xvEPA
           United States
           Environmental Protection
           Agency
             Office of Research and
             Development
             Washington DC 20460
EPA/540/AR-92/014
August 1993
Membrane Treatment  of
Wood Preserving
Site  Groundwater  by
SBP Technologies,  Inc.

Applications Analysis Report
                 SUPERFUND INNOVATIVE
                 TECHNOLOGY EVALUATION

-------
                                              EPA/540/AR-92/014
                                                   August 1993
  Membrane Treatment  of Wood  Preserving
Site Groundwater by  SBP Technologies, Inc.

        Applications Analysis Report
       Risk Reduction Engineering Laboratory
        Office of Research and Development
       U.S. Environmental Protection Agency
              Cincinnati, Ohio 45268
                                          "1^  Printed on  Recycled  Paper

-------
                                                       Notice

The information in this  document has been funded by the U.S. Environmental Protection Agency under the auspices of the
Superfund  Innovative Technology Evaluation (SITE) Program  under  Contract  No.   68-CO-0048   to  Science Applications
International Corporation.  It has been subjected  to  the Agency's peer  and administrative review, and it has been approved
for publication as an EPA document. Mention of trade names or commercial products  does  not constitute an endorsement or
recommendation for use.

-------
                                                     Foreword
The  Superfund Innovative Technology Evaluation (SITE)  Program was  authorized in the  1986 Superfund Amendments.  The
Program is a joint  effort between EPA's  Office of Research and  Development and Office of Solid Waste  and Emergency
Response. The  purpose  of the program  is to assist the development  of hazardous waste treatment technologies  necessary to
implement new cleanup standards  which require  greater reliance  on permanent  remedies.  This  is  accomplished through
technology demonstrations  designed  to  provide engineering  and cost data  on selected technologies.

This project was designed to evaluate  the  effectiveness of  SBP's  formed-in-place membrane process on wood preserving
waste contaminated groundwater  and establish the potential applicability at other  Superfund  or hazardous waste  sites. This
process separates contaminated water into a large volume of relatively uncontaminated permeate potentially suitable for
discharge and a smaller volume of a  contaminant-rich concentrate suitable for biodegradation or other means of final
destruction. The study was  carried  out at the  American  Creosote Works  facility  in Pensacola.  Florida,  a site where wood
preserving operations  had been carried  from 1902 to  1981  using creosote  and. more recently, pentachlorophenol.  The study
is summarized  in this Applications  Analysis Report and described in more detail  in the  companion Technology  Evaluation
Report.

Additional copies of this  report may be obtained at no charge from  EPA's Center for Environmental  Research  Information,
26 West Martin Luther King Drive, Cincinnati, Ohio  45268. using the  EPA document number found on  the report's front
cover. Once this supply is exhausted, copies can be purchased from  the National Technical Information Service, Ravensworth
Bldg., Springfield, VA, 22161,  703-487-4600. Reference copies will  be available at EPA libraries in their Hazardous  Waste
Collection.
                                                   E. Timothy  Oppelt. Director
                                                   Risk Reduction  Engineering  Laboratory
                                                   U.S. Environmental Protection Agency
                                                           111

-------
                                                     Abstract
This document provides an evaluation of the SBP Technologies, Inc.  (SBP) formed-in-place membrane hyperfiltration process.
The  purpose of the technology  is  to reduce  the volume of waste requiring  further treatment through  such techniques as
immobilization or destruction. This volume reduction technology,  when  coupled with other technologies,  may  reduce total
treatment  costs and minimize off-site transportation  of hazardous materials. Using  cross-flow filtration to minimize fouling,
the membrane filtration system separates contaminated groundwater and other waste waters into a small, concentrated stream
that can be treated biologically or otherwise, and a relatively clean permeate that can be discharged,  reinjected.  or reused with
little or no additional treatment. In  hyperfiltration,  pollutants are separated on the basis of molecular weight, molecular size,
polarity,  or charge.

This report summarizes the  utility  and application  of   SBP's membrane system  to  the  treatment of  organic  contaminated
wastewater. This  analysis utilizes  information  from the  Superfund Innovative Technology Evaluation (SITE)  Program's
demonstration at the American Creosote Works  wood preserving site  in Pensacola, Florida as well as data  from other  SBP
investigations. Conclusions were reached  concerning the technological  effectiveness and  economics of  the  process and its
suitability for  use  at other sites  and with  other waste waters.

During the SITE  demonstration,  operations were carefully monitored to establish a database against which the vendor's claims
for the technology could be evaluated.  These claims  were that the filtration system would (1) provide an 80% volume
reduction  for  the  contaminants in the feed stream;  and (2) achieve  90% removal  of semivolatile contaminants, based on  a
comparison of the concentrations in the feed  stream and those  in the permeate from  the filtration system.

Based on the demonstration study using the system as configured and as used, an 83% reduction in the volume  of the
contaminated feed water was achieved.  However, the system achieved a 74% overall  removal of the designated semivolatile
components, which included  low molecular weight  phenols and polynuclear aromatic hydrocarbons  (PAHs).  The filtration
system was much  more effective at removing the PAHs  than phenols. The average  removal efficiencies were 92% for  PAHs
and  18%  for  phenols.

Capital and operating costs for the  system are estimated to  be between $220 and $1,740  per thousand gallons (on an annual
basis), dependent  on the type and magnitude of contamination  encountered in the waste stream.

-------
                                                     Contents
Section                                                                                                         Page

Notice   	   	           ii
Foreword  	           iii
Abstract   	           iv
Figures   	          vii
Tables   	           viii
Abbreviations  and Symbols  	           ix
Conversion  Factors   	          xii
Acknowledgements    	          xiii

    Executive   Summary   	             1
      1.1 Inlroduction   	              1
      1.2 Conclusions  	              1
      1.3 Discussion  of  Conclusions   	             2

2.   Introduction   	               5
      2.1 The Site Program   	              6
      2.2 Site Program Reports	•	           6
      2.3  Purpose of the  Applications  Analysis Report   	           6
      2.4 Process Description  	              6
      2.5 Key Contacts  	            7

3.   Technology  Applications  Analysis   	           8
      3.1 Introduction  	              8
      3.2 Mechanisms  of Membrane  Separations  	           8
      3.3 Applications of Membrane Processes for the  Treatment of Hazardous Waste    	        9
      3.4 Features of SBP's Hyperfiltration System   	          H
      3.5 Demonstration Results  	            11
      3.6 Discussion of Demonstration Results  and  Applications    	         12
      3.7 Applicable Wastes	             16
      3.8 Site Characteristics  	           16
      3.9 Environmental  Regulations  Requirements  	          16
      3.10 Materials Handling  Requirements  	           17
      3.11 Personnel Issues   	             19
      3.12 Potential  Community Exposures  	           19

-------
4.  Economic Analysis 	            20
      4.1  Introduction 	            20
      4.2  Conclusions  of Economic Analysis   	          20
      4.3  Issues and  Assumptions  	           21
      4.4  Basis for Economic Analysis   	           21
      4.5  Results   	             26
      4.6  Remediation of a Hypothetical Site  	          26

Appendix  A. Process  Description  	            29
      A.I Introduction	             29
      A.2  Process Description   	           29

Appendix  B. Vendor's Claims 	            31
      B.I  Introduction - Hyperfiltration  System  	          32
      B.2  Results BTEX/Petroleum Hydrocarbons Hyperfiltration  	         32
      B.3  Conclusions - Hyperfiltration  System  	          32
      B.4  Introduction -  Bioremediation  System  	          32
      B.5  PAH and PCP Contamination and Bioremediation   	         32
      B.6 Bioremediation Results  	            34
      B.7 Technology Applications 	           34

Appendix  C. SITE Demonstration  Results	           36
      C.I    Introduction..  	            36
      C.2  Field Activities   	            37
      C.3  Test  Procedures  	            37
      C.4 Results  	             38
                                                        VI

-------
                                                     Figures






                                                                                                               Page




Figure 1. Cross-Flow Filtration                                                                                  30




Figure 2.   Schematic  of SBP Filtration System   	        30




Figure 3.  Simplified  Process Flow for SBP  Technologies  Inc.'s  Bioreactor  System 	       34




Figure 4.  American  Creosote Works  Site  	        36




Figure 5.  Feed Pressure vs.  Run  Time  	       38




Figure6.  Flux vs. Run Time  	        38




Figure 7.  Molecular  Weight vs.  Order of Magnitude Reduction (ORD)  	        39
                                                         vn

-------
                                                      Tables





                                                                                                                Page




 Table  1.  Feed and Permeate  Semivolatiles  - Total  Concentrations and Contaminant Reductions    	      13




 Table  2.  Individual  Semivolatile  Concentrations and  Rejections  	         14




 Table  3.  Conventional Parameters  	            14




 Table  4.  Estimated Costs for  SBP Filtration Unit   	          23




 Table  5.  SBP Labor  Requirements and Rates  	         24




 Table  6. Hypothetical Site Cost Analysis  	           28




Table  7.  Classes and Characteristics of PAHs	           33




 Table  8.  Summary of Bioremediation Results of PAH and PCP Removal   	        34




Table  9.  Average Characteristics  of Feed Stream to SBP Unit  	          37




 Table  10. Overall Semivolatile Rejection Efficiency  for SBP Filtration Unit  	         39




 Table  11. Mass Contribution of Semivolatiles  in  all  Streams   	          39




Table  12. Behavior on Extended  Filtration   	            40




Table 13.  Dioxins/Furans  in Process Streams   	            41




Table 14.  Volatiles in  Process Streams	,	       41
                                                         vni

-------
                                     Abbreviations  and Symbols
BOD          biochemical oxygen demand (mg oxygen/liter)




BTEX        benzene,  toluene, ethyl benzene, and xylenes




CERCLA    Comprehensive Environmental Response, Compensation,  and Liability Act of 1980




COD        chemical  oxygen demand (mg oxygen/liter)




GCMS        gas chromatograph/mass  spectrometer




gpd           gallons per day




gpm          gallons per minute




HSWA      Hazardous and Solid Waste  Amendments to RCRA  - 1984




kwh          kilowatt-hour




Mgd          million gallons per  day




mg/L          milligrams per liter




ng/kg         nanograms per kilogram




ng/L          nanograms per titer




NPL         National  Priorities List




O&G          oil and grease




ORD         Office of Research and Development




OSHA        Occupational Safety and Health Administration or Act




OSWER     Office  of Solid  Waste  and  Emergency  Response




PAHs         polynuclear aromatic hydrocarbons






                                                     ix

-------
PCBs           polychlorinated  biphenyls




PCP            pentachiorophenol




PEL            Permissible Exposure  Limit




POTW         publicly owned treatment works




ppb             parts  per  billion




ppm            parts  per  million




ppt             parts  per  trillion




psi             pounds per square  inch




PVC            polyvinyl   chloride




QA/QC         quality assurance/quality  control




RCRA         Resource  Conservation and  Recovery Act  of 1976




RI/FS           Remedial  Investigation/Feasibility Study




RREL          Risk  Reduction Engineering  Laboratory




SAIC           Science Applications  International  Corporation




SARA          Super-fund  Amendments and Reauthorization  Act  of 1986




SITE           Superfund Innovative  Technology Evaluation




SBP            SBP  Technologies, Inc., formerly  Southern Bio Products, Inc.




TDS            total  dissolved solids (mg solids/liter)




TOC            total  organic carbon (mg carbon/liter)




TPH            total  petroleum hydrocarbons




TSS            total  suspended solids (mg  solids/liter)




VOCs          volatile organic  compounds

-------
                                           Conversion  Factors
               English (US)     x
.Factor
Metric
 Area:
 Flow Rate:
Ift2
Iin2
1 gal/min
1 gal/min
1 Mgal/d
1 Mgal/d
1 Mgal/d
1ft
lin
lyd
lib
lib
1ft3
1ft3
Igal
-M- ^»-M.
Igal
x
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
9.2903 x 10-2 =
6.4516
6.3090 xlO'5 =
6.3090 xlO'2 =
43.8126
3.7854xl03
4.3813 xlO'2 =
0.3048
2.54
0.9144
4.5359 xlO2
0.4536
28.3168
2.8317 xlO'2 =
3.7854
3.7854xlO-3 =
m2
cm2
m3/s
L/s
L/s
m3/d
m3/s
m
cm
m
g
kg
L
m3
L
m3
 Length:
 Mass:
 Volume:
 ft = foot, ft2 = square foot, ft3 = cubic foot
 in = inch, in  = square inch
 yd = yard
 Ib = pound
 gal = gallon
 gal/min (or gpm) = gallons per minute
Mgal/d (or MGD) = million gallons per day
m = meter, m2 = square meter,  m3 = cubic meter
cm = centimeter,  on2 = square  centimeter
L = liter
o =gram
kg  =  kilogram
m3& = cubic meters per second
Us  =  liters/sec
     = cubic meters per day
                                                      XI

-------
                                              Acknowledgements
This report was directed and coordinated by  Ms..Kim Lisa Kreiton, EPA SITE Project Manager in the Risk Reduction
Engineering Laboratory  - Cincinnati,  Ohio.

This report was prepared for EPA's  Superfund Innovative Technology  Evaluation (SITE) Program by Dr. Scott W. Beckman,
'Dr. Herbert S.  Skovronek, Mr. Omer  Kitaplioglu  and a cast of thousands at Science Applications International Corporation
for  the  U.S.  Environmental  Protection  Agency under  Contract No. 68-CO-0048. The  Work  Assignment Manager for  this
project was Dr. Scott W. Beckman.

The cooperation and participation  of Ms.  Heather M. Ford. Dr. David J. Drahos. Dr. Ron Thomas, and supporting staff of
SBP Technologies, Inc.  throughout the  course of the project and in review of this  report are gratefully acknowledged.

Mr. Charles Logan of the Florida  Department of Resources  (DER)  and Ms. Madolyn  Streng, the Remedial Project Manager
of  USEPA's Region IV, provided invaluable assistance and guidance in initiating the project and in interpreting and
responding to regulatory needs of the project.

Finally, the project could not have been carried out without the efforts of the many  SAIC  and S-Cubed personnel who were
responsible for the  actual  sample  collection and  analyses.
                                                         xu

-------
                                                 SECTION  1
                                        EXECUTIVE  SUMMARY
 1.1      Introduction

The  SBP Technologies, Inc. membrane filtration system,
using a formed-in-place  hyperfiltration membrane, has been
used to treat  a creosote and  pentachlorophenol  (PCP)
contaminated groundwater stream at a Pensacola,  Florida  site
on the Superfund National Priorities List. Operational and
cost data were collected for that investigation  and are the
basis for the  evaluation of this process.

SBP's  membrane technology can be used as an integral part
of a remediation system to significantly reduce the  volume
and toxicity  of contaminated wastewater.  The technology  is
particularly  suited  for the treatment  of contaminated
groundwater as part of a pump and treat system. The
technology   reduces risks  to   human health  and  the
environment  by  transferring  the  contaminants  to a smaller
volume facilitating destruction or detoxification by other
technologies. The technology is  particularly  applicable to
the  treatment- of dilute waste  streams,  where  the
concentration of the contaminants into a reduced volume
would  result in significant  cost savings as well as minimize
off-site treatment. The reduced-volume concentrated residual
could be further treated  on-site.  or transported off-site for
treatment  and disposal.

The system  is simple to operate, reliable and requires a
minimum of operator attention or maintenance once the
membrane has  been formed. The stability of the system
makes  it particularly  suitable  for long-term  use  as is
necessary for extended pump and treat remedial programs

The demonstration at the American Creosote Works was
designed to evaluate the two most  critical process parameters
for membrane systems;  volume reduction and  contaminant
 reduction. A summary of the demonstration results  for these
 critical processes parameters are presented below.

1.2    Conclusions

Based on the results of the SITE demonstration project at the
American Creosote Works site in Pensacola, Florida and
information concerning other studies provided by the vendor,
SBP Technologies,  Inc.,  for different  wastes at other sites,
several conclusions can be drawn.  The conclusions  are
organized  based  on the evaluation factors  of volume
reduction and contaminant reduction.   In  addition, unit
operability and waste applicability are  also presented. These
factors are  critical  in applying  the technology to other sites
and wastes.

 1. Contaminant Reduction -  The SBP filtration unit  (as
    configured) effectively removed high molecular weight
    compounds from the feed  stream, but smaller molecular
    weight compounds were not removed.

  • The  formed-in-place membrane system is quite effective
    (92%)  at  removing  polynuclear aromatic hydrocarbons
    (PAHs) found in creosote from the feed  water and
    producing  a  permeate  with little of these materials.

  • However,  the membrane - as used in the demonstration
    - is not very efficient at removing phenolics. Rejections
    were in the range of 18%  for phenolics. (see Appendix
    B  for vendor  discussion.)

  • Overall, based on a comparison  of total concentrations
    of a predesignated list of creosote-derived PAH and
    phenolic semivolatile contaminants in the permeate
    versus the feed water, the system did not meet the

-------
    claimed rejection efficiency  of  90%. On the combined
    basis, rejection was 74% over the six days  of tests.

  •  On the basis of the PAH rejections of over 90%, the
    permeate would  be expected to be  acceptable for
    discharge  to  POTWs  with little or no polishing.

  •  Other pollutants found in contaminated waters at wood
    treatment facilities (e.g., polychlorinated  dioxins and
    furans) also are  concentrated in the reject stream based
    on results of this and other  SBP studies.  The permeate
    retains little of these species.

  •  Other constituents commonly encountered at such sites
    including colloidal oils and suspended solids are also
    extensively removed by the membrane process. Removal
    efficiencies for oil and grease were 93%. Suspended
    solids were removed to non-detectable levels. These
    materials did not appear to have an adverse effect on
    the filtration  process.

2.  Volume Reduction  - The  system effectively  concentrates
    organic  contaminants into a concentrate of much smaller
    volume.

  • The  volume of wood preserving waste contaminated
    waste water was reduced by over 80%. This means that
    only 20% of the volume of the feed water would require
    further treatment to immobilize or destroy  the  organic
    contaminants.

