United States
                   Environmental Protection
                   Agency	
Risk Reduction Engineering
Laboratory
Cincinnati OH 45268
                   Research and Development
EPA/600/S2-88/064  Feb. 1990
&EPA         Project Summary
                   Treatment of Hazardous  Landfill
                   Leachates and  Contaminated
                   Groundwater
                   Robert C. Ahlert and David S. Kosson
                    The objective of this study was to
                  assess pilot-  and  field-scale
                  application of separate or combined
                  biological and physical/chemical
                  treatment to high-strength hazardous
                  dump site  or landfill leachates,
                  extracted sludges and wastes, and
                  land-spilled  hazardous substances.
                  Four types  of  real-world toxic,
                  concentrated, complex wastes served
                  as  prototype  waste  streams:
                  industrial landfill  leachate, leachate
                  from an industrial sludge Impound-
                  ment, forced extract from impounded
                  industrial sludge,  and groundwater
                  contaminated with  a chlorinated
                  solvent Techniques evaluated during
                  bench-  and  laboratory-scale experi-
                  ments were: dispersed and fixed-film
                  aerobic  and anaerobic mixed
                  mlcroblal systems, flocculation/pre-
                  cipitation, ultrafiltratlon  (UF), and
                  reverse  osmosis  (RO). In general,
                  some physical/chemical pretreatment
                  (liming, floe/settle) was required  to
                  obtain high efficiency biodegrada-
                  tion. The biodegradation effluent
                  could be "polished- by UF, RO, or ion
                  exchange when  required to  meet
                  National Pollutant Discharge Elimina-
                  tion System (NPDES) standards. Soil-
                  based  field  pilot  plants  were
                  constructed  and  operated to dem-
                  onstrate sequential  aerobic/an-
                  aerobic mlcrobial  treatment  for
                  leachates from  two CERCLA-NPL
                  sites and for groundwater contam-
                  inated  with  1,1,1-trichloroethane
                  (TCA) at a third  site. Even in  the
presence  of high concentrations of
inorganic  salts, organic carbon
reductions  of  95% to 99% were
achieved.  The  levels of chlorinated
solvents  extracted  In  subsurface
waters were reduced from  5 mg/L to
less than 20  pg/L (greater than
99.6%).  The contaminant reductions
attained are not the maximum that
could be achieved  but reflect
pragmatic,  cost effective  treatment
levels.
  This Project Summary was devel-
oped  by  EPA's  Risk  Reduction
Engineering Laboratory, Cincinnati,
OH, to announce key findings of the
research  project  that  Is fully
documented In a separate report of
the same  title (see Project Report
ordering information at back).

Introduction
  The major objective of this  study  was
to demonstrate pilot- and field-scale
treatability  of high-strength  dump or
landfill leachates, extracted sludges or
wastes,  and land-spilled hazardous
substances. Two specific treatment types
were examined: biological treatment and
physical/chemical treatment.
  Biological  investigations involved
dispersed  and  fixed-film mixed  and
indigenous  cultures for  treating
representative concentrated  hazardous
aqueous wastes  that could contain both
organic and inorganic  compounds.  The
need for sequential process  steps  was
recognized. At the outset, only bench-
scale methodologies were undertaken.

