was found in the class 1 material. For this
reason during this trail it was not possible to coarse
material back into the columns.
During the entire treatment the material was completely
self—contained so that no extra measure for emmision
and protections of workers and maschinery were
necessary.
The results of core—drilling and following tests with
this core material is shown on diagram 10. As these re—
suits show the levels of phenol and cresol found in the
soil were well below the required levels. Thus the pro-
cess described here confirms the sucess of a procedure
using the combination of a jet stream process with an
immediate direct decontamination and recycling the clean
material.
197
-------
0
Column
mass for
production
m 3
mass for
treabnent
m 3
Separation
m 3
heno1
Input
mg/i
concentration
output
mg/i
Time
reactio
m m
foi
B
Drilling
17,5
0,8
24
Coarse grain
0,6
760
20
32 mg/kg
120
Washing
12,7
Fine grain
2,4
24 mg/kg
Filling
4,0
Filtration
21,0
5 mg/i
C
25,0
37
1560
20
180
Drilling
1,1
Coarse grain
0,4
i T mg/kg
washing
5,5
Fine grain
3,6
60 mg/kg
Filling
8,5
+
10
Filtration
33,0
9,8 mg/i
D
Drilling
11,4
0,6
22,2
Coarse grain
0,2
2035
20
53 mg/kg
35
Washing
Filling
5,8
5,0
Fine grain
Filtration
3,0
19,0
35 mg/kg
19,8 mg/i
E
11,8
24
1585
20
120
Drilling
05
Coarse grain
0,4
417 mg/kg
Washing
Filling
5,3
6,0
Fine grain
Filtration
3,6
20,0
150 mg/kg
12,4 mg/i
A
12,0
24
633
20
150
Drilling
1,1
SOILCRETE
Coarse grain
1,0
125 mg/kg
Water
2,9
+
4,2
Fine grain
3,6
159 mg/kg
Suspension
3,75
Filtration
15,4
9,7 mg/i
Diagram 9 Results of Treatment
-------
4. Conclusions
4.1 Applicability and limitations
This method of high pressure soil iashing and treatment
by extraction can be used to treat any soil which can be
cut with high pressure, that means nearly all types of
soil and also soil with a high clay contents or high
contents of organic material. The method of treatment
depends on the kind of contamination and has to be
checked for each different application.
4.2 Prognosis
With the technique of water—jet—cutting, chemical treat-
ment and separation of contaminated soil and ground
water, a lot of contamination sites can be purified.
The advantages of this technique and process are:
— no pits have to be build to treat contaminated
soils,
— no additional earth work is required,
— no ground water—lowering is necessary,
— no smell because all air is absorbed,
— no changes of process underneath buildings,
— process can be used for any form of contamination
with a uniform chemical treatment,
— no contaminated soil or water has to leave the site,
— all steps of the process are self—contained, no
contamination can escape into the atmosphere,
— stabisisation yriderneath foundation without changes
of technique,
— the amount of soil material whLch has to removed is
minimised.
199
-------
With these advantages, a wide field of application is
developed. The technique is applicable for a wide range
of contaminated soil and allows treatment also on sites
with buildings.
5. Contacts for more Inforation
5.1 Client
Freie und Hansestadt Hamburg, Umweltbehörde
— Amt für Altiastensanierung —
HermannstraBe 40
D—2000 Hamburg 1
Herr Wolf D — 040/34913— 445
Herr Schnittker —3473
Herr £1arg — 760
Herr Dr. Zarth —3476
5.2 Contractor
GKN Keller GmbH, Spezialtiefbau
Hardenstr. 51 Kaiserleistr. 44
D—2000 Hamburg 28 D—6050 Offenbach 12
D — 040/78 17 51 0 — 069/8051—213
Herr Pielsticker Herr Dr. Sondermann
S + I Schlaminentwasserungs GmbH ÷ Co. KG
Kruppstr. 9
D—4047 Doriuagen—Hackenbroich
D — 02106/651—18
Herr Dörner
WUE Umwelt Engineering GmbH
Stresemannstr. 80
D—4100 Duisburg 1
0 — 0203/3004—171
Herr Dr. Decker
200
-------
Diagra. I: Plan view of site and buildings
M 1:500
Plan view
0
N
U
K
N
G
F
I
0
C
e
A
Diagra. 2: Subsoil conditions
Drill-Logs M_J:100
OHRUNGIN AUSGEFUHRI 0M DIR FA DR N PiEIES HAMBURG AM 33 7 • AS -Ill ISIS
Z%12.ZL F RMALBOIIRUNGEN • 23 RAMI4KIRNSONO I(RBOMRUNG
zi Z2
0
z
0
0
I
Gol dbek-Caua 1
( GKN kELLER =
Z3
Z4
1.