  • During  an extended run of the system, the volume of
    contaminated waste water was  reduced  by  96.3%.  This
    represents the maximum  volume reduction  capability  of
    the unit for  the waste  stream tested.

3.  Operability -The filtration unit  operated consistently and
    reliably over the six day testing period. The unit was
    easy to  operate and maintain.

  • The filtration unit operated in a batch mode for six
    hours  each  day,  for  six   days,  and processed
    approximately  1,000 gallons of feed  per  day. Over the
    six day test period, permeate flux was  a relatively
    constant 0.0085 gallons/min/ft  (coefficient of variation
    < 10%). Based on a total membrane  area of 300 ft2 for
    the system, the permeate flow  rate for the four module
    filtration unit averaged 2.6 gpm.

  . Excessive fouling  of  the membrane, necessitating
    frequent  cleaning or regeneration, was not  encountered.
    However, the membrane system did exhibit a gradual
    and  controllable  fouling which  required periodic
    cleaning.

  • The  operating cost for the membrane process as used at
    American Creosote Works is in the range of $220 to
    $1,740/1,000 gallons, depending on system size. Major
    conaibutors are labor and residuals disposal. Labor costs
    decrease  significantly as the  scale of the process
    increases.

  • Auxiliary equipment that could be needed to support
    this  process is comparable  to  that which would be
    needed for other  above-ground treatment  systems such
    as oil/water  separators  and  clarifiers  for  pretreatment,
    and  filters, carbon  adsorbers, etc. for effluent polishing
    as required.

4. Waste Applicability

  • With membranes similar to those manufactured for the
    American Creosote Works site, the system could be well
    suited for the concentration of polynuclear aromatic
    hydrocarbons from wastewaters (groundwater.  process
    wastes, lagoon leakage, etc.) found at coke plants, wood
    preserving  sites, and some chemical plants.

  • Based on the expected mechanisms of membrane
    filtration  and results from this study, the technology also
    may be useful for waste waters containing other  large
    molecules such as polychlorinated biphenyls (PCBs) and
    polychlorinated  dioxins  and   furans, particularly  where
    these are associated with oil or particulate matter. It
    probably is  also  highly effective for oils, colloidal
    solids, and greases.

  • According,  to  the  developer, the  formed-in-place
    membrane can be easily modified to  conform to waste
    characteristics and  degree  of  contaminant removal
    desired. Therefore,  the membrane can be tailored to the
    unique characteristics  of the  waste stream.

 7.3 Discussion of Conclusions

A small scale Filtration Unit with a nominal  5-10  gpm
capacity  was tested at the American Creosote Works site
under the Superfund Innovative Technology Evaluation
(SITE) program. Extensive data were collected over six days
(six hours each day) of operation to assess (a) the removal
of wood preserving wastes  from  contaminated groundwater
at the  site; (b) the operational requirements of the  system:
and (c) the cost of operation. The  data from this study serve
as  the  primary basis  for the  foregoing  conclusions.

-------
Additional supporting evidence  was provided  by SBP  in the
form of results from other field studies.

A Quality Assurance (QA) program was conducted by  SAIC
in conjunction with EPA's QA program,  including audits and
data review  along  with corrective action procedures and
special studies to resolve  specific data quality problems. This
program is the basis for the high quality of the data derived
from the SITE project. Discussion of the QA program  and
the results of audits, data  reviews, and special studies can be
found in the  Technology  Evaluation Report for this project.

Extensive  data were collected  on primary pollutants  (PCP,
other phenols, and PAHs) and on secondary pollutants (oil,
suspended and dissolved  solids, COD, dioxins. and VOC's).

The results of this SITE  project demonstrated the ability of
the formed-in-place membrane, operating in a cross-flow
mode, to minimize fouling,  and to remove polynuclear
aromatic hydrocarbons from the contaminated feed water. As
operated rejection of the  PAHs appears to increase with the
number of aromatic rings, from  78% for naphthalene to
94+% for the 4-ringPAHs.  However, similar correlations
appear to exist with molecular weight as well as with the
partition coefftcient  reflecting  hydrophobicity. The permeate,
accounting for  approximately  80%  of the feedwater,
contained only  about  12%  of the  predominant  PAHs,
naphthalene  and  phenanthrene.

The removal of phenol and methyl  phenols was not
comparably high under the conditions  of the demonstration,
with an average rejection of 18%. The concentrations of
phenolics  in the permeate  could  present  a  regulatory
problem, depending on the concentrations in the feedwater
and the final disposition of the permeate, However, the
vendor  states   that  different  membranes  and  tube
configurations could  resolve  this  issue.

Secondary  constituents, such  as oil,  suspended solids, and
dissolved solids, did not appear to  interfere with operation of
the process at the concentrations present in the waste water
studied during the demonstration. Decreases in chemical
oxygen  demand (COD), total organic carbon (TOC)  and oil
and  grease (O&G)  indicated  that the  system removes other
organic  species as well as PAHs, but not necessarily with the
same efficiency.

The SBP membrane process  would be most  applicable to
wastewaters  containing large molecular weight organic
compounds   (PAHs,  dioxins/furans,   polychlorinated
oiphenyls. and certain pesticides/herbicides). The system
can  remove smaller weight molecular compounds (phenols,
benzene,  toluene,  ethylbenzene,  xylenes) if larger molecular
weight compounds are  not  abundantly present.  Removal  of
smaller weight  molecular compounds can be  accomplished
by modifying the  structure of the formed-in-place membrane.
For these  applications, the  pores of the membrane are
reduced, resulting  in higher  retentions of smaller components
as well as a reduction in the flux (throughput) of the  system.
To compensate  for  the reduced  flux, either  additional
membrane modules can be added or more time  will be
required to accomplish  the  remediation. In either case, the
overall cost may be higher.

The SBP system may be most suitable to treating relatively
dilute, but toxic, wastestreams  in which the percent reduction
of contaminants will allow discharge  of the permeate without
further treatment. This feature makes the unit highly  suitable
for polishing  effluents as part  of a multi-technology
treatment  train.    In this  system,  the primary treatment
technology can be utilized  to remove the bulk of the
contamination, with the  filtration unit being used as a final
polishing   step.

A major  attribute of  the  SBP system is its  ability  to
minimize fouling.  SBP effectively controlled excessive
fouling, in spite of the problematical nature of the wood
preserving waste feed, through a combination  of cross-flow
operation and membrane cleaning.  The membrane  cleaning
process effectively regenerated the membrane to its  original
clean  permeate  flux conditions.  This enabled the membrane
to be  reused, without the necessity to reformulate.

The  ability to repeatedly regenerate the flux after the
cleaning procedure is a  good  indication that the formed-in-
place  membrane is stable and can be used over an extended
length of time. In the unlikely event  of an irreversible
fouling, the membrane can be cost-effectively and easily
reformed  on-site with a minimum of downtime.

SBP  uses  a  proprietary formed-in-place  membrane
technology. The membrane is formed on porous  sintered
stainless steel tubes by depositing microscopic layers of
inorganic and  polymeric  chemicals.  The properties  of the
formed-in-place membrane can be varied  by controlling the
type of membrane chemicals used,  their thickness,  and the
number of layers.   This important feature allows for
customization of the membrane  system to a wide variety  of
waste  characteristics and clean-up  criteria. The formed-in-
place  membrane can be quickly  and  economically
reformulated in the field to accommodate  changes in waste
characteristics or  treatment  requirements.

-------
The formed-in-place membrane is compatible with a wide
variety of contaminants often encountered in hazardous
wastewater streams.  The  SBP formed-in-place  membrane is
stable under most  chemical environments and will not
degrade  even at high  contaminant concentrations.

The extent of contaminant reduction required (overall and
for  individual pollutants) can  also be an important factor in
system design and operation. This will impact membrane
selection, and operational requirements such as the  number
of cycles necessary to achieve the targeted volume reduction.
Generally,   as  the  desired level  of  volume-reduction
increases, the overall quality  of the permeate decreases, so
a balance must be maintained between  throughput and
permeate quality.   This will also  affect the throughput
capability (as permeate) for a particularly  sized system.

Other factors that could affect the removal of PAHs or other
contaminants (e.g., PCP) may include the presence of other
organics, oil and grease, suspended solids, and dissolved
solids in the  feed  water. While the  levels  of  such
contamination encountered in the demonstration project had
no  apparent  adverse effect, it is  unclear how much rejection
(of PAHs) was due to molecular size or weight and how
much was  due to  solubility   in oil  that was rejected and
coalesced by the membrane.   Additional or alternative
mechanisms  also  may be operative.

-------
                                                 SECTION  2
                                             INTRODUCTION
2.7 The SITE Program

The EPA's Office  of Solid Waste and Emergency Response
(OSWER)  and the Office of Research and Development
(ORD) established the Superfund Innovative Technology
Evaluation  (SITE) Program  in  1986  to  promote  the
development and use of innovative technologies to clean up
Superfund sites across the country. Now in its sixth  year.
the SITE  Program is helping  to provide  the  treatment
technologies necessary to meet new federal and state  cleanup
standards  aimed at  permanent, rather  than  temporary,
remedies. The SITE Program is composed of two major
elements: the Demonstration  program and an Emerging
Technologies Program. In addition, the Program includes
research on analytical methods that can  expedite cleanups at
Superfund  sites.

The Demonstration  Program  is  designed  to  provide
engineering and cost data on selected  technologies.  EPA
and the developers participating in the program share the
cost of demonstrating their innovative systems  at chosen
sites, usually Superfund  sites. Developers are responsible
for the operation of their equipment (and related costs). EPA
is  responsible for sampling, analyzing,  and evaluating all  test
results and comparing these results to claims  originally
defined by the developer.  The result is an assessment of  the
technology's performance,   reliability, and  cost.  In  addition
to providing the  deveioper with carefully documented
information  useful  in   marketing,  the information,  in
conjunction with other data, also  will  be used to  select the
most appropriate  technologies  for the cleanup of  other
S  uperfund sites.

Developers  of innovative technologies  apply  to  the
Demonstration Program by responding to EPA's annual
solicitation.  To qualify for  the program, a new  technology
must  have  a pilot  or  full scale unit and  offer  some
measurable advantage over existing technologies. Mobile
technologies  are of particular  interest to EPA.

Once  EPA has accepted a proposal, EPA  and the developer
work  with the EPA Regional  offices and  state  agencies to
identify a site containing wastes suitable for testing the
capabilities  of the  technology. EPA's contractor designs a
detailed sampling  and  analysis plan that will  thoroughly
evaluate the technology and ensure that the resulting data are
reliable. The duration of a  demonstration  varies  from a few
days to several months,  depending on the type  of process
and the quantity of waste needed to assess the  technology.
While meaningful results  can be obtained in a demonstration
lasting one  week with some technologies, such as the SBP
formed-in-place membrane  process, others,  such as in-situ
bioremediation of contaminated  soil, may require  months.
On completion of a demonstration, EPA prepares  two reports
which are explained in more detail below. Ultimately, the
Demonstration Program  leads to  an  analysis of  the
technology's  overall applicability to  Superfund  problems.

The second principal element of the SITE Program is the
Emerging  Technologies  Program, which  fosters  the
investigation and development of treatment technologies
which are  still at the laboratory scale. Successful validation
of these technologies could lead to the development of
systems ready for  field demonstration. A  third component of
me SITE Program, the Measurement and  Monitoring
Technologies  Program,  provides  assistance  in   the
development and demonstration of innovative techniques and
methods for better  characterization of Supetfund  sites.

-------
2.2 SITE Program Reports

The results of  the SITE  Demonstration Program  are
incorporated in  two basic documents,  the Technology
Evaluation Report and the Applications Analysis Report. The
former provides a  comprehensive  description of  the
demonstration and its results. The anticipated audience is
engineers  responsible  for  detailed  evaluation  of  the
technology  for  application to  other  sites and wastes. These
technical evaluators will  need to  understand thoroughly  the
performance of the technology  during the demonstration and
the advantages, risks, and costs  of the  technology  for the
given  application.

The Applications Analysis Report is  directed to  decision-
makers  responsible for selecting  and implementing  specific
remedial actions at other sites. This report provides sufficient
information  to enable  them to determine whether  the
technology merits further consideration as an option in
cleaning up specific  sites. If the candidate technology
described in the  Applications Analysis Report appears to
meet the needs of the site, more thorough analysis of the
technology based on  the  Technology Evaluation Report and
site-specific information from remedial investigations for  the
site will be made. Thus, me Applications Analysis Report
helps in determining whether the  specific technology should
be considered further  as  an option for  a particular  cleanup
situation and  the Technology Evaluation Report provides the
detailed data  for  such  further  evaluation.

2.3 Purpose of the Applications Analysis
        Report

Each SITE demonstration evaluates the  performance of a
technology while treating  a portion of the particular waste at
the demonstration site. Additional data  from  other projects
also will be presented where available, although such results
often lack the quality  control and quality assurance imposed
during the SITE  evaluation and, consequently, cannot be
used with the same confidence.

Usually the waste  at other  sites being considered  for
remediation will differ in some way from the waste  tested in
a  demonstration. These  differences  may affect waste
treatability  and use of the demonstrated technology at such
other sites.  Successful demonstration of a technology at one
site does not  assure that a technology will work equally well
at other  sites.   To  the  extent possible  during  the
demonstration, the  operating  range  over  which  the
technology  performs  satisfactorily  is  established  by
examining  a broad range  of wastes and sites. To a limited
extent, this report provides  an indication of the breadth of
applicability of the SBP membrane filtration system by
examining not only the demonsuation test data but also data
available from other  applications of the technology.

To  encourage  the  widespread  use  of demonstrated
technologies, EPA will evaluate the probable applicability of
each technology to  sites and wastes in addition to those
tested, and will study the technology's likely costs in these
applications, identifying unique characteristics that make the
demonstrated technology either attractive or unattractive. The
results of these analyses will  be summarized and distributed
to potentially interested parties through the Applications
Analysis Report.

2.4 Process Description

SBP's membrane technology can be used as an integral part
of a remediation system to significantly reduce the volume
and toxicity of contaminated wastewater.  The technology  is
particularly  suited  for the  treatment of contaminated
groundwater  as part  of a pump and treat system. The
technology reduces  risks  to  human  health  and  the
environment by transferring the  contaminants  to  a  smaller
volume facilitating destruction or detoxification by other
technologies.

SBP  uses  a proprietary  formed-in-place  membrane
technology. The membrane is formed  on  porous sintered
stainless steel tubes by depositing microscopic layers of
inorganic and  polymeric chemicals. The properties  of the
formed-in-place membrane can be varied by controlling the
type of membrane  chemicals  used, their thickness, and the
number of laj'ers.    This important feature allows for
customization of the  membrane system to a wide variety of
waste characteristics  and clean-up criteria.  The  formed-in-
place  membrane  can  be   quickly  and economically
reformulated in the field to accommodate changes in waste
characteristics  or treatment  requirements.

Contaminated feedwater is recirculated through  the filtration
unit  until the desired level  of volume reduction is attained.
The  filtration  unit generates  two  process waste streams. A
relatively clean stream,  called the "permeate", passes through
me membrane while a smaller portion of the feedwater.
retaining those species  that do  not  pass  through the
membrane,  is retained  in a stream called the "concentrate".
The permeate stream should be clean enough for disposal as
a non-hazardous waste  with little or no additional treatment.
The  concentrate  would  require  further  treatment to
immobilize  or  destroy the  contaminants.

-------
2.5 Key Contacts

 For more information on the demonstration  of the SBP
 Technologies Filtration System for contaminated
 groundwater,  please contact:

  1. Vendor concerning the process:

   SBP  Technologies,  Inc.
   2155 West Park Court
   Stone  Mountain, Georgia 30087
   404-498-6666
   Dr. David J. Drahos,  Director of Research & Development
 2.  EPA  Project  Manager concerning  me SITE
     Demonstration:

   Ms. Kim Lisa Kreiton
   U.S. EPA -ORD
   Risk Reduction  Engineering  Laboratory
   26 West Martin  Luther King Drive
   Cincinnati, OH 45268
   513-569-7328

 3. State contact concerning the American Creosote Works
     site:

   Mr. Doug Fitton
   Florida  Department  of Environmental Regulation
   2600 Blair Stone Road
   Tallahassee, FL  32399
   904-488-0190
 4.  EPA  Regional  contact  concerning  the  American
     Creosote  Works site:

   Mr. Mark Fite
   (4WD-SSRB)
   U.S. EPA, Region IV
   345 Courtland  Street,  NE
   Atlanta. GA 30365
   404-347-2643

-------
                                                SECTIONS

                        TECHNOLOGY  APPLICATIONS ANALYSIS
3.1     Introduction

This section of the report addresses the potential applicability
of the  SBP filtration technology as a means of concentrating
organic contaminants in aqueous  waste streams. The prime
benefit of concentrating contaminants is to minimize costly
treatment  of the  entire  wastestream.  In addition, by
concentrating the  organic contaminants into a  smaller
volume, alternative  treatment technologies may be  feasible
based on technical  and/or economic criteria.

The  ability of the filtration unit to  concentrate organic
contamination from  aqueous waste streams was demonstrated
on a groundwater contaminated with wood preserving wastes
(phenolics, PAHs,   and  PCP).  The results from  the
demonstration, in conjunction with information  supplied by
the vendor, were used to assess the applicability of the
technology  for a variety of waste types and site conditions.

The process uses a formed-in-place hyperfiltration membrane
on a stainless steel support to separate and concentrate
higher molecular weight contaminants.    Contaminated
groundwater (feed)  is  pumped  through the  modules under
pressure. A portion  of the feed  passes  through the   formed-
in-place membrane forming a permeate. The membrane
retains  certain contaminants resulting in a permeate  that is
clean relative to the feed. The bulk of the contamination
remains in the "concentrate" fraction. The concentrate is
recycled through the  unit until the desired concentration  or
level  of volume reduction is attained,  or the  level  of
contaminants  in the recycling  concentrate  inhibits the
filtration process (fouling).  The system  relies on a technique
called  "cross-flow filtration" to minimize fouling of the
membrane   and thus maximize  throughput.
The  properties of the two process  streams (permeate  and
concentrate) are of particular  importance since these
characteristics define waste disposal options.  The permeate
stream should exhibit significant reductions in contamination
so as to allow economical discharge to local wastewater
ueatment  facilities  without   extensive   pretreatment
requirements.      The  concentrate  stream  should  be
volumetrically small, relative to the original feed,  in order to
minimize the volume of waste requiring further treatment
prior  to disposal. Furthermore, the filtration process should
enable the  use of additional  disposal  options for the
concentrate (as compared to the raw feed).