-------
Various sources of microbial seed were
used,  and  evidence  for  long-term
adaptations  and  responses  were
considered  of primary  importance.
Experiments were designed to fill a gap
in  existing literature  that  focuses  on
degradation of single hazardous species
and  binary  or very  simple  multiple
solutes in solution with water. Studies of
complex—mixed organic  and inorganic-
solute systems have not been reported.
  Physical/chemical treatment of influent
to  and effluent  from bioreactors  was
identified  as important, but was given a
lower priority since it  is  less  novel.
Potential  physical/chemical  treatments
included:  RO, UF,  flocculation/sedi-
mentation, membrane  separations, and
adsorption.  As with biological treatment,
feed streams were  mixed solutions  of
multiple organic  species  and  heavy
metals in water. Emphasis was placed on
high-strength  and complex solutions and
on the need to inactivate  or detoxify toxic
metal species, as  well as biogenic,
biodegradable,  and refractory  organic
compounds.  Investigations of biological
treatment techniques were marked  by
two concerns: (1) product water  quality
suitable for discharge to the environment,
and (2) minimum volume and maximum
strength of  concentrated residues  for
ease of handling and/or recovery.
  During  the  program,   microbial
treatment for four types of high-strength,
complex aqueous  wastewaters  was
investigated. These  included: industrial
landfill  leachate,  leachate from  an
industrial  sludge impoundment,  forced
extract from impounded industrial sludge,
and  groundwater  contaminated with
chlorinated solvent. Pretreatment  by
physical/chemical  methods  was  soon
shown to be essential to  remove colloids
and  fouling agents.  Ultimately, total
organic carbon (TOC) reductions of 95%
to  99% were achieved, even  in  the
presence  of  high  concentrations  of
inorganic  salts.  Acclimated,  mixed
microbial populations were employed in
all cases;  no genetically engineered  or
specifically  cultured  organisms  were
used. Aerobic and  anaerobic  regimes
were developed, in  sequence, in packed-
bed (soil-based) bioreactors.
  Biological treatment  was  evaluated in
bench-scale  experiments  with laboratory
column reactors. Much common ground
was  discovered with respect to organic
carbon loading rates,  nutrient require-
ments, buffering, and  co-substrates for
co-metabolism. Design criteria for scaled-
up investigations were derived.
  The laboratory  column  work pro-
gressed  so well  that resources were
available to  treat  three  additional
hazardous waste liquors treated in pilot
plants  designed specifically for the
purpose.  Out-of-door,  pilot-scale  treat-
ment of  landfill leachate was  first
demonstrated in a  pilot  unit at the
University. Subsequently,  sludge extract
solution  and chlorinated solvent solution
were treated in field-site pilot plants built
and  operated at the  actual  sites  of
contamination.  In these  cases,   field
performance  matched or  exceeded
bioreactor performance in  the laboratory.
In conjunction with Enviresponse, Inc., a
conceptual design for  a  transportable,
mixed  microbial, hazardous organic
solute treatment system was developed.
In  short,  the ability  to treat complex,
organic  waste  solutions  has  been
demonstrated in the field and is available
for  application  to numerous  CERCLA
sites.
  Physical/chemical  treatment is gener-
ally  necessary  for  pre- and  post-
biological  reactor  processing.   Unit
processes of this type are required for
separation of dispersed oil and particulate
matter and for quantitative reductions in
heavy  metal  and  inorganic   salt
concentrations.  Through  membrane
separations and selective adsorption, it is
possible to reduce solute  concentrations
to levels compatible with direct discharge
of treated effluent. Several process tech-
nologies  in  this  category   were
demonstrated to function very effectively
in conjunction with biological treatment.

Methods,  Discussion, and
Results

Leachate  Characteristics
  For the initial  phases  of this  work,
leachate  was obtained from  a  large
commercial  landfill receiving substantial
quantities of industrial wastes.  Leachate
emanating from this landfill is collected in
open basins or lagoons and allowed  to
settle by gravity. A floating oily layer is
removed  and stored  in  drums. The
underflow from the  collection  basin  is
pumped to API separators for further  oil
removal.  Bulk oil separated by gravity
has  been  found  to  contain   high
concentrations  of  PCBs. The   bulk
aqueous phase has high residual turbidity
consisting of colloids, stable emulsions,
and fine  particles coated  with oil,  which
must be  removed by pretreatment. After
pretreatment, the clarified aqueous  phase
has the following range of properties:

  dissolved organic
   carbon (DOC)    8,000 - 12,000 mg/L
  chemical oxygen
  demand (COD)  23,000 - 30,000 mg
  total dissolved
  solids (TDS
  @ 103c C)       15,000 -17,000 mg

Other  typical  values  include  TK
ammonia-N,  and sulfate  at 1,450,  1,0t
and 3,400 mg/L, respectively.  A lari
number of heavy metals, i.e.,  nickel, lee
chromium, copper, etc., are  present
concentrations above 0.1  mg/L.
  Later,  effluent  was  obtained  fro
natural  leaching and by forced extracti
of an  industrial sludge disposal lagoc
Primary and secondary sludges  frc
diverse chemical manufacturing opei
tions had been impounded in the lago
for several decades. The natural leache
has the following properties:
  TOC
  TDS @ 103°C
  TKN
 170-5,000 mg
2,700 4,300 mg
    25 - 820 mg
Extraction of the sludges with sodiu
hydroxide  solution,  at  pH  10  to 1
increases these values by factors of 3
5. Trace  metal concentrations  a
increased also: aluminum and zinc at 1
10 mg/L, and, numerous others at 0.1
1 mg/L.
  During the period  of  investigation,
problem  of  groundwater contaminati
with  chlorinated  solvent was  suggest
for study. Concentrations of I to 15 mg
of TCA had   been  encountered
otherwise  high-quality  groundwate
Lesser  concentrations  of  byprodu
species  were also observed.   Based '
environmental impacts and groundwal
quality criteria,  this situation  could I
defined  as involving a high-streng
wastewater. A  microbial  remediatii
process comparable to  the renovatii
processes  for  the  very high-streng
leachates  described  above   w,
developed,  tested in  the  laboratory, ai
operated at pilot scale in the field.