201
FGXN KELiU]
-------
Mud/Pe d
— -I
—-t——
H14
Building B North-Wing Phenol Concentration (mg/kg)
zi
Filling
Z2
Wells
W .
Sand J(1
619
8663
6828
1600
I i0400
500
16
Diagra.
/
/
3: SubsoIl conditions and contamination
Gold bek - Canal
0
0
CD
-)
B
.
‘pa
e
0
B
‘I
0
Cl )
\ V
CD
0
/
C)
0
0,
0)
0
‘
l x i
‘ X I
lxi
-D
CD
C)
LL0
C,
3CD
c0
-‘ -‘ _& _& _& . -‘
o O 00 00;0 0
0 0000 0
o 00 0
0 0
202
-------
Dlagra. 5 :Process of treat.ent
Technology of High Pressure Washing and Treatment
Mixture
•Fiiiing
-
Building B
0
U,
Plan View of
Test Area
o A Test Column
$BP Gaugs
• B • Bore Holes
r Concrete
L Slab
BPI
6,35
• B 5692
— 5,00 —
B 5693
B 5481
Building
A
BPiI
203
-------
Test - Result Column C
Core-Drilling and Core - Material
B 5693 Phenol Concentration (mg/kg)
431mNN Core Drilling 10 100 1000__10000 100000
— I
Filling ‘Stand-Pipe 238 09
j ,Q • 1 Q 4P 5
4
128’
20 Lost Core 2 ,9 _____ ______
2487!
2 ____ _____ ____ ____
30 _____ ______ _____ “3100 ‘ ‘
3.5 qu 2 N/mm 2 :. ______ ____ _—Z
.z.1d
2930
3.8 k ‘1.8 1O 40 ______ ,_ -63 —4. 8664
_______ _______ 5670;
50 ____ ____
____ 5.4 qu:11N/mm 2
______ 900 -- --
05 !,30
— .. ___________ __________
.Pit
.-.
O P’ A 56 k :1,7 i 8m,s 6 ,Q ____
70- Core Mortar 70 _____ ______ 121
-- ____
Lost Core 7,7 ______ ______ ________
0 Test Boring B 5693 A Back-Flow
o Core Sample
( GIN ki
204
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTERIZATION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On—/Off—site Treatment Location:
Pre- and Post-treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Steen Vedby, Kim Broholm
Dept. of Environmental Engineering
Technical University of Denmark
Building 115
2800 Lyngby
De nina rk
45-2-884200
BiologiCal
Aerobic/anaerobic degradation
Technical University of Denmark
University Of Copenhagen
Kemp & Laurltzen
Experimental - Bench scale
Halogenated aliphatic hydrocarbons
Soil, groundwater
On—site/In—situ
Skrydstrup Chemical Waste Disposal
site, Denmark
ICE, TCA, PCE
Sandy soils from site varying In
extent of contamination
Some degradation of TCE and TCA
Inhibited effect at higher con-
centrations
Not available
11 / 89
7-a-i
205
-------
Paper Not Received by
Time of Printing
206
-------
NATO/CCMS Cover Sheet
TREATMENT CHARACTER! ZAT ION
General Type:
Specific Type:
Manufacturer/Researcher:
Status:
Treatable Contaminants:
Treatable Waste Matrices:
On—/Off-site Treatment Location:
Pre— and Post-treatment Requirements:
SITE DEMONSTRATION
Site Location:
Contamination:
Site Characteristics:
Observed Treatment Limits:
Costs:
CONTACTS
Troels Wenzel
Hojgaard & Schultz A/S
Jaegersborg 4
DK 2920 Charlottenlund
Denmark
01-67-20-00
Microbial treatment
On—site/In—situ reclamation:
membrane filtering and
biodegradation
Hojgaard & Schultz A/S
Former gas works, Fredensborg,
Denmark
Polycyclic aromatic hydrocarbons,
phenols, cynides, small amounts of
heavy metals
11/89
7-b-2
207
-------
HØJGAARD
& SCHULTZ
DANISH
GEOTECHNICAL
INSTITUTE
DOS
FILTRATION
On-site / in-situ Reclamation
Membrane filtering and Biodegradation
208
-------
97 10/TWE/BW/810 1989-10—01
On-site/in-situ Reclamation
Membrane Filtering and Biodegradation
1. Introduction
Sites where spills, leakage and dumping of organic chemicals
have occured are numerous and widely distributed. Several of
these substances may pose hazards which might restrict the
choice of future use.