The  following  subsections summarize observations  and
conclusions drawn  from  the  current  study and  supporting
information. Included in the discussion  are factors such as
the application of membrane processes for waste water
reduction, benefits of the SBP  system, other applicable waste
waters, site characteristics  and constraints,  applicability  and
impact of state and federal environmental  regulations,   unique
handling requirements, and personnel factors.  Additional
information on the  technology,  including  a process
description, vendor claims,  a summary of the  demonstration
test  results  and case studies of other investigations is
provided in the  Appendices.

3.2 Mechanisms of Membrane Separations

Membranes are semi-permeable barriers that are used to
isolate and separate  constituents  from a fluid stream. The
separation process can be accomplished through a number of
physical and chemical properties of the membrane as well as
the  material  being separated.  Separation can  occur through
processes such as size,  ionic  charge, solubility,  and
combinations of several processes. Membranes can remove

-------
materials ranging from large visible particles to molecular
and ionic chemical species.  Membrane materials are diverse
and can consist of synthetic  polymers, natural fabrics, porous
metals, porous ceramics, or liquids. The surface of the
membrane can be  chemically or biologically altered  to
perform separations  on specific chemical compounds. The
interaction of the  components of the fluid stream  with the
membrane is the mechanism controlling the outcome of the
separation process.

There are two basic modes of membrane separation.  In
dead-end filtration specific species are trapped within the
matrix of the membrane material. The membrane "filters-
out" these species producing a relatively clean effluent.  In
dead-end filtration the components  that are trapped  are
usually not recovered and remain within the membrane
matrix.   In addition, the membrane eventually becomes
plugged necessitating the replacement of the membrane.
Dead-end filtration is  principally  utilized to purify a fluid  in
applications where the removed species is relatively dilute.

In cross-flow filtration the fluid stream  is  directed parallel  to
the surface  of the membrane. This action inhibits the
accumulation of components  within the matrix of the
membrane. The cross-flow action of the fluid keeps the
surface  of the  membrane  clean allowing  for the passage  of
species  smaller than the pores of the membrane. Cross-flow
filtration produces two effluent streams.  The permeate is the
stream that passes through the  membrane and is relatively
depleted in species  larger than the pore  size  of the
membrane.  The concentrate is  the  cross-flow stream  that
contains the larger species that are unable to pass through
the membrane and accumulate.  The concentrate can  he
recycled allowing  for progressive  concentration of species
over time. Due to the ability of the cross-flow system  to
concentrate  components  from  the  feed stream,  it  is
commonly used as a method to separate  and  recover these
components.  Furthermore, the cross-flow  action minimizes
plugging of the membrane (fouling) by constantly  sweeping
the membrane's surface. This cleaning  action extends the
life of the membrane and minimizes degradation of flow
through the membrane.

Membrane systems  have many applications for the pre-
treatment and treatment of hazardous wastes. Membrane
separation is  a  volume  reduction technology.    This
technology  can   separate and   concentrate    specific
contaminants from a  waste  stream,  resulting in a significant
reduction in  the volume  of waste requiring  treatment. The
concentrated contaminants can then be destroyed or rendered
non-toxic. The utility of a membrane based technology is
based on its ability to reduce the volume of waste  by
removing contaminants from the feed stream and producing
an effluent stream that would require little or no further
treatment.  The greater the  volume reduction, the more
effective the technology is in reducing ultimate  disposal
costs. However, there is a balance between the magnitude
of the volume reduction, the  quality  of the effluent stream.
and the size  and operation of the unit. A higher  volume
reduction would  require additional  recycling, reducing  the
overall flow through the system.  In  addition, higher levels
of contaminant removal will usually  result in lower fluxes
through  the membrane  requiring either more  membrane area
or longer processing time.  The balance between throughput
and  effluent quality  is dictated by clean-up  standards  and
treatment costs. This balance will impact such factors as the
size  and type of the equipment, mode of operation, time
required for  remediation, treatment requirements for the
permeate,   and  ultimate  disposal mechanism  for  the
concentrated  contaminants.

3.3 Applications of Membrane Processes for
        the Treatment of Hazardous Wastes

Membrane processes have many applications  in the treatment
of contaminated waste  streams.    The most  common
applications  involve  the removal and  concenuation  of
organic and inorganic contaminants from liquid waste
streams. The  waste  streams  can originate from industrial
processes, contaminated  groundwater. contaminated surface
water bodies,  or as by-products of other treatment processes.

Membrane and filtration processes have historically been
utilized  for the treatment and purification of  drinking water.
For  this application, filtration is used to remove a wide
variety  of constituents, ranging from visible particulates
(sand filters)  to ionic species (reverse osmosis). From these
conventional applications, new uses of membrane separations
have recently been applied to the treatment of hazardous
waste streams.

Membranes can be used to separate and concentrate organic
contaminants  from waste streams. In these applications, the
organic contaminants are removed based on their  size
(molecular weight) or polarity. Size separations  rely on
membranes with specific pore  size distributions. The smaller
the pores,  the greater will be the removal of  small molecular
weight compounds. However,  as  the membrane's pore  size
decreases, the flux (flow per unit membrane area)  also
decreases impacting  the overall economics  and efficiency of
the process.  The polarity of an organic constituent is a
measure of it's ability to ionize in solution. Examples of
polar molecules are water,  alcohols,  and  compounds  with
hydroxyl (e.g. phenols) and  carboxyl groups  (e.g.   organic

-------
acids). Aliphatic hydrocarbons and polynuclear aromatic
hydrocarbons are  examples on non-polar organic molecules.
The chemical  characteristics of the  membrane can be used
to  separate  non-polar constituents in a waste  stream from
polar constituents. For  example, a membrane whose surface
is hydrophilic will allow passage of  polar components while
retaining the non-polar components.  These membranes can
be used to separate dissolved and emulsified oils from
aqueous waste  streams.

Inorganic contaminants, such as salts and heavy metals, can
be removed and concentrated from waste streams  by
membrane  processes.  Suspended inorganics  can be  easily
removed through the use of microfiltration  membranes.
These membranes have pore sizes ranging from as low as
0.01 \un to several microns.  Dissolved inorganics can be
removed  either through the use of  hyperfiltration  (reverse
osmosis)  membranes, or by  precipitation  followed by
microfiltration.  Conventional  reverse  osmosis  membranes
may require extensive prefiltration to avoid fouling, and
therefore can only be used on  relatively clean feed solutions.
Chemical precipitation, followed by microfiltration, allows
for the  use  of microfilters which exhibit higher fluxes and
are not  as sensitive  to  fouling.

Membrane  processes can be helpful in solving many
remediation problems at hazardous  waste sites.

Contaminated   Groundwater

Containment and/or remediation of contaminated aquifers
typically utilizes pump  and treat technologies to control
contaminant plume migration and ultimately restore the
quality of the  groundwater. The recovered groundwater
usually requires  treatment prior to discharge.  Treatment
alternatives for  the recovered  groundwater  are dependent on
the nature and extent of the contamination.   Membrane
systems can be effectively used to  significantly reduce the
quantity of  groundwater  requiring  costly  treatment

The  contaminants  of  concern  can  be   isolated and
concentrated into a reduced volume which can  be more
easily handled.     Another  potential  benefit   of  the
concentration process is that additional destructive treatment
alternatives may become feasible.    For  example, the
concentration  of  hydrocarbons  from  a  contaminated
groundwater can produce a reduced  volume waste  with a
high BTU value  allowing for fuel  blending  as  a  disposal
alternative.     This  not  only reduces   the  quantity  of
groundwater that must be  treated, but also  produces a more
easily treatable  final waste product.    As  another  example,
heavy metals can be concentrated from  an aqueous stream
by   membrane   processes    and   immobilized   by
solidification/stabilization   technologies.

Membrane processes  can be potentially used to recover
organic  and inorganic constituents for recycle/reuse.  In these
applications, the separation scheme  must be  developed  to
produce a high quality concentrate.

Membrane processes can be applied to the removal of many
organic contaminants  from waste  streams.     Organic
contaminants that can be removed include petroleum derived
hydrocarbons (benzene, toluene, ethylbenzene, xylenes),
polynuclear aromatic hydrocarbons,   PCBs, dioxins/furans,
pesticides,  and chlorinated hydrocarbons.    Generally,
membrane process are more easily applied to removing
larger molecular weight, non-polar organic components
because larger pored membranes  can be utilized and surface
chemistry  interactions  can augment size separations.

Removal of hazardous  inorganic  species from contaminated
groundwater requires  a detailed knowledge of the water
chemistry in order to optimize the separation.   In many
cases, addition of precipitating  chemicals must be added  in
order to  induce  particulate  formation.    Furthermore,
groundwater  containing  high  concentrations  of innocuous
inorganic constituents such as iron  and divalent cations (egs.
potassium and calcium)  may compete  with and interfere with
the removal of toxic heavy metals.  Conventional reverse
osmosis membranes are fragile and must be protected from
the  corrosive  nature of many  highly  contaminated  aquifers.
Integration  With Other Technologies

Membrane processes are particularly amenable to integration
with other remedial technologies enabling applications to
additional waste matrices.  Ease of integration is  facilitated
by  the  modular  and  scalable  properties of membrane
systems.  These  systems can be readily integrated with other
remedial process equipment to enhance the effectiveness  and
economy of these systems.

Membrane processes can be used  as a final polishing tool for
remedial technologies involving discharge of process water.
In this capacity, the membrane system is utilized to remove
contaminants from a relatively dilute waste stream. The
benefit  of using this  polishing step is to  avoid costly
overdesign  of the primary  remedial technology.    For
example, a membrane system  can be impiemented as a final
polishing step on a bioreactor.  The bioreactor can  be
designed to cost-effectively  treat the bulk of the organic
                                                         10

-------
contamination,  while the  polishing  membrane  can  be
designed to  treat the aqueous phase prior to discharge.

Membrane processes can be used as a pie-treatment step for
other remedial technologies.    The purpose of the pre-
treatment would be to concentrate the contaminants to a
level that is amenable for specific remedial technologies.
For  example,  organic  contaminants  in dilute  aqueous
streams (e.g.  groundwater, leachate) can be concentrated to
a  level  that  could  support  an  efficient biomass  for
bioremediation  technologies.

Membranes can be integrated with remedial technologies as
a component in the process. For example,  membranes can
be used to recycle and recover extraction fluids used to
concentrate organic and  inorganic contaminants in soil
extraction technologies.

3.4 Features ofSEP's Hyperfiltration System

The  SBP Hypetliltration system has several unique features
which provides advantages over conventional membrane
processes  in  wastewater  treatment   applications.  The
demonstration was  designed to  evaluate  these features  under
actual  remediation  conditions.

Formed-In-Place Membrane
SBP  uses  a  proprietary  formed-in-place  membrane
technology. The membrane is formed on porous sintered
stainless steel tubes by depositing microscopic layers of
inorganic and polymeric chemicals. The properties of the
formed-in-place membrane can be varied by controiling the
type of membrane chemicals used, their thickness,  and the
number of layers.    This  important feature allows for
customization of  the membrane system to a wide variety of
waste characteristics and clean-up criteria.  The   formed-in-
place  membrane  can  be  quickly  and  economically
reformulated in the  field to  accommodate  changes in waste
characteristics  or  treatment  requirements.

Conventional  membranes rely on rigid polymeric,  ceramic,
or porous  stainless steel  membranes.  These membranes are
available in discrete pore sizes and cannot be customized to
the characteristics of the feed.  Furthermore,  once  installed
on-site it is difficult and costly to modify their separation
properties  in response to variable feed characteristics.

The  formed-in-place membrane  is compatible  with a wide
variety of contaminants often encountered in hazardous
wastewaterstreams.   Many  conventional   reverse osmosis
membranes are made  from materials such  as cellulose
acetate and exhibit poor compatibility with  reactive
substances  often encountered in hazardous wastes.  These
conventional  membranes  will  degrade and  become
inoperative  when challenged with many organic compounds.
The compatibility problem becomes more critical as the level
of concentration increases.   The SBP  formed-in-place
membrane is stable under most chemical environments  and
will not degrade even at high contaminant concentrations.

Fouling Control

A major limitation of many membrane systems is their
propensity  to irreversibly  foul.  Fouling is  the uncontrolled
build-up of materials on the surface of the membrane.
Fouling leads to a loss of flux and eventually results in
cessation of flow. If a membrane fouls, it must be cleaned
in order to  restore flux. If cleaning is unsuccessful,  then the
membrane is replaced.

SBP  utilizes  a cross-flow  filtration   mechanism  to
continuously  clean the surface of the membrane,  hence
minimizing fouling.   In this mode,  the feed stream is
directed parallel to the membrane's surface resulting in  a
cleaning action which minimizes  the buildup of materials on
the membrane's  surface.

Since  all membranes eventually foul, a cleaning cycle is
necessary to restore flux and operability.  Many membrane
systems have limited abilities  to  be  regenerated due to
restrictions  in the choice of cleaning chemicals. The SBP
formed-in-place  membrane is compatible with a wide  range
of chemical cleaning methods, enabling  in-place regeneration
of flux.    In situations where  the membrane becomes
irreversibly  fouled,  the formed-in-place membrane  can be
stripped and reformulated  on-site.

3.5 Demonstration Results

The  results of the  demonstration test  program  at the
American Creosote Works  site provides information relevant
to the application of  SBP's  technology to other waste types
and sites.

SBP's  membrane technology can be used as an integral part
of a  remediation system  to significantly reduce the volume
and toxicity  of  contaminated wastewater. The technology is
particularly suited  for  the  treatment  of contaminated
groundwater as part of a pump  and treat system. The
technology  reduces  risks to  human  health and the
environment by transferring the  contaminants  to a smaller
volume facilitating destruction or detoxification by other
technologies.
                                                        11

-------
 The system  is simple to operate, reliable and requires a
 minimum of operator attention or maintenance once the
 membrane has been formed. The stability of the system
 makes  it  particularly suitable for  long-term  use as is
 necessary for extended pump and treat remedial  programs.

 The demonstration at the American Creosote Works was
 designed to evaluate the two most critical process  parameters
 for membrane systems;  volume  reduction  and contaminant
 reduction. A  summary of the demonstration results for these
 critical  processes  parameters  are  presented  below. A
 discussion of the  demonstration  results  and  process
 performance,  as they relate to  applicability to other wastes
 and  sites, follows in the  subsequent chapter.

 Volume  Reduction
The  claim that the system can be  operated to recover 80%
of the feedwater volume as permeate was achieved. Average
water recovery (volume reduction) for the first five runs was
83%. The volume reduction for the extended run (day six)
was  96%,  and represents the  maximum volume reduction
capability of the unit for the waste stream tested.

Contaminant  Reduction
The process did  not achieve the  developer's claim of 90%
overall removal of the semivolatiles present in the feedwater
(on the average,  a 74%  reduction was achieved). However,
the process does effectively  remove polynuclear  aromatic
hydrocarbons from the feedwater and place them in the
concentrate.   Overall,  removal  of polynuclear aromatic
hydrocarbons  averaged 92%. Removals  of individual PAHs
range  from 78%  to well  over 94% for individual two, three,
and four ring PAHs.

Other high molecular weight pollutants, such as oils and
dioxins, are  also rejected from  the permeate with high
efficiency  (93%  for oils and  >99% for dioxins). However,
removal of low molecular weight phenols  is much less
effective, with values between 15 and  21%.

Depending on how a system is used, i.e.. level of volume
reduction and quality of permeate,  operating plus capital cost
could be as low as $200/1,000 gallons. Capital cost for an
averaged size system is approximately  $300,000.
3.6 Discussion of Demonstration Results and
        Applications

The  results and observations from the SITE demonstration,
coupled with  information supplied   from the vendor, is used
to discuss the applicability  of SBP's  membrane technology
to the treatment of organic  contaminated wastewater. The
SITE demonstration was designed to evaluate the innovative
features of  SBP's  process as  a volume reduction technology.

The  SITE demonstration took  place at  the American
Creosote Works  in  Pensacola,  Florida  and  utilized
groundwater   contaminated    with   creosote    and
pentachlorophenol.    Creosote  was  chosen as a testing
material for two reasons.

  1.     Creosote  is  a complex  mixture of over 250
        individual  compounds,  dominated by  polynuclear
        aromatic hydrocarbons  and phenolics,  and  exhibits
        a wide range of chemical  and physical properties.
        The  wide molecular weight distribution of the
        organic contaminanta  is  an excellent challenge
        material for a membrane process, allowing  for
        analysis of removal  efficiencies over a wide range
        of feed characteristics.

 2.     Wood preserving waste  contaminated  aquifers
        represent  a  significant    and    widespread
        environmental  problem.     Results  from this
        demonstration could  be  directly applicable  to other
        wood preserving waste  sites.

A  pumping  well  recovered  the   creosote   and  PCP
contaminated  groundwater from  the site. The  groundwater.
which  contained aqueous and dense free product fractions.
was allowed to settle and the  aqueous phase retained for the
study.  The  aqueous phase  was  diluted  with carbon-treated
potable water in order to  adjust the concentration  of the
semivolatiles in the feed to fully test the concentrating
capabilities  of the tiltration unit.

Contaminant  Reduction

The utility of a membrane system is its ability to remove
contaminants from a wastewater stream and concentrate
them into a reduced  volume. The contaminant reduction is
the percent decrease in specific  contaminants from  the feed
to the permeate (discharge).    The higher the percent
contaminant reduction, the  more effective  is the membrane-
at removing contaminants from  the  waste stream.
                                                        12

-------
 It is  important to note that the applicability of the technology
 cannot be made solely  on the percent contaminant reduction.
 Since contamination  is  reduced as a percentage of the
 concentration in the feed, the quality of the permeate  is
 dependent on  feed concentrations. In order to assess
 applicability, the predicted quality of the permeate can  be
 estimated by calculating contaminant reductions from the
 feed. The  estimated permeate quality can  then be compared
 to  site  specific  discharge standards.