Pretreatment
  Physical/chemical pretreatment of hig
strength  leachates   with   coag
lating/flocculating  agents  produces
minimal decrease in DOC, while reduci
turbidity  and the level of heavy met!
present  in the wastewater. Lime gave t
best results with the leachate samp
involved in  this study.
  Utilizing lime  to a  solution equilibria
pH  of  12  produced  greater  than 98
reduction  in turbidity.  Reductions
dissolved TOC  and COD  averaged le
than 5%. Polyelectrolytes  were found

-------
ncrease the rate but not the final level of
urbidity  reduction.  A  pretreatment
Drotocol  was  developed;  it  included
iddition of lime in sufficient  quantity,
ipproximately 6 g/L, to achieve a  pH of
12 with subsequent floe settlement. The
 upernatant was decanted and sparged
with C02 until the pH was reduced to 9 or
ess. At this point, the supernatant was
Jecanted  and a second recarbonation
tep was  used  to  decrease  final  pH to
7.5.

  icrobial Treatment

Biochemical  Process
  Aerobic  biological studies revealed that
i dispersed mixed microbial population,
icclimated to landfill  leachate,  degraded
30% to 90% of  the  organic species
Dresent in  the hazardous industrial waste
iquor,  with or  without the  addition  of
glucose or other nutrients. The decrease
n DOC  was not due  to  stripping,
evaporation, and/or sorption; it was due to
Diological  oxidation. Biostabilization was
rapid. Mixed microbial cultures exhibited
a two-rate  (diauxic) growth pattern (Figure
1). As  signaled  by the increase  in pH
during  the first exponential growth phase,
it is likely  that the mixed culture  utilizes
the fatty  acid  fraction  of  the organic
;olutes in  this first phase. Further, it is
probable  that at  least two groups  of
organisms participate in the  biooxidation
process.   Fatty   acid metabolizing
organisms have a higher specific growth
rate than  the  other  organisms that
contribute  to the  second,  lower rate
portion of the  growth  pattern.  Good
agreement in carbon balances provides
clear evidence for  biodegradation  of the
organic species  present in the leachate.
  Low  sludge yield was observed  in  this
study;  this  implies a  reduced  sludge
problem  associated  with  aerobic
treatment. Oxygen  uptake rates,  as
reflected in the volumetric flow of air-per-
reactor volume, were quite nominal. If it is
assumed  that  microbial maintenance
requirements  are negligible, as has been
reported  frequently  for wastewater
systems, it is possible to quantify the role
of co-metabolism in   the  biological
oxidation of anthropogenic  compounds.
The  possibility  of  oxidative  assimilation
(nonproliferation) is ruled out because of
the quantitative  evolution  of carbon
dioxide; increase in cell mass and protein
content were not measured. The ability of
the acclimated population to  utilize
organic carbon and other nutrients solely
'rom  the  leachate  further improves
.jrocess prospects. It was   possible to
treat highly  concentrated  waste liquor,
i.e., up to 10,000 mg/L of organic carbon.
The absence of  highly fluctuating  DOCs
indicates a  stable and well-acclimated
microbial population.
  Anaerobic biological studies  demon-
strated a DOC reduction of 64% for a
culture  grown   on   leachate  and  a
reduction of 69% for  a culture  selected
for the  degradation  of  acetate,
propionate, and  butyrate. Specific  DOC
utilization rates of 0.15 and 0.21  day1
were observed for the leachate and these
volatile fatty  acid digesting cultures,
respectively. Cell  growth was  not
observed, to any  significant extent, during
these  batch  experiments.  Leachate
effects  on the  cultures  were  studied
through examination of individual volatile
fatty acids  in  the course  of the exper-
iments (Figure 2). Large concentrations of
acetate were accumulated  before overall
removal  was  observed.  The  butyrate
profile  demonstrated responses similar to
that of  acetate removal.
  Propionate and isobutyrate were more
difficult to  remove,  as  both  left  an
appreciable amount  of unmetabohzed
acid. However, their concentrations were
small  relative  to  acetate  and  butyrate.
Acetate and butyrate are the major fatty
acids  contributing to  methanogenesis.
Also, the fraction of the nonvolatile fatty
acid contaminants in  the  leachate that
were converted to volatile fatty acids  by
the acid  formers  ended up primarily  as
acetate, with a small fraction converted to
butyrate,  propionate,  and isobutyrate.
Reactor  failures  were encountered  in
studies with  20 volume-% leachate.
These  failures were probably the result of
overloading the system with volatile fatty
acids.  At leachate concentrations of 5%
and  10%, no toxicity problems due to
nonvolatile  fatty  acids were observed.
Methane was produced at  levels of 0.95
to 0.99 L/g (m3/kg) DOC removed.
  The  data  from this study  clearly
indicated  that aerobic  and anaerobic
biological treatment  can  be  used  in
conjunction  to  stabilize  organic
compounds  found  in high-strength
hazardous waste  residues.
Microbiology of Chlorinated
Hydrocarbons
  This phase examined mixed anaerobic
population degradation of the haloorganic
compounds  dichloromethane (DCM), l,l-
dichloroethane (DCA), and TCA. Half-kill
doses,  determined  from   batch
experiments, defined a relative degree of
toxicity for each  of the compounds. The
microorganisms exhibited great tolerance
for DCA; there was no apparent inhibition
at concentrations up to 35 mg/L.
  Acclimation  studies  with  TCA
demonstrated that  continued periods of
zero gas production  do not necessarily
reflect the  death of  the organisms.
Reactors dosed  with  2  and 4  mg/L
recovered after  20  days of zero  gas
production. After this lag period, daily gas
production was greater  than or equal to
control reactors.  The overall acclimation
period  was 33 days, less than half the
acclimation  period of  10  wk cited in
reports of  previous work.  However,
acclimation  periods tend  to vary greatly
for anaerobic seed cultures.
  Studies with separate anaerobic  popu-
lations  indicated that the methanogens
were most  likely responsible for degra-
dation  of  the chlorinated compounds
considered.
Microbial Reactor Design
  Soil-based  sequential aerobic/an-
aerobic  microbial  degradation  was
investigated  as a potential onsite  or  in-
situ treatment process.  Laboratory soil
column experiments were carried out  for
initial evaluation of the proposed process
using three different contaminant streams
and several soil types.
  Treatment  of  landfill  leachates
achieved reductions in DOC in excess of
90% (Figure 3). Influent  leachate-derived
organic carbon  (LOC)  concentration,
organic carbon removal,  effluent pH, and
long-term  system  permeability are
interrelated. Microbial acclimation periods
increased  in  length with  increased
influent  leachate concentration.  In-
creasing influent LOC  concentration
resulted in greater overall  removal
efficiency for DOC. In addition, increased
influent LOC concentration is associated
with reduced  incidence  of plugging,
allowing  the  system to operate more
closely to target volumetric flux.
  Treatment of TCA using the soil-based
microbial  system,   operated  under
anaerobic  conditions,  resulted  in
quantitative removal  of  solvent. Influent
concentrations of 20  mg/L corresponded
to effluent concentrations of less than  20
p,g/L.   Failure  to  observe  any
breakthrough  phenomena,  over  long
periods of  operation, confirmed that
sorption onto  soil constituents was  not
controlling TCA  removal. Gas  evolution
rates and gas  analyses indicated that
methanogenesis  was the most probable
rate-controlling step.