If these hazards are recognized, and the site investigation pro-
perly carried out, it will as a rule be possible to restore the
site to beneficial use for a suitable purpose.
The remedial action needed should be designed and carried out in
such a way as to provide adequate protection for the lifetime of
a future development.
2. Situation in Denmark
In Denmark we have about 8-10,000 contaminated sites, several of
them being situated in the town centers.
Therefore, many of these sites are now developed and that often
in the form of residential housing.
3. The Concept of the In-situ Technique
The bioremediation of soil contaminated by organic components
depends on the ability of fungi and bacteria to utilize the
contaminants as a source of energy.
Several projects have documented that watersoluble tarcomponents
can be broken down to harmless components by microorganisms.
On that basis a joint-venture of Højgaard & Schultz a/s, DDS-
Filtration A/S and the Danish Geotechnical Institute has
developed a concept of a combined reclamation technique for
contaminated sites and aquifers.
209
-------
9710 / TWE / BW / 810
1989—10—01
The three companies jointly cover a large field comprising con-
struction, advanced cleaning technology, biotechnique, geotech-
nique and hydrology.
The basic idea of this combined reclamation technology is that
the reclamation should take place on the contaminated site
within a relatively short period of time.
The method is based on the following four main processes.
Fig.
1.
a) washing out of contaminants by enhanced recirculation of
purified groundwater.
b) Purification of the pumped-up groundwater by removal of
iron and sandfiltering followed by Reverse Osmosis (RO).
c) Biological purification of the concentrate from the RO-
filtering.
d) Biological degradation of the organic contaminants in the
soil. The degradation is enhanced by stimulating the bac-
terial activity by adding nutrients and oxygen to the
infiltration water.
The main processes in the Technique
R•co .sT*Ilq of
,sr.. ...o.1.
P e: S It O O
Ilelsqlc.1 purlilcatlrl
asc1rc st1o
Po11uUo, p1i
witer 1,vel
Fig. 1
210
-------
9710/TWE/BW/810
1989—10—01
The present concept and the technical equipment will be tested
in a large scale pilot project on a gaswork site, situated in
North Zealand, Denmark.
4. Description of the Site
The site covers an area of approximately 5.000 rn2. Five residen-
tial houses have been built on the area. Fig. 4.
The geology of the site mainly consists of filling, glacial
deposits and post glacial deposits, varying from sand to clay-
sand down to a depth of 5-7 rn where the primary sandy aquifer is
situated.
Fig. 2
The Geology of the Site
sDu1
NOATI4
— — — Groundwster eve1
211
-------
9710/TWE/BW/8l0 1989—10—01
A secondary aquifer is divided from the primary aquifer by a
peatlayer about 2.5 - 4 m beneath ground level.
Over the natural deposits a heterogenous 1-1,5 in deep filling
layer has been laid on the site where the former gaswork was
established.
After closing down the gasproduction plant 0.5-1.5 m clean soil
has been brought in on which the residential houses are situ-
ated.
The geotechnical investigations on the site have revealed the
following main pollutants and concentrations in the soil:
Coal tars 100 - 37000 mg/kg
Naphtalen 5 - 200 mg/kg
Cyanides (total) 100 - 3500 mg/kg
Sulphur 5000 - 40000 mg/kg
The highest concentrations are found in connection with the
previous location of the gashoiders and the retorts. The remain-
ing area, reproduced contamination concentrations in the lower
range of the concentration scale.
The main pollution components and concentrations in the ground-
water are:
Volatile aromatics 100 - 1600 ugh
Phenols 10 - 2300 ugh
Naphthaienes 300 - 18000 ugh
Sulfur 30 - 1700 ug/l
Ainmoniuin 20 - 200 ugh
Total CN- 0,01 - 1,9 mg/i
Iron 100 - 500 mg/i
Calcium 100 - 750 mg/i
The secondary aquifer is primarily contaminated with less mobile
compounds like methylbenzenes and naphthalenes contrary to the
primary aquifer which to a large degree is contaminated with
phenols and benzenes.