For   the   demonstration  at  the  ACW site,  the   total
 concentrations of semivolatile contaminants for each run are
 summarized in Table 1 for the feedwater and permeate.  The
 system  was evaluated by  comparing  the total concentrations
 of  these compounds in the feedwater against the permeate.
 Over the six day period, an average overall rejection of 74%
 was achieved. Thus, starting with a feedwater containing on
 the average 90  mg/L   of total designated semivolatile
 components,  the  composited permeate, accounting  for  80%
 of  the original  feedwater  volume, contained on the  average
 23 mg/L.  This  did not meet the vendor's claim for  90%
 removal, largely because of the noted inefficiency  with
 phenolics.  This  is  not totally  unexpected  since the
 membrane,  as  formulated, was not expected  to remove
 species  with molecular weights  less  than 200.
 TABLE 1.  Feed and Permeate Semreototites - Total Concentrations and
         Contaminant Reductions

                Total  Semivolatile Concentrations
                           mg/L

        Feed   Permeate  Contaminant Reduction
Run1
Run2
Run3
Run4
Run5
Run6
104
91
92
104
85
60
18
24
26
22
23
24
83
74
72
79
73
60
 A summary of the average concentrations for individual
 semivolatile compounds in the feed and permeate, along with
 the associated rejections,  for the six day demonstration are
 presented in  Table 2.  The results of the demonstration
 indicated that the pilot unit was capable of removing over
 94% of some PAHs but only  15 - 21% of the phenoiics.
 The permeate  generated  during the process was discharged
 directly to the local POTW.

 These results indicate, as expected, that the membrane is
 more  effective  in  removing  larger molecular weight
components (PAHs) than the smaller molecular weight
molecules  (phenoiics).    With  a complex feed  such as
creosote, it is difficult to achieve high reductions of all
components  and at  the same  time deliver adequate
throughput.    In this  application, the  membrane  was
formulated to maximize reduction of  the  more  toxic
poiynuclear aromatic  hydrocarbons. Passage of  the phenolic
compounds into  the permeate did not pose a significant
disposal problem since the local POTW  could accept the
phenols in their treatment system. At other sites, careful
attention should be made to local discharge  requirements and
available  treatment facilities.

The SBP membrane process would be most applicable to
wastewaters  containing large  molecular weight  organic
compounds  (PAHs,  dioxins/furans,   polychlorinated
biphenyls, and certain pesticides/herbicides). The system
can remove smaller weight molecular compounds (phenols,
benzene, toluene,  ethylbenzene,  xylenes) if larger molecular
weight compounds are  not abundantly present.  Removal of
smaller weight  molecular  compounds  can  be accomplished
by modifying the  structure of the formed-in-place membrane.
For  these applications the pores of the membrane are
reduced,  resulting  in higher retentions of smaller components
as well as a reduction in the flux (throughput) of the  system.
To compensate  for the  reduced  flux, either  additional
membrane modules can  be added or more time will be
required  to accomplish the remediation. In either case, the
overall cost may be higher.

The  SBP system may be most  suitable to treating relatively
dilute, but toxic, wastestreams in which the  percent reduction
of contaminants will allow discharge of the permeate without
further treatment. This feature makes the unit highly suitable
for polishing  effluents  as part of a multi-technology
treatment tram.    In this system,  the  primary treatment
technology can  be  utilized  to remove  the  bulk of the
contamination, with the  filtration unit being used  as  a  final
polishing  step.

If the  concentration of contaminants  in  the permeate does
not meet  clean-up requirements, then the permeate  can be
recycled  back through the  membrane to achieve the  targeted
effluent  quality.    Recycling  of the  permeate has  the
disadvantage of requiring additional  membrane  modules, or
additional time, both  of which increase treatment costs.

A number of mechanisms could explain the contaminant
reduction results, including rejection by the membrane on the
basis  of molecular weight or  molecular  size, rejection and
coalescence of dispersed oil in which  specific components
                                                         13

-------
are soluble, or even rejection simply by adsorption of the
PAHs  on inert suspended solids.

Examination of the results for the conventional parameters
tested in the feed and permeate (Table 3) provides some
insight into  the  separation mechanism.  High concentrations
of oil and grease found in the  feedwater suggests  that
considerable oil remained in a  dispersed or colloidal  form.
This  oil would be removed  by  a  membrane  with
ultrafiltration or hyperfiltration characteristics. Since the
PAHs  are more  soluble in oil than in water, concurrent
removal of the  PAHs entrained within the oil may have
occurred. The phenols  with  relatively high solubility in
water are, also as expected, removed  more  poorly.  This also
is reflected in the poor rejections calculated for TOC and
COD. Other contaminants, not  quantified by  the semivolatile
analysis,  also may contribute to the high TOC and COD in
the permeate.

TABLE 2. Individual Semivolatile Concentrations and Rejections (average of
         six daily runs)
TABLE 3. Conventional Parameters
ANALYTE
Phenol
2-Methyl phenol
4-Methyl phenol
2,4-Dimethyl phenol
Benzole Acid
Pentachlorophenol
Naphthalene
2-Methyl Naphthalene
Acenaphthylene
Acenaphthene
Dibeazofuran
Fhiorene
Fhenanthrene
Anthracene
Flnoranthene
Pyrene
Benzo(a) anthracene
Cluyseue
Benzo(b) flnoranthene
Benzo(k) flnoranthene
Benzo(a)pyrene
FEED
(mg/L)
4.90
2.31
6.92
1.82
(1 .42)
(2.42)
12.67
4.52
(0.14)
6.64
4.86
5.92
17.06
1.96
7.01
4.70
1.24
1.13
(0.«D
«M9
(0.31)
PERMEATE
(mg/L)
3.66
1.93
5.75
1.54
2.16
1.88
2.87
0.46
(0.02)
0.51
0.41
0.37
0.59
0.07
0.10
0.05
'0.03
'0.03
'0.03
'0.03
'0.03
REJECTION
(%)
20.6
16.5
16.9
15.4
a
a
77.7
89.6
a
91.7
91.6
93.6
96.6
36.5
96.6
36.9
>97.6
>97.4
a
a
a
Values in parentheses represent analytes with estimated values that are above instrument I limits
but below quantitation limits

' • Analytes not delected are presented ty an ', and the values represent one-fialf fle
quantitation limit.

a -Individual refections not calculated due to estimated values.
Analyte

Feed
(mq/L)
Permeate
(mg/L)
Rejection
%
Average of six runs
IDS
TSS
237
34
OIL/GREASE 191
TOC
COD
Volume
121
379
Reduction
190
<4
14
92
362

20
>88
94
24
7

The utility of a membrane separation system for treating
hazardous wastestreams is also dependent on the magnitude
of volume reduction. The volume reduction is  a measure of
the percent of the feed water  that can be generated as
cleaner permeate.  The  higher the volume reduction, the
greater the potential utility of  the  membrane system.
Volume reduction cannot be  solely used  as  an indicator of
membrane performance.  The quality of the permeate must
also be considered when  evaluating the applicability  of  the
technology. A high volume reduction with low permeate
quality is not acceptable since the permeate will not be
dischargeable and will require further treatment. When
designing  a  membrane separation system, volume reduction
and permeate quality must be balanced in order to develop
a cost-effective treatment meeting site-specific clean-up
criteria.

For the six-day demonstration  at the American Creosote
Works an 80% volume reduction  was achieved each day.
This level of volume reduction was set as a target prior to
the demonstration  and was easily attained. The level of
volume reduction was achieved by  continuously recirculating
the concentrate through the system. On the last day of
operation the process was allowed to run until the unit could
no  longer function,  representing the maximum volume
reduction for that feed.  The maximum volume reduction
was 96%.

The  relationship between volume  reduction and permeate
quality is exemplified by results  from the demonstration.
During the  six day demonstration,  grab samples  of the
permeate stream  were collected at the beginning, middle,  and
end of each run.  The  purpose  of  these  samples  is  to
document  changes  in permeate quality  during the course of
the batch filtration.  The analysis of the data reveals an
increase in total  semivolatile  content  of the permeate from
the beginning to  the end of each run.  Six day average
permeate concentrations  of  total semivolatiles  were  19.24
                                                          14

-------
mg/L at the beginning of the run, 24.17  mg/L in the middle,
and 29.95 mg/L  at the end of the run. In addition, on day
six,  when the  unit was allowed to run to a maximal volume
reduction  of  96%,   the  final  permeate  semivolatile
concentration  was 47.25 mg/L.

These  changes in permeate quality  during the filtration  are
due  to  increasing semivolatile contents  of the recirculating
concentrate. As the batch filtration  proceeds, the  surface of
the  membrane is challenged with  progressively higher
concentration of contaminants. Since the membrane can
only reject a certain proportion of the feed stream, the
concentration  of contamination in the permeate will increase.

When applying  a membrane  solution  to  a wastewater
problem  it is crucial to  evaluate the balance between
permeate quality and volume reduction.   Maximizing volume
reduction  is  important since it impacts  economics by
minimizing  the volume of wastewater  requiring  treatment.
However, the  quality of the discharged water is critical and
must be  maintained during the filtration  process. Treatability
testing  is necessary  to  determine  the optimal  balance
between  permeate quality  and volume  reduction.
Fouling Control

Fouling is the  loss of flux due to the buildup of components
on the surface of the  membrane. All membranes exhibit
some degree of fouling and eventually require cleaning to
restore flux.  Many membranes  foul readily and are not
amenable to cleaning for flux  restoration. If flux  cannot be
restored,  then  the membrane must be replaced resulting in
considerable expense and  downtime.

A major attribute of the  SBP system is its  ability  to
minimize fouling. SBP  effectively controlled  excessive
fouling,  in spite  of the problematical nature of the wood
preserving waste feed,  through a combination of  cross-flow
operation and membrane cleaning. Flux and pressure data
collected during  the demonstration indicated gradual and
slight  fouling of the membrane.  This slight fouling was
reversed  after  each two-run  cycle by a membrane cleaning
procedure. Analysis of the washwaters from the cleaning
process indicated that approximately 8% of the mass of
semivolatiles  remained in the system and were removed
during  the washing process.  The membrane cleaning process
effectively regenerated the  membrane to its original clean
permeate flux  conditions.  This enabled the  membrane  to be
reused, without the necessity to reformulate.
The ability  to  repeatedly regenerate the flux after the
cleaning  procedure is  a  good indication that the  formed-in-
place membrane is stable and can be used over an extended
length of time. In the unlikely event of an irreversible
fouling, the membrane can be cost-effectively and easily
reformed on-site with a minimum of downtime.
Operational  Reliability and  Implementability

Operational reliability and implementability is important in
deciding  the applicability  of the  technology  to  other
wastestreams and sites. Experiences from the  Demonstration
represent the most extensive compilation of operability data
for the SBP system at a hazardous waste site.

The  system proved to be  quite stable  and required a
minimum of attention over the six days of study. System
performance was relatively constant during  the six day test.
With feed concentrations of total semivolatiles ranging from
60.4 to 103.8 mg/L, the percent rejection averaged 74%.,
with  a narrow standard deviation of 7.5. SBP Technologies
was able to reproducibly achieve targeted volume reductions
of 80%.  Other than adjustment of  the pressure to maintain
flux and  the cleaning  of the unit, which consumed  about 2
hours  every  other day, there  was little need for an operator.
In a  commercial  installation some  means  of  on-line
monitoring  (eg.,  changes  in  pressure,   contaminant
concentration,  etc.)   could  alert  the operator to out-of-
specification operation or  out-of-compliance permeate.  It  is
estimated that the unit could be run by two operators (health
and safety requirements). Additional units could easily be
operated  by  the  existing personnel.

Other than the cleaning operation every other day, there was
no downtime during the  demonstration.  With the  exception
of the pump there  are no moving parts to break down or
require service.

The  process equipment  and supplies for the system  are
commercially available. This includes  the filtration modules.
membrane  forming chemicals,  pumps, tanks, process
controls,  gauges,  and  flowmeters.

The membrane formation procedure requires a high level  of
expertise  and may require trial  and  error  methods to achieve
the desired  separation characteristics. However, these  are
not obstacles  to implementation  since  the process  is
inexpensive  and  rapid.

The process is easily scalable and can  be modified by adding
or deleting modules in response to processing requirements.
                                                          15

-------
The addition of modules does  not affect the mode of
operation,  except  for additional  support equipment (pumps,
tanks,  plumbing).

Based on the observations from the demonstration,  it is
feasible that the membrane system can be effectively and
reliably operated over an extended time period as would be
necessary  for pump and treat remediations.

3.7 Applicable Wastes

While  this study of the SBP Hype&ration Unit was limited
to a single wastewater, the groundwater available at the
American Creosote Works site, the  results of the study along
with other results provided by  the vendor suggest that  the
technology would  have applicability  to other contaminated
groundwaters  and process waters. The  developer believes the
system can treat wastes with  100 to 500 mg/L of COD
where the molecular weight  of the contaminants to be
concentrated  are   over  about  200.  However,   the
characteristics  of the membrane  can be modified to treat
smaller molecular weight compounds.  More dilute feedwater
will necessitate additional cycles to achieve the  desired
concentrations in  permeate  and  concentrate streams.
However, the more dilute feedwaters  would also allow  for
higher fluxes.  Other than having an impact on cost and
throughput,  this should  not  adversely  affect  operation.

Wastestreams exceeding the target concentration range  (100
- 500  mg/L COD) would require reduced cycling to achieve
the required level of concentration. The effect of elevated
feedwater concentrations on  the rejection of individual
components may also need to be  determined by laboratory
resting. Data from this study indicates  ,a reduction in
permeate quality as the concentration  of the feed increases.

Based on the results obtained at the ACW site with the
membrane formulated to treat wood  preserving wastes,   the
composition of the  groundwater   can determine   the
applicability of the system. Groundwater rich in  PAHs would
probably  be   suitable  while  feedwater  where  smaller
molecular weight compounds  are a major pollutant would
probably not be appropriate. However, the vendor claims
that membranes could be formulated  to  separate small
molecular weight species (BTEX) such as those found in
hydrocarbon  contaminated  wastewaters (see  Appendix   B).
Since the former wood-preserving  facility had not used  the
newer  chromated  copper  arsenate  preservative,   no
information on the removal of metallic contaminants was
obtained.
 Cross-flow  filtration using the formed-in-place membrane
 may also be applicable to other wastestreams containing
 different high molecular weight organic  contaminants.  This
 might include polychlorinated biphenyls  (PCBs)  as might be
 encountered from a spill from a PCB transformer leak,
 particularly since the same preferential solubility in oil noted
 earlier  may prevail.  On the same basis,  the system may be
 useful for separating other emulsified or  dispersed  organics
 which do not lend themselves to simple physical phase
 separation. The system is also well suited to significantly
 reduce  the concentration of dioxins and furans  in wastewater.
 Reduction   of   dioxins/furans   encountered  in   this
 demonstration was greater than 99.9%.

 The developer believes  the membrane can be  customized to
 achieve different rejection characteristics that  could be
 applied  to a wide  range of  contaminants  (see  Appendix B).

 3.8 Site Characteristics

 The pilot-plant unit used in the  demonstration program
 required a level base large enough to accommodate the unit,
 and storage tanks  for the  feed, concentrate, and washwater.
 As was  used for the  SITE demonstration, a covered concrete
 pad is  recommended to  protect the equipment from the
 elements as well as  contain the  accidental release of
contaminated  materials.

 Clean water and  power were the only utilities needed. If
 necessary, the relatively small amount of clean water needed
 for washing of the membrane could be trucked  in and
 power for the compressor could be  provided by an  on-site
 generator. While it was not studied,  it may be practical to
 use permeate for washing.  Where the unit was  being used to
treat groundwater,  power also would  have to be provided for
 the well pumps.

Acquisition of groundwater for the unit may require the
development of an extraction well network, consisting of the
appropriate  pumps, regulators, and  plumbing.

Permit requirements and the mode in which the filtration
 unit is  operated may make it necessary to have additional
space for storage tanks for equalization of the permeate until
 analyses can confirm  acceptability for the POTW or  surface
 water body discharge.

 3.9 Environmental  Regulation Requirements

 Anticipating that the SBP  system would  be  used on
groundwater at a contaminated site, concerns would be local
                                                        16

-------
well-drilling   requirements    and   storage  of  pumped
groundwater. Depending on the size of these wells, their
productivity, and the capacity of the treatment  system being
installed, storage tanks may be needed as a reservoir, to
separate free oil (if needed), and to provide equalization of
feed  and  permeate. Such tanks  must  meet  regulatory
requirements  for design (permits, materials, etc.), consistent
with their  size and placement.

Discharge of the permeate to  a POTW would certainly
require a  NPDES/pretreatment   permit (or state  equivalent).
Depending on the operational mode used, the permit may
require continuous monitoring.    This would  prevent
averaging of high concentrations of contaminants in the
permeate late in a cycle with the low concentrations during
the  early  stages. Permits that  allow equalization  or
compositing  of permeate  during a  cycle  would be less
restrictive.

The reject  stream  (concentrate)   from  the  filtration  process
will contain elevated concentrations  of the higher molecular
weight contaminants. In the case  of creosote, these will
largely  be PAHs.  several of which  are  considered
carcinogenic.  Consequently,  the  concentrate  will probably
have  to be  considered hazardous and any subsequent
treatment could require that  the  facility have a "Treatment,
Storage, and  Disposal"  permit.

The washwater generated will also need to be addressed.
Since  it contains me  same PAH  (and phenolic) constituents,
it too will be hazardous. Because of the chemicals added,
recycle into the concentrate or disposal with the concentrate
may  not be possible unless  pretreated.