-------
     .o
     I
        2400
        2000
        1600
        7200
         800
         400
                                                    Cell Mass
                                                     DOC
                                                                   2.0
                                                                   1.0
                                                                   0.6
                                                                         IB


                                                                   0.4    *
                                            12
                                                       16
                                                                 20
                                    Time.h
 Figun 1.    Fate of organic carbon and microbial responses observed during a study with
             20% leachate (with pH control @ 7.5 ±0.1).
Microbial Process
Demonstration

  Positive results  from  laboratory soil
column experiments were followed by the
design,  construction,  and  operation  of
three field pilot plants  to demonstrate the
process  feasibility   for  large-scale
applications. One of the pilot plants was
located at Rutgers and the other two  at
the actual sites of differing contamination
problems.
  The  first  pilot  plant  in the  series
consisted of six self-contained lysimeters
(bioreactor columns packed with soil), 2 ft
in diameter and 6 ft in depth. These units
were implanted  in the ground, at a
location  on  campus,  and operated  in
simulation of site  conditions. Pretreated
leachate  from  a  CERCLA-NPL  industrial
landfill was treated. TOO mineralization to
CO2 and CH4 was obtained with a single-
pass efficiency of 90% to 97%. Operation
was carried  out for two  consecutive
spring-through-fall operating  periods  of
approximately 160 days  each. Laboratory
results were readily  transferable to field
operation.
  The second pilot plant was designed to
demonstrate  in  situ sludge extraction
coupled  with   soil-based  microbial
treatment of  recovered extract;  it was
constructed and operated at the sludge
disposal  site for 140 days. The cleanup
process  consisted  of  two  steps  in
sequence (Figure 4). The first  step was
removal  of contaminants  from  mixed
primary and secondary industrial sludges
through in situ extraction with aqueous
sodium  hydroxide.  Extractant  solution
was  injected into  the sludges  throu
well-points  or  applied to the  surfa
through a  perforated pipe distribut
network. Extract was recovered by mee
of two  wells screened near the bottom
the sludge deposits.  The  results
system operation indicated that as mi
as a  15-fold increase in  removal  ra
relative to  natural processes  driven
infiltrating  rainfall, could  be obtain
through controlled alkaline extraction.
additional  4-fold increase  in  the rate
site renovation  can be obtained throi
increased  hydraulic  flux  through  I
sludges.
  In  this  second  pilot plant,  extr;
recovered from the sludges was trea
to remove TOG  with  a  soil-bas<
sequential  aerobic/anaerobic microt
reactor.  Treatment  occurred  onsi