The inorganic components are equally distributed in the two
aquifers.
212
-------
9710/TWE/BW/810 1989—10—01
Proj ect Description and Status
Hot Spots :
Before establishing the recovering/infiltration system and the
on-site equipment, the worst contaminated soil is excavated from
the site. The excavated areas are refilled with coarse sand to
enhance the water infiltration rate on the site.
The heaviest tar contaminated areas are excavated to approxirn-
ately 2000 ppm which in this case we have chosen as a suitable
initial concentration in relation to the in-situ degradation,
bearing the complicated geology in mind.
Ferricyanid contaminated topsoil is entirely substituted by
clean soil to 1 rn.u.t.
Infiltration and Recovery System
The Fredensborg in situ project will consist of 2 systems:
_________
DRAIN SYSTEM
UPPER RECOVERY SYSTEM
LOWER IMJECTION WELLS
® RECOVERY WELLS
WATER MOVEMENT. SECONDARY AQUIFER
WATER MOVEMENT, PRIMARY AQUIFER
ROuNO WATER LEVEL
Fig 3 Schematic Wiev of the Infiltration
and Recovery System
213
-------
9710/TWE/BW/810 1989-10—01
- an upper infiltration system with a drain outlet network
and well points between the drain strings,
- a lower well system with 2 pumping wells and 5 injections
wells.
The upper infiltration system will be established in the fill
deposits approximately 1 rn.u.t., the pipes will have a diameter
of approximately 100 mm and a slope of 0.5-1%. The total outlet
from the infiltration system will be pumped back by 40 suction
pipes, which will be functioning with an overcapacity to ensure
a closed circuit.
A leakage loss to the lower circulation system is assumed to be
negligable due to the low gradient between the aquifers.
The lower circulation system consists of filters and welipoints
which will be placed at a depth of 5-6 n.u.t. in the glacial
sand, which have perrneabilities in the range of 1 x 10-5 - 1 x
10-4 rn/sec. The location of the wells and their yields, are
based on the following conditions and assumptions:
- The injection wells are dimensioned as gravity injection
wells.
- The yields can be increased/decreased at the pumping wells
and injection wells in case of unexpected conditions, such
as occurance of ohcre.
- The average transmission is 1.4 x 10-4 m2/sec in the well
area.
- The number of pumping wells and their yields are reduced to
a minimum, to reduce the costs of cleaning the filters in
case of occurrance of ohcre.
- The pumping wells are placed in the middle of the polluted
area with a ring of injection wells, surrounding them. Fig.
4.
- The lowering of the groundwater table is as far as possible
kept inside the area.
- The horizontal water velocity is increased from 20-30
m/year to 100-200 rn/year in the poluted area.
214
-------
9710 / TWE / BW / 810
1989—10—01
The system will be operated with a surplus recovery yield to
prevent further contamination of the aquifer in case of mal-
function of the system.
Fig. 4
Well locations for the lower system.
Welt locations f the lower system
V Pumping walls 1 2
• lnject on wells 3 -7
(jncIudir water vel contour lines).
215
-------
9710/TWE/BW/810 1989-10—01
Pretreatment of Recovered Groundwater :
The RD-membrane is very sensitive to the composition of the
feedwater, especially silt and colloids can clog the feedchan-
nels in the RD-unit and reduce the production rate. This is
especially of significance on a gasworksite, where high levels
of iron in the groundwater as Fe++ are common to be found. Fe#+
will be oxidized in the RO—unit to Fe+++ and precipitate on the
membrane as Fe(OH)3.
High levels of Fe++, Ca++ and HCO3 have caused problems in the
laboratory RD-units due to precipitation on the membrane.
The raw groundwater will be treated in a treatment train in the
following order: oxidation, basification, precipitation, and
filtration. The treated water will be acidified to an acceptable
level in regard to the RD-membrane.
The sludge from the pretreatment will mainly consist of fern-
hydroxyd Fe(oH)3, smaller amounts of calcium-carbonate CaCo3,
and to some degree contaminants sorbed to the sludge. The
dewatered material will be deposited at a disposal site.
Reverse Osmosis :
The RD-unit consists of a spiraiwound membrane system, loaded
with a HR-thinfiim polyamidcoxnposit membrane.
The unit which will be driven at a pressure of approximately 20
bar can withstand a suspended solid concentration of about 3.
mg/i and a Fe+++ content of 0,1 mg/i.