Additional regulations  approved in 1991 under the "Third
third" rule of the  Resource  Conservation and Recovery  Act
of 1976  (RCRA)  and the Hazardous  and  Solid Waste
Amendments of 1984 (HSWA) lists certain wood preserving
wastes as hazardous  (F032, F034.  F035).  Required
treatment  for these wastes has not yet been defined. In
addition,  RCRA and  HSWA  could  also designate the
permeate, the concentrate, the  washwater. and any other
wastes subsequently  generated as  a residual  from  a
hazardous  waste  under the "derived  from  regulation".
Consequently, these streams  also  would  be  considered
hazardous.  A recent  court decision leaves the  interpretation
of this matter unclear at this time.

Under  the  Comprehensive  Environmental  Response,
Compensation,  and Liability Act of 1980  (CERCLA)  and the
Superfund Amendments and Reauthorization Act of 1986
(SARA), EPA is responsible for determining the methods
and  criteria for the removal of contamination  from  a
Superfund site. The  utility  and cost-effectiveness of the SBP
filtration unit as  one segment of a treatment system would,
to an extent, be  dependent on the  final level of remediation
deemed appropriate and necessary at a particular site by
EPA. However, since the use  of remedial  actions  by
treatment  that ".  ..permanently  and significantly reduces  the
volume, toxicity, or mobility of hazardous substances" is
strongly recommended (Section  121  of SARA), inclusion of
the SBP filtration system could be attractive as part of the
remediation for  a site  contaminated with wood  preserving
chemicals.

Chlorinated dioxins/furans  are known to be  produced during
pentachlorophenol manufacture.  Limited testing  during this
demonstration indicated that low concentrations  of certain of
these  materials, particularly  the  octa isomers, were  present
in the groundwater and were concentrated in the concentrate.
The highly toxic 2.3,7,8-tetrachlorodioxin    (TCDD) was  not
found. Subsequent treatment or disposal  of the concentrate
would need to address  this  matter in more detail.

3.10   Materials Handling Requirements

Materials  handling requirements for the SBP unit involve
 1) the acquisition of feed material for the unit,
2) pretreatment.  and 3) residuals (permeate  and concentrate)
management.

Acquisition  of Feed

If the SBP filtration unit is part of a system used to treat
groundwater, the first need  is a well drilling rig  to provide
the well(s) from which the feedwater is to be obtained. Once
the wells  are drilled and developed, each  must be equipped
with a pump to draw up the necessary feed water.  Local well
drilling  requirements  would  have  to   be   taken into
consideration.

Pretreatment
At  some sites pretreatment may  be  necessary (as  it was at
this site) to remove  free oil and even suspended solids. Since
the developer has indicated that the filtration unit is most
effective when  operating with  a  feed  water having  a COD
range of 100 -500  mg/L  and  is most effective  in rejecting
materials with molecular weights greater than about 200.
pre-testing   will  be   necessary   to  assure  that  these
requirements  are consistently met. If the vendor's  system is
provided with relatively clean  ground- or process  water,  no
pretreatment  may be necessary.
                                                         17

-------
Permeate  Disposal  Options

The applicability of this membrane technology at a site is
dependent on the  quality of the permeate,  site-specific
discharge criteria, and the availability and accessibility of
local  public or industrial  wastewater treatment  facilities.  It
is important to conduct a treatability study to assess the
quality of the permeate and  to  determine  options for
disposal.    If the permeate quality is  not amenable for
discharge to surface waters or local  treatment facilities, then
the technology is  not applicable  to the site.

Prior to the initiation  of the demonstration at the ACW site
in Pensacola, a limited filterability test  was conducted on
contaminated groundwater  to determine if the permeate
would be accepted  by the local  POTW.  The permeate was
subjected to biological testing  (Ceriodaphnia) and chemical
analysis to determine  its  suitability for  discharge. The
permeate passed the local  POTW's  criteria and was directly
discharged to a local  sewer hook-up.

Several additional options  are available for permeate disposal
and are dependent on waste and site conditions, as well as
local  discharge  regulations and treatment  options.

   . The permeate quality may meet local standards for
     direct discharge to local surface water bodies. This
     would occur only if the level  of contaminants in the
     permeate was extremely low  and  meeting the strict
     requirements  for surface  discharge.

  . The permeate could be treated on-site with additional
     treatment equipment  to  reduce  contaminant levels for
     either surface water body discharge or sewer discharge.
     Treatment, such as with activated carbon, may  be
     necessary to  reduce  contamination to acceptable  limits.
     The use  of additional treatment equipment will increase
     remediation  costs  and  may   necessitate additional
     disposal   requirements.

   . The permeate may be recycled through the filtration
     unit, or processed through a  smaller unit, to further
     reduce  contaminants  for surface water body or  sewer
     discharge. The  secondary  filtration  unit may  have
     different membrane  characteristics as the primary unit
     to remove species  that were  not retained or require
     greater  reductions. This  option  would also add to the
     overall  cost  of the  remediation since  additional
     equipment and time would  be  required.

   .  If it is not feasible to reduce contaminant concentrations
     to levels adequate for on-site discharge, and if no local
     sewer hook-up is  accessible, then it may be necessary
     to transport the permeate  by  tanker  truck to  an
     acceptable treatment  facility.  This option  would only
     be economically feasible  if the membrane process
     drastically reduced the volume of  a wastestream  that is
     very  costly or difficult to treat (e.g. dioxin contaminated
     wastewater).
Concentrate  Disposal  Options

The SBP membrane process minimizes the quantity of waste
requiring  extensive  treatment  by  concenuating   the
contaminants into a reduced volume while producing a
cleaner permeate  for discharge. Since the contaminants are
not desuoyed by the process it is  necessary to consider
disposal options for the reduced volume  concentrate  stream.
If the treatment options for the concentrate stream do not
reduce overall treatment  costs or provide  a reduction in risk
to human health  and the  environment, then the  membrane
system is  not a feasible remedial technology.

Optimally, a disposal option that can  permanently destroy or
immobilize the contaminants in the concentrate   stream  on-
site is  preferable  to off-site transportation and  disposal.

A portion of the  concentrate from the SITE demonstration
was utilized by  SBP to develop a bioremediation technology
that could be coupled  co the filtration unit to produce a
treatment  system  for on-site destruction  of a major  portion
of the waste. The  system uses  a  two-stage  bioreactor
containing several naturally occurring  strains of soil bacteria
capable of mediating PAH and PCP contamination. SBP
uses the  membrane system'  to reduce the quantity of
wastewater input into  the bioreactors  and  to optimize
contaminant concentrations to support the biomass.  The use
of the  concentrate as  a  feed to the bioreactors extends the
utility of this volume limiting technology by reducing the
volume of wastewater that must be processed, therefore
reducing  equipment costs  and  site  space requirements. The
results  from SBP's bioremediation studies are  presented in
Appendix  B.

Additional disposal options available for the concentrate
stream are discussed below.

  • The concentration process  enhances  the  calorific  value
    of most organic wastes. This enables the  utilization of
    thermal technologies as  a means of destroying  the
    organic contaminants. The feasibility  of using and
    choosing a thermal technology is based on the nature of
    the organic  contaminants.  Concentrates from petroleum
                                                         18

-------
     based contamination could be readily used for fuel
     blending,  while  concentrates from other sources (such
     as wood preserving wastes) would require careful
     testing to determine selection of the appropriate thermal
     technology.  Thermal destruction could be accomplished
     on-site  (mobile units) or transported  off-site.

   •   Concentrates  containing  highly toxic constituents, such
     as PCBs  and dioxins/furans.  which are not amenable to
     biodegradation or thermal treatments, can be chemically
     neutralized  by  processes such as dechlorination. The
     neutralized waste could  then be disposed  of in  a
     conventional manner.

 3.71 Personnel Issues

The  system  requires  little  attention once operation  is
underway. Flux  must  be maintained by manually adjusting
pressure.  In  a larger-scale  system,  this   could  be
accomplished automatically.  Similarly,  the  transfer  of
concentrate  back to the feed side of the unit would be
automated,  requiring  an operator only  to  monitor the
permeate quality and flux. Washing of the system with clean
water every two days required  about   2  hours.  The
frequency of washing can vary with the waste  and the
operating mode. It is estimated that  two operators would be
necessary, primarily  for health  and  safety  concerns.

In order  to  assure protection of workers  during the
remediation,  all on-site personnel should have an OSHA 40
hour  health  and safety training  and an annual 8-hour
refresher course.

3.72 Potential Community  Exposures

Contaminant exposure to the community from the SBP
filtration unit is minimal. Potential community exposure
may  occur  during  the  filtration of  volatile  organic
compounds and can be contained through the implementation
of emission  control  equipment.  Potential exposure  of
contaminants could occur from  the accidental discharge  of
waste and process water from the equipment and well
distribution network. This could be minimized by placing
all tanks and equipment on containment structures, and
utilizing leak  detection  systems  for plumbing.
                                                        19

-------
                                                SECTION 4
                                         ECONOMIC ANALYSIS
4.7  Introduction

The primary purpose of this economic analysis is to estimate
costs  (excluding  profit)  for  commercial-scale  remediation
using the  SBP filtration unit. With realistic costs and  a
knowledge  of the bases for their  determination,  it should be
possible to estimate the economics for operating similar-
sized systems at other sites utilizing scale-up cost formulas.
Among such scale-up  cost formulas available in the literature
for chemical process plant  equipment is the "six-tenths rule".
The six-tenths rule  is an exponential method  for estimating
capital costs from existing  equipment costs. If the cost of  a
piece of equipment of size or capacity q, is C,, then the cost
of a similar piece of equipment of size or capacity  q2 can be
calculated  from:
The value for n in this study was taken as 0.6.

it  was assumed that the performance of commercial-scale
equipment will  be the same as that demonstrated here. This
economic analysis  is based on  assumptions and  costs
provided by SBP,  on results and experiences from this  SITE
demonstration, and  on best  engineering  judgement as
practiced by the authors. The results are presented in such
a  manner that if the reader disagrees with  any of the
assumptions  made  here, other conclusions can  be  derived
irom  such other assumptions.

Certain actual or potential costs were  omitted because  site-
specific engineering aspects beyond the  scope of this  SITE
project would be required.  Certain functions were  assumed
to be the obligation of the responsible party or site owner
and  also were not included in the  estimates.
Cost  figures  provided  here  are  "order-of-magnitude"
estimates, generally +50%/-30%,  and are representative of
charges typically assessed to  the  client by  the vendor,
exclusive of profit.

An economic analysis for the remediation of a hypothetical
site  is  presented  in  sub-section  4.6.    This exercise
demonstrates the application of the costing information
derived from this  study to a realistic remediation scenario.


4.2  Conclusions

  The total annual cost to  operate a 12-module  filtration unit
  ranges between 35  14.180 and  $1,209,700,   depending on
  whether  effluent treatment and  costs are  considered, the
  flow rate through the unit, the cleanup requirements, and
  the cost of effluent treatment  and  disposal  (if required).

  Effluent treatment and disposal costs, if considered,  could
  account for up to 60% of the total cost. Labor can
  account for up  to  40% of total annual  costs. Processing
  costs are  more  dependent on  labor costs than equipment
  costs.

  The cost  per  1,000 gallons  can be broken  down by flow
  rate as follows (for with  and without  effluent  treatment
  and disposal costs):
             With Effluent  Treatment Costs

  24 cpm                    12 gpm           7.2 gpm

  $228-522/1,000  gal      $456-1,044/1,000 gal    $760-1,739/1,000 gal
                                                         20

-------
             Without Effluent Treatment Costs

    24qpm                12gpm           7.2 qpm

     $222/1.000 gal        $444/1,000 gal       $739/1,000 gal


    As  expected, the cost category  having the largest impact
     and variability on  total cost was effluent treatment and
     disposal.

4.3  Issues and Assumptions

This section summarizes the major issues  and assumptions
used to evaluate the cost of SBP's filtration unit. In general,
assumptions  are based on information provided by SEP.
Certain assumptions were made to  account for variable site
and waste  parameters  and will, undoubtedly, have to be
refined to reflect site-specific conditions.

System Design and Performance Factors

The SITE demonstration  used a four-module filtration unit.
For a  full-scale remediation,  twelve  of the same modules
instead of four would be used with a portable generator for
power, a mix tank,  and a single pump and motor.

No assumptions as to the site size or volume of waste to be
treated were made. It was assumed that the  same unit would
be operated at different flow rates for a one year period to
obtain the  desired  results. For  example,  at the maximum
assumed flow rate  of 24  gpm,  2.6  million  gallons of waste
would be treated in 230 days of operation. The annual cost
was then divided by the volume  of waste that would be
treated at a particular flow rate to obtain $/1000 gal.
days per week. The extra hour each day will be used for
cleaning and maintaining the unit.  A site supervisor will
visit the site for approximately two to three days each month
for oversight purposes. The two-person crew could operate
up to  three 12-module  systems.   If  more modules are
required, additional manpower would be needed.

 Utilization Rates and IMaintenance Schedules

The filtration unit was assumed to be utilized for 230  days
out of  a possible 365 days a year. Scheduled maintenance
was assumed to  be performed  during normal operating hours.

Financial Assumptions

For the purpose of this analysis, capital equipment  costs
were amortized over a  7-year period with no salvage value.
Interest rates, time-value of money, etc. were not taken into
account.

The following is a list of additional assumptions used in this
study.

   •       Access to the site is available.
   •       Utilities, such as electricity, water, telephone, is
          easily accessible.
   •       The  permeate stream  will  not  require further
          treatment.
   •       A hook-up  to  the  appropriate outlet (sanitary
          sewer,  storm  sewer, surface water  body) is
          available on or near the site.
   •       There  are no  wastewater  pre-treatment
          requirements.

4.4  Basis for Economic Analysis
System Operating Requirements

No  assumptions  regarding  percent rejection  or  outlet
contaminanr concentrations were made.  Based on results
from the  SITE demonstration, a volume  reduction of 80%
between waste and concentrate was assumed.  Costs  per
 1000 gal. treated were calculated for 24. 12 and 7.2 gpm
flow rates: the last corresponding to what was demonstrated
in the SITE program. Flow rates, the amount of recycle,
and  the initial  concentration of contaminants may impact
costs significantly.

One equipment operator/supervisor and one technician will
operate the unit and be on-site eight hours per day, although
the system will be operated   only seven hours per day, five
In order to compare the cost-effectiveness of technologies in
the  SITE program,  EPA  breaks  down  costs  into  12
categories shown in Table 4 using the assumptions already
described. The assumptions used for each cost factor are
described in more detail beiow.

Site Preparation Costs

The amount of preliminary preparation will depend on the
site and is assumed to be performed by  the responsible  party
(or site owner). Site preparation responsibilities include site
design  and layout, surveys and site logistics, legal searches.
access rights and  roads,  and  preparations for support
facilities, decontamination facilities, utility connections, and
auxiliary buildings. These preparation activities are assumed
                                                        21

-------
 to be completed in 500 staff hours,  At a labor rate of
 $50/hr.   this  would equal $25,000.

 Other significant costs associated with site preparation
 include  construction of a pad and cover, well  drilling  as well
 as buying and installing  a groundwater pump, holding tanks,
 and  associated  plumbing.

 The cost to construct a concrete pad and cover to support the
 unit and protect the unit  from the elements is  estimated to be
 520,000.

 Based on the SITE demonstration, the cost to drill a well
 was  assumed to be $5,000. To achieve the appropriate
 maximum groundwater extraction rate of 24 gpm. three
 recovery  wells  are required,  resulting  in a  cost of
 approximately  $15,000. A 5200 gal.  holding tank  cost
 55,000.  Using the "six-tenths rule" to  scale-up,  the cost of
 a 10,000 gal. tank for a full-scale remediation was assumed
 to  cost  $7,400.  Three tanks will  be required, resulting  in a
 cost of  $22,200. A  1/2  horse-power pump  cost $1,035  for
 the SITE demonstration.  A pump for each well would  cost
 a total of $3,105. These additional costs amount to about
 $40,000.

 Therefore, the total site preparation costs for a full-scale
 remediation would be about 385,000 as shown in Table 4.

 Permitting and Regulatory  Costs

 Permitting and regulatory costs include actual permit costs,
 system  health/safety monitoring, and analytical protocols.
 Permitting and  regulatory costs can vary greatly because
 they are very site- and waste-specific. For this cost estimate,
 permitting and regulatory costs are assumed to be 5% of the
 equipment costs. This assumption is based on   operation at
 a Superfund site.    At RCRA corrective  action  sites
 permitting and regulatory costs may  be higher  and an
 additional 5% of the equipment cost should be  added.

Equipment  Costs

 Capital equipment costs are for a twelve-module filtration
 (unit equipped with a portable generator for power, a mix
 tank,  and a single pump and motor  ail mounted on a trailer
 with  associated instrumentation, alarms and controls.
 Variation in  equipment costs  from site to site should not be
 significant.  However, based on the cleanup requirements
 and the material being treated,  the flow rate  through the
 system may  vary dramatically  resulting in a wide range of
 costs per unit treated.
Based on SBP's  capital cost estimate of $300,000 for 12
modules, each module  would cost $25,000.

As recommended by SBP, equipment costs were  amortized
over 7 years, with no salvage value at  the end of that time
period, giving an  annual cost  of $42,850 as shown in Table
4, without any  interest  factor.
SBP's  filtration units are mobile and are designed to  move
from site to site. Transportation costs are only charged to
the client for one direction of travel and are usually included
with mobilization  rather than demobilization.  Transportation
costs are variable  and dependent on site location  as well as
on applicable oversize/overweight load  permits, which vary
from  state to state.  The total cost will depend on how many
and which state lines are crossed.

The system is designed to be ready to operate as mounted on
the trailer so mobilization costs  should be primarily the cost
of travel and the time to connect the plumbing  and adjust the
membranes if  necessary. The startup labor  cost  is included
in the total labor cost component and includes relocation and
or hiring expenses.

The  cost of health monitoring programs has been broken
down into  two components - OSHA  training, estimated  at
$l,000/person    and medical   surveillance,  estimated  at
$500/person for a  total cost  of  $l,500/person. For  two
people, on-site, this would be  $3,000 Depending on the
site,  however, local  authorities  may impose  specific
guidelines  for monitoring  programs.  The stringency and
frequency  of  monitoring  required  may have significant
impact on the project cost. A conservative estimate  of
$5,000 was assumed as  shown in Table 4.