-------
                                                                                                   - 250
                     5% Leachate
                     Leachate Culture
                                                                                           450
                                                          500
  Figure 2.   Acetate and propionate concentrations as functions of batch reaction time.
immediately adjacent  to  the  extraction
unit.  A  diverse,  mixed  microbial
population  was  developed in the  soil
system. Neutralized extract was applied
to the surface of the treatment bed  and
allowed to percolate  through the  soil
column. Aerobic and anaerobic microbial
populations metabolized  organic  con-
taminants  to  C02  and CH4.  Treatment
efficiencies in excess of 95%  were
observed.
  A  third  pilot  plant,  designed  to
demonstrate  the   treatment   of
groundwater contaminated with TCA,  was
operated at the site of a solvent spill. This
pilot plant consisted of two self-contained
lysimeters,  3  ft in  diameter and  6 ft in
depth.  The soil  packing  employed  was
excavated  from the site.  Influent to the
two lysimeters was contaminated ground-
water, recovered from a monitor well, with
TCA concentrations of 5 to  20  mg/L.
Reactors were operated  under entirely
anaerobic  conditions.  At  steady  state,
effluent  TCA  concentrations  were less
than 20 u.g/L.

Sorption  and Extraction
  Sorption  and extraction are  processes
that influence  the distribution  of solutes
between  separate phases. The absolute
performance  of  these  processes  is
dictated  by thermodynamic  equilibria.
Thus, the ultimate distribution of a solute
between otherwise homogeneous  phases
(steady  state  in  batch  contact
circumstances)  is  that  at  which  the
chemical potential  of  the  solute  is  the
same in the several phases.  As used in
this study,  sorption is  the exchange of
dissolved  organic  solutes between
aqueous  solutions and a variety of solid
substances. The direction of exchange is
dictated by  thermodynamic  considera-
tions, and  the rate is influenced  by
physical  transport  and the strength of
binding forces. Extraction is the removal
of constituents from a solid  phase by
contact with solvent. Distribution and  rate
are  controlled   by  thermodynamic
equilibria and,  also, by physical barriers
to exchange.

Granular Activated Carbon
  The time  required  to  approach
thermodynamic  equilibrium must be
determined experimentally for each
carbon (GAC) and organic solute  system
combination.  In  general, greater than
95% of equilibrium sorptive  capacity is
attained in several days. However, it may
take  much  longer periods to reach final
equilibrium  because of slow diffusion in
carbon  particle pores   and/or slow
changes in surface binding states.
  Pretreatment with lime  to remove  oily
phases from otherwise aqueous leachate
was  necessary before the determination
of carbon sorption isotherms. Interactions
of GAC with leachate were independent
of the primary  pretreatment  process.
Control of pH through recarbonation with
C02, air  stripping,  or  sulfuric acid
addition had little or no detectable effect
on subsequent sorption of organic solutes
onto  GAC. The treatability  of two leachate
samples  (EPA #01  and  EPA  #02) with
GAC was identical.
  The  relationship between  equilibrium
organic  sorbate  loading  and equilibrium
solute concentrations, in  multiple  solute
systems,  is  dependent  on  initial
conditions.  Type  A isotherms were
obtained  for full-strength  leachate  and
varying  masses  of GAC.  Type  Bn
isotherms  were constructed with  a
constant mass of GAC; the subscript "n"
represents  the ratio of carbon-m ass-to-
volume of full-strength leachate. Finally,

-------
    4500
    4000
    3500-
    3000-
    2500.
 O
 0  2000
    7500-
    7000-
    500.
                                                                                  Influent
                                                                                   fffluent
                     I
                    20
                                                                               720
                                                  740
                                                      SO         700

                                                        Time, days

Figure 3.    Typical influent and effluent responses for sequential aerobic/anaerobic soil-based microbial treatment.
                                                                                                      160
                                                                                                                 180
Type  C isotherms used  data for  a fixed
mass  of GAC and  varying  leachate
dilutions. With this conceptual separation
of driving  forces for sorption,  it was
possible to  construct  a mechanistic
description for the process within classes
(types).  Plots of equilibrium  loading
versus equilibrium concentrations of TOC
for these three types of isotherms can be
used  for leachate  sample  character-
ization.
  There is a weak pH effect on sorptive
capacities of GAC for TOC in leachate. It
appears desirable to carry out measure-
ments and  treatment at  near-neutral  pH
rather than at higher pH levels.