The RD-unit will be able to concentrate the feedwater 10 times,
depending on the effectiveness of the pretreatment. This wtll
result in about 90% permeate and 10% concentrate which goes to
the bioreactor for degradation before discharge to the sewer.
The previous outlined pretreatment, in concert with a special
cleaning in place membrane regeneration technique (CIP) should
be an effective way of dealing with membrane malfunctions during
the largescale trial.
216
-------
9710/TWE/BW/810 1989—10—01
The RO has varying rejection rates in relation to varying con-
tainination components.
Due to interactions between the membrane and contaminating
components, the rejection rates are on an average 10% lower for
phenol and 0-cresol compared to other compounds. In spite of
that the overall rejection efficiency is in the area of 96-99%.
Metal ions and cyanid ions have a rejection rate of 98% or more.
Table 1.
Component Approx. rejection rate in %
Benzene 98 *
Tolune 98 *
Xylene 98 *
Phenol 89
0-cresol 88
M—cresol 95
P-cresol 97
Naphthalene 98 *
Cyanid 99
Cloride 99
Metal ions 98
* Estimated values
Measured values
Tabel 1
Rejection rates for the Ro-unit under laboratory and
large scale operation.
The membrane chosen is designed for production of drinking water
from saltwater and has proved to be very stable during
operation, providing the pretreatment of the feedwater can be
done at a high standard level. If that can be accomplished on
site, we expect the lifespan of the membrane to be 1-2 years.
217
-------
9710/T JE/BW/Bl0 1989—10—01
Bioreactor :
To examine the degradation potential of the components found in
the polluted groundwater, a number of batch experiments has been
carried out.
The biomass used in the laboratory degradation experiments, ori-
ginates from naturally adapted microorganisms from soil and pore
water taken at the contaminated areas on the gasworksite.
The organisms which were grown in the batch reactor were
gradually confronted with increasing pollution concentrations in
the feedwater of the system.
The experiments were carried out at 10 °C under aerobic con-
ditions. After an adaption period of approx. 3 weeks all the
monitored components were degraded to an acceptable concen-
tration. The xylenols were more difficult to degrade in the
feedwater than the rest of the components, especially 2,6
xylenol has caused some nuisance in the experiments.
During the laboratory work we have experienced significantly
increasing decomposition rates, due to specialization of the
biomass.
After work on degradation of watersoluble tar components in the
laboratory, we have decided to test a rotating disc biological
contractor (RBC) system.
The RBC which is proposed to be used on-site will have a
specific area of 700 m2 and will fit into a container of 3.5 m3.
The reactor will be divided into a number of compartments and
covered to get hold of the eventual emission of volatile strip-
ped components.
6. Further Research and Development :
Several projects have used laboratory adapted bacterial strains
to inoculate the contaminated soil and thereby enhance the
degradation - often with moderate success, to a large degree
due to the laboratory adapted bacterial strains’ inability to
compete with the natural population.
218
-------
9710/TWE/BW/810 1989—10—01
Therefore, one of the main criterias of success in this project
depends on the possibility of transfering nutrients and oxida-
tion components to the native in-situ adapted microflora and in
that way optimize the microbial activity and biodegradation,
which already is taking place in the soil and groundwater.
Pumping tests and work in the laboratory have shown that more
than 99% of the contaminants in the water are phenolic- and
naphtalene compounds. This indicates that the higher molecular
PAR’S will accumulate in the soil -as the leaching goes on. It
is, therefore, of vital importance to construct a system which
will be able to transfer the above mentioned components to the
contaminated areas in the soil.
The biological degradation in soils together with the leaching
scale are important factors when working with in-situ treatment
of polluted soils and aquifers.
An additional research programme has been initiated with the
object to study the degradation rate of tarcompounds, by using
nutrients, oxidation medium (02, KNO3) and different levels of
contaminants.
The research programme includes microbiological tests as DEFT,
platecount etc. and the leaching rate of different tar com-
pounds.
The laboratory setup consists of a number of sterilized bottles
with soil material taken from the gaswork site. The degradation
rate is followed by regular sampling. Groundwater discharge is
simulated by rotating the test bottles with a velocity of 1
rev/mm. All experiments are carried out at 10 degrees Celsius
in darkness to simulate natural conditions.
7. Acknowledgements
The pilot project is partly founded by grants from the Danish
Industrial Research and Development Fund, under the National
Agency of Trade and Industry.
Troels Wenzel
M.Sc.
219
------- |