Labor
Labor costs may be  broken down into two major categories:
salaries  and living  expenses.   SBP  estimates  that the
equipment will  require two on-site  personnel for  operation
and maintenance. Due to the extended time requirements for
major groundwater restoration projects, SBP plans for hire
local operators or relocate  personnel to the site. These
actions  would minimize costs  associated with living
expenses. A cost  of $5,000  is estimated for hiring and or
relocation.
                                                          22

-------
TABLE  4  Estimated Costs for SBP  Filtration  Unit
      COST  COMPONENT                                                        TOTAL



       1.  Site Preparation Costs  *                                                      $85,000

       2.  Permitting & Regulatory  Costs *                                             $15,000

       3.  Equipment Costs  (amortized over  7 years)                                      $42.850

       4. startup*                                                                    $5,000

       5. Labor                                                                      $199,080

       6.  Consumables and Supplies
            Health & Safety Gear                                                      $3,000
            Maintenance   Supplies                                                       $500

       7.  Utilities
            Telephone                                                                 $6,600
            Electricity                                                                 $2,000
            Sewer/Water                                                              $2,000

       8.  Effluent Treatment &  Disposal (Concentrate)                               513,9154695,520

       9.  Residuals/Waste  Shipping, Handling                                           $46,000
            and Transport  Costs

       10. Analytical Costs                                                           $60,000

       11. Facility Modification,  Repair &  Replacement                                  $37,150

        12. Demobilization  Costs  *                                                      $10,000

       TOTAL (WITHOUT  CONCENTRATE DISPOSAL)                              $514,180

       TOTAL (WITH CONCENTRATE   DISPOSAL)                             $528,095-$!,209,700

         * one-time costs
                                                      23

-------
 Site supervision will require periodic visits from the main or
 regional office to oversee the progress of the remediation.
 Per diem is assumed to be $125 per day per person, but may
 vary widely by location. This rate is a liberal estimate
 assuming that cleanups may occur in some of the more
 expensive areas of the country. Travel to and from the site
 (periodic supervision) is estimated to be  $800/visit. One
 rental car is assumed to be obtained at a rate of  $55/day.

 Supervisory and  administrative staff will consist of an  off-
 site program manager  at  $75/hour.   The  SBP filtration
 system  will operate 7 hours per day, 5 days per week. One
 equipment operator/ supervisor  at  $50/hr.  and one technician
 at $35/hr. will be on-site  8 hr./day. The labor requirements
 and rates are detailed in Table 5.

 Consumables  and Supplies

 There are two items to consider  under this  cost category.
 The first is health and safety   gear which include hard hats,
 safety glasses,  respirators  and  cartridges,  protective clothing,
 gloves,  safety boots, and  a photoionization  detector monitor,
 all estimated at  $l,500/person. For two people  this totals
 $3,000.

 The second  item  is maintenance supplies (spare  parts, oils,
 greases and  other  lubricants,  etc.) estimated at  1%  of the
 annual  amortized  capital  costs or approximately $500. The
 cost of membrane forming chemicals are inconsequential
 (less  than  $200).

Utilities

 Telephone charges are  estimated at $500/month plus an
 additional 10% for fax  service or $550/month.  This will
 total $6,600   annually.

 Electric usage  is  estimated by  SBP to cost about  $10/day  or
 52.000  annually.  Combined sewer and  water usage  costs is
 assumed to  be about $0.05/1000  L ($0.20 per  1000 gal).
 Based on the  SITE demonstration results, approximately  150
 gallons of water were used  to flush a 4-module system.
 Hence  a 12-module system was assumed to use three times
 as much water or about  500  gallons/day. This  would  cost
 about  $10/day or $2,000 a year  as well.  This does not
 consider discharge of permeate, which may incur additional
 cost.
TABLES. SBP Liter Requirement!vtd Rate*
Living and Travelling Expanses:  3 days/month for 12 months
Per Diem $125/day/person x 1 person x
  3 days/week  x 12 weeks

Rental Car $55/day x 7 days/week x
  52 weeks

Travel $800/trip x 12 months

Salaries:

Program Manager - S75/hr(1) x 8 hr/day x
  36 days
  $4,500


  $1,980

  $9,600





 $21,600
Operator/Supervisor -$50/hr(1) x 8 hr/day x
  230 days                          =      $92,000
Technician -$35/hr(1) x 8 hr/day  x
  230 days

Relocation/Hiring
Total Labor
 $64,000

  $5,000


$199,080
(1)         Includes  salary, benefits, and administration/overhead
           costs but excludes profit.

Effluent  Treatment  and Disposal

Two process  streams are produced by the filtration unit.  The
permeate is considered to be essentially free of contaminants
and is assumed to meet  standards appropriate  for discharge
to a POTW.  The concentrate is the  reduced-volume portion
of the waste stream containing the  enriched contaminants.
This  stream would require further treatment such  as
biological  degradation,  incineration,  fuel-blending,  or  some
other  process appropriate to the type  and  concentration of
contaminants.

The  filtration system being evaluated is  a volume  reduction
technology, and as such minimizes the volume of wastewater
that would require treatment. For this demonstration, the
technology was demonstrated as a method to  reduce the
volume of wood preserving waste contaminated groundwater.
Therefore,  treatment of  the concentrate is not part of the
demonstrated technology  and it is not necessarily  appropriate
to consider costing for this parameter. However, the cost for
ueating  these  effluents can  be a substantial  factor in
designing a remediation program.  Based on these issues,
overall costing wiil be  calculated both with and without
effluent treatment  and disposal costs.
                                                           24

-------
Two concentrate disposal options are considered in this
exercise. The first, bioremediation, is a system developed by
SBP Technologies. SBP claims that their biodegradation
system  provides  on-site  destruction  of PAH  contaminants.
At this  time, the bioremediation system has not been tested
as a  full-scale  process.  SBP has provided a projected cost
estimate  of  10-40  cents  per  gallon  of groundwater
contaminated with 100-2000 ppm of PAHs for its full-scale
bioremediation   system.

The second disposal  option for the  concentrate is more
conventional  and was derived from  the demonstration.
Based on the characteristics of the concentrate, fuel blending
is considered a  viable disposal option, resulting  in a cost of
$1.50/gallon.

It is  important  to note that effluent treatment costs can be
very high and  are dependent on specific waste and site
conditions.  Cost estimates for this  exercise were based  on
waste and  site  characteristics  of the demonstration.

Based on the  SITE  demonstration, the  concentrate accounts
for  20% by  volume of the contaminated groundwater
influent  stream  to  the  filtration unit.   The volume of
concentrate generated each day and the range of costs  for the
three different flow rates suggested by SBP are  shown in me
following   table  for  the  bioremediation   system  and
conventional  disposal:
Gallons of Waste Treated/Day
                         24gpm

                         10.060
Gallons of Concentrate
Generated/Day (assumes 20%)    2.016
Annual Treatment Costs
Eioremediation

Annual Treatment Costs
Conventional
$46,370
$185,470
12gpm      7.2gpm

 5,040       3,024


 1,008       605

 $23,169    $13,915
 $92,736
$66,660
                          $696.520   $347,760   $208,725
Effluent treatment  and disposal costs can range  from
$14,000-$700,000   depending on the flow rate through the
filtration  unit, the  mode of treatment, and the cost of
treatment in SBP's bioremediation system.  The reader is
cautioned  that SBP's bioremediation process was not
investigated here and the effectiveness and costs of this
process have not been independently verified.

Residuals/Waste Shipping. Handling and Transport Costs

Waste  disposal costs including  storage, transportation  and
treatment costs are assumed to be the obligation of the
responsible party (or site owner),  it is assumed that residual
or solid  wastes  generated from  this process would  consist
only of contaminated health  and  safety gear, used materials,
etc. Landtilling is the anticipated disposal method for this
material  and costs were once again  derived from the
demonstration test. Twenty-six 208 L  (55 gal) drums of
waste  were generated  during  the  demonstration test. This
resulted in  approximately  four  drums  of  solid  waste
generated  each day of operation. However, due to intensive
sampling activities  during  the demonstration, excessive solid
waste was generated. Under  actual remediation conditions,
substantially  less waste  would be  generated. It is estimated
that approximately one  drum  of solid waste would be
generated each day of operation. At a disposal cost of
$200/drum,   the total yearly cost of disposal  is  estimated to
be $46,000.

Analytical Costs

Standard operating procedures  do not require planned
sampling and analytical  activities. Periodic spot checks may
be executed at SBP's discretion to verify that equipment is
performing properly and that  cleanup criteria are being met,
but costs incurred from these actions are not  assessed to the
client. The client  may elect, or may be required by local
authorities, to initiate a  sampling and analytical program at
their own expense.

For this  cost analysis, one sample per day  for  100  days at
$600/sample was assumed to be  required by  local authorities
for monitoring  and  permitting purposes. This  would total
approximately   $60.000.

Facility Modification   Repair and Replacement Costs

Since site preparation costs were assumed to be borne by the
responsible  patty  (or site  owner),  any modification,  repair,
or replacement to the site was also assumed to be done by
the responsible party  (or site owner). The annual cost of
repairs and maintenance was  estimated  by SBP  to  be
$37,150.

Demobilization Costs

Site  demobilization will include  shutdown of the operation,
final  decontamination and removal of  equipment,  site
cleanup and restoration, permanent storage  costs,  and site
security.  Site demobilization  costs  will  vary depending on
whether the treatment operation  occurs at a Superfund site
or at a RCRA-corrective  action  site. Demobilization at the
latter type of site will require detailed closure and post-
closure plans and permits. Demobilization at a Superfund
                                                          25

-------
 site does  not require  as  extensive post-closure  care; for
 example,  30-year monitoring is  not  required. This analysis
 assumed site demobilization costs are limited to the removal
 of all equipment and facilities from the site. It is estimated
 that demobilization would take about  two  weeks and consist
 primarily of labor charges. Labor costs  include salary and
 living expenses.  See "Labor Costs" for information  on labor
 rates.  Demobilization is estimated to be  $10,000.

 Grading or recompaction requirements of  the  soil will vary
 depending on the future use of the site and are assumed to
 be the obligation of the responsible  party  (or  site owner).

4.5  Results

 Table  4  shows the total  annual cleanup  cost  to  range
 between  $514,180 and  $1,209,700. This is based on me
 assumption that the remediation will take one year. Most
 applications for this technology will require several years, as
 in pump-and-treat remedial projects. Since many of the cost
 factors  are one-time, the overall  $/gallon cost will go down
 as the length of the project increases. This is illustrated in
 the hypothetical site example  in the subsequent sub-section.
 The  total  cost  is  also   highly dependent on  whether
 concentrate treatment and  disposal is considered  as part of
 SBP's  technology  and  responsibility.  Concentrate  disposal
 costs can vary widely,  and are dependent on technical and
 regulatory issues related to the  waste  characteristics.
 Therefore, if concentrate disposal costs are considered, this
 category could account for up to 60% of the total costs.
 Without concentrate disposal, labor is the dominant cost,
 accounting for approximately  40%  of the cost.  Equipment
 costs represent a relatively minor component. Furthermore,
 the system can be easily  scaled-up by adding  12-module
 units. SBP indicates that up to three  12-module units could
 be operated without adding additional labor. This would
 significantly reduce overall treatment costs. The smallest
 cost categories appear to  be those associated with startup,
 and consumables and  supplies.  All other cost categories
 appear to contribute to the total cost about equally (i.e. 5-
 10%).

 The costs  per 1,000 gal is dependent on  the flow rate,  the
 duration of the project,  whether concentrate disposal is being
 considered, and the  cost of effluent treatment  and disposal.
 These ranges are shown below and are based on a one year
 project.
With Concentration
Disposal
24 gpm     12gpm   7.2 gpm

   Cost par thousand gallons of feed

    $228    $456    $760
    $522   $1,044    $1,739
Without Concentrate
Disposal
    $222    $444    $739
In all of the  above analyses,  it should be  remembered that
costs  for 10 out of the  12 cost components were considered.
One  of the cost  components  not included  here was
permitting and  regulatory expenses.  Additional effluent
treatment and disposal  for the permeate was assumed to  be
not required.  If these factors  are  taken into account, costs
could  significantly increase.

4.6 Remediation of a  Hypothetical Site

The economic analysis presented  in the preceding sections
is based on costs for a one year remedial project.    The
dominant application of the membrane system is expected to
be for groundwater  restoration projects.  Since groundwater
restoration  projects can last for ten to twenty years, a
hypothetical  economic analysis is  presented to  illustrate the
application of the twelve factors in  developing a multi-year
project.

The hypothetical site contains groundwater contaminated
with  wood  preserving wastes  (creosote and PCP)  in
composition and concentrations similar to the feedwater
tested in this demonstration. The remedial plan calls for
containment  of the groundwater plume, with eventual aquifer
restoration.       A  hypothetical   model  predicts    that
approximately two  million gallons  of  groundwater is
contaminated, and that 10 pore  volumes of groundwater
(twenty million  gallons) must be treated to restore  the
aquifer. The groundwater will be extracted from me shallow
aquifer (ten to thirty feet below surface) through three wells.

The remedial design will utilize three 12-module filtration
units operating at  7.2 gpm/unit for a combined throughput of
2 1.6  gpm. Treatability testing identified that an 80% volume
reduction could be achieved, with the permeate meeting
discharge standards to me local POTW. The concentrate
from  the process  will be treated on-site by  a bioremediation
technology similar to SBP's proposed system at a cost of 40
cents/gallon,   [see Appendix B.) Based  on these  conditions.
and the economic  assumptions  previously  stated,  the
remedial time-frame will be  ten years.   Approximately
2.100.000 gallons of groundwater  will be treated  each year
                                                         26

-------
by the filtration unit, and 420,000 gallons of concentrate by
the bioremediation system. The total volume of groundwater
fo  be treated for the ten year project is 21 million gallons.

Table 6 is a summary of the costs for each of the twelve
criteria as they relate to the conditions set forth  in the
hypothetical analysis. Based on the requirements  of the
hypothetical  site,  the overall treatment  costs  for the
remediation is $300/1,000 gallons. It is important to note
the overall  $/gallon treatment cost is highly dependent on the
length of the remediation project. The longer the project.
the lower the  $/gallon treatment cost.
                                                        27

-------
TABLE 6. Hypothetical Site Cost Analysis




                  Ten Year  Project (1993  Dollars)










1.  Site  Preparation                                      $85,000




2.  Permitting and  Regulatory                              $15,000




3.  Equipment                                          $900,000




4.  Startup                                                55,000




5.  Labor                                             $1,990,800




6.  Consumables  and Supplies                             535,000




7.  Utilities                                            $106,000




8.  Effluent Treatment and Disposal                    81,669.248




9.  Residuals                                           $460,000




10. Analytical                                          $600,000




11. Facility  Modification Repair  and Replacement        $371,500




12. Demobilization                                        310.000




                                                      $6,247,548
                                 28

-------
                                                APPENDIX  A
                                       PROCESS  DESCRIPTION
A.1 Introduction
This section of the report presents  a  concise description of
the SBP Filtration Unit and its operation at the American
Creosote Works site. Factors involved in site and waste
selection are presented  in this section or in the body  of the
report to assist engineers and scientists in evaluating the
suitability of the  process for their own needs  at Superfund
and  other  hazardous  waste   sites.  Results   of  the
demonstration, including summaries of analytical test  results,
are presented in Appendix C.

A.2  Process Description

Membranes are being used increasingly for  the removal  of
dissolved and colloidal contaminants in wastewater streams.
Reverse osmosis (hyperfiltration)   is  well  known  for  its
ability  to concentrate  ionic  species  while ultrafiltration has
found broad utility for the removal of dispersed colloidal  oil,
non-settlable suspended  solids, and  larger  organic  chemical
molecules. One of the major problems  these  processes have
faced is the  fouling or blinding of the membranes after
limited use.  Various approaches have been developed in an
effort to  minimize this  deterrent.  Cross-flow  filtration, where
the contaminants are  constantly flushed or washed from the
membrane surface by  the  feedwater stream,  is one of these
approaches. The SBP  unit goes further. Rather than  a thin
polymeric membrane  requiring careful handling  to avoid
perforation,  a  membrane is  created  on a stainless steel
microfilter  support by the  introduction of a mixture  of
carefully selected  (and  proprietary) chemicals. This  approach
imparts  special   properties  and  allows   a  degree  of
customization  that  may be  difficult to  achieve with
conventional  membranes.  The resulting  "formed-in-place"
membrane  can  be designed  to provide properties  similar to
conventional hyperfiltration  or ultrafiltration.   as needed for
a  specific application.

The  filtration unit consists of porous  sintered stainless  steel
tubes arranged  in a  modular,  shell-and-tube configuration.
Multi-layered inorganic and polymeric "formed-in-place"
membranes are  coated at microscopic thickness on the inside
diameter of the stainless steel  tubing by the recirculation of
an aqueous  slurry of membrane formation chemicals.  This
"formed-in-place"   membrane  functionally  acts  as  a
hyperfilter.  rejecting species with molecular  weights as low
as 200. in addition,  surface chemistry interactions between
the membrane matrix  and the components in the  feed  play
a  role in the separation process. A relatively clean stream,
called the "permeate", passes  through the membrane while
a  smaller portion of the feedwater, retaining those species
that do not pass through the membrane, is retained in a
stream called the "concentrate" or "reject".

For efficient operation of a membrane filtration system,  it is
necessary to prevent the buildup of dissolved and particulate
species  on the  surface  of  the  membrane and in  the
membrane pores. The buildup of contaminants, termed
"fouling", can lead to a steady decline in the permeate   flux
(flow per unit area of membrane surface), eventually causing
cessation of flow.  To prevent or retard excessive fouling, the
SBP filtration unit is operated in a cross-flow mode (Figure
1) In cross-flow mode the feed stream is directed  parallel to
the surface of the  membrane. Material larger than the surface
                                                         29

-------
porosity  is temporarily retarded  on the membrane  surface
and  then swept clean by the cross-flow action  - if the fluid
velocity is sufficient. Meanwhile,  the portion of the stream
containing the smaller species passes through the  membrane.
The  goal  of cross-flow filtration  is not to  trap components
within the pore structure  of the membrane.