Soil
  Soil is an  active  sorbent  because of
organic matter (humic) and clay fractions.
These fractions interact with  organic
solutes  in groundwater  and  infiltrating
surface  water to  exchange organic
substances  and dissolved  inorganic
species.  Thus, the presence  of organic
contaminants in groundwater  and  soil
water infers  the  distribution  of these
substances  in stable  or transient
equilibria. As an example of the role of
the organic carbon fraction, the mass of
2,4-dichlorophenol sorbed  onto a loam
containing 4.7% organic matter  has been
observed to be approximately   5 times
greater than the  mass  sorbed onto  a
sandy loam containing  0.84% organic
matter.
  The sorption  of  a series of  phenolic
compounds onto a loam  increases in the
order:

    phenol  < o-chlorophenol < 2,4-
    dichlorophenol.

This order is preserved for the sorption of
these solutes onto  cupric and calcium
salts of commercially  available humic
acid, which  is a primary component of
soil organic matter.
  The mass of TCA sorbed onto a loam
containing  4.7%  organic  matter  was
                                                                                observed to be 2.5 times greater than tli
                                                                                mass sorbed  onto a sandy  clay loai
                                                                                containing  l.4% organic matter.  In  th
                                                                                presence of a mixed solvent (10 volum
                                                                                % ethanol), the mass of  TCA sorbed onl
                                                                                a loam decreased  by 40%, as compare
                                                                                to the single solute in aqueous solution.

                                                                                Solvent Extractions
                                                                                 The presence of components of
                                                                                landfill leachate solute matrix altered th
                                                                                distribution of phenol and  o-chlorophem
                                                                                between a bulk organic phase and th
                                                                                aqueous  wastewater  phase.  Th
                                                                                alteration  was more  pronounced  i
                                                                                experiments  involving  phenol than  i
                                                                                experiments in which o-chlorophenol wa
                                                                                the  distributed solute.  Since phen<
                                                                                favors  the  aqueous  phase,  it  i
                                                                                reasonable  to  expect solute-sa
                                                                                interactions  in  the  aqueous  phase  t
                                                                                produce  a  larger change  in  th
                                                                                distribution  of  phenol  than in  th
                                                                                distribution of  o-chlorophenol,   whic
                                                                                favors  the   organic  phase.  Th

-------
     To
                NaOH Prep.
                 pH 10-13
                  (Tank 1)
                                                                                               Extract
                                                                                               Ballast
                                                                                               Tank
                                                                                               (Tank 2)
                                      ////
                                           Extraction Bed
                              Effluent
                              Ballast
                              Tank
                              (Tank 4)
Discharge



Figure 4.    Pilot plant process flow diagram.
                                                  Soil
                                            Treatment Bed
                                                    flQJt
Hflft

                                                                                                        Nutrients
                                                                                                        COi
                                                                               Mixing Tank
                                                                                  (Tank 3)
components  of a leachate solute matrix
that contribute most to the salting-out of
phenol appear to be the  larger inorganic
anions,  i.e.,  C03-2 and  S04-2.  High
concentrations of acetic,  propionic, and
butyric acids in  leachate also contribute
to observed changes in phenol as a result
of solute-salt interactions in the aqueous
or bulk organic phases.