The  SBP  test  unit used in this demonstration operates with
four modules aligned in parallel. The  filtration  unit  is
approximately  13 feet long, 5 feet  wide, and 7 feet high and
contains an estimated total membrane area of up to 300
square feet.  Automatic  level controls provide for  unattended
operation with continuous feed  to the tank. Concentrate
recycle  flow also can be  controlled automatically. Figure  2
provides a schematic of the filtration unit.

At the American Creosote Works site, groundwater was
pumped from  a well to an aboveground storage tank where
a  quiescent period of  several  hours  allowed oil  and
suspended solids  to  coalesce  and separate. The  feedwater
stream to  the filtration unit was drawn from the mid-section
of the  storage  tank to  minimize introduction of  these
materials.  The pump that  drew material from the  tank  also
provided the compression for  the   system to  operate,
approximately  750 psig.

The  permeate leaving the filtration unit  was sampled as
required and  then discharged in accordance with permit
requirements. The concentrate was collected in a  smaller
tank until the  desired volume was accumulated.  It is then
recycled as feed until  the desired final concentration and
volume  are achieved. This mode  of operation  was  selected
for the  demonstration in anticipation of a  companion  study
of a proprietary  biodegradation process for the concentrate.
Alternate operating modes can be used  to achieve other
goals, depending on disposal plans  and options  for the
permeate  and  the concentrate.
Influent
Permeate
  o
    °o  T  o       Concentrate
              •   O Species > 200 Molecular Weight
               0    Species < 200 Molecular Weight



            Figure  1.  Cross-Flow  Filtration
Groundwater

V
1

F«ed

Membrane
~ Filtration -O
Feed Tank
*

i


*»• Membrane
Filtration Unit
Permeate I*
Disposal *



^ Cotic,enJtr.ate. Recycle* i
Concentrate
— ^ Storage
Tank

r
                                 Sampling  Points
 Figure 2. Schematic of  SBP  Filtration System
                                                         30

-------
                                             APPENDIXB

                          VENDORCLAIMSANDCASESTUDIES

            Note: Information contained in this  appendix was provided by SBP Technologies, Inc. and
            has not been independently verified by the SITE  Program.
B. 1   Introduction - HyperfiltrationSystem

SBP's hyperfiltration  system, which consists of porous
stainless steel  tubes  internally coated  with specially
formulated chemical  membranes,  has been demonstrated to
successfully treat water contaminated with a number of
hazardous or toxic  materials.  In  this system,  contaminated
ground and  surface waters are pumped through the filtration
system tubes and contaminants are collected inside the tube
membrane while "clean" water permeates the membrane and
tubes. The  system has been shown to be highly versatile
and able to effectively remove a variety of materials from
contaminated    waste    streams,   including   petroleum
hydrocarbons,  benzene, toluene,  ethylbenzene.  xylene.
chlorinated dioxins. chlorinated  furans, and heavy metals.
This extent of  versatility  is  provided largely  by the
application of several types of chemical membranes with
distinct permeability  and  ion exchange capabilities.

As discussed in this  Applications Analysis  Report, SBP used
this hyperfiltration technology in a pilot-scale field study,
performed  in conjunction with  the U.S. EPA Superfund
Innovative Technology  Evaluation (SITE)  Program. At the
American Creosote Works (ACW) site in Pensacola,  Florida.
groundwater  has  been  encountered  with  significant wood
preserving  waste contamination, of which a significant
percentage   is   high   molecular   weight  carcinogenic
polynuclear  aromatic  hydrocarbons  (PAHs)  and
pentachlorophenol (PCP), typically associated  with the wood
treatment industry.
In this field study, one type of chemical membrane was used
to coat the porous stainless steel tubing, with an aim to
provide the optimal separation efficiencies for the higher
molecular weight contaminants.    These  are generally
;egarded to be most "hazardous" molecules and are generally
more resistant to degradation in the environment. The
results confirmed that this membrane system was very
effective  in removing >95% (w/v) of the high molecular
weight  polynudear aromatic hydrocarbon  (PAH)
contaminants, the  most carcinogenic components, during the
monitored demonstration run. However, on the average,
approximately 20% of the  lower molecular weight phenolic
contaminants  were removed by  this  membrane type.  SBP
feels that this percentage can be improved by adjusting
membranes and flows in the field. The latitude to tinker
with the  system  was not available during the  formal
demonstration  period.

In the ACW study, the relatively heavy concentrations of
contaminants in the feed  material for the  hyperfiltration unit
precluded effective use  of other "tighter"  types  of
membranes as the  first pass barrier due to the potential for
fouiing. Optimally,  a combination system employing the
initial membrane used here to remove high molecular weight
contaminants, followed by  a "tighter" membrane to remove
lower weight phenoiics from the permeate of the first
membrane would have been more likely to provide the full
spectrum  of contaminant removal desired.
                                                     31

-------
 8.2 Results BTEX/Petroleum Hydrocarbon
         Hyperfiltration

 A. BTEX/Gasoline  Contaminated Waste Stream

 Other waste  streams,  highly  contaminated  with lower
 molecular weight molecules,  have  been  successfully
 hyperfiltered  and concentrated using alternate chemical
 membrane types  than  that used in the ACW field  site
 demonstration.  One example  has  been the hyperfiltration of
 waste water from  certain oil refining operations containing
 significant amounts of benzene,  toluene, ethylbenzene.  and
 xylenes (BTEX),  as well as other petroleum hydrocarbons.

 A summary of the hyperfiltration  results  of this  treatment is
 as  follows:
Contaminant Original Feed Sample
Permeate Sample    % Removal
Benzene
Toluene
Ethyl-benzene
Xylenes
TPH
2200 ppb
7640 ppb
2590 ppb
12200 ppb
483 ppm
34 ppb
92 ppb
15 ppb
100 ppb
3.5 ppm
98.5%
98.9%
99.4%
99.2%
99.3%
B.  Contaminated Landfill Liquids

In a separate  study, condensate  from a methane-recovery
operation at a municipal landfill was treated for the removal
and concentration of a large spectrum of contaminants,
including naphthalene, heavy metals, and BTEX. The results
of this  are:
Contaminant  Original Feed Sample
 Permeate Sample    % Removal
Toluene
Ethyl-benzene
Xylenes
Naphthalene
177 ppm
268 ppm
561 ppm
90.4 ppm
<5 ppm
<5 ppm
<10 ppm
<10 ppm
>97.2%
>98.1%
>98.2%
>89.0%
'All permeate readings were below the detectable limits for the analysis method
used in this study.

B.3 Conclusions - Hyperfiltration System
These results demonstrate  the capacity and versatility of the
hyperfiltration system in treating a variety of waste  streams
and  achieving  effective  volume reduction in removal of
contamination    from  groundwater.  municipal  landfill
leachates, or contaminated petroleum waste streams. The
vaiue of this technology can be  further enhanced by coupling
it with a biodegradation process. This  can  be  achieved by
usmg the hyperfiltration concentrate  as a biorcactor feed
stream, as well as by using the hyperfiltration system to
polish bioreactor effluent to yield two streams: one, a clean
stream suitable for discharge;  and the other, a polished
concentrate to  feed to the bioreactor. This creates  a closed
loop for targeted contaminants, and provides for an efficient
continuous  flow remediation design.

B.4 Introduction - Bioremediation System

In addition to the application of hyperfiltration technology to
the remediation of creosotecontaminated  groundwater from
the American Creosote  Works  (ACW)  site,  effective
biodegradation of creosote  and pentachlorophenol  has also
been achieved at this site by SBP using specially selected,
non-engineered  microorganisms  in  a pilot-scale bioreactor
system.     The  combination  of  these two  systems,
hyperfiltration and bioremediation. provides a novel  and
reliable means to first concentrate the waste feed to the
bioreactor to an optimal level for efficient bioremediation
activity, as well as to  provide for a final polishing step using
hyperfiltration  of bioreactor effluent.

For this  study on the ACW site, a bi-phasic bioreactor
design was utilized, operating in a semi-continuous flow
process,  having a hydraulic retention time of four days.
Groundwater, with contaminant  concentrations  as  high  as
7.000 ppm     creosote,   was  ueated  on-site.     This
demonstration  achieved a removal efficiency of greater than
99% for total  polynuclear  aromatic  hydrocarbons  (PAHs).
This  includes a removal rate of 98%  for the most
recalcitrant, and most hazardous fraction of the PAHs and
88% for PCP.  The  field  test proved  that SBP's
biotechnology  application for  hazardous  waste  remediation
can be effective at an actual waste site. SBP is ready to
apply this bioremediation approach  for on-site treatment of
PCP and  PAH-contaminated soil,  sludge and water.

B.5 PAH and PCP Contamination and
      Bioremediation

PAHs are  a  widespread contaminant of soil and groundwater
typical of creosote wood treating  facilities, manufactured  gas
plants, refineries and  related industries.  Bioremediation  has
been attempted for  PAH constituents in several studies  and
field  applications, but  until  now.  biodegradation using
indigenous bacterial strains has been able to achieve only
50% to  75% removal dFAHs. The untreated portion is
generally comprised of the recalcitrant  high molecular
weight (HMW) PAHs which are those compounds with  4. 5
or 6 fused rings and are the PAH components which are
known or suspected  carcinogens  (see Table  7).  Similarly.
                                                        32

-------
 PCP, a common wood  preservative,  has proven difficult to
 biodegrade  under  field  conditions.

TABLE  7.  Classes  and  Characteristics of PAHs

 Low Molecular Weight PAHs - relatively non-hazardous,
 relatively easy to  degrade.

 naphthalene
 2-methylnaphthalene
  1-methylnaphthalene
 biphenyl
 2.6-dimethylnaphthalene
 2.3-dimethylnaphthalene
 Medium  Molecular Weight  PAHs  - some  potentially
 hazardous  to  human  health,  more  complex but  still
 biodegradable  by many bioremediation  systems.

 acenaphthylene
 anthacene
 fluorene
 phenanthrene
 2-methylanthmcene
 anthmquinone
 High Molecular Weight PAHs  - several carcinogens, very
 slow rates of biodegradation without  specialized microbes.

 fluoranthene
 pyrene
 benzo(b)fluorene
 chtysene
 benzo(a)pyrene
 benzo(a)anthmcene
 bento(b)fluomnthene
 benzo(k)fluomnthene
 indeno(l,2.3-c,d)pyrene
 Since  1987, SBP,  in collaboration with  USEPA  scientists at
 the  Gulf  Breeze  (Florida)  Environmental  Research
 Laboratory (GBERL), has isolated and identified several
 strains of  naturally  occurring  soil  bacteria  capable of
 mediating PAH and PCP degradation at rates in excess of
 those  achieved  by   undifferentiated  communities  of
 indigenous microbes.  The  strains are identified  as:

     CRE 1-13: low and medium weight PAH degraders
     comprised  of  an assemblage of  13  Pseudomonads.
  EPA  505: a strain of Pseudomonas paucimobilis   capable
  of high rate degradation of HMW PAH constituents.

  SR3:  a strain of Pseudomonas sp.  which degrades PCP.

All organisms have been shown to mineralize their target
contaminants.  Additionally, EPA 505 has  been patented  for
use in the degradation of high molecular weight PAH.
Through  a  Cooperative  Research  and  Development
Agreement with  EPA, SBP has the exclusive license to
market the PAH degraders along with rights to certain other
environmental  biotechnology  developed  at GBERL.

The SBP bioremediation process was tested on groundwater
from the ACW Superfund Site in Pensacola, Florida. The
ACW site was an active wood treatment facility  from 1902
to 1981.   Both  creosote  and PCP were used as wood
preserving agents  during the site's operation.  Waste liquids
from  the  process were  placed  in  unlined  surface
impoundments on-site.      These  impoundments  often
overflowed into drainage ditches which discharged into local
waterways.  In addition, wastes have migrated into  the
shallow  aquifer, contaminating both soil  and  groundwater.

The ACW Superfund  site  has large volumes of shallow
groundwater contaminated by creosote and PCP. In order to
prove the capability of the SBP organisms to degrade these
contaminants  under field  conditions, a highly contaminated
groundwater was chosen as  the test matrix.

Groundwater was  pumped from the aquifer via an  existing
monitoring  well. The extracted  groundwater  was stored in
an equalization tank prior to the test From this tank, the
contaminated  feed  was  pumped  to the two-stage bioreactor
treatment system (see Figure 3). Each bioreactor had a
hydraulic capacity of 200 gallons and was designed to
provide  mixing and up-lift type aeration.

The bioreactors   were  operated  sequentially,  i.e.,  the
contaminated water was transferred to Bioreactor  1  (BR1) at
a pre-set flow rate of four days. After four days, when BR1
was full, the water was allowed to overflow into Bioreactor
2  (BR2).  Laboratory grown  concentrates  of CRE  1-13
(specialized degraders  for the low  and  medium weight
PAHs) were added to BR1,  along with nutrients and  sparged
air, and the tank was mixed. Similarly, BR2  was inoculated
with EPA 505  (HMW PAH-degradmg  strain) and SR3 (PCP
degrader) during its eight days  of  operation. Treated flow
from BR2 was held in a tank for testing; after testing, the
water was discharged to the city  sanitary  sewer.
                                                         33

-------
    Contaminated
    Groundwater
         CRE 1-13
1
r
Bioreactor
No. 1
Nutrients &
    Air
                                       Nutrients &
                                           Air
  Discharge to City
  Sanitary Sewer
     Figure 3.  Simplified  Process  Flow for  SBP
       Technologies  Inc.'s Bioreactor System
 During  operation  of the bioreactors,  samples  were collected
 for  the analysis  of key operating parameters, such as
 dissolved  oxygen,  nutrient levels, total organic  carbon  and
 suspended  solids. Microbial analysis was  performed to
 assess the cell concentration of the specialized  bacteria being
 added. All cultures were prepared in advance  and added to
 the   bioreactors  to  obtain  a final  concentration of
 approximately 1  X  106cfu/mL. Additional  samples were
 collected  to measure  the  contaminant  concentration across
 the bioreactor, as well as in rhe various portions of the
 treatment  system, in order to  calculate a  mass balance.

B.6 Bioremediation Results

 The  overall PAH and PCP degradation performance  of
 SBP's treatment system is shown below.

 TABLE 8. Summary of Bioremediation Results of PAH and PCP Removal
Contaminant Influent
(mg/L)
Low Molecular Weight PAHs 31
Med Molecular Weight PAHs 539
High Molecular Weight PAHs 368
Pentachlorophenol 256
Effluent
(mg/L)
8.1
1.6
52
31
% Removal
<99
>98
98
88
These results represent a significant advancement in PAH
bioremediation. Not only has the total PAH been reduced to
<1% of its  original concentrations, the HMW PAH. a class
of compounds which  are typically not removed in biological
remediation systems, have been reduced  by  98-plus percent.
This removal has the additional significant  characteristic of
having  been mediated  by  an organism  specially and
specifically  isolated, cultured and  introduced  into the
bioremediation system to  remove this class of compounds,
and this  has  been  accomplished under typical field
remediation  conditions.

To  further  validate  the  treatment  technology,  the
contaminated  groundwater was subjected to tests  for toxicity
and teratogenicity before and after treatment (sampled before
carbon treatment). As determined by  McrotoxTM,  embryonic
Menidia beryllina, Mysydopsis   bahia,  and Ceriodaphnia
dubia bioassays,  these indicators of potential threat  to human
health and the  environment were significantly reduced.
These  data show that the treatment system was effective in
removing the hazardous properties of the waste  material,
while simultaneously  demonstrating  that no  metabolic by-
products or toxic intermediates were created  by the  microbial
biotransformation of  the waste  constituents.

A mass balance was performed by comparing the total
influent  loading  of PAHs to the residual PAHs left in the
treatment system at shutdown. The HMW  PAHs were found
to be 80% removed (20% remained in the residual). In a
scaled-up system with higher volume  of waste water mated.
the removal efficiently at steady state would approach the
percent removal from  influent  to effluent (98%).

B. 7  Technology Application

SBP's approach  to bioremediation of hazardous  waste sites
is a "multi-phase" strategy, meaning  that  the technology is
adaptable  to the treatment  of soil,  sludges,  leachates,
groundwater and surface water.    By  the multi-phase
approach,  the waste matrix is modified and managed  :o
support the greatest  level  of specialized organisms survival
and highest biodegradation rates. Examples are given below.

Bioremediation  is applicable to a wide range of organic
contamination. SBP has focused on  research, development
and field implementation  of biotechnology-based approaches
to  some of the more difficult hazardous waste constituents.
in  particular  wood  preserving   wastes  and  solvent
contamination. SBP has bioremediation  solutions  for:

    PAHs (e.g.,  creosote, coal tar)
    Chlorinated  Aliphatic (e.g.. TCE)
                                                         34

-------
    Pentachlorophenol
    Petroleum  Hydrocarbons

Treatment of contaminated liquids such as water, leachate,
filtrate,  groundwater, storm water, surface  water  and
industrial process waste water can be accomplished by
several of SBP's  technologies. Certain waste streams can be
concentrated by  our hyperfiltration units; the permeate is
clean water and the "concentrate" contains the reduced
volume  of the  pollutants. These concentrated contaminants
can  often be bioremediated. thus minimizing the waste
stream. Other  liquids may be  treated directly, either by
biological processes,  or in the  case of volatiles,  air stripping
with biotreatment  of the pollutants in a gas phase bioreactor,
or "biofilter".

SBP has developed  several approaches for the  handling of
contaminated soil.   Solid phase systems are used when
biostimulation  of  indigenous  bacteria  is  appropriate.  Slurry
phase reactors, either in-vessel or in situ are used for  soils
or sludges requiring  better control of microbial  systems. A
multi-phase  approach  combining  soil   washing  and
hyperfiltration to  reduce volume  and  bioreactor  treatment of
the washwater is appropriate  for  heavily contaminated
wastes.
                                                         35

-------
                                               APPENDIX C
                                SITE DEMONSTRATION  RESULTS
 C.I   Introduction

 The goal of this demonstration project was to study the
 effectiveness  of the SBP formed-in-place  hypefiltration
 membrane unit,  operating in a cross-flow mode, for the
 concentration  of PAHs (and  PCP)  from  contaminated
 groundwater  found at a wood treating site. The original plans
 called for a companion  study of a proprietary  biodegradation
 process  for the concentrate: however, a decision was made
 during this study not to carry  out that portion of the  study.