Membrane Separations
  Thin semi-permeable  films  become
important in  the separation of dissolved
species,  especially from aqueous
solution. Membranes have been used for
the  production of  process water and/or
drinking water from  sea water and natural
brines. In addition,  membranes  have
been adopted for commercial  separations
and  concentration of  products  and
valuable process constituents  such as
catalysts  and recyclable  intermediates.
Membrane techniques have a role to play
  high-strength wastewater renovation.
                                      Reverse Osmosis
                                        Several  hazardous  wastewaters, e.g.,
                                      industrial  landfill  leachates,  lagoon
                                      wastes, pesticide  wastewaters,  and
                                      synthetic organic manufacturing effluent
                                      were successfully renovated with an  RO
                                      system Moderate  and  high-strength
                                      industrial landfill leachates, pretreated by
                                      physical and  chemical  methods  and
                                      treated biologically, were  separated and
                                      concentrated in semi-batch, steady-state,
                                      and  unsteady modes  of operation.
                                      Physical/chemical pretreatment  with  lime
                                      was  found  necessary to  remove
                                      suspended and colloidal  matter, heavy
                                      metals,  and dispersed oil phases in  the
                                      raw  leachate; these  constituents  can
                                      cause  membrane fouling  and  rapid flux
                                      loss.
                                        All   physically,  chemically,  and
                                      biologically  treated  leachates were
                                      separated  to produce clear, turbidity-free
                                      permeates.  Eight  landfill  leachate
                                      experiments  were   conducted over
       extensive periods of time. For example,
       high-strength,  pretreated leachate EPA
       #07  with  conductivity  of  30,000
       micromho/cm, IDS of 27,000 mg/L, and
       TOG of  6,700 mg/L was separated and
       concentrated in  a semi-batch, unsteady
       experiment. At  room temperature  and
       moderate feed pressure (approximately
       400  psig),  inorganic solute  rejection
       exceeded   95%.   During   feed
       concentration  in the retentate (concen-
       trate) recycle  mode,  operation was
       limited  to  feed  IDS concentrations of
       less  than  30,000  mg/L.  Flux  became
       unacceptably low and pressure had to be
       increased to overcome the high osmotic
       pressure  of  the  feed.  In addition,
       concentration  polarization  appeared to
       occur at high recoveries  accompanied by
       high-feed  IDS  concentrations.  When
       feed was too  concentrated, solutes may
       have exceeded solubility limits  and
       caused  additional deposition  of  mem-
       brane surfaces.

-------
  Moderate-strength  leachates were
renovated to a greater  extent. In a study
with  a recovery  of 75%  at  170 hr of
operation, initial and final rejections  for
inorganic species were  in excess of 99%.
Initial  TOC  rejection  was  low  but
increased, with time, to over 70%. This
phenomenon of increasing TOC rejection
is explained by the fact that  the more
permeable solutes were purged from the
system early in the process, thus leaving
the feed with organic species  that were
rejected continually at higher efficiencies
as time/recovery advanced.  No  fouling
was  observed in any  of the moderate-
strength leachate studies.
  Biologically-treated  landfill  leachate
was  renovated successfully. TDS  and
conductivity rejection were in excess of
98%; maximum TOC rejection was 94%.
Bulk  permeate  TDS  and  TOC  con-
centrations  of  47  and   14  mg/L
respectively, were obtained at recoveries
of over 60%. No fouling was observed in
this study; flux averaged 0.29 m3/m2d.
  Lagoon sludge extracts,  resembling
industrial landfill  leachate,  were also
treated with an RO process. In a  steady-
state experiment,  conductivity  and TDS
rejections  averaged  97%  and 99%,
respectively; TOC rejection was 79%. A
slight increase in rejection and decrease
in  flux appeared evident,  possibly
because of  the formation of a nonfouling
gel layer.

Ultrafiltration
  Results of UF experiments  indicated
that  a high proportion of the organic
matter  in  leachate   samples  had  a
molecular weight below about  500. This
agrees well with  results in the literature
from  similar experiments.  For  raw
leachate and lime-treated  leachate,
approximately 80% and 85%  of the
organic solutes  had a  molecular  weight
below  500,  respectively. These  results
suggest that  leachate  samples  contain
primarily synthetic organic compounds
and  lesser  amounts of biogenic matter,
such as proteins  and  humic and fulvic
substances. These observations  were
expected,  since  the  leachate was
generally of an industrial rather than a
biochemical origin.
  Since most organic solutes in leachates
studies do not  pass  the  500 MW
membrane,  the utility of UF for efficient
removal  of organic  contaminants is
subject to question. At best, UF can be
used as a pretreatment process to
remove  high  molecular  weight
compounds that  interfere  with other
treatment technologies, i.e., RO.
  This  UF  study  confirmed  the
pretreatment  study  results;  lime
treatment did not remove  a significant
mass of the organic  solutes  from
industrial landfill  leachate  samples. The
UF investigation  suggested that lime
treatment may remove  organic  species
with  molecular weights  greater than
10,000; these represent a relatively small
fraction of  the  total organic  matter
present in solution.
  Although  membrane UF  may  not  be
effective for removing organic matter of
low molecular weight from leachate, it is
a valuable tool for evaluating the nature
of  a  leachate  sample  and  the
effectiveness of other pretreatment
processes.  UF  is  not  an effective
treatment technique, since it does not
eliminate any of the low molecular weight
organic  matter  that  is  destroyed
efficiently by biological renovation.

Conclusions
  Several important conclusions  can be
drawn based on the study results. These
are summarized in the same order as the
sections of this summary.

Pretreatment
  - Treatment of  aqueous wastes  with
    significant   concentrations  of
    dispersed  or  suspended  phases —
    high  levels of turbidity—by mem-
    brane or biological techniques either
    was not possible or inefficient without
    pretreatment  to  produce  a
    homogeneous aqueous phase.

  - Lime addition to pH 12, followed by
    floe separation and  recarbonation to
    pH  7 (addition of CO2),  was  an
    effective  pretreatment  process that
    resulted  in  greater  than  98%
    reduction in turbidity.

Microbial Treatment
  - Aerobic mixed microbial populations
    degraded a significant fraction of the
    dissolved organic solutes present in
    some leachates (80% to 90%), while
    being inherently limited in degrading
    dissolved organic  solutes  in  other
    leachates (<  50%).

  - Aerobic mixed microbial populations
    may require a readily biodegradable
    co-substrate  to   mineralize
    anthropogenic compounds  (co-
    metabolism)  and may exhibit two
    distinctive growth regimes.

  -Anaerobic  mixed   microbial
    populations degraded  a significant
    fraction of the dissolved  organic
   solutes  present in some leachc
   (60% to 70%).

 - Anaerobic   mixed  microb
   populations  degraded sevc
   chlorinated hydrocarbons presen
   aqueous  solution  from  ini
   concentrations up to 20 mg/L to I
   concentrations below  the detec
   limit of  20 pg/L. Half-kill  doses
   toxicity  inhibition   levels  w
   determined.

 - Methanogens were  most likely
   group of  organisms responsible
   degradation  of  the  chlorina
   compounds considered.

 - Soil-based sequential aerobic/
   aerobic  microbial  degradat
   mineralized between 90% and 9
   of the  dissolved organic  solt
   present in all  leachates  test
   Carbon  dioxide and  methane w
   principal end products.

 - Soil-based anaerobic microbial C
   radation  reduced   TCA
   contaminated groundwater from
   mg/L to  less than 20 pg/L. (
   production   rates   and   c
   composition  indicated  tl
   methanogenesis  was the control
   process.

 - Three pilot plants were  design
   constructed, and operated in the f
   to demonstrate process feasibility
   the following wastes:

   (a) An industrial landfill leachate '
       treated  using  a soil-bas
       sequential aerobic/anaero
       microbial process with  sim
       pass  organic  destruct
       efficiencies between 90%  ,
       97% .

   (b) In-situ sludge  extraction  \
       coupled  with   soil-bas
       microbial treatment  of the
       covered extract to demonsti
       controlled, rapid  removal  ;
       mineralization  of extracta
       organic species, with treatrr
       efficiencies greater than 95%

   (c) TCA   in  groundwater  v
       reduced  from 5 to 20  mg/L
       less  than  20  pg/L, utiliz
       anaerobic soil-based micro
       treatment.

Pilot  plants (a) and (b) treated  waj
from  CERCLA-NPL sites. Pilot studies
and (c) were conducted on site.

-------
Sorption and Extraction
- Activated  carbon  adsorption  was
  inefficient when dispersed oily  phases
  were present in the aqueous waste.
- Relationships  between equilibrium
  organic  loading and  equilibrium
  concentrations were  dependent on
  initial conditions.
- The mass fraction of organic  matter
  strongly influenced  the  sorptive
  capacity of a soil.
- A landfill leachate solute matrix can
  significantly alter  aqueous/organic
  phase solute partitioning.

Membrane Separations
- RO was  effective  for  removing
  inorganic  species after  biological
  treatment.
- UF  was  an  effective  tool  for
  characterizing leachates, but was  of
  limited value for treatment  because  a
  majority of the solutes present had  a
  molecular weight less than 500.

-------
  Robert C. Ahlert and David S. Kosson are with the Department of Chemical and
    Biochemical Engineering, Rutgers University, Piscataway, NJ 08855-0909.
  John E. Brugger is the EPA Project Officer (see below).
  The complete report, entitled "Treatment of Hazardous Landfill  Leachates and
    Contaminated Groundwater," (Order No. PB 89-124  6481 AS; Cost: $28.95,
    subject to change)  will be available only from:
           National Technical Information Service
           5285 Port Royal Road
           Springfield, VA 22161
           Telephone: 703-487-4650
  The EPA Project Officer can be contacted at:
           Releases Control Branch
           Risk Reduction Engineering Laboratory
           U.S. Environmental Protection Agency
           Edison, NJ 08837
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
      BULK RATE
POSTAGE & FEES PAK
         EPA
   PERMIT No. G-35
Official Business
Penalty for Private Use $300

EPA/600/S2-88/064
            000085836    HHERL
            USEPA   REGI01I  ¥  LIBRARI
            230  S  DEA880BM  ST

            CHICAGO               IL  6060H

-------