 Based  on available  information, including a Remedial
 Investigation/Feasibility  Study  (RI/FS),  the  American
 Creosote Works  site in Pensacola, Florida was selected for
 the investigation. The site was included on the National
 Priorities List on  the basis of the  RI/FS  study which
 suggested that the groundwater  under the site was  heavily
 contaminated with both PAHs  and  PCP.

 The  American Creosote Works site  had been used for wood
 treatment  from  1902 to 1981.  Creosote  was the preservative
 used until  the  1970's when  a shift  was  made  to
 pentachlorophenol  in a light oil.  Over the years, wood
 impregnation was carried out in open  troughs, resulting in
 spills and drippage  which seeped into  the ground. Figure 4
 presents the  general layout of the facility and indicates the
 location of the well  that  was  drilled to  assess  the suitability
 of the site for the  SITE demonstration project. This well was
 selected as the source of groundwater for this  demonstration.

The well yielded sufficient groundwater voiume  and  flow but
the output was contaminated with free  product. Field COD
tests  were used  to  determine  that  five-fold dilution  of the
groundwater with city water was necessary to meet the 100  -
 500 mg/L COD input guideline for feedwater to the pilot
filtration unit.  Approximately 700  gallons of the groundwater
was pumped  from  the  well  on alternate  days.  The
groundwater was allowed to settle for several hours, and the
aqueous phase pumped to a separate tank.  The contaminated
aqueous phase was diluted with 2,800 gallons of carbon-
treated  city water to bring the contaminated level of the feed
in  the desired  range   of   100  to  500  mg/L  COD.
Approximately  1,000 gallons of this diluted  stream was  used
as the  feedwater to the filtration unit per day. The average
characteristics of this feed stream are shown in Table 9, and
are based on composite samples taken on each day over the
six days of the demonstration.
         Figure  4.  American  Creosote Works  Site
                                                       36

-------
TABLE 9. Average Characteristics of Feed Stream to the SBP Filtration Unit
 Analyte
Average
IDS (ppm)
TSS (ppm)
TOC (ppm)
COO (ppm)
Oil/Grease (ppm)
Semivolatiles (PPD)
Phenol
2-Melhylphenol
4-Methylphenol
2,4-Qimethylphenol
Pentachlorophenol
Benzole Acid
Naphthalene
2-Mettiylnapttialene
Acenaphthylene
Acenaphthene
Dibenzofuran
fluorene
Phenanthrene
Anthracene
fluoranttwne
Pyrene
3enzo(a)Anttiracane
Chrysena
8enzo(b)Fluoranthene
8enzo(k)Fluoranttiene
8enzo(a)Pyren9

237
34
121
379
199

4,900
2,308
6.917
1,817
(2,425)
(1,421)
12,867
4.525
(138)
6.842
4,875
5.925
17.083
1,983
7,008
4,700
1,235
1.127
(460)
(428)
(312)
( ) estimated value
 To achieve the desired volumes  and concentrations, the
 feedwater was introduced during  the first two hours
 (approximately) of each day's run. For the remainder of each
 day's run,  the  reject  (concentrate) was  recycled,  becoming
 the feed, while the permeate was continuously removed  and
 discharged  to the POTW after samples   were composited for
 analysis.   Each day's  run was  terminated  when  the
 flowmeters indicated that  the  desired  800 gallons  of
 permeate had been discharged, leaving approximately 200
 gallons  of concentrate.

Samples were analyzed for oil  and grease, dissolved and
 suspended  solids, volatile  organics,  TOC.  COD,  and
 dioxins/furans   in  addition to  the   list  of  designated
 semivolatile organics used in evaluating the  developer's
 claims.

 C2  Field Activities

 SBP personnel were responsible for preparing the membrane,
 operating the  system over the test period (six days), and
 washing of the unit on alternate days. Separation  of the oil
and  other readily separable material and dilution  of the
groundwater with city water were field decisions that were
agreed  to  by  SBP  and EPA to meet the  operational
requirements of  the filtration unit.  EPA's  contractor
personnel monitored flow rates and sampled  feed, permeate,
concentrate,  and  washwater while the system was operating.

While  some  leaks were encountered  during  the shakedown
of the  filtration unit,  no breakdowns occurred during the  six
days of testing. The flux (transport of liquid across the
membrane) did have  a tendency to decrease during  each  day
of operation, but this  was  accommodated by manually
increasing the pressure.

C.3 Test Procedures

As  noted  earlier, the system was  operated  by  introducing
approximately 1,900  gallons  of the diluted feedwater (about
1 volume of groundwater to 4 volumes of city  water) over
a two hour period at a rate of about 8 gpm. At that time,
recycle  of the concentrate was initiated and continued until
approximately 800 gallons  of permeate was generated and
approximately 200 gallons of concentrate was  collected.

Based  on  operational parameters  (flux,  pressure),  the
developer determined that washing of the unit was necessary
every other  day.  At the end of the  second day, the  fourth
day, and the sixth day approximately 200 gallons of city
water,  to which cleaning  chemicals had  been  added, was
flushed  through the system and collected.  This material was
also  analyzed.

Grab samples of the feed stream were collected  every 1.5
minutes during the initial two-hour single pass phase, while
permeate grab samples were collected every 45 minutes over
the  whole  run.  Laboratory samples were prepared by
compositing  the  grab  samples on a flow proportional  basis.
After the initial single pass, the concentrate was  recycled for
approximately four hours in each test and one composite
sample  was  taken at the end of each run.  All samples were
transferred to bottles, inhibited or preserved as called  for by
the individual test methods, labelled,  sealed, and shipped in
ice-tilled  coolers to  off-site  laboratories by overnight
express.

Instantaneous flow and accumulated  volumes  of feed and
permeate were  measured automatically using calibrated
 flowmeters. Volume  of the final  concentrate  was determined
by the  difference between the feed  and permeate  volumes.
The  temperature and  pH in the three streams were measured
in the  field  and  recorded to assure that there was no  gross
change  in   characteristics  that  could  affect  filtration

-------
effectiveness. Sampling points are indicated on the schematic
of the system,  shown in Figure 2. The  sampling schedule,
analytical protocols and results, and QA/QC protocols are
described fully in the Technology Evaluation Report. The
ongoing QA  program  allowed for the collection and
reporting of high  quality data.

At the end of the sixth day of operation, after all the
required samples had been taken, the process was allowed  to
continue  for several hours, further  concentrating  the
contaminants. Additional samples of the permeate were taken
about every  25  gallons to observe whether there was a  fall-
off in contaminant  rejection rate.

 C4  Results

The following  results  are all based  on the composite
analytical results and  field data obtained on  each day.  The
Technology Evaluation Report presents the complete  data for
grab  and composite samples taken each day of feed,
permeate,  reject,  and washwater.

Operation of Unit

The filtration unit operated in a batch mode for six hours
each day,  for six days,  and  processed approximately 1,000
gallons  of feed per day. Over the  six day test  period,
permeate   flux  was   a   relatively   constant   0.0085
gak'ons/min/ft2  (coefficient of variation  < 10%). Based  on
a total membrane  area of 300  ft2  for the system, the
permeate  flow  rate  for the  four-module  filtration  unit
averaged 2.6 gpm.

The unit was cleaned every two days with approximately
200 gallons  of city water and membrane  cleaning materials.
Over the two day period, the surface  of the membrane
gradually fouled, requiring an increase  in feed pressure in
order  to maintain a constant  permeate flux. The cleaning  at
the  end of each  two-day period was sufficient to restore the
original flux. Figure  5 depicts  the pressure adjustment  over
each two day period necessary to maintain the constant flux
(Figure  6).
     MO*
                   #ws        *«       MWI
                              i (min)
            Figure 5, feed Pressure vs. Run Time
  |3
  K  3-.«* f~
  «  «G7 f*
  S ,ao«>.
  M     '
  2  \1QOS f»
  A.  «J*4 i-
a   **
 a  31
                          TilH«<

                Figure 6. Flux vs.  Run Time

Semivolatile  Contaminant  Analysis

Contaminant  Reduction

It was  agreed between SBP and EP.4 that a  selected list of
semivolatile organic compounds would be used to assess the
effectiveness of  ihe  Filtration  Unit. The  analytes  consisted
of all quantifiable semivolatile compounds (Method 8270)
detected in the feed.

Rejection efficiency was defined  as the change in total
concentration for these  contaminants between the feed and
the permeate:
                                                        38

-------
 % Rejection = 10Q-Cr X
                  Q

 where
     C, =  total  concentration of contaminants in  feed
     Cp = total concentration of  contaminants in permeate

 On the basis of composite samples, rejection for the total
 designated contaminants averaged 74% over the six days,  as
 shown in  Table 10.

 Analysis of the data reveals that the system is very effective
 for the rejection of polynuclear aromatic hydrocarbons  and
 much less effective for  phenols. Separated  in this fashion,
 rejection for the  PAHs averaged 92% over the six days
 while rejection for the phenols  averaged  18%.
TABLE 10. Overall Semivolatile Rejection for the SBP Filtration Unit
                                Ju-
                                I 1.4
                                  M
                                          11*
                                         190
               13«   13*   170
               Motocafcr Weigfct
Figure 7. Molecular Weight vs. Order of
          Magnitude  Reduction  (ORD)
                       Mass  Balance
 Day    Feed   Permeate  Concentrate
       (ppm)   (ppm)        (ppm)
Rejection
1
2
3
4
5
6
104
91
92
104
85
60
18
24
26
22
23
24
206
585
248
242
199
538
83
74
72
79
73
60
 A more accurate means  of quantifying and graphically
 representing rejection is to calculate the reductions on a
 logarithmic basis.    For exatnple.  attempting to graph
 differences between  90% and 99% rejections  does  not fully
 depict the fact that a ten-fold reduction has occurred.  In
 order to accurately represent this process, it is necessary to
 calculate the order of magnitude of rejection  (ORD) as the
 log value of the ratio of the contaminants concentration in
 the feed to the concentration in the permeate. The equation
 is  written as:

      ORD  = Log  OCp)

 Therefore, an ORD of 2.0 means that a  100 fold decrease in
 contamination  has occurred  in the permeate  relative to the
 feed.    An  ORD of  1.0  means a  10-fold  decrease has
 occurred.

 For the semivolatile constituents, it appears that rejection
 follows the molecular weight of the chemical. Figure 7
 presents the average  ORD over the six days versus the
 molecular  weight of the  individual  compounds.
Combining the masses of pollutants in the  permeate  and the
concentrate on a daily basis  did not provide a good material
balance, relative  to the feed. Only when the material in the
washwater was also included was all the material accounted
for. Based on  the volume of washwater  used  in  each
washing operation, approximately  200  gallons,  about  8% of
the masses of the designated pollutants are retained  on the
membrane  or in  the liquid in the  system. Table  11 provides
a  summary of the contribution of the  different streams for
each two-day period between wash operations. The relative
distribution of pollutants  in  the washwater  was very  similar
to that in  the concentrate, with PAHs far exceeding the
phenols. However, the concentration of pollutants in the
washwater was similar to the feedwater.

TABLE 11. Mass Contribution of Total Semivolatiles in all Streams (2 day
        totals)

Days    Feed   Perm   Cone   Wash  Total  Recovery
        (Influent)                       Effluent
        gms   gms     gms   gms    gms    %
1 &2
3&4
5&6
777
778
581
140
159
151
566
320
i 11
46
84
32
742
563
710
95
72
122



6 day recovery = 94%
                       Extended   Operation

                       After the system yielded the desired  200 gallons  of
                       concentrate on the sixth (last day)  of the test and   all
                       necessary  samples had been taken, operation of the system
                       was continued with further recycle of the concentrate as long
                                                           39

-------
 as possible to observe the behavior of the unit (i.e., the rate
 of fouling and the quality of the permeate). The extended
 operation resulted in only  thirty seven gallons of concentrate
 remaining, representing a 97% volume reduction. However,
 during this time  the  rejection  efficiency  decreased,  as
 expected, as concentration continued (Table  12).

TABLE 12 Behavior  on  Extended Fiibation
Additional Gallons of
Permeate Produced
25
50
75
100
150
Semivolatile Rejection %
(calculated from initial Feed)
47.5
48.5
60.0
32.3
21.7
Conventional  Parameters

 In addition to the designated semivolatile pollutants, the
 removal of several  conventional parameters from the
 feedwater was also evaluated.  These included total dissolved
 solids  (TDS), total suspended solids (TSS), chemical oxygen
 demand (COD), total organic carbon (TOC), and oil and
 grease (O&G).  Comparison of feed and  permeate   confirm
 that the  membrane is capable  of removing  larger molecules
 such as oil  and suspended solids, but is not nearly as
 efficient in the removal of dissolved material (TDS) or
 otherwise-unidentified organic species (COD and TOC).
 Such  results, summarized in Table  3, suggest that the
 membrane is operating predominantly as a hyperfiltration
 membrane.   In addition, considerable concentrations of
organics  remain in the permeate even though the  PAHs are
efficiently    removed.    For  example,   the  ratio  of
 semivolatiles/COD  decreases significantly, from about 0.2 in
 the feedwater to  about 0.07 in the permeate, demonstrating
 that semivolatiles are not the major contributor to the COD,
 and that  significant  unidentified,  smaller  molecular weight
compounds can  pass through the membrane.
Polvchlorinated  Dibenzo-p-Dioxins/Dibenzofurans

Polychlorinated dioxins (PCDDs)  and furans (PCDFs)  are
often  encountered in the manufacture of pentachlorophenol
and can remain with the commercial  product.  To investigate
this matter, selected  samples of the three streams taken  on
the first, second,  and sixth day of the demonstration were
scanned for  the  various dioxins and  furans  using high
resolution GC coupled with low resolution MS. A number of
the more highly chlorinated dioxin and furan species were
found in the feed water and in the reject at significantly
higher concentrations, but analyses of the permeate  indicated
efficient removal of these pollutants.  The  2,3,7,8-TCDD
isomer was not  found above the detection limit in any
sample of feed, permeate, or concentrate. The  2,3,7,8-TCDF
isomer was found in  only one  sample of  feed  and one
sample of concentrate, in both cases the value was <0.5 ng/L
(ppt).  The major dioxins and furans were the  octa  species.
These results are consistent with rejection by molecular
weight or  oil  solubility,  but not by number of rings.
                                                         4fl

-------
TABLE 13. Qioxins/furans in Process Streams

Dayl
              Feed  Perm. Cone.     Removal
(ng/L)  (ng/L)   (ng/L)    %
Day
2378 TCCD
12378PeCDD
123478 HxCDD
123678 HxCDD
123789 HxCDD
1 234678 HpCDD
OCDD
2378 TCDF
12378 PeCDF
23478 PeCDF
123478 HxCDF
123678 HxCDF
234678 HxCDF
123789 HxCDF
1234678 HpCDF
1234789 HpCDF
OCDF
4 »
1 •
4 »
53.0
74.6
1980 1.6
14670 38.9
• «
f *
4 *
11.2
3.2
8.5
.
346
28.3
1030 3.3

,
.
277.7
13.3
3836
42365
.21
.
.
52
8.7
30.4
.
1066
76
2596





99.9
99.7









99.7
.  An '.'indicatesttieisomer was absent ortwlowdetection IJmitsinstream.
Day3
              Feed   Pwm.   Cone.     Removal
              (ng/L)    (ng/L)    (ng/L)    %
2378 TCDD
12378 HxCDD . .
123478 HxCDD 2.4 '
123678 HxCDD 47.7 .
123789 HxCDD 6.7 •
1234678 HpCOO 1460 •
OCDD 13250 7 . 9
2378 TCDF .37
12378 PeCDF . .
23478 PeCDF
123478 HxCDF 8.1 •
123678 HxCDF .
234678 HxCDF
123789 HxCDF .
1234678 HpCDF 300
1234789 HpCDF .
OCDF 224



116
9.9
3630
34135 99.9



28.4
4.2
17.0

689
40.1
1223
       An '•' indicates the isomer was absent or below detection limits in stream.
Feed
(ng/L)
2378 TCDO
12378PCDD
123478 HxCDD 1.9
123678 HxCDD 30.6
123789 HxCDD 4.4
1234678 HpCOO 1370
OCDD 9200
2378 TCDF .39
12378 PeCDF
23478 PeCDF .
123478 HxCDF 5.1
123678 HxCDF 1.5
234678 HxCDF 5.1
123789 HxCDF
1234678 HpCDF 1 9 6
1 234789 HpCOF 14.1
OCDF 473
Perm.
(ng/L)
,
4
.
*
.
.36
8.8
.
.
4
4
4
4
t>
.
t
.50
Cone.
(ng/L)
„
*
8.1
334
27.6
8950
90400
1.5
.
1.7
111.7
34.2
42.6
,
2055
126.2
5235
Removal





99.97
99.90









99.89
                                                                 .  An V indicates the isomer was absent or bebw detection limits in stream.
Volatile Organics

Analyses for volatile organics  (VOCs) were  carried  out
during the  runs on  days  1, 3,  and 6  because the  developer
was somewhat concerned that the high pressures used in  the
system  might force  volatiles through  the membrane into  the
permeate.  All  VOC  concentrations  were  low  in  the
feedwater;  the  same species were also found  in  the permeate
and the concentrate but at lower  concentrations.  The
principal VOC was  acetone, but the expected BTEX species
were also  present.
                                                                 TABLE 14. Volatiles in procen Streams
                                                                                Feed
                                                                                        \tgfl.
                                    ng/L
Mettwtene Cfitonde
Acetone
Caiton Oisuifide
2-Butanone
Benzene
Toluene
Eihvt Benzene
Styrene
Xylenes
*

2
32
16
20
12
18
43
t

<5
31
7
16
(7)
7
26
*

<5
32
<5
<5
<5
<5

